Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure...

13
Goals of Computer Goals of Computer Vision Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Transcript of Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure...

Page 1: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Goals of Computer VisionGoals of Computer Vision

• To make useful decisions based on sensed images

• To construct 3D structure from 2D images.

Page 2: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Related Areas of Computer Related Areas of Computer VisionVision

• Computer Graphics

• Image Processing

• Pattern Recognition

• Artificial Intelligence

• Virtual Reality

Page 3: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

1. Human Face Identification2. Image Database Query3. Inspecting Products4. Examining the Inside of a Human Head (MRI)5. Optic Character Recognition (OCR)6. Analyze Satellite Image7. Unmanned Autonomous Vehicle (UAV)

Applications of Computer Applications of Computer VisionVision

Page 4: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Human Visual PerceptionHuman Visual Perception

Page 5: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Human Human eyeeye

Page 6: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

How do we see an object?How do we see an object?

Light

Object

Eye

• Luminance Lightness Rods• Chrominance Color Cones

Human eye is more sensitive to luminance than to chrominance

Page 7: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Cones & RodsCones & Rods( day & night )( day & night )

Page 8: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Machine Visual PerceptionMachine Visual Perception

Page 9: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Digital ImageDigital Image

Page 10: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Sampling & Sampling & QuantizationQuantization

Page 11: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Types of ImagesTypes of Images

• Analog Image

• Digital Image

1. Binary Image2. Gray-scale Image3. Color (Multispectral) Image

Page 12: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Vector Image

Bitmap Image• RAW no header• RLE (Run-Length Encoding)• PGM,PPM,PNM (Portable Gray Map)• GIF (Graphics Interchange Format) no more than 256 colors• TIF (Tag Image File Format) Scanner• EPS (Encapsulated Postscript) Printer• JPEG (Joint Photographic Experts Group) Compression ratio• MPEG (Motion Picture Experts Group) Video

Image FormatsImage Formats

Page 13: Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.

Comparison of Image Comparison of Image FormatsFormats