SUK · Third Year B.Tech. Computer Science and Engineering (Artificial Intelligence and Data Science) · NEP 2020
Computer Vision
AI&DS3104 · Elective I · Semester V · 3 credits · ISE 10 + mid-semester 30 + ESE 60
One of the Elective I options.
Unit 1: Introduction to Computer Vision
5 hoursHuman versus computer vision; applications; image formation and camera models; image acquisition and representation; basics of digital image processing.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Computer vision pipeline
- Camera models and calibration
- Digital image representation
- Sampling and quantisation
Unit 2: Image Processing Fundamentals
7 hoursImage enhancement; spatial and frequency-domain filtering; Sobel and Canny edge detection; smoothing and sharpening; morphological operations.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Intensity transformations
- Frequency domain filtering
- Gradient edge detectors
- Canny edge detector
- Sharpening filters
- Morphological operations
Unit 3: Feature Detection and Description
6 hoursHarris and FAST corner detection; SIFT, SURF and ORB descriptors; feature matching; image registration and alignment.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Corner detection
- SIFT, SURF and BRISK features
- Hough transform and RANSAC
Unit 4: Image Segmentation and Object Detection
6 hoursThresholding, region-based segmentation and contour detection; traditional object detection; an introduction to CNN-based detectors.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Thresholding and Otsu's method
- Region-based segmentation
- Active contours
- Object detection
- R-CNN family, SSD and YOLO
Unit 5: Motion Analysis and Tracking
6 hoursOptical flow, background subtraction and object tracking; Kalman and particle filters; video analytics.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Optical flow
- Background subtraction and motion segmentation
- Object tracking with Kalman filters
- Particle filters and tracking under occlusion
Unit 6: Deep Learning for Computer Vision
6 hoursConvolutional neural networks and transfer learning; YOLO and SSD detectors; face recognition systems; ethical issues in vision-based AI.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- CNNs for image classification
- Transfer learning
- Face detection and recognition
- Facial recognition and public surveillance
Unit contents are summarised from the official SUK syllabus . AiCanCode.org is not affiliated with Shivaji University, Kolhapur.