Mahesh Kumar JatMahesh Kumar JatDepartment of Civil EngineeringDepartment of Civil Engineering

Malaviya National Institute of Technology, Malaviya National Institute of Technology, JaipurJaipur

Mahesh Kumar JatMahesh Kumar JatDepartment of Civil EngineeringDepartment of Civil Engineering

Malaviya National Institute of Technology, Malaviya National Institute of Technology, JaipurJaipur

Image Enhancement and InterpretationImage Enhancement and InterpretationImage Enhancement and InterpretationImage Enhancement and Interpretation

Image EnhancementImage Enhancement

Reduction Magnification Spatial Profiles Spectral Profiles Ratioing Contrast Stretching Frequency Filtering Edge Enhancement Vegetation Indices Texture

Reduction Magnification Spatial Profiles Spectral Profiles Ratioing Contrast Stretching Frequency Filtering Edge Enhancement Vegetation Indices Texture

Integer Image Reduction

Integer Image Reduction

Integer Image Reduction

Integer Image Reduction

Integer Image Magnification

Integer Image Magnification

Integer Image Magnification

Integer Image Magnification

Integer Image Magnification

Integer Image Magnification

Band RatioingBand Ratioing

lji

kjiratioji BV

BVBV

,,

,,,,

lji

kjiratioji BV

BVBV

,,

,,,,

where: - BVi,j,k is the original input brightness value in band k - BVi,j,l is the original input brightness value in band l- BVi,j,ratio is the ratio output brightness value

where: - BVi,j,k is the original input brightness value in band k - BVi,j,l is the original input brightness value in band l- BVi,j,ratio is the ratio output brightness value

Band RatioingBand Ratioing

1127,,,, rjinji BVIntBV 1127,,,, rjinji BVIntBV

2128 ,,

,,rji

nji

BVIntBV

2128 ,,

,,rji

nji

BVIntBV

Ratio values within the range 1/255 to 1 are assigned values between 1 and 128 by the function:

Ratio values within the range 1/255 to 1 are assigned values between 1 and 128 by the function:

Ratio values from 1 to 255 are assigned values within the range 128 to 255 by the function:

Ratio values from 1 to 255 are assigned values within the range 128 to 255 by the function:

Band Ratioing of Charleston,

SC Landsat Thematic

Mapper Data

Band Ratioing of Charleston,

SC Landsat Thematic

Mapper Data

Band Ratio ImageBand Ratio Image

Landsat TMBand 4 / Band 3

Landsat TMBand 4 / Band 3

Band Ratio

Band Ratio

SPOT HRVBand 3 / Pan

SPOT HRVBand 3 / Pan

Band Ratio

Band Ratio

Landsat TMBand 40 / Band 15

Landsat TMBand 40 / Band 15

Spatial Profile -Transect-

Spatial Profile -Transect-

20042004

Spatial Profile -Transect-

Spatial Profile -Transect-

20042004

Spectral Profile of SPOT 20 x 20 m

Multispectral Data of Marco Island, Florida

Spectral Profile of SPOT 20 x 20 m

Multispectral Data of Marco Island, Florida

20042004

Spectral Profile of HYMAP 3 x 3 m

Hyperspectral Data of Debordieu Colony near

North Inlet, SC

Spectral Profile of HYMAP 3 x 3 m

Hyperspectral Data of Debordieu Colony near

North Inlet, SC

20042004

Standard Deviation Contrast Stretch Standard Deviation Contrast Stretch Standard Deviation Contrast Stretch Standard Deviation Contrast Stretch

Common Common Symmetric and Symmetric and

Skewed Skewed Distributions in Distributions in

Remotely Sensed Remotely Sensed DataData

Common Common Symmetric and Symmetric and

Skewed Skewed Distributions in Distributions in

Remotely Sensed Remotely Sensed DataData

Min-Max Contrast Stretch

Min-Max Contrast Stretch

+1 Standard Deviation Contrast Stretch

+1 Standard Deviation Contrast Stretch

Linear Contrast Enhancement: Minimum- Maximum Contrast Stretch

Linear Contrast Enhancement: Minimum- Maximum Contrast Stretch

kkk

kinout quant

BVBV

minmax

mink

kk

kinout quant

BVBV

minmax

min

where: - BVin is the original input brightness value - quantk is the range of the brightness values that can be displayed on the CRT (e.g., 255),- mink is the minimum value in the image,- maxk is the maximum value in the image, and- BVout is the output brightness value

where: - BVin is the original input brightness value - quantk is the range of the brightness values that can be displayed on the CRT (e.g., 255),- mink is the minimum value in the image,- maxk is the maximum value in the image, and- BVout is the output brightness value

Linear Contrast Enhancement: Minimum- Maximum Contrast Stretch

Linear Contrast Enhancement: Minimum- Maximum Contrast Stretch

2552554105

4105

02554105

44

minmax

min

inout

inout

BV

BV

2552554105

4105

02554105

44

minmax

min

inout

inout

BV

BV

All other original brightness values between 5 and 104 are linearly distributed between 0 and 255.

All other original brightness values between 5 and 104 are linearly distributed between 0 and 255.

Min-Max Contrast Stretch

Min-Max Contrast Stretch

+1 Standard Deviation Contrast Stretch

+1 Standard Deviation Contrast Stretch

Contrast Stretch of Charleston, SC Landsat

Thematic Mapper Band 4 Data

Contrast Stretch of Charleston, SC Landsat

Thematic Mapper Band 4 Data

OriginalOriginal

Minimum-maximum

Minimum-maximum

+1 standard deviation

+1 standard deviation

Contrast Stretching of Charleston, SC Landsat Thematic Mapper Band 4 Data

Contrast Stretching of Charleston, SC Landsat Thematic Mapper Band 4 Data

Specific percentage linear contrast stretch designed to highlight

wetland

Specific percentage linear contrast stretch designed to highlight

wetland

Histogram Equalization Histogram Equalization

Contrast Stretching of Predawn Thermal Infrared Data of the

the Savannah River

Contrast Stretching of Predawn Thermal Infrared Data of the

the Savannah River

OriginalOriginal

Minimum-maximum

Minimum-maximum

+1 standard deviation

+1 standard deviation

Specific percentage linear contrast stretch

designed to highlight the thermal plume

Specific percentage linear contrast stretch

designed to highlight the thermal plume

Histogram Equalization Histogram Equalization

Contrast Stretching of Predawn Thermal Infrared Data of the the Savannah River

Contrast Stretching of Predawn Thermal Infrared Data of the the Savannah River

Non-linear Contrast StretchingNon-linear Contrast Stretching

PiecewisePiecewise

Piecewise Linear Contrast Stretching

Piecewise Linear Contrast Stretching

Non-linear Contrast StretchingNon-linear Contrast Stretching

Logarithmic and Inverse Log

Logarithmic and Inverse Log

Spatial Filtering to Enhance Low- and Spatial Filtering to Enhance Low- and High-Frequency Detail and Edges High-Frequency Detail and Edges

Spatial Filtering to Enhance Low- and Spatial Filtering to Enhance Low- and High-Frequency Detail and Edges High-Frequency Detail and Edges

A characteristics of remotely sensed A characteristics of remotely sensed images is a parameter called images is a parameter called spatial spatial frequencyfrequency, defined as the number of , defined as the number of changes in brightness value per unit changes in brightness value per unit distance for any particular part of an distance for any particular part of an image.image.

A characteristics of remotely sensed A characteristics of remotely sensed images is a parameter called images is a parameter called spatial spatial frequencyfrequency, defined as the number of , defined as the number of changes in brightness value per unit changes in brightness value per unit distance for any particular part of an distance for any particular part of an image.image.

Spatial frequencySpatial frequency in remotely sensed imagery may be in remotely sensed imagery may be enhanced or subdued using two different approaches:enhanced or subdued using two different approaches:

- - Spatial convolution filteringSpatial convolution filtering based primarily on the based primarily on the use of convolution masks, and use of convolution masks, and

- - Fourier analysisFourier analysis which mathematically separates an which mathematically separates an image into its spatial frequency components.image into its spatial frequency components.

Spatial frequencySpatial frequency in remotely sensed imagery may be in remotely sensed imagery may be enhanced or subdued using two different approaches:enhanced or subdued using two different approaches:

- - Spatial convolution filteringSpatial convolution filtering based primarily on the based primarily on the use of convolution masks, and use of convolution masks, and

- - Fourier analysisFourier analysis which mathematically separates an which mathematically separates an image into its spatial frequency components.image into its spatial frequency components.

Spatial Filtering to Enhance Low- and Spatial Filtering to Enhance Low- and High-Frequency Detail and Edges High-Frequency Detail and Edges

Spatial Filtering to Enhance Low- and Spatial Filtering to Enhance Low- and High-Frequency Detail and Edges High-Frequency Detail and Edges

A linear A linear spatial filterspatial filter is a filter for which the brightness is a filter for which the brightness value (value (BVBVi,j,outi,j,out) at location ) at location i,ji,j in the output image is a in the output image is a

function of some weighted average (linear function of some weighted average (linear combination) of brightness values located in a combination) of brightness values located in a particular spatial pattern around the particular spatial pattern around the i,ji,j location in the location in the input image. input image.

The process of evaluating the weighted neighboring The process of evaluating the weighted neighboring pixel values is called pixel values is called two-dimensional convolution two-dimensional convolution filteringfiltering. .

A linear A linear spatial filterspatial filter is a filter for which the brightness is a filter for which the brightness value (value (BVBVi,j,outi,j,out) at location ) at location i,ji,j in the output image is a in the output image is a

function of some weighted average (linear function of some weighted average (linear combination) of brightness values located in a combination) of brightness values located in a particular spatial pattern around the particular spatial pattern around the i,ji,j location in the location in the input image. input image.

The process of evaluating the weighted neighboring The process of evaluating the weighted neighboring pixel values is called pixel values is called two-dimensional convolution two-dimensional convolution filteringfiltering. .

Spatial Convolution FilteringSpatial Convolution FilteringSpatial Convolution FilteringSpatial Convolution Filtering

The size of the neighborhood The size of the neighborhood convolution maskconvolution mask or or kernel kernel ((nn) is usually 3 x 3, 5 x 5, 7 x 7, or 9 x 9. ) is usually 3 x 3, 5 x 5, 7 x 7, or 9 x 9.

We will constrain our discussion to We will constrain our discussion to 3 x 33 x 3 convolution convolution masks with masks with ninenine coefficients, coefficients, ccii, defined at the , defined at the

following locations:following locations:

cc1 1 cc2 2 cc33

Mask templateMask template = = cc4 4 cc5 5 cc66

cc7 7 cc8 8 cc99

The size of the neighborhood The size of the neighborhood convolution maskconvolution mask or or kernel kernel ((nn) is usually 3 x 3, 5 x 5, 7 x 7, or 9 x 9. ) is usually 3 x 3, 5 x 5, 7 x 7, or 9 x 9.

We will constrain our discussion to We will constrain our discussion to 3 x 33 x 3 convolution convolution masks with masks with ninenine coefficients, coefficients, ccii, defined at the , defined at the

following locations:following locations:

cc1 1 cc2 2 cc33

Mask templateMask template = = cc4 4 cc5 5 cc66

cc7 7 cc8 8 cc99

Spatial Convolution FilteringSpatial Convolution FilteringSpatial Convolution FilteringSpatial Convolution Filtering

11 11 1111 11 1111 1111

The coefficients, The coefficients, cc11, in the mask are multiplied by the , in the mask are multiplied by the

following individual brightness values (following individual brightness values (BVBVii) in the ) in the

input image:input image: cc1 1 x BVx BV1 1 cc2 2 x BVx BV2 2 cc3 3 x BVx BV3 3

Mask templateMask template = = cc4 4 x BVx BV4 4 cc5 5 x BVx BV5 5 cc6 6 x BVx BV6 6

cc7 7 x BVx BV7 7 cc8 8 x BVx BV8 8 cc9 9 x BVx BV99

The primary input pixel under investigation at any one The primary input pixel under investigation at any one time is time is BVBV55 = = BVBVi,ji,j

The coefficients, The coefficients, cc11, in the mask are multiplied by the , in the mask are multiplied by the

following individual brightness values (following individual brightness values (BVBVii) in the ) in the

input image:input image: cc1 1 x BVx BV1 1 cc2 2 x BVx BV2 2 cc3 3 x BVx BV3 3

Mask templateMask template = = cc4 4 x BVx BV4 4 cc5 5 x BVx BV5 5 cc6 6 x BVx BV6 6

cc7 7 x BVx BV7 7 cc8 8 x BVx BV8 8 cc9 9 x BVx BV99

The primary input pixel under investigation at any one The primary input pixel under investigation at any one time is time is BVBV55 = = BVBVi,ji,j

Spatial Convolution FilteringSpatial Convolution FilteringSpatial Convolution FilteringSpatial Convolution Filtering

Various Convolution

Mask Kernels

Various Convolution

Mask Kernels

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Low Frequency FilterLow Frequency Filter

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Low Frequency FilterLow Frequency Filter

9

...int

int

9321

9

1,5

BVBVBVBV

n

BVxcLFF

ii

i

out

9

...int

int

9321

9

1,5

BVBVBVBV

n

BVxcLFF

ii

i

out

1

1

1

1

1

1

1

1

1

Low Pass FilterLow Pass Filter

9

273

9

364

9

455

Spatial Frequency Filtering

Spatial Frequency Filtering

Spatial Convolution Filtering: Median Filter

Spatial Convolution Filtering: Median Filter

A A median filtermedian filter has certain advantages when compared has certain advantages when compared with weighted convolution filters, including: 1) it does with weighted convolution filters, including: 1) it does not shift boundaries, and 2) the minimal degradation to not shift boundaries, and 2) the minimal degradation to edges allows the median filter to be applied repeatedly edges allows the median filter to be applied repeatedly which allows fine detail to be erased and large regions which allows fine detail to be erased and large regions to take on the same brightness value (often called to take on the same brightness value (often called posterization). posterization).

A A median filtermedian filter has certain advantages when compared has certain advantages when compared with weighted convolution filters, including: 1) it does with weighted convolution filters, including: 1) it does not shift boundaries, and 2) the minimal degradation to not shift boundaries, and 2) the minimal degradation to edges allows the median filter to be applied repeatedly edges allows the median filter to be applied repeatedly which allows fine detail to be erased and large regions which allows fine detail to be erased and large regions to take on the same brightness value (often called to take on the same brightness value (often called posterization). posterization).

Spatial Frequency Filtering

Spatial Frequency Filtering

Spatial Convolution Filtering: Minimum or Maximum Filters

Spatial Convolution Filtering: Minimum or Maximum Filters

Operating on one pixel at a time, these filters examine Operating on one pixel at a time, these filters examine the brightness values of adjacent pixels in a user-the brightness values of adjacent pixels in a user-specified radius (e.g., 3 x 3 pixels) and replace the specified radius (e.g., 3 x 3 pixels) and replace the brightness value of the current pixel with the brightness value of the current pixel with the minimumminimum or or maximummaximum brightness value encountered, respectively. brightness value encountered, respectively.

Operating on one pixel at a time, these filters examine Operating on one pixel at a time, these filters examine the brightness values of adjacent pixels in a user-the brightness values of adjacent pixels in a user-specified radius (e.g., 3 x 3 pixels) and replace the specified radius (e.g., 3 x 3 pixels) and replace the brightness value of the current pixel with the brightness value of the current pixel with the minimumminimum or or maximummaximum brightness value encountered, respectively. brightness value encountered, respectively.

Spatial Frequency Filtering

Spatial Frequency Filtering

Spatial Spatial ConvolutionConvolution Filtering: Filtering: High Frequency FilterHigh Frequency Filter

Spatial Spatial ConvolutionConvolution Filtering: Filtering: High Frequency FilterHigh Frequency Filter

High-pass filteringHigh-pass filtering is applied to imagery to remove the is applied to imagery to remove the slowly varying components and enhance the high-slowly varying components and enhance the high-frequency local variations. One high-frequency filter frequency local variations. One high-frequency filter ((HFFHFF5,out5,out) is computed by subtracting the output of the ) is computed by subtracting the output of the

low-frequency filter (low-frequency filter (LFFLFF5,out5,out) from twice the value of ) from twice the value of

the original central pixel value, the original central pixel value, BVBV55::

High-pass filteringHigh-pass filtering is applied to imagery to remove the is applied to imagery to remove the slowly varying components and enhance the high-slowly varying components and enhance the high-frequency local variations. One high-frequency filter frequency local variations. One high-frequency filter ((HFFHFF5,out5,out) is computed by subtracting the output of the ) is computed by subtracting the output of the

low-frequency filter (low-frequency filter (LFFLFF5,out5,out) from twice the value of ) from twice the value of

the original central pixel value, the original central pixel value, BVBV55::

outout LFFBVxHFF ,55,5 )2( outout LFFBVxHFF ,55,5 )2(

Spatial Frequency Filtering

Spatial Frequency Filtering

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Unequal-weighted smoothing Filter

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Unequal-weighted smoothing Filter

0.250.25 0.500.50 0.250.25

0.500.50 11 0.500.50

0.250.25 0.500.50 0.250.25

11 11 11

11 22 11

11 11 11

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Edge EnhancementEdge Enhancement

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Edge EnhancementEdge Enhancement

For many remote sensing Earth science applications, the For many remote sensing Earth science applications, the most valuable information that may be derived from an most valuable information that may be derived from an image is contained in the image is contained in the edgesedges surrounding various surrounding various objects of interest. objects of interest. Edge enhancementEdge enhancement delineates these delineates these edges and makes the shapes and details comprising the edges and makes the shapes and details comprising the image more conspicuous and perhaps easier to analyze. image more conspicuous and perhaps easier to analyze. Edges may be enhanced using either Edges may be enhanced using either linear linear or or nonlinear nonlinear edge enhancementedge enhancement techniques. techniques.

For many remote sensing Earth science applications, the For many remote sensing Earth science applications, the most valuable information that may be derived from an most valuable information that may be derived from an image is contained in the image is contained in the edgesedges surrounding various surrounding various objects of interest. objects of interest. Edge enhancementEdge enhancement delineates these delineates these edges and makes the shapes and details comprising the edges and makes the shapes and details comprising the image more conspicuous and perhaps easier to analyze. image more conspicuous and perhaps easier to analyze. Edges may be enhanced using either Edges may be enhanced using either linear linear or or nonlinear nonlinear edge enhancementedge enhancement techniques. techniques.

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Directional Directional First-Difference Linear Edge EnhancementFirst-Difference Linear Edge Enhancement

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Directional Directional First-Difference Linear Edge EnhancementFirst-Difference Linear Edge Enhancement

KBVBVDiagonalSE

KBVBVDiagonalNE

KBVBVHorizontal

KBVBVVertical

jiji

jiji

jiji

jiji

1,1,

1,1,

,1,

1,,

KBVBVDiagonalSE

KBVBVDiagonalNE

KBVBVHorizontal

KBVBVVertical

jiji

jiji

jiji

jiji

1,1,

1,1,

,1,

1,,

The result of the subtraction can be either negative or possible, therefore a constant, K (usually 127) is added to make all values positive and centered between 0 and 255

The result of the subtraction can be either negative or possible, therefore a constant, K (usually 127) is added to make all values positive and centered between 0 and 255

Spatial Spatial ConvolutionConvolution Filtering: Filtering: High-pass Filters that Accentuate or Sharpen Edges

Spatial Spatial ConvolutionConvolution Filtering: Filtering: High-pass Filters that Accentuate or Sharpen Edges

-1-1 -1-1 -1-1

-1-1 99 -1-1

-1-1 -1-1 -1-1

11 -2-2 11

-2-2 55 -2-2

11 -2-2 11

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Linear Edge Enhancement - Embossing

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Linear Edge Enhancement - Embossing

00 00 00

11 00 -1-1

00 00 00

Emboss East Emboss East

00 00 11

00 00 00

-1-1 00 00

Emboss NW Emboss NW

Spatial Frequency Filtering

Spatial Frequency Filtering

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Compass Gradient Masks

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Compass Gradient Masks

11 11 11

11 -2-2 11

-1-1 -1-1 -1-1

North North

11 11 11

-1-1 -2-2 11

-1-1 -1-1 11

NortheastNortheast

-1-1 11 11

-1-1 -2-2 11

-1-1 11 11

EastEast

Spatial Frequency Filtering

Spatial Frequency Filtering

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Edge Enhancement Using

Laplacian Convolution Masks

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Edge Enhancement Using

Laplacian Convolution Masks

The Laplacian is a second derivative (as opposed to the gradient which is a first derivative) and is invariant to rotation, meaning that it is insensitive to the direction in which the discontinuities (point, line, and edges) run.

The Laplacian is a second derivative (as opposed to the gradient which is a first derivative) and is invariant to rotation, meaning that it is insensitive to the direction in which the discontinuities (point, line, and edges) run.

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Laplacian Convolution Masks

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Laplacian Convolution Masks

00 -1-1 00

-1-1 44 -1-1

00 -1-1 00

-1-1 -1-1 -1-1

-1-1 88 -1-1

-1-1 -1-1 -1-1

11 -2-2 11

-2-2 44 -2-2

11 -2-2 11

Spatial Frequency Filtering

Spatial Frequency Filtering

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Edge Enhancement Using Laplacian Convolution Masks

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Edge Enhancement Using Laplacian Convolution Masks

The following The following LaplacianLaplacian operator may be used to subtract operator may be used to subtract the Laplacian edges from the original image: the Laplacian edges from the original image: The following The following LaplacianLaplacian operator may be used to subtract operator may be used to subtract the Laplacian edges from the original image: the Laplacian edges from the original image:

11 11 11

11 -7-7 11

11 11 11

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Edge Enhancement Using Laplacian Convolution Masks

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Edge Enhancement Using Laplacian Convolution Masks

By itself, the By itself, the LaplacianLaplacian image may be difficult to interpret. image may be difficult to interpret. Therefore, a Laplacian edge enhancement may be added back to the Therefore, a Laplacian edge enhancement may be added back to the original image using the following mask: original image using the following mask:

By itself, the By itself, the LaplacianLaplacian image may be difficult to interpret. image may be difficult to interpret. Therefore, a Laplacian edge enhancement may be added back to the Therefore, a Laplacian edge enhancement may be added back to the original image using the following mask: original image using the following mask:

00 -1-1 00

-1-1 55 -1-1

00 -1-1 00

Spatial Frequency Filtering

Spatial Frequency Filtering

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Non-linear Non-linear Edge Enhancement Using the Sobel OperatorEdge Enhancement Using the Sobel OperatorSpatial Spatial ConvolutionConvolution Filtering: Filtering: Non-linear Non-linear

Edge Enhancement Using the Sobel OperatorEdge Enhancement Using the Sobel Operator

987321

741963

22,5

22

22

BVBVBVBVBVBVY

BVBVBVBVBVBVX

where

YXSobel out

987321

741963

22,5

22

22

BVBVBVBVBVBVY

BVBVBVBVBVBVX

where

YXSobel out

1

4

7 8

2

6

9

3

order

The Sobel operator may also be computed by simultaneously applying the following 3 x 3

templates across the image:

The Sobel operator may also be computed by simultaneously applying the following 3 x 3

templates across the image:

-1-1 00 11

-2-2 00 22

-1-1 00 11

11 22 11

00 00 00

-1-1 -2-2 -1-1

X = X = Y = Y =

Spatial Frequency Filtering

Spatial Frequency Filtering

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Non-linear Edge Non-linear Edge Enhancement Using the Robert’s Edge DetectorEnhancement Using the Robert’s Edge Detector

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Non-linear Edge Non-linear Edge Enhancement Using the Robert’s Edge DetectorEnhancement Using the Robert’s Edge Detector

86

95

,5

BVBVY

BVBVX

where

YXRoberts out

86

95

,5

BVBVY

BVBVX

where

YXRoberts out

The Robert’s edge detector is based on the The Robert’s edge detector is based on the use of only four elements of a 3 x 3 mask. use of only four elements of a 3 x 3 mask.

The Robert’s edge detector is based on the The Robert’s edge detector is based on the use of only four elements of a 3 x 3 mask. use of only four elements of a 3 x 3 mask.

1

4

7 8

2

6

9

3

order

5

Spatial Frequency Filtering

Spatial Frequency Filtering

The Robert’s Edge operator may also be computed by simultaneously applying the

following 3 x 3 templates across the image:

The Robert’s Edge operator may also be computed by simultaneously applying the

following 3 x 3 templates across the image:

00 00 00

00 11 00

00 00 -1-1

00 00 00

00 00 11

00 -1-1 00

X = X = Y = Y =

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Non-linear Edge Non-linear Edge Enhancement Using the Kirsch Edge DetectorEnhancement Using the Kirsch Edge Detector

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Non-linear Edge Non-linear Edge Enhancement Using the Kirsch Edge DetectorEnhancement Using the Kirsch Edge Detector

The The KirschKirsch nonlinear edge enhancement nonlinear edge enhancement

calculates the gradient at pixel location calculates the gradient at pixel location BVBVi,j i,j

. To apply this operator, however, it is first . To apply this operator, however, it is first necessary to designate a different 3 x 3 necessary to designate a different 3 x 3 window numbering scheme. window numbering scheme.

BVBV0 0 BVBV1 1 BVBV22

Kirsh windowKirsh window = = BVBV7 7 BVBVi,j i,j BVBV33

BVBV6 6 BVBV5 5 BVBV44

The The KirschKirsch nonlinear edge enhancement nonlinear edge enhancement

calculates the gradient at pixel location calculates the gradient at pixel location BVBVi,j i,j

. To apply this operator, however, it is first . To apply this operator, however, it is first necessary to designate a different 3 x 3 necessary to designate a different 3 x 3 window numbering scheme. window numbering scheme.

BVBV0 0 BVBV1 1 BVBV22

Kirsh windowKirsh window = = BVBV7 7 BVBVi,j i,j BVBV33

BVBV6 6 BVBV5 5 BVBV44

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Non-linear Edge Non-linear Edge Enhancement Using the Kirsch Edge DetectorEnhancement Using the Kirsch Edge Detector

Spatial Spatial ConvolutionConvolution Filtering: Filtering: Non-linear Edge Non-linear Edge Enhancement Using the Kirsch Edge DetectorEnhancement Using the Kirsch Edge Detector

76543

21

7

0, 35max,1max

iiiiii

iiii

iii

ji

BvBVBvBVBVT

BvBVBVS

where

TSAbsBV

76543

21

7

0, 35max,1max

iiiiii

iiii

iii

ji

BvBVBvBVBVT

BvBVBVS

where

TSAbsBV

The subscripts of The subscripts of BVBV are evaluated modulo are evaluated modulo 88, meaning that the computation moves , meaning that the computation moves around the perimeter of the mask in eight steps. The edge enhancement computes around the perimeter of the mask in eight steps. The edge enhancement computes the maximal compass gradient magnitude about input image points although the the maximal compass gradient magnitude about input image points although the input pixel value input pixel value BVBVi,ji,j is never used in the computation is never used in the computation

The subscripts of The subscripts of BVBV are evaluated modulo are evaluated modulo 88, meaning that the computation moves , meaning that the computation moves around the perimeter of the mask in eight steps. The edge enhancement computes around the perimeter of the mask in eight steps. The edge enhancement computes the maximal compass gradient magnitude about input image points although the the maximal compass gradient magnitude about input image points although the input pixel value input pixel value BVBVi,ji,j is never used in the computation is never used in the computation

Spatial Frequency Filtering

Spatial Frequency Filtering

Histogram Equalization Histogram Equalization Histogram Equalization Histogram Equalization

• evaluates the individual brightness values in a band of evaluates the individual brightness values in a band of imagery and imagery and assigns approximately an equal number of assigns approximately an equal number of pixels to each of the user-specified output gray-scale classespixels to each of the user-specified output gray-scale classes (e.g., 32, 64, and 256). (e.g., 32, 64, and 256).

• applies the greatest contrast enhancement to the most applies the greatest contrast enhancement to the most populated range of brightness values in the image. populated range of brightness values in the image.

• reduces the contrast in the very light or dark parts of the reduces the contrast in the very light or dark parts of the image associated with the tails of a normally distributed image associated with the tails of a normally distributed histogram. histogram.

• evaluates the individual brightness values in a band of evaluates the individual brightness values in a band of imagery and imagery and assigns approximately an equal number of assigns approximately an equal number of pixels to each of the user-specified output gray-scale classespixels to each of the user-specified output gray-scale classes (e.g., 32, 64, and 256). (e.g., 32, 64, and 256).

• applies the greatest contrast enhancement to the most applies the greatest contrast enhancement to the most populated range of brightness values in the image. populated range of brightness values in the image.

• reduces the contrast in the very light or dark parts of the reduces the contrast in the very light or dark parts of the image associated with the tails of a normally distributed image associated with the tails of a normally distributed histogram. histogram.

Statistics for a 64 x 64 Hypothetical Image Statistics for a 64 x 64 Hypothetical Image with Brightness Values from 0 to 7with Brightness Values from 0 to 7

Statistics for a 64 x 64 Hypothetical Image Statistics for a 64 x 64 Hypothetical Image with Brightness Values from 0 to 7with Brightness Values from 0 to 7

4096 total4096 total4096 total4096 total

Histogram EqualizationHistogram Equalization

Transformation Function, ki for each individual brightness value

Transformation Function, ki for each individual brightness value

kquant

i

ii n

BVfk

0

kquant

i

ii n

BVfk

0

For each brightness value level BVi in the quantk range of 0 to 7 of the original histogram, a new cumulative frequency value ki is calculated:

For each brightness value level BVi in the quantk range of 0 to 7 of the original histogram, a new cumulative frequency value ki is calculated:

where the summation counts the frequency of pixels in the image with brightness values equal to or less than BVi, and n is the total number of pixels in the

entire scene (4,096 in this example).

where the summation counts the frequency of pixels in the image with brightness values equal to or less than BVi, and n is the total number of pixels in the

entire scene (4,096 in this example).

Histogram EqualizationHistogram Equalization

The histogram equalization process iteratively compares the The histogram equalization process iteratively compares the transformation function transformation function kkii with the original values of with the original values of llii, to determine , to determine

which are closest in value. which are closest in value. The closest match is reassigned to the The closest match is reassigned to the appropriate brightness value.appropriate brightness value.

For example, we see that For example, we see that kk0 0 = 0.19= 0.19 is closest to is closest to LL11 = 0.14 = 0.14. Therefore, . Therefore,

all pixels in all pixels in BVBV00 (790 of them) will be assigned to (790 of them) will be assigned to BVBV11. Similarly, . Similarly,

the 1023 pixels in the 1023 pixels in BVBV11 will be assigned to will be assigned to BVBV33, the 850 pixels in , the 850 pixels in BVBV22

will be assigned to will be assigned to BVBV55, the 656 pixels in , the 656 pixels in BVBV33 will be assigned to will be assigned to

BVBV66, the 329 pixels in , the 329 pixels in BVBV44 will also be assigned to will also be assigned to BVBV66, and all 448 , and all 448

brightness values in brightness values in BVBV5–75–7 will be assigned to will be assigned to BVBV77. The new image . The new image

will not have any pixels with brightness values of 0, 2, or 4. This is will not have any pixels with brightness values of 0, 2, or 4. This is evident when evaluating the new histogram. evident when evaluating the new histogram. When analysts see such When analysts see such gaps in image histograms, it is usually a good indication that gaps in image histograms, it is usually a good indication that histogram equalization or some other operation has been applied.histogram equalization or some other operation has been applied.

The histogram equalization process iteratively compares the The histogram equalization process iteratively compares the transformation function transformation function kkii with the original values of with the original values of llii, to determine , to determine

which are closest in value. which are closest in value. The closest match is reassigned to the The closest match is reassigned to the appropriate brightness value.appropriate brightness value.

For example, we see that For example, we see that kk0 0 = 0.19= 0.19 is closest to is closest to LL11 = 0.14 = 0.14. Therefore, . Therefore,

all pixels in all pixels in BVBV00 (790 of them) will be assigned to (790 of them) will be assigned to BVBV11. Similarly, . Similarly,

the 1023 pixels in the 1023 pixels in BVBV11 will be assigned to will be assigned to BVBV33, the 850 pixels in , the 850 pixels in BVBV22

will be assigned to will be assigned to BVBV55, the 656 pixels in , the 656 pixels in BVBV33 will be assigned to will be assigned to

BVBV66, the 329 pixels in , the 329 pixels in BVBV44 will also be assigned to will also be assigned to BVBV66, and all 448 , and all 448

brightness values in brightness values in BVBV5–75–7 will be assigned to will be assigned to BVBV77. The new image . The new image

will not have any pixels with brightness values of 0, 2, or 4. This is will not have any pixels with brightness values of 0, 2, or 4. This is evident when evaluating the new histogram. evident when evaluating the new histogram. When analysts see such When analysts see such gaps in image histograms, it is usually a good indication that gaps in image histograms, it is usually a good indication that histogram equalization or some other operation has been applied.histogram equalization or some other operation has been applied.

Statistics of How a a 64 x 64 Hypothetical Image with Brightness Values from 0 to 7 is Histogram Equalized

Statistics of How a a 64 x 64 Hypothetical Image with Brightness Values from 0 to 7 is Histogram Equalized

Principal Components Analysis Principal Components Analysis Principal Components Analysis Principal Components Analysis

• transformation of the raw remote sensor data using PCA transformation of the raw remote sensor data using PCA can result in can result in new principal component imagesnew principal component images that may be that may be more interpretable than the original data. more interpretable than the original data.

• may also be used to compress the information content of a may also be used to compress the information content of a number of bands of imagery (e.g., seven Thematic Mapper number of bands of imagery (e.g., seven Thematic Mapper bands) into just two or three transformed principal bands) into just two or three transformed principal component images. The ability to reduce the component images. The ability to reduce the dimensionalitydimensionality (i.e., the number of bands in the dataset that must be (i.e., the number of bands in the dataset that must be analyzed to produce usable results) from analyzed to produce usable results) from nn to two or three to two or three bands is an important economic consideration, especially if bands is an important economic consideration, especially if the potential information recoverable from the transformed the potential information recoverable from the transformed data is just as good as the original remote sensor data. data is just as good as the original remote sensor data.

• transformation of the raw remote sensor data using PCA transformation of the raw remote sensor data using PCA can result in can result in new principal component imagesnew principal component images that may be that may be more interpretable than the original data. more interpretable than the original data.

• may also be used to compress the information content of a may also be used to compress the information content of a number of bands of imagery (e.g., seven Thematic Mapper number of bands of imagery (e.g., seven Thematic Mapper bands) into just two or three transformed principal bands) into just two or three transformed principal component images. The ability to reduce the component images. The ability to reduce the dimensionalitydimensionality (i.e., the number of bands in the dataset that must be (i.e., the number of bands in the dataset that must be analyzed to produce usable results) from analyzed to produce usable results) from nn to two or three to two or three bands is an important economic consideration, especially if bands is an important economic consideration, especially if the potential information recoverable from the transformed the potential information recoverable from the transformed data is just as good as the original remote sensor data. data is just as good as the original remote sensor data.

The spatial relationship between the first two principal components: (a) Scatter-plot of data points collected from two remotely bands labeled X1 and X2 with the means of the distribution labeled µ1 and µ2. (b) A new coordinate system is created by shifting the axes to an Xsystem. The values for the new data points are found by the relationship X1 = X1 – µ1 and X2 = X2 – µ2. (c) The X axis system is then rotated about its origin (µ1, µ2) so that PC1 is projected through the semi-major axis of the distribution of points and the variance of PC1 is a maximum. PC2 must be perpendicular to PC1. The PC axes are the principal components of this two-dimensional data space. Component 1 usually accounts for approximately 90% of the variance, with component 2 accounting for approximately 5%.

The spatial relationship between the first two principal components: (a) Scatter-plot of data points collected from two remotely bands labeled X1 and X2 with the means of the distribution labeled µ1 and µ2. (b) A new coordinate system is created by shifting the axes to an Xsystem. The values for the new data points are found by the relationship X1 = X1 – µ1 and X2 = X2 – µ2. (c) The X axis system is then rotated about its origin (µ1, µ2) so that PC1 is projected through the semi-major axis of the distribution of points and the variance of PC1 is a maximum. PC2 must be perpendicular to PC1. The PC axes are the principal components of this two-dimensional data space. Component 1 usually accounts for approximately 90% of the variance, with component 2 accounting for approximately 5%.

Statistics Used in the Principal Components AnalysisStatistics Used in the Principal Components AnalysisStatistics Used in the Principal Components AnalysisStatistics Used in the Principal Components Analysis

Statistics Used in the Principal Components AnalysisStatistics Used in the Principal Components AnalysisStatistics Used in the Principal Components AnalysisStatistics Used in the Principal Components Analysis

1. The n  n covariance matrix, Cov, of the n-dimensional remote sensing data set to be transformed is computed. Use of the covariance matrix results in an unstandardized PCA, whereas use of the correlation matrix results in a standardized PCA.

2. The eigenvalues, E = [1,1, 2,2, 3,3, …n.n], and eigenvectors EV = [akp … for k = 1 to n bands, and p = 1 to n components] of the covariance matrix are computed such that:

where EVT is the transpose of the eigenvector matrix, EV, and E is a diagonal covariance matrix whose elements, i,i, called eigenvalues, are the variances

of the pth principal components, where p = 1 to n components.

where EVT is the transpose of the eigenvector matrix, EV, and E is a diagonal covariance matrix whose elements, i,i, called eigenvalues, are the variances

of the pth principal components, where p = 1 to n components.

EE

Eigenvalues Computed for the Covariance MatrixEigenvalues Computed for the Covariance Matrix

p = 1

p = 1

7

Eigenvectors (Eigenvectors (aapp) (Factor Scores) Computed ) (Factor Scores) Computed for the Covariance Matrixfor the Covariance Matrix

Eigenvectors (Eigenvectors (aapp) (Factor Scores) Computed ) (Factor Scores) Computed for the Covariance Matrixfor the Covariance Matrix

Correlation, Correlation, RRk,pk,p, Between Band , Between Band kk and Each Principal Component and Each Principal Component ppCorrelation, Correlation, RRk,pk,p, Between Band , Between Band kk and Each Principal Component and Each Principal Component pp

where:where:aak,p k,p = eigenvector for band = eigenvector for band kk and component and component pp

pp = = ppth eigenvalueth eigenvalueVarVarkk = variance of band = variance of band kk in the in the covariance matrixcovariance matrix

where:where:aak,p k,p = eigenvector for band = eigenvector for band kk and component and component pp

pp = = ppth eigenvalueth eigenvalueVarVarkk = variance of band = variance of band kk in the in the covariance matrixcovariance matrix

where kp = eigenvectors, BVi,j,k = brightness value in band k for the pixel at row i, column j, and n = number of bands.

where kp = eigenvectors, BVi,j,k = brightness value in band k for the pixel at row i, column j, and n = number of bands.

n

kkjikppji BVanewBV

1,,,,

n

kkjikppji BVanewBV

1,,,,

It is possible to compute a new value for pixel 1,1 (it has 7 bands) in principal component number 1 using the following equation:

It is possible to compute a new value for pixel 1,1 (it has 7 bands) in principal component number 1 using the following equation:

original remote sensor data for pixel

1,1

original remote sensor data for pixel

1,1

3,1,11,32,1,11,21,1,11,11,1,1 BVaBVaBVanewBV 3,1,11,32,1,11,21,1,11,11,1,1 BVaBVaBVanewBV

7,1,11,76,1,11,65,1,11,54,1,11,4 BVaBVaBVaBVa 7,1,11,76,1,11,65,1,11,54,1,11,4 BVaBVaBVaBVa

)22(204.0)30(127.0)20(205.01,1,1 newBV )22(204.0)30(127.0)20(205.01,1,1 newBV

)50(106.0)62(376.0)70(742.0)60(443.0 )50(106.0)62(376.0)70(742.0)60(443.0

53.119 53.119

Principal Components

Analysis (PCA)

Principal Components

Analysis (PCA)

Vegetation Transformation (Indices) Vegetation Transformation (Indices) Vegetation Transformation (Indices) Vegetation Transformation (Indices)

Spectral Spectral CharacteristicsCharacteristics

Spectral Spectral CharacteristicsCharacteristics

Remote Sensing of Vegetation Remote Sensing of Vegetation Remote Sensing of Vegetation Remote Sensing of Vegetation

Dominant Factors Controlling Leaf ReflectanceDominant Factors Controlling Leaf ReflectanceDominant Factors Controlling Leaf ReflectanceDominant Factors Controlling Leaf Reflectance

Jensen, 2004Jensen, 2004

Water Water absorption bands:absorption bands:

0.97 0.97 mm1.19 1.19 mm1.45 1.45 mm1.94 1.94 mm2.70 2.70 mm

Water Water absorption bands:absorption bands:

0.97 0.97 mm1.19 1.19 mm1.45 1.45 mm1.94 1.94 mm2.70 2.70 mm

Cross-section Through A Cross-section Through A Hypothetical and Real Hypothetical and Real

Leaf Revealing the Major Leaf Revealing the Major Structural Components Structural Components

that Determine the that Determine the Spectral Reflectance Spectral Reflectance

of Vegetationof Vegetation

Cross-section Through A Cross-section Through A Hypothetical and Real Hypothetical and Real

Leaf Revealing the Major Leaf Revealing the Major Structural Components Structural Components

that Determine the that Determine the Spectral Reflectance Spectral Reflectance

of Vegetationof Vegetation

Jensen, 2004Jensen, 2004

Jensen, 2004Jensen, 2004

Jensen, 2004Jensen, 2004

• Chlorophyll Chlorophyll aa peak absorption is at 0.43 and 0.66 peak absorption is at 0.43 and 0.66 m.m.• Chlorophyll Chlorophyll bb peak absorption is at 0.45 and 0.65 peak absorption is at 0.45 and 0.65 m.m.• Optimum chlorophyll absorption windows: 0.45 - 0.52 Optimum chlorophyll absorption windows: 0.45 - 0.52 m and 0.63 - 0.69 m and 0.63 - 0.69 m m

• Chlorophyll Chlorophyll aa peak absorption is at 0.43 and 0.66 peak absorption is at 0.43 and 0.66 m.m.• Chlorophyll Chlorophyll bb peak absorption is at 0.45 and 0.65 peak absorption is at 0.45 and 0.65 m.m.• Optimum chlorophyll absorption windows: 0.45 - 0.52 Optimum chlorophyll absorption windows: 0.45 - 0.52 m and 0.63 - 0.69 m and 0.63 - 0.69 m m

Absorption Spectra of Chlorophyll Absorption Spectra of Chlorophyll aa and and bb, , --carotene, Pycoerythrin, and Phycocyanin Pigments carotene, Pycoerythrin, and Phycocyanin Pigments

Absorption Spectra of Chlorophyll Absorption Spectra of Chlorophyll aa and and bb, , --carotene, Pycoerythrin, and Phycocyanin Pigments carotene, Pycoerythrin, and Phycocyanin Pigments

lack of lack of absorptionabsorption

lack of lack of absorptionabsorption

Litton Emerge Spatial, Inc., CIR image Litton Emerge Spatial, Inc., CIR image (RGB = NIR,R,G) of Dunkirk, NY, at 1 x (RGB = NIR,R,G) of Dunkirk, NY, at 1 x

1 m obtained on December 12, 1998.1 m obtained on December 12, 1998.

Litton Emerge Spatial, Inc., CIR image Litton Emerge Spatial, Inc., CIR image (RGB = NIR,R,G) of Dunkirk, NY, at 1 x (RGB = NIR,R,G) of Dunkirk, NY, at 1 x

1 m obtained on December 12, 1998.1 m obtained on December 12, 1998.

Natural color image (RGB = RGB) of a N.Y. Natural color image (RGB = RGB) of a N.Y. Power Authority lake at 1 x 1 ft obtained on Power Authority lake at 1 x 1 ft obtained on

October 13, 1997.October 13, 1997.

Natural color image (RGB = RGB) of a N.Y. Natural color image (RGB = RGB) of a N.Y. Power Authority lake at 1 x 1 ft obtained on Power Authority lake at 1 x 1 ft obtained on

October 13, 1997.October 13, 1997.

Spectral Reflectance Spectral Reflectance Characteristics of Characteristics of Sweetgum Leaves Sweetgum Leaves

((Liquidambar Liquidambar styracifluastyraciflua L.) L.)

Spectral Reflectance Spectral Reflectance Characteristics of Characteristics of Sweetgum Leaves Sweetgum Leaves

((Liquidambar Liquidambar styracifluastyraciflua L.) L.)

1 2

a

3

4

0

5

10

15

20

25

30

35

Blue (0.45 - 0.52 m)

Per

cen

t R

efle

ctan

ce

Green leaf

Yellow

Red/orange

Brown 4

2

1

3

45

40

Green (0.52 - 0.60 m)

Red (0.63 - 0.69 m)

Near-Infrared (0.70 - 0.92 m)

a.

b.

c.

d.

1 2

a

3

4

0

5

10

15

20

25

30

35

Blue (0.45 - 0.52 m)

Per

cen

t R

efle

ctan

ce

Green leaf

Yellow

Red/orange

Brown 4

2

1

3

45

40

Green (0.52 - 0.60 m)

Red (0.63 - 0.69 m)

Near-Infrared (0.70 - 0.92 m)

a.

b.

c.

d.

Spectral Reflectance Spectral Reflectance Characteristics of Characteristics of Selected Areas ofSelected Areas of

Blackjack Oak LeavesBlackjack Oak Leaves

Spectral Reflectance Spectral Reflectance Characteristics of Characteristics of Selected Areas ofSelected Areas of

Blackjack Oak LeavesBlackjack Oak Leaves

Jensen, 2004Jensen, 2004

Hypothetical Hypothetical Example of Example of

Additive Additive Reflectance from Reflectance from A Canopy with A Canopy with

Two Leaf LayersTwo Leaf Layers

Hypothetical Hypothetical Example of Example of

Additive Additive Reflectance from Reflectance from A Canopy with A Canopy with

Two Leaf LayersTwo Leaf Layers

Jensen, 2004Jensen, 2004

Jensen, 2004Jensen, 2004

Distribution of Pixels in a Scene in Distribution of Pixels in a Scene in Red and Near-infrared Multispectral Feature Space Red and Near-infrared Multispectral Feature Space

Distribution of Pixels in a Scene in Distribution of Pixels in a Scene in Red and Near-infrared Multispectral Feature Space Red and Near-infrared Multispectral Feature Space

Jensen, 2004Jensen, 2004

Reflectance Response of a Single Magnolia Leaf Reflectance Response of a Single Magnolia Leaf ((Magnolia grandifloraMagnolia grandiflora) to Decreased Relative Water Content ) to Decreased Relative Water Content

Reflectance Response of a Single Magnolia Leaf Reflectance Response of a Single Magnolia Leaf ((Magnolia grandifloraMagnolia grandiflora) to Decreased Relative Water Content ) to Decreased Relative Water Content

Jensen, 2004Jensen, 2004

Airborne Visible Infrared Imaging

Spectrometer (AVIRIS) Datacube of Sullivan’s

Island Obtained on October 26, 1998

Airborne Visible Infrared Imaging

Spectrometer (AVIRIS) Datacube of Sullivan’s

Island Obtained on October 26, 1998

Imaging Spectrometer Data of Healthy Green Vegetation in the San Imaging Spectrometer Data of Healthy Green Vegetation in the San Luis Valley of Colorado Obtained on September 3, 1993 Using AVIRIS Luis Valley of Colorado Obtained on September 3, 1993 Using AVIRIS

Imaging Spectrometer Data of Healthy Green Vegetation in the San Imaging Spectrometer Data of Healthy Green Vegetation in the San Luis Valley of Colorado Obtained on September 3, 1993 Using AVIRIS Luis Valley of Colorado Obtained on September 3, 1993 Using AVIRIS

Jensen, 2000Jensen, 2000224 channels each 10 nm wide with 20 x 20 m pixels224 channels each 10 nm wide with 20 x 20 m pixels224 channels each 10 nm wide with 20 x 20 m pixels224 channels each 10 nm wide with 20 x 20 m pixels

Hyperspectral Analysis Hyperspectral Analysis of AVIRIS Data of AVIRIS Data

Obtained on September Obtained on September 3, 1993 of San Luis 3, 1993 of San Luis Valley, ColoradoValley, Colorado

Hyperspectral Analysis Hyperspectral Analysis of AVIRIS Data of AVIRIS Data

Obtained on September Obtained on September 3, 1993 of San Luis 3, 1993 of San Luis Valley, ColoradoValley, Colorado

Jensen, 2000Jensen, 2000Jensen, 2000Jensen, 2000

Temporal Temporal (Phenological) (Phenological) CharacteristicsCharacteristics

Temporal Temporal (Phenological) (Phenological) CharacteristicsCharacteristics

Remote Sensing of Vegetation Remote Sensing of Vegetation Remote Sensing of Vegetation Remote Sensing of Vegetation

Predicted Percent Cloud Cover in Four Areas in the United StatesPredicted Percent Cloud Cover in Four Areas in the United StatesPredicted Percent Cloud Cover in Four Areas in the United StatesPredicted Percent Cloud Cover in Four Areas in the United States

Jensen, 2000Jensen, 2000

Phenological Cycle of Hard Red Winter Wheat in the Great PlainsPhenological Cycle of Hard Red Winter Wheat in the Great PlainsPhenological Cycle of Hard Red Winter Wheat in the Great PlainsPhenological Cycle of Hard Red Winter Wheat in the Great Plains

JULJUNMAY AUGAPRMARFEBJANDECNOVOCTSEP

crop establishment

10 14

greening up heading mature

14 14 21 13 425 7 9 5 21 29 34 28 108 days 50

26

Sow Tillering

Emergence

Dormancy Growth resumes

Heading Boot

Dead ripe

Hard doughSoft dough

Harvest

Jointing

Maximum Coverage

Winter Wheat Phenology

snow cover

JULJUNMAY AUGAPRMARFEBJANDECNOVOCTSEP

crop establishment

10 14

greening up heading mature

14 14 21 13 425 7 9 5 21 29 34 28 108 days 50

26

Sow Tillering

Emergence

Dormancy Growth resumes

Heading Boot

Dead ripe

Hard doughSoft dough

Harvest

Jointing

Maximum Coverage

Winter Wheat Phenology

snow cover

Jensen, 2000Jensen, 2000

Phenological Cycles of Phenological Cycles of San Joaquin and San Joaquin and Imperial Valley, Imperial Valley,

California Crops and California Crops and Landsat Multispectral Landsat Multispectral

Scanner Images of One Scanner Images of One Field During A Field During A

Growing SeasonGrowing Season

Phenological Cycles of Phenological Cycles of San Joaquin and San Joaquin and Imperial Valley, Imperial Valley,

California Crops and California Crops and Landsat Multispectral Landsat Multispectral

Scanner Images of One Scanner Images of One Field During A Field During A

Growing SeasonGrowing Season

Jensen, 2000Jensen, 2000

Landsat Thematic Landsat Thematic Mapper Imagery of Mapper Imagery of

the Imperial the Imperial Valley, California Valley, California

Obtained on Obtained on December 10, 1982December 10, 1982

Landsat Thematic Landsat Thematic Mapper Imagery of Mapper Imagery of

the Imperial the Imperial Valley, California Valley, California

Obtained on Obtained on December 10, 1982December 10, 1982

Jensen, 2000Jensen, 2000

Band 1 (blue; 0.45 – 0.52 m) Band 2 (green; 0.52 – 0.60 m) Band 3 (red; 0.63 – 0.69 m)

Band 4 (near-infrared; 0.76 – 0.90 m) Band 5 (mid-infrared; 1.55 – 1.75 m) Band 7 (mid-infrared; 2.08 – 2.35 m)

Band 6 (thermal infrared; 10.4 – 12.5 m)

Sugarbeets

Alfalfa

Cotton

Fallow

feed lot

fl

Ground Reference

Landsat Thematic Mapper Imagery of

Imperial Valley, California, December 10, 1982

Band 1 (blue; 0.45 – 0.52 m) Band 2 (green; 0.52 – 0.60 m) Band 3 (red; 0.63 – 0.69 m)

Band 4 (near-infrared; 0.76 – 0.90 m) Band 5 (mid-infrared; 1.55 – 1.75 m) Band 7 (mid-infrared; 2.08 – 2.35 m)

Band 6 (thermal infrared; 10.4 – 12.5 m)

Sugarbeets

Alfalfa

Cotton

Fallow

feed lot

fl

Ground Reference

Landsat Thematic Mapper Imagery of

Imperial Valley, California, December 10, 1982

Landsat Thematic Landsat Thematic Mapper Color Mapper Color

Composites and Composites and Classification Map of a Classification Map of a Portion of the Imperial Portion of the Imperial

Valley, CaliforniaValley, California

Landsat Thematic Landsat Thematic Mapper Color Mapper Color

Composites and Composites and Classification Map of a Classification Map of a Portion of the Imperial Portion of the Imperial

Valley, CaliforniaValley, California

Jensen, 2000Jensen, 2000

Phenological Cycles Phenological Cycles of Soybeans and of Soybeans and Corn in South Corn in South

CarolinaCarolina

Phenological Cycles Phenological Cycles of Soybeans and of Soybeans and Corn in South Corn in South

CarolinaCarolina

Jensen, 2000Jensen, 2000

JUN MAY APR MAR FEB JAN DEC NOV OCT SEP

Initial growth Harvest Maturity

snow cover 25 cm height

50

75

100

125

Dormant or multicropped

Soybeans

100% ground cover

JUL AUG

Development

50%

JUN MAY APR MAR FEB JAN DEC NOV OCT SEP

Dent/Harvest Tassle 8-leaf

snow cover

25 cm height

50

75

125

Dormant or multicropped

JUL AUG

Dormant or multicropped

100

150

200

250

300

10 - 12 leaf

12-14 leaf

Blister

Corn

100%

50%

a.

b.

JUN MAY APR MAR FEB JAN DEC NOV OCT SEP

Initial growth Harvest Maturity

snow cover 25 cm height

50

75

100

125

Dormant or multicropped

Soybeans

100% ground cover

JUL AUG

Development

50%

JUN MAY APR MAR FEB JAN DEC NOV OCT SEP

Dent/Harvest Tassle 8-leaf

snow cover

25 cm height

50

75

125

Dormant or multicropped

JUL AUG

Dormant or multicropped

100

150

200

250

300

10 - 12 leaf

12-14 leaf

Blister

Corn

100%

50%

a.

b.

SoybeansSoybeansSoybeansSoybeans

CornCornCornCorn

Phenological Cycles Phenological Cycles of Winter Wheat, of Winter Wheat,

Cotton, and Tobacco Cotton, and Tobacco in South Carolinain South Carolina

Phenological Cycles Phenological Cycles of Winter Wheat, of Winter Wheat,

Cotton, and Tobacco Cotton, and Tobacco in South Carolinain South Carolina

Jensen, 2000Jensen, 2000

JULJUNMAY AUGAPRMARFEBJAN DECNOVOCTSEP

SeedTillering Booting HarvestJointing

snow cover 25 cm

50

75

100

100% ground cover

50%

Head Dormant or multicropped

Winter Wheat

JUN MAY APR MAR FEB JAN DEC NOV OCT SEP

Seeding Maturity/harvest Boll

Winter Wheat Phenology

snow cover 25 cm height

50

75

100

125

150

Dormant or multicropped

Cotton

100% ground cover

JUL AUG

Fruiting

Pre-bloom

50%

JUN MAY APR MAR FEB JAN DEC NOV OCT SEP

Development Maturity/harvest Transplanting

snow cover 25 cm height

50

75

100

125

Dormant or multicropped

Tobacco

100%

JUL AUG

Topping

50%

Dormant or multicropped

a.

c.

b.

JULJUNMAY AUGAPRMARFEBJAN DECNOVOCTSEP

SeedTillering Booting HarvestJointing

snow cover 25 cm

50

75

100

100% ground cover

50%

Head Dormant or multicropped

Winter Wheat

JUN MAY APR MAR FEB JAN DEC NOV OCT SEP

Seeding Maturity/harvest Boll

Winter Wheat Phenology

snow cover 25 cm height

50

75

100

125

150

Dormant or multicropped

Cotton

100% ground cover

JUL AUG

Fruiting

Pre-bloom

50%

JUN MAY APR MAR FEB JAN DEC NOV OCT SEP

Development Maturity/harvest Transplanting

snow cover 25 cm height

50

75

100

125

Dormant or multicropped

Tobacco

100%

JUL AUG

Topping

50%

Dormant or multicropped

a.

c.

b.

Winter WheatWinter WheatWinter WheatWinter Wheat

CottonCottonCottonCotton

TobaccoTobaccoTobaccoTobacco

Vegetation Indices Vegetation Indices Vegetation Indices Vegetation Indices

Reflectance Curves for Selected PhenomenaReflectance Curves for Selected Phenomena

Soil LineSoil Line

Infrared/Red Ratio Vegetation IndexInfrared/Red Ratio Vegetation IndexInfrared/Red Ratio Vegetation IndexInfrared/Red Ratio Vegetation Index

The near-infrared (NIR) to red simple ratio (SR) is the first true vegetation The near-infrared (NIR) to red simple ratio (SR) is the first true vegetation index:index:

It takes advantage of the inverse relationship between chlorophyll absorption of It takes advantage of the inverse relationship between chlorophyll absorption of red radiant energy and increased reflectance of near-infrared energy for healthy red radiant energy and increased reflectance of near-infrared energy for healthy plant canopies (Cohen, 1991) .plant canopies (Cohen, 1991) .

The near-infrared (NIR) to red simple ratio (SR) is the first true vegetation The near-infrared (NIR) to red simple ratio (SR) is the first true vegetation index:index:

It takes advantage of the inverse relationship between chlorophyll absorption of It takes advantage of the inverse relationship between chlorophyll absorption of red radiant energy and increased reflectance of near-infrared energy for healthy red radiant energy and increased reflectance of near-infrared energy for healthy plant canopies (Cohen, 1991) .plant canopies (Cohen, 1991) .

SR NIR

redSR

NIR

red

Normalized Difference Vegetation IndexNormalized Difference Vegetation IndexNormalized Difference Vegetation IndexNormalized Difference Vegetation Index

The generic normalized difference vegetation index (NDVI):The generic normalized difference vegetation index (NDVI):

has provided a method of estimating net primary production over varying has provided a method of estimating net primary production over varying biome types (e.g. Lenney et al., 1996), identifying ecoregions (Ramsey et biome types (e.g. Lenney et al., 1996), identifying ecoregions (Ramsey et al., 1995), monitoring phenological patterns of the earth’s vegetative al., 1995), monitoring phenological patterns of the earth’s vegetative surface, and of assessing the length of the growing season and dry-down surface, and of assessing the length of the growing season and dry-down periods (Huete and Liu, 1994).periods (Huete and Liu, 1994).

The generic normalized difference vegetation index (NDVI):The generic normalized difference vegetation index (NDVI):

has provided a method of estimating net primary production over varying has provided a method of estimating net primary production over varying biome types (e.g. Lenney et al., 1996), identifying ecoregions (Ramsey et biome types (e.g. Lenney et al., 1996), identifying ecoregions (Ramsey et al., 1995), monitoring phenological patterns of the earth’s vegetative al., 1995), monitoring phenological patterns of the earth’s vegetative surface, and of assessing the length of the growing season and dry-down surface, and of assessing the length of the growing season and dry-down periods (Huete and Liu, 1994).periods (Huete and Liu, 1994).

NDVI NIR red

NIR redNDVI

NIR red

NIR red

Vegetation Indices

Vegetation Indices

Time Series of 1984 and 1988 NDVI Measurements Derived from AVHRR Global Time Series of 1984 and 1988 NDVI Measurements Derived from AVHRR Global Area Coverage (GAC) Data for the Region around El Obeid, Sudan, in Sub-Saharan Area Coverage (GAC) Data for the Region around El Obeid, Sudan, in Sub-Saharan

Africa Africa

Time Series of 1984 and 1988 NDVI Measurements Derived from AVHRR Global Time Series of 1984 and 1988 NDVI Measurements Derived from AVHRR Global Area Coverage (GAC) Data for the Region around El Obeid, Sudan, in Sub-Saharan Area Coverage (GAC) Data for the Region around El Obeid, Sudan, in Sub-Saharan

Africa Africa

Jensen, 2000Jensen, 2000

Jensen, 2004Jensen, 2004

Vegetation IndicesVegetation Indices

34

3457

577

56

566

TMTM

TMTMNDVI

MSSMSS

MSSMSSNDVI

MSSMSS

MSSMSSNDVI

NIR

redratioSimple

TM

34

3457

577

56

566

TMTM

TMTMNDVI

MSSMSS

MSSMSSNDVI

MSSMSS

MSSMSSNDVI

NIR

redratioSimple

TM

Infrared IndexInfrared IndexInfrared IndexInfrared Index

An Infrared Index (II) that incorporates both near and middle-infrared An Infrared Index (II) that incorporates both near and middle-infrared bands is sensitive to changes in plant biomass and water stress in smooth bands is sensitive to changes in plant biomass and water stress in smooth cordgrass studies (Hardisky et al., 1983; 1986):cordgrass studies (Hardisky et al., 1983; 1986):

Healthy, mono-specific stands of tidal wetland such as Healthy, mono-specific stands of tidal wetland such as SpartinaSpartina often often exhibit much lower reflectance in the visible (blue, green, and red) exhibit much lower reflectance in the visible (blue, green, and red) wavelengths than typical terrestrial vegetation due to the saturated tidal wavelengths than typical terrestrial vegetation due to the saturated tidal flat understory. In effect, the moist soil absorbs almost all energy flat understory. In effect, the moist soil absorbs almost all energy incident to it. This is why wetland often appear surprisingly dark on incident to it. This is why wetland often appear surprisingly dark on traditional infrared color composites.traditional infrared color composites.

An Infrared Index (II) that incorporates both near and middle-infrared An Infrared Index (II) that incorporates both near and middle-infrared bands is sensitive to changes in plant biomass and water stress in smooth bands is sensitive to changes in plant biomass and water stress in smooth cordgrass studies (Hardisky et al., 1983; 1986):cordgrass studies (Hardisky et al., 1983; 1986):

Healthy, mono-specific stands of tidal wetland such as Healthy, mono-specific stands of tidal wetland such as SpartinaSpartina often often exhibit much lower reflectance in the visible (blue, green, and red) exhibit much lower reflectance in the visible (blue, green, and red) wavelengths than typical terrestrial vegetation due to the saturated tidal wavelengths than typical terrestrial vegetation due to the saturated tidal flat understory. In effect, the moist soil absorbs almost all energy flat understory. In effect, the moist soil absorbs almost all energy incident to it. This is why wetland often appear surprisingly dark on incident to it. This is why wetland often appear surprisingly dark on traditional infrared color composites.traditional infrared color composites.

Jensen, 2000Jensen, 2000

II NIRTM 4 MIRTM 5

NIRTM 4 MIRTM 5

II NIRTM 4 MIRTM 5

NIRTM 4 MIRTM 5

Kauth-Thomas TransformationKauth-Thomas Transformation

Kauth-Thomas “Tasseled Cap” TransformationKauth-Thomas “Tasseled Cap” Transformation

Kauth-Thomas Transformation

Kauth-Thomas Transformation

Landsat Thematic Mapper November 9, 1982

Landsat Thematic Mapper November 9, 1982

Soil Adjusted Vegetation Index (SAVI)Soil Adjusted Vegetation Index (SAVI)Soil Adjusted Vegetation Index (SAVI)Soil Adjusted Vegetation Index (SAVI)

Recent emphasis has been given to the development of improved vegetation indices that Recent emphasis has been given to the development of improved vegetation indices that may take advantage of calibrated hyperspectral sensor systems such as the moderate may take advantage of calibrated hyperspectral sensor systems such as the moderate resolution imaging spectrometer - MODIS (Running et al., 1994). The improved indices resolution imaging spectrometer - MODIS (Running et al., 1994). The improved indices incorporate a incorporate a soil adjustment factor soil adjustment factor and/or a and/or a blue band for atmospheric normalizationblue band for atmospheric normalization. . The soil adjusted vegetation index (SAVI) introduces a soil calibration factor, The soil adjusted vegetation index (SAVI) introduces a soil calibration factor, LL, to the , to the NDVI equation to minimize soil background influences resulting from first order soil-NDVI equation to minimize soil background influences resulting from first order soil-plant spectral interactions (Huete et al., 1994):plant spectral interactions (Huete et al., 1994):

An An LL value of 0.5 minimizes soil brightness variations and eliminates the need for value of 0.5 minimizes soil brightness variations and eliminates the need for additional calibration for different soils (Huete and Liu, 1994). additional calibration for different soils (Huete and Liu, 1994).

Recent emphasis has been given to the development of improved vegetation indices that Recent emphasis has been given to the development of improved vegetation indices that may take advantage of calibrated hyperspectral sensor systems such as the moderate may take advantage of calibrated hyperspectral sensor systems such as the moderate resolution imaging spectrometer - MODIS (Running et al., 1994). The improved indices resolution imaging spectrometer - MODIS (Running et al., 1994). The improved indices incorporate a incorporate a soil adjustment factor soil adjustment factor and/or a and/or a blue band for atmospheric normalizationblue band for atmospheric normalization. . The soil adjusted vegetation index (SAVI) introduces a soil calibration factor, The soil adjusted vegetation index (SAVI) introduces a soil calibration factor, LL, to the , to the NDVI equation to minimize soil background influences resulting from first order soil-NDVI equation to minimize soil background influences resulting from first order soil-plant spectral interactions (Huete et al., 1994):plant spectral interactions (Huete et al., 1994):

An An LL value of 0.5 minimizes soil brightness variations and eliminates the need for value of 0.5 minimizes soil brightness variations and eliminates the need for additional calibration for different soils (Huete and Liu, 1994). additional calibration for different soils (Huete and Liu, 1994).

SAVI (1 L) NIR red

NIR red LSAVI

(1 L) NIR red NIR red L

Soil and Atmospherically Adjusted Vegetation Index (SARVI)Soil and Atmospherically Adjusted Vegetation Index (SARVI)Soil and Atmospherically Adjusted Vegetation Index (SARVI)Soil and Atmospherically Adjusted Vegetation Index (SARVI)

Huete and Liu (1994) integrated the Huete and Liu (1994) integrated the LL function from SAVI and a blue-band function from SAVI and a blue-band normalization to derive a soil and atmospherically resistant vegetation index normalization to derive a soil and atmospherically resistant vegetation index (SARVI) that corrects for both soil and atmospheric noise:(SARVI) that corrects for both soil and atmospheric noise:

wherewhere

The technique requires prior correction for molecular scattering and ozone The technique requires prior correction for molecular scattering and ozone absorption of the blue, red, and near-infrared remote sensor data, hence the term absorption of the blue, red, and near-infrared remote sensor data, hence the term p*p*. .

Huete and Liu (1994) integrated the Huete and Liu (1994) integrated the LL function from SAVI and a blue-band function from SAVI and a blue-band normalization to derive a soil and atmospherically resistant vegetation index normalization to derive a soil and atmospherically resistant vegetation index (SARVI) that corrects for both soil and atmospheric noise:(SARVI) that corrects for both soil and atmospheric noise:

wherewhere

The technique requires prior correction for molecular scattering and ozone The technique requires prior correction for molecular scattering and ozone absorption of the blue, red, and near-infrared remote sensor data, hence the term absorption of the blue, red, and near-infrared remote sensor data, hence the term p*p*. .

SARVI p * nir p * rb

p * nir p * rbSARVI

p * nir p * rb

p * nir p * rb

p * rb p* red p * blue p * red p * rb p* red p * blue p * red

Enhanced Vegetation Index (EVI)Enhanced Vegetation Index (EVI)Enhanced Vegetation Index (EVI)Enhanced Vegetation Index (EVI)

The MODIS Land Discipline Group proposed the The MODIS Land Discipline Group proposed the Enhanced Vegetation IndexEnhanced Vegetation Index (EVI) for use (EVI) for use with MODIS Data:with MODIS Data:

The EVI is a modified NDVI with a soil adjustment factor, The EVI is a modified NDVI with a soil adjustment factor, LL, and two coefficients, , and two coefficients, CC11 and and CC22

which describe the use of the blue band in correction of the red band for atmsoperhic aerosol which describe the use of the blue band in correction of the red band for atmsoperhic aerosol scattering. The coefficients, scattering. The coefficients, CC11 , , CC22 , and , and LL, are empirically determined as 6.0, 7.5, and 1.0, , are empirically determined as 6.0, 7.5, and 1.0,

respectively. This algorithm has improved sensitivity to high biomass regions and improved respectively. This algorithm has improved sensitivity to high biomass regions and improved vegetation monitoring thorugh a de-coupling of the canopy background signal and a vegetation monitoring thorugh a de-coupling of the canopy background signal and a reduction in atmospheric influences (Huete and Justice, 1999). reduction in atmospheric influences (Huete and Justice, 1999).

The MODIS Land Discipline Group proposed the The MODIS Land Discipline Group proposed the Enhanced Vegetation IndexEnhanced Vegetation Index (EVI) for use (EVI) for use with MODIS Data:with MODIS Data:

The EVI is a modified NDVI with a soil adjustment factor, The EVI is a modified NDVI with a soil adjustment factor, LL, and two coefficients, , and two coefficients, CC11 and and CC22

which describe the use of the blue band in correction of the red band for atmsoperhic aerosol which describe the use of the blue band in correction of the red band for atmsoperhic aerosol scattering. The coefficients, scattering. The coefficients, CC11 , , CC22 , and , and LL, are empirically determined as 6.0, 7.5, and 1.0, , are empirically determined as 6.0, 7.5, and 1.0,

respectively. This algorithm has improved sensitivity to high biomass regions and improved respectively. This algorithm has improved sensitivity to high biomass regions and improved vegetation monitoring thorugh a de-coupling of the canopy background signal and a vegetation monitoring thorugh a de-coupling of the canopy background signal and a reduction in atmospheric influences (Huete and Justice, 1999). reduction in atmospheric influences (Huete and Justice, 1999).

EVI p * nir p * red

p * nir C1p * red C2 p * blue LEVI

p * nir p * red

p * nir C1p * red C2 p * blue L

Murrells Inlet

Murrells Inlet

Murrells Inlet

Murrells Inlet

Murrells Inlet in South CarolinaMurrells Inlet in South Carolina

Phenological Cycle of Smooth Cordgrass Phenological Cycle of Smooth Cordgrass ((Spartina alternifloraSpartina alterniflora) Biomass in South Carolina) Biomass in South Carolina

Phenological Cycle of Smooth Cordgrass Phenological Cycle of Smooth Cordgrass ((Spartina alternifloraSpartina alterniflora) Biomass in South Carolina) Biomass in South Carolina

Jensen, 2000Jensen, 2000J A S O N0

250

500

750

1000

1250

1500D

ry W

eigh

t B

iom

ass,

g/

m2

F M A M J J D

Live Biomass

Dead Biomass

Smooth Cordgrass (Spartina alterniflora )

J A S O N0

250

500

750

1000

1250

1500D

ry W

eigh

t B

iom

ass,

g/

m2

F M A M J J D

Live Biomass

Dead Biomass

Smooth Cordgrass (Spartina alterniflora )

Phenological Cycle of Cattails and Waterlilies in Par Pond, S.C.Phenological Cycle of Cattails and Waterlilies in Par Pond, S.C.Phenological Cycle of Cattails and Waterlilies in Par Pond, S.C.Phenological Cycle of Cattails and Waterlilies in Par Pond, S.C.

Jensen, 2000Jensen, 2000

Characteristics of the NASA Calibrated Airborne Multispectral Characteristics of the NASA Calibrated Airborne Multispectral Scanner (CAMS) Mission of Murrells Inlet, S.C. on August 2, 1997Scanner (CAMS) Mission of Murrells Inlet, S.C. on August 2, 1997

Characteristics of the NASA Calibrated Airborne Multispectral Characteristics of the NASA Calibrated Airborne Multispectral Scanner (CAMS) Mission of Murrells Inlet, S.C. on August 2, 1997Scanner (CAMS) Mission of Murrells Inlet, S.C. on August 2, 1997

AltitudeAltitude CAMS CAMS MissionMission Relative above- Relative above- Spatial Spatial CAMS CAMS DateDate VisibilityVisibility HumidityHumidity ground-levelground-level ResolutionResolution Spectral ResolutionSpectral Resolution 8/2/978/2/97 clear clear 45% 4000’ 45% 4000’ 3.08 x 3.08 Band 1 (0.42 - 0.52 3.08 x 3.08 Band 1 (0.42 - 0.52 m); bluem); blue

Band 2 (0.52 - 0.60 Band 2 (0.52 - 0.60 m); greenm); green Band 3 (0.60 - 0.63 Band 3 (0.60 - 0.63 m); redm); red

Band 4 (0.63 - 0.69 Band 4 (0.63 - 0.69 m); redm); red Band 5 (0.69 - 0.76 Band 5 (0.69 - 0.76 m); near-m); near-

IRIR Band 6 (0.76 - 0.90 Band 6 (0.76 - 0.90 m); near-m); near-

IRIR Band 7 (1.55 - 1.75 Band 7 (1.55 - 1.75 m); mid-IRm); mid-IR Band 8 (2.08 - 2.35 Band 8 (2.08 - 2.35 m); mid-IRm); mid-IR Band 9 (10.5 - 12.5 Band 9 (10.5 - 12.5 m); TIRm); TIR

AltitudeAltitude CAMS CAMS MissionMission Relative above- Relative above- Spatial Spatial CAMS CAMS DateDate VisibilityVisibility HumidityHumidity ground-levelground-level ResolutionResolution Spectral ResolutionSpectral Resolution 8/2/978/2/97 clear clear 45% 4000’ 45% 4000’ 3.08 x 3.08 Band 1 (0.42 - 0.52 3.08 x 3.08 Band 1 (0.42 - 0.52 m); bluem); blue

Band 2 (0.52 - 0.60 Band 2 (0.52 - 0.60 m); greenm); green Band 3 (0.60 - 0.63 Band 3 (0.60 - 0.63 m); redm); red

Band 4 (0.63 - 0.69 Band 4 (0.63 - 0.69 m); redm); red Band 5 (0.69 - 0.76 Band 5 (0.69 - 0.76 m); near-m); near-

IRIR Band 6 (0.76 - 0.90 Band 6 (0.76 - 0.90 m); near-m); near-

IRIR Band 7 (1.55 - 1.75 Band 7 (1.55 - 1.75 m); mid-IRm); mid-IR Band 8 (2.08 - 2.35 Band 8 (2.08 - 2.35 m); mid-IRm); mid-IR Band 9 (10.5 - 12.5 Band 9 (10.5 - 12.5 m); TIRm); TIR

Nine Bands of 3 x 3 m Nine Bands of 3 x 3 m Calibrated Airborne Calibrated Airborne

Multispectral Scanner Multispectral Scanner (CAMS) Data of Murrells (CAMS) Data of Murrells

Inlet, SC Obtained on Inlet, SC Obtained on August 2, 1997August 2, 1997

Nine Bands of 3 x 3 m Nine Bands of 3 x 3 m Calibrated Airborne Calibrated Airborne

Multispectral Scanner Multispectral Scanner (CAMS) Data of Murrells (CAMS) Data of Murrells

Inlet, SC Obtained on Inlet, SC Obtained on August 2, 1997August 2, 1997

Jensen, 2000Jensen, 2000

Band 1 (blue; 0.45 – 0.52 m) Band 2 (green; 0.52 – 0.60 m) Band 3 (red; 0.60 – 0.63m)

Band 4 (red; 0.63 – 0.69 m) Band 5 (near-infrared; 0.69 – 0.76 m) Band 6 (near-infrared; 0.76 – 0.90 m)

Band 7 (mid-infrared; 1.55 – 1.75 m) Band 9 (thermal-infrared; 10.4 – 12.5 m)Band 8 (mid-infrared; 2.08 – 2.35 m)

Band 1 (blue; 0.45 – 0.52 m) Band 2 (green; 0.52 – 0.60 m) Band 3 (red; 0.60 – 0.63m)

Band 4 (red; 0.63 – 0.69 m) Band 5 (near-infrared; 0.69 – 0.76 m) Band 6 (near-infrared; 0.76 – 0.90 m)

Band 7 (mid-infrared; 1.55 – 1.75 m) Band 9 (thermal-infrared; 10.4 – 12.5 m)Band 8 (mid-infrared; 2.08 – 2.35 m)

Calibrated Airborne Multispectral Scanner Data of Calibrated Airborne Multispectral Scanner Data of Murrells Inlet, S.C. Obtained on August 2, 1997Murrells Inlet, S.C. Obtained on August 2, 1997

Calibrated Airborne Multispectral Scanner Data of Calibrated Airborne Multispectral Scanner Data of Murrells Inlet, S.C. Obtained on August 2, 1997Murrells Inlet, S.C. Obtained on August 2, 1997

Natural Color Natural Color Composite Composite

(Bands 3,2,1 = RGB)(Bands 3,2,1 = RGB)

Natural Color Natural Color Composite Composite

(Bands 3,2,1 = RGB)(Bands 3,2,1 = RGB)

Masked and Contrast Masked and Contrast Stretched Color Stretched Color

CompositeComposite

Masked and Contrast Masked and Contrast Stretched Color Stretched Color

CompositeComposite

Calibrated Airborne Multispectral Scanner Data of Calibrated Airborne Multispectral Scanner Data of Murrells Inlet, S.C. Obtained on August 2, 1997Murrells Inlet, S.C. Obtained on August 2, 1997

Calibrated Airborne Multispectral Scanner Data of Calibrated Airborne Multispectral Scanner Data of Murrells Inlet, S.C. Obtained on August 2, 1997Murrells Inlet, S.C. Obtained on August 2, 1997

Color Infrared Composite Color Infrared Composite (Bands 3,2,1 = RGB)(Bands 3,2,1 = RGB)

Color Infrared Composite Color Infrared Composite (Bands 3,2,1 = RGB)(Bands 3,2,1 = RGB)

Masked and Contrast Masked and Contrast Stretched Color Stretched Color

CompositeComposite

Masked and Contrast Masked and Contrast Stretched Color Stretched Color

CompositeComposite

In Situ In Situ Ceptometer Leaf-Area-Index MeasurementCeptometer Leaf-Area-Index MeasurementIn Situ In Situ Ceptometer Leaf-Area-Index MeasurementCeptometer Leaf-Area-Index Measurement

•• LAI may be computed using a LAI may be computed using a Decagon Accupar CeptometerDecagon Accupar Ceptometer™ that consists of ™ that consists of a linear array of a linear array of 8080 adjacent 1 cm adjacent 1 cm22 photosynthetically active radiation (PAR) photosynthetically active radiation (PAR) sensors sensors along a bar. along a bar.

•• Incident sunlight above the canopy, QIncident sunlight above the canopy, Qaa, and the amount of direct solar energy , and the amount of direct solar energy

incident to the ceptometer, Qincident to the ceptometer, Qbb, when it was laid at the bottom of the canopy , when it was laid at the bottom of the canopy

directly on the mud is used to compute LAI. directly on the mud is used to compute LAI.

•• LAI may be computed using a LAI may be computed using a Decagon Accupar CeptometerDecagon Accupar Ceptometer™ that consists of ™ that consists of a linear array of a linear array of 8080 adjacent 1 cm adjacent 1 cm22 photosynthetically active radiation (PAR) photosynthetically active radiation (PAR) sensors sensors along a bar. along a bar.

•• Incident sunlight above the canopy, QIncident sunlight above the canopy, Qaa, and the amount of direct solar energy , and the amount of direct solar energy

incident to the ceptometer, Qincident to the ceptometer, Qbb, when it was laid at the bottom of the canopy , when it was laid at the bottom of the canopy

directly on the mud is used to compute LAI. directly on the mud is used to compute LAI.

In Situ In Situ Ceptometer Leaf-Area-Index MeasurementCeptometer Leaf-Area-Index MeasurementIn Situ In Situ Ceptometer Leaf-Area-Index MeasurementCeptometer Leaf-Area-Index Measurement

Relationship Between Relationship Between Calibrated Airborne Calibrated Airborne

Multispectral Scanner Multispectral Scanner (CAMS) Band 6 Brightness (CAMS) Band 6 Brightness

Values and Values and in situin situ Measurements of Measurements of Spartina Spartina

alternifloraalterniflora Total Dry Total Dry Biomass (g/mBiomass (g/m22) at ) at

27 Locations in Murrells 27 Locations in Murrells Inlet, SC Obtained on Inlet, SC Obtained on August 2 and 3, 1997August 2 and 3, 1997

Relationship Between Relationship Between Calibrated Airborne Calibrated Airborne

Multispectral Scanner Multispectral Scanner (CAMS) Band 6 Brightness (CAMS) Band 6 Brightness

Values and Values and in situin situ Measurements of Measurements of Spartina Spartina

alternifloraalterniflora Total Dry Total Dry Biomass (g/mBiomass (g/m22) at ) at

27 Locations in Murrells 27 Locations in Murrells Inlet, SC Obtained on Inlet, SC Obtained on August 2 and 3, 1997August 2 and 3, 1997

Jensen, 2000Jensen, 2000

NASA Calibrated NASA Calibrated Airborne Multispectral Airborne Multispectral

Scanner Imagery Scanner Imagery (3 x 3 m) and Derived (3 x 3 m) and Derived

Biomass Map of a Biomass Map of a Portion of Murrells Portion of Murrells

Inlet, South Carolina Inlet, South Carolina on August 2, 1997on August 2, 1997

NASA Calibrated NASA Calibrated Airborne Multispectral Airborne Multispectral

Scanner Imagery Scanner Imagery (3 x 3 m) and Derived (3 x 3 m) and Derived

Biomass Map of a Biomass Map of a Portion of Murrells Portion of Murrells

Inlet, South Carolina Inlet, South Carolina on August 2, 1997on August 2, 1997

CAMS Bands 1,2,3 (RGB) CAMS Bands 6,4,2 (RGB)

TM Bands 5,3,2 (RGB)

Biomass in a Portion of Murrells Inlet, SC Derived from 3 x 3 m

Calibrated Airborne Multispectral Scanner (CAMS) Data Obtained on

August 2, 1997

Total Biomass (grams/m 2)

500 - 749

750 - 999

1000 - 1499

1500 - 1999

2000 - 2499

2500 - 2999

CAMS Bands 1,2,3 (RGB) CAMS Bands 6,4,2 (RGB)

TM Bands 5,3,2 (RGB)

Biomass in a Portion of Murrells Inlet, SC Derived from 3 x 3 m

Calibrated Airborne Multispectral Scanner (CAMS) Data Obtained on

August 2, 1997

Total Biomass (grams/m 2)

500 - 749

750 - 999

1000 - 1499

1500 - 1999

2000 - 2499

2500 - 2999 Jensen, 2000Jensen, 2000

Total Above-ground Biomass Total Above-ground Biomass in Murrells Inlet, S. C. in Murrells Inlet, S. C.

Extracted from Calibrated Extracted from Calibrated Airborne Multispectral Scanner Airborne Multispectral Scanner

Data on August 2, 1997Data on August 2, 1997

Total Above-ground Biomass Total Above-ground Biomass in Murrells Inlet, S. C. in Murrells Inlet, S. C.

Extracted from Calibrated Extracted from Calibrated Airborne Multispectral Scanner Airborne Multispectral Scanner

Data on August 2, 1997Data on August 2, 1997

Total Biomass (grams/m2)

500 - 749

1000 - 1499

750 - 999

1500 - 1999

2000 - 2499

2500 - 2999

Remote Sensing Univariate Statistics - VarianceRemote Sensing Univariate Statistics - Variance

The variance of a sample is the average squared deviation of all possible observations from the sample mean. The variance of a band of imagery, vark, is computed using the equation:

The numerator of the expression is the corrected sum of squares (SS). If the sample mean (k) were actually the population mean, this would be an accurate measurement of the variance. The standard deviation is the positive square root of the variance.

The variance of a sample is the average squared deviation of all possible observations from the sample mean. The variance of a band of imagery, vark, is computed using the equation:

The numerator of the expression is the corrected sum of squares (SS). If the sample mean (k) were actually the population mean, this would be an accurate measurement of the variance. The standard deviation is the positive square root of the variance.

n

BVn

ikik

k

1

2

var

TextureTransformation

TextureTransformation

Second-Order Statistics in the Spatial DomainSecond-Order Statistics in the Spatial Domain

Second-Order Statistics in the Spatial DomainSecond-Order Statistics in the Spatial Domain

Original image

0 1 1 2 3

0 0 2 3 3

0 1 2 2 3

1 2 3 2 2

2 2 3 3 2

Original image

0 1 1 2 3

0 0 2 3 3

0 1 2 2 3

1 2 3 2 2

2 2 3 3 2

h c

0 1 2 3

0 1 2 1 0

1 0 1 3 0

2 0 0 3 5

3 0 0 2 2

h c

0 1 2 3

0 1 2 1 0

1 0 1 3 0

2 0 0 3 5

3 0 0 2 2

Spatial Dependency MatrixSpatial Dependency Matrix

Second-Order Statistics in the Spatial DomainThe Angular Second Moment (ASM):

Second-Order Statistics in the Spatial DomainThe Angular Second Moment (ASM):

2

00

,

kk quant

j

quant

i

jihASM 2

00

,

kk quant

j

quant

i

jihASM

where,

quantk = quantization level of band k (e.g., 28 = 0 to 255)

hc(i, j) = the (i, j)th entry in one of the angular brightness value spatial-dependency matrices,

where,

quantk = quantization level of band k (e.g., 28 = 0 to 255)

hc(i, j) = the (i, j)th entry in one of the angular brightness value spatial-dependency matrices,

Texture MeasurementTexture Measurement

Texture MeasurementTexture Measurement

Moisture Vegetation IndexMoisture Vegetation IndexMoisture Vegetation IndexMoisture Vegetation Index

Rock et al (199) utilized a Moisture Stress Index (MSI):Rock et al (199) utilized a Moisture Stress Index (MSI):

based on the Landsat Thematic Mapper near-ifnrared and middle-infrared bandsbased on the Landsat Thematic Mapper near-ifnrared and middle-infrared bands

Rock et al (199) utilized a Moisture Stress Index (MSI):Rock et al (199) utilized a Moisture Stress Index (MSI):

based on the Landsat Thematic Mapper near-ifnrared and middle-infrared bandsbased on the Landsat Thematic Mapper near-ifnrared and middle-infrared bands

MSI MidIRTM 5

NIRTM4

MSI MidIRTM 5

NIRTM4