Project Description

This project involves preprocessing images, extracting HOG features, and employing machine learning algorithms for gender and vehicle classification. The HOG (Histogram of Oriented Gradients) algorithm excels in capturing local gradients and orientations within image patches, providing a rich representation of shape and texture.

Through systematic experimentation, we showcase the efficacy of HOG features in training a classifier for diverse image recognition tasks. We present compelling evidence of the method's success in identifying objects with intricate shapes and textures, demonstrating its versatility across various datasets. The proposed approach not only achieves high accuracy but also exhibits efficiency in real-world scenarios.