Computer Vision
(ECE 3235 )
L T P C
3 1 1 5
(ECE 3235 )
L T P C
3 1 1 5
Instructor: Anjan Kumar Talukdar
email id: anjantalukdar@gauhati.ac.in
Teaching assistant: Mrs. Ananya Choudhury
email id: a.choudhury50@gmail.com
Class timing:
Monday: (9:30 AM - 11:30 AM) (Practical) (SPC lab) and
(2:30 PM - 3:30 PM) (Tutorial) (Room no, 3)
Tuesday : (1:30 PM - 2:30 PM) (Room no. R3)
Wednesday : ( 1: 30 PM - 2: 30 PM) (Room no. 3)
Thursday : ( 1: 30 PM - 2: 30 PM) (Room no. 3)
Course objective:
The course provides an insight into different aspects of Computer Vision and Machine learning, working principles, systems associated, and applications.
Course outcomes:
At the end of the semester, students can
Explain the working of the camera, behavior of various sources, surfaces, shadows, human visual systems, etc.
Perform various image analysis operations on the images and videos such as segmentation, counting objects, shape determination, feature extraction, etc.
Perform mid-level vision analysis of images and videos such as segmentation using clustering, graph-based, etc using advanced algorithms.
Write algorithms for high-level vision analysis such as object detection and classifications using input features and classifiers.
Grading policy:
First-class test : 20 marks
Second class test: 20 marks
Third class test/assignment : 10 marks
End semester examination: 50%
Grading (absolute):
Mark obtained Letter grade Grade point
91 - 100 O 10
81 - 90 A+ 9
71 - 80 A 8
61 - 70 B+ 7
51 - 60 B 6
41 - 50 C 5
31 - 40 D 4
<31 F 0
Programming:
Programming will be done using OpenCV library in Python language.
Prerequisites:
A good background in linear algebra, probability theory, calculus, image processing, and working with Python.
Lectures:
Topic 1 : Introduction to Computer Vision
Topic 2 : Camera model and calibration
Topic 3: Human Visual System
Topic 4: Radiometry
Topic 5: Color Image Processing
Topic 6 : Image analysis
Topic 7 : Boundary descriptor
Topic 8 : Clustering techniques for image segmentation
Topic 8: Stereo Vision
Topic 9 : Features extraction
Topic 10 : Pattern classification
Syllabus:
Module 1: Introduction
Camera- Pinhole and Lens Types; Human Eye; Sensing; geometric Camera Models; Geometric Camera Calibrations; Radiometry; Projections; Transforms- Fourier, Hough and Radon; Sources, Shadows and Shading; Colour- Generation, Human Perception, Representation, Model for an Image Colour; Surface Colour.
Module 2: Image Analysis
Scene Segmentation and Labeling; Counting Objects; Perimeter Measurements; Following and Representing Boundaries; B-Splines; Least Squares and Eigen Vector Line Fitting; Shapes of Regions.
Module 3: Shape Representation and Description
Introduction; Statistical Decision Theory; Pattern Recognition Principles; Clustering Approach- K- Means Clustering; Parametric Approach- Bayes’ Classifier; Relaxation Approach; Shape Similarity-Based Recognition; Expert System.
Module 4: Mid-level Vision
Image Segmentation using K-means clustering and Graph-Theoretic Clustering; Segmentation by fitting a model; Segmentation and fitting using probabilistic methods; Tracking with linear dynamic models.
Module 5: High-Level Vision
Probabilistic and inferential methods- templates using classifiers, building classifiers from class histograms, feature selection, neural networks, support vector machines; Recognition by relations between templates; Geometric templates from spatial relations.
Suggested readings:
Mubarak Shah, “Fundamentals of Computer Vision“.
Forsyth and Ponce, “ Computer Vision-A Modern Approach”.
Richard Szeliski, “Computer Vision: Algorithms and Applications”.
E. Gose and R. Johnsonbaugh and S. Jost, “Pattern Recognition and Image Analysis”.
Simon J.D. Prince, “Computer vision: models, learning and inference”.
E. R. Davies, “Computer and Machine Vision”.
Richard Hartley and Andrew Zisserman “ Multiple View Geometry in Computer Vision”.
Gary Bradski and Adrian Kaehler, “Learning OpenCV”.
Jan Erik Solem, “Programming Computer Vision with Python”.
Leading Journals and Conferences in Computer Vision: