CourseName:Computer Vision Lab
Course Code: CSP-422
Name: Sahil
UID: 20BCS2574
Experiment:1.1
Aim:
Write a program to implement various feature extraction techniques for image
classification.
Software Required:
Any Python IDE e.g., PyCharm
Description:
Here's a concise description of the various feature extraction techniques for
image classification and an outline of the experiment:
Feature Extraction Techniques for Image Classification:
• SIFT (Scale-Invariant Feature Transform): Identifies keypoints and extracts local invariant
descriptors, robust to scale, rotation, and illumination changes.
• SURF (Speeded-Up Robust Features): Detects and describes local features, computationally
efficient, suitable for real-time applications.
• HOG (Histogram of Oriented Gradients): Computes gradient orientations' distribution,
effective for shape and edge information, useful in object detection.
• CNN (Convolutional Neural Networks): Deep learning model that learns hierarchical
features, revolutionized image classification, excels in various tasks.
• Color Histograms: Captures color distribution, quantizes pixel colors into bins, effective for
certain image classification problems.
• LBP (Local Binary Patterns): Encodes texture by comparing pixel intensities with neighbors,
useful for texture analysis and classification.
• Gabor Filters: Captures localized frequency and orientation information using linear filters,
applied in texture and recognition tasks.
• Deep Convolutional Features: Extracts features from pre-trained CNN models' intermediate
layers, retains high-level semantics, generalizes well.
