Machine Vision 1392 10 The aim of course

Machine Vision 1392. 10. The aim of course: To give sufficient theoretical depth on important topics on Machine vision and using the Open. CV programming developing environment to implement programming assignments and course project. After graduating this course, it is expected that the students have enough theoretical and practical knowledge to start industrial applications of Machine Vision.

Machine Vision Text Books: 1. Computer Vision Linda G. Shapiro, George C. Stockman Prentice Hall, 2001 2. Digital Image Processing (3 rd Edition) Rafael C. Gonzalez, Richard E. Woods Prentice Hall, 2008 3. Learning Open. CV, Computer Vision with the Open. CV Library Gary Bradski and Adrian Kaebler, O’REILLY, 2008

Course Syllabus Chapter-1 Introduction, What is Machine Vision Its applications Relation of Machine Vision with related fields such as Image Processing, Computer Graphics and Artificial Intelligence

Chapter-2 Digital Image Fundamentals Elements of visual perception Structure of the Human eye Image formation in the Eye Brightness adaption and discrimination Light and Electromagnetic spectrum

Chapter-2 Digital Image Fundamentals Image sensing and acquisition Image acquisition using a single sensor Image acquisition using sensor strips Image acquisition using sensor arrays A simple image formation model Image sampling and quantization Basic concepts in sampling and quantization Representing digital images Spatial and intensity resolution Image interpolation

Chapter-2 Digital Image Fundamentals Some basic relationships between pixels Neighbors of pixels Adjacency, Connectivity, Regions, and Boundary Distance measures An introduction to mathematical tools used in digital image processing Array versus matrix operations Linear versus non-linear operations Arithmetic operations

Chapter-3 Binary Image Analysis Pixels and neighborhoods Applying masks to images Counting the objects in an image Connected component labeling Binary Image Morphology Structuring element Basic morphological operations Dilation and Erosion Opening and Closing The Hit-or-Miss transformation

Chapter-3 Binary Image Analysis Some basic morphological algorithms – – – – Boundary extraction Hole filling Extraction of connected components Convex Hull Thinning Thickening Skeleton Pruning

Chapter-3 Binary Image Analysis Region Properties Region adjacency graphs Thresholding gray-scale images (some basic methods)

4. Image Segmentation Fundamentals Point, Line, and edge detection Background Detection of isolated points Line detection Edge models Basic edge detection More advanced techniques for edge detection Edge linking and boundary detection

4. Image Segmentation Thresholding Foundation Basic Global thresholding Optimum global thresholding and Otsu’s Method Multiple thresholding Variable thresholding Multivariable thresholding Thresholding in un-even illumination

4. Image Segmentation Region-based segmentation Region growing Region splitting and merging Segmentation using morphological watersheds The use of motion in segmentation

5. Color and Shading Color Fundamentals Color Models The RGB color model The CMY and CMYK color model The HIS, YIQ and YUV color models Pseudo color image processing Basics of full color image processing

5. Color and Shading Color Transformation Color slicing Tone and color corrections Color histograms Image segmentation based on color Noise in color images

5. Color and Shading Radiation from one light source Diffuse reflection Specular reflection Darkening with distance Phong model of shading

5. Color and Shading Color constancy The color of objects taken under different lighting conditions (lights different from white) look different from their real color. How can we convert these colors to their real colors as if the image was taken under normal white color.

6. Texture, Texels (Texture element) and Statistics Texture descriptions Quantitative (statistical) texture measures - Co-occurrence matrices - Laws texture energy measures - Tamura texture measure * In computer graphics, textures are represented by array of texles.

6. Texture segmentation Structural approaches Spectral approaches

7. Content based image retrieval Image database examples Database organization - Standard indexing - Spatial indexing - Indexing for content based image retrieval with multiple measures Image database Queries Image distance measures - Color, Texture and Shape similarity measures Precision and Recall measures to evaluate the performance of a CBIR system

8. Representation and description Representation Boundary following Chain codes Signatures Boundary descriptors Regional descriptors Use of Principal components for description An example for object recognition

9. Motion from 2 D image sequences Motion phenomena and applications Image subtraction Computing motion vectors Using point correspondences MPEG compression of video Computing image flow Computing the path of moving points Detecting significant changes in video

10. Perceiving 3 D from 2 D images Intrinsic images Labeling of line drawing from blocks world 3 D cues available in 2 D images The perspective imaging model Depth perception from stereo Estimating correspondences using cross correlation Correspondences using epipolar constrains

10. Perceiving 3 D from 2 D images The thin lens equation Lens distortion

11. Tracking Different vision systems used for motion detection Reference image A control traffic application Corner finding Invariant Features SIFT (Scale Invariant Feature Transform)

11. Tracking Mean-Shift segmentation and Tracking Cam-shift tracking Kalman Filter Particle filters Introducing tracking systems based on machine learning approaches

12. Camera model and calibration Intrinsic camera parameters Extrinsic camera parameters A calibration example

13. Omni-directional mirrors and vision system Applications of Omni-Vision system Design parameters for omni-mirrors Camera calibration in omni-vision Omni-vision application in soccer robots

14. Introducing some industrial applications of machine vision in Iran 1. How to detect defected eggs 2. Selecting the best stones for asphalt

Course Evaluation 1. Mid term exam-1 (2. 5) 2. Mid term exam-2 (2. 5) 2. Final Exam (5) 3. Seminar (Oral presentation of a research paper or a book chapter by students) ( 2 ) 4. Course work (Implementation Machine Vision algorithms using Open. CV) ( 6 ) 5. Course Project (Programming a research topics that may be related to the seminar) ( 2 )