per person /


Home Courses

ML for Image Analytics

ML for Image Analytics


N.V.S. Shankar


Course Attendees

Still no participant

Course Reviews

Still no reviews

Course Name: ML for Image Analytics
Code (Credits): CUML2011 (0-2-2)

Course Objectives:

  1. Understand the fundamental concepts and techniques in image preprocessing and manipulation
  2. Apply machine learning algorithms and techniques to analyze and interpret image data
  3. Evaluate and analyze the performance of machine learning models for image analytics

Learning Outcomes

  1. Demonstrate an understanding of the importance of image preprocessing steps
  2. Apply machine learning algorithms for classification and extraction of data based on input images
  3. Assess the robustness and generalization capabilities of trained models through cross-validation and testing on unseen data

Image Pre-processing

  • Accessing individual pixels using matrix concept
  • Image resize, grey scale conversion, Colour channel splitting
  • Histogram equalisation (CLAHE)

Image Feature Extraction

  • Edge detection (Sobel, Canny), Morphological operations
  • Image segmentation, Image Thresholding, Binary conversion
  • Cluster based segmentation
  • Feature extraction based on size, shape and colour
  • Feature extraction using predefined functions: SIFT, SURF, STAR, ORB.
  • Feature Extraction using convolutional neural network (CNN).

Creation of Feature Matrix by combining Extracted Features

  • Matrix flattening, Horizontal stacking, Vertical stacking, padding.
  • Splitting the feature matrix (training/testing) and labelling.

Classification algorithms

  • Support vector machine (SVM)
  • Different kernels of SVM (linear, polynomial, radial basis function).
  • Gradient Boosting (GB)
  • Multi-layer Perceptron (MLP), Deep Learning.

Complete Lab Manual: ImageAnalyticsLabManual

Each session is of 3hrs
Session 01: Loading an Image, Color Models, Accessing Pixel Intensity values, Color Space conversions
Session 02:Splitting and Merging Channels, Histograms
Session 03: Histogram Equalization, CLAHE
Session 04: Image Resizing and Rotation, Image Arithmetic, Image Bitwise Operations
Session 05: Morphological Operations
Session 06: Image Smoothening, Sharpening and Edge detection using Sobel, Robert Cross-Gradient
Session 07: Laplacian filters, Canny Edge detection
Session 08: Frequency domain: converting to frequency domain, Low pass, High pass, band pass and High boost filtering
Session 09: Thresholding
Session 10: Image Segmentation: Edge Based, Region Based and Cluster Based
Session 11: Matrix flattening, Horizontal stacking, Vertical stacking, padding, Splitting the feature matrix and labelling
Session 12: Feature extraction based on Color, shape and size
Session 13: Feature extraction using predefined functions: SIFT, SURF, STAR, ORB
Session 14: Support vector machine (SVM)- Linear and Kernel
Session 15: Gradient Boosting (GB)
Session 16: Multi-layer Perceptron (MLP) - CNN and Deep Learning

Our Main Teachers

N.V.S. Shankar

Associate Professor