IMAGE PROCESSING AND PATTERN RECOGNITION CT 725 04
Lecture : 3
Tutorial : 1
Practical : 3/2
Course Objectives:
To be familiar with processing of images, pattern recognition and their applications.
- Introduction to digital image processing
(4 hours)
- Digital image representation
- Digital image processing: Problems and applications
- Elements of visual perception
- Sampling and quantization, relationships between pixels
- Two-dimensional systems
(5 hours)
- Fourier transform and Fast Fourier Transform
- Other image transforms and their properties: Cosine transform, Sine transform, Hadamard transform, Haar transform
- Image enhancement and restoration
(8 hours)
- Point operations, contrast stretching, clipping and thresholding, digital negative, intensity level slicing, bit extraction
- Histogram modeling: Equalization, Modification, Specification
- Spatial operations: Averaging, directional smoothing, median, filtering, spatial low pass, high pass and band pass filtering, magnification by replication and interpolation
- Image coding and compression
(4 hours)
- Pixel coding: run length, bit plane coding, Huffman coding
- Predictive and inter-frame coding
- Introduction to pattern recognition in images
(3 hours)
- Recognition and classification
(5 hours)
- Recognition and classification
- Feature extraction
- Models
- Division of sample space
- Grey level features edges and lines
(6 hours)
- Similarity and correlation
- Template matching
- Edge detection using templates
- Edge detection using gradient models, model fitting
- Line detection, problems with feature detectors
- Segmentation
(3 hours)
- Segmentation by thresholding
- Regions based Segmentation, edges, line and curve detection
- Frequency approach and transform domain
(3 hours)
- Advanced Topics
(4 hours)
- Neural networks and their application to pattern recognition
- Hopfield nets
- Hamming nets, perceptron
Practical:
Laboratory exercises using image processing and pattern recognition packages.
References:
- R. C. Gonzalez and P. Wintz, "Digital Image Processing", Second Edition, Addison-Wesley Publishing.
- K. Castlemann. "Digital Image Processing", Prentice Hall of India Ltd.
- A. K. Jain, "Fundamentals of Digital Image Processing", Prentice Hall of India Pvt. Ltd..
- Sing Tze Bow, M. Dekker, "Pattern Recognition and Image Processing",
- M. James, "Pattern Recognition", BSP professional books.
- P. Monique and M. Dekker, "Fundamentals of Pattern Recognition".
Evaluation Scheme:
The questions will cover all the chapters in the syllabus. The evaluation scheme will be as indicated in the table below:
Chapter |
Hours |
Marks Distribution* |
1, 2 |
4, 5 |
16 |
3, 5 |
8, 3 |
16 |
4, 6 |
4, 5 |
16 |
7, 8 |
6, 3 |
16 |
9, 10 |
3, 4 |
16 |
Total |
45 |
80 |
*Note: There could be minor deviation in mark distribution.
|