More Related Content
Similar to PPT s01-machine vision-s2
![ANPARA THERMAL POWER STATION[1] sangam.pdf](https://cdn.slidesharecdn.com/ss_thumbnails/anparathermalpowerstation1sangam-251121115219-9261cde4-thumbnail.jpg?width=640&height=640&fit=bounds)
PPT s01-machine vision-s2
- 1. Course: Machine Vision Introductionto Machine Vision Session 01 D5627 – I Gede Putra Kusuma Negara, B.Eng., PhD
- 2. Outline • Course introduction •Related fields • Optical Illusions • Application areas • Software • Resources
- 3.
- 4. Course Description Computing withimages is no longer just for the realm of the sciences, but also for the arts and social sciences. Machine Vision extracts descriptions of the world from picture or sequence of pictures and relies on a solid understanding of cameras and of the physical process of image formation. This course begins with introducing the concept of vision system, image formation, and image filtering. It then addresses some aspects of machine vision such as feature detection, feature extraction, pattern recognition, and image matching. Through this course, students will also explore some applications of machine vision.
- 5. Learning Outcome • LO1:Understand various aspects of machine vision algorithms • LO2: Comprehend the importance of machine vision applications • LO3: Apply the concept of machine vision to solve real-world engineering problems • LO4: Analyze the enabling technologies of machine vision applications • LO5: Evaluate the advances and research challenges in machine vision • LO6: Design new approaches that can improve the machine vision applications
- 6. Textbooks Forsyth. (2011). ComputerVision a Modern Approach (2nd Edition). Prentice Hall. New Jersey. ISBN-10: 013608592X. ISBN-13: 978-0136085928 Szeliski. (2010). Computer Vision: Algorithms and Applications. Springer. London. ISBN-13: 978-1848829343. ISBN-10: 1848829345 Gonzales. (2011). Digital Image Processing (3rd Edition). Prentice Hall. New Jersey. ISBN-10: 013168728X. ISBN-13: 978-0131687288
- 7. Syllabus Session Topics SubTopics 1 Introduction to Machine Vision Course introduction Related fields Optical Illusions Application areas Software Resources 2 Human Visual System and Digital Camera Human Visual System Digital Cameras Image Formation Camera and Image Plane Coordinates Image Sensing 3 Light and Color The Physics of Color Human Color Perception Representing Color A Model of Image Color
- 8. Syllabus Session Topics SubTopics 4 Image Filtering (1) Filtering in Spatial Domain First derivative filters Second derivative filters 5 Image Filtering (2) Filtering in Frequency Domain Fourier Transform Wavelet Transform 6 Feature Detection Edge Detection Canny Edge Detector Interest Point and Corner Harris Corner Detector 7 Topics review and midterm examination
- 9. Syllabus Session Topics SubTopics 8 Shape Features Thresholding Identifying Boundary Chain Code Fourier Descriptor Identifying Region Moments 9 Texture Features Texture Analysis Structural Approaches Statistical Approaches LBP, GLCM, Laws Texture Energy, Fourier Power Spectrum, Wavelet Texture Features Texture Segmentation 10 Recognition (1) Pattern Recognition Supervised Learning Classification
- 10. Syllabus Session Topics SubTopics 11 Recognition (2) Unsupervised Learning Clustering K-Means Clustering Agglomerative Hierarchical Clustering 12 Image Matching Definition Invariant Features Scale Invariant Feature Transform Feature Matching 13 Sample Applications Content Based Image Retrieval Face Recognition 14 Topics review and final examination
- 11. Evaluation Assessment Activity Weight LearningOutcomes 1 2 3 4 5 6 Attendance (F2F) 10% Discussion Forum Activity 10% Assignment 20% Midterm Examination 30% - - Final Examination 30% - -
- 12. Class Policies • Studentmust active in classroom discussion forum, responding to lecturer’s questions and discussing with classmates • Student must active in team room, especially discussing team assignment • Student must read learning material and other references before class, reading/case will be distributed before class, team/group and group discussion/presentation will be notified before class • Student must complete and submit all personal assignment and team assignment • Do not rely on handout distributed by lecturer, student can use other references • Achieve a satisfactory average grade on assignments and examinations. • Penalty for cheating and plagiarism will be extremely severe. If you are not sure about certain activities, consult the instructor. Standard academic honesty procedure will be followed and active cheating and plagiarism automatically results FAIL in the final grade.
- 13. What is MachineVision? • A process that produces from images of the external world a description that is useful to the viewer and not cluttered with irrelevant information (Marr) • Construction of explicit, meaningful descriptions of physical objects from images (Ballard and Brown) • To make useful decisions about real physical objects and scenes based on sensed images (Shapiro and Stockman) Some industrial applications of computer vision
- 14.
- 15. Related Fields: ImageProcessing • Image Processing – Imagein, imageout – Usually low level techniques (e.g., image enhancement, compression, edge detection) – Quantitative measurements • Computer Vision – Extracting symbolic descriptions – Higher level techniques (e.g., object recognition) – Semantic (quantitative or qualitative) output • Image processing techniques are often used in computer vision
- 16. Related Fields: ComputerGraphics • Computer vision is the inverse of computer graphics Computer Vision • The forward process is unique, the inverse process is not! 3D Models of objects, locations Lighting information Camera parameters Images Computer graphics Computer vision
- 17. Related fields • Patternrecognition – Recognition of patterns (classification) – Inputs often represented as feature vectors – Techniques useful for 2D and constrained 3D image recognition problems, but usually too limited for general 3D problems • Photogrammetry – Concerned with accurately measuring properties from images – An older field - historically focused on remote sensing (e.g., images from airplanes or satellites) – Computer vision concerned with more than just measuring – However, many techniques are the same or similar
- 18. Vision is noteasy for computer • Objects can be highly variable in shape – E.g., trees, cars, animals, ... • Loss of information in sensing process – 3D objects projected onto 2D images • Missing data – Occlusions and hidden surfaces – Shadows and noise obscure signal • Confounding effects – Observed color may be due to – object albedo or scene lighting
- 19. Approach to Solution •Apply assumptions and a priori knowledge to recover the most likely description • Use knowledge of object shape and the lighting that is present (if available) • Use information from multiple images (stereo, motion sequences) • Guess based on cues – Shading, texture, geometry – Knowledge about typical real world objects
- 20.
- 21. Optical Illusions • Humanvisual system is good at picking out structure from noisy, incomplete, and missing data • We make and use assumptions about the real world to do this • Optical illusions occur when these assumptions are incorrect
- 22. Optical Illusions • BrightnessAdaptation & Discrimination An example of simultaneous contrast
- 23. Optical Illusions • BrightnessAdaptation & Discrimination (cont…)
- 24.
- 25.
- 26. Application Areas (example) •Industrial inspection – Find known objects in the scene – Measure dimensions, verify features • Optical character recognition – Processing scanned text pages – Detect and identify characters
- 27. Application Areas (example) •Scene modeling – Find and identify objects (ships, buildings, roads) – Create models of scenes from multiple images
- 28. Application Areas (example) •Target recognition – Find enemy vehicles (which are trying not to be found!) • Human Interfaces – Detect faces and identify people – Recognize gestures, activities
- 29. Application Areas (example) •Robotics – Recognize objects – Estimate motion and position • Medical modeling – Surgical planning and navigation
- 30. Software • Matlab – Commercialpackage (but only $99 for students) – Interpreted language – easy to prototype – Image processing toolbox; computer vision toolbox; user contributed toolboxes • OpenCV – Free – www.opencv.org • Several applications have simple image processing and computer vision functions – Image – CVIP tools
- 31. Resources • Some keyjournals – IEEE Trans. Pattern Analysis & Machine Intelligence – International Journal of Computer Vision • Some key conferences – Computer Vision and Pattern Recognition (CVPR) – International Conference on Computer Vision (ICCV) • Some good websites – Extensive collection of material - http://www.cs.cmu.edu/~cil/vision.html – Indexed bibliography - http://iris.usc.edu/Vision-Notes/bibliography/contents.html
- 32. Acknowledgment Some of slidesin this PowerPoint presentation are adaptation from various slides, many thanks to: 1. Professor William Hoff, Department of Electrical Engineering & Computer Science (http://inside.mines.edu/~whoff/) 2. Dr. Brian Mac Namee, School of Computing at the Dublin Institute of Technology (http://www.comp.dit.ie/bmacnamee/gaip.htm)
- 33.