Learning OpenCV 4 Computer Vision with Python 3 - Third Edition

Using HOG to describe regions of an image

For each HOG cell, the histogram contains a number of bins equal to the number of gradients or, in other words, the number of axis directions that HOG considers. After calculating all the cells' histograms, HOG processes groups of histograms to produce higher-level descriptors. Specifically, the cells are grouped into larger regions, called blocks. These blocks can be made of any number of cells, but Dalal and Triggs found that 2x2 cell blocks yielded the best results when performing people detection. A block-wide vector is created so that it can be normalized, compensating for local variations in illumination and shadowing. (A single cell is too small a region to detect such variations.) This normalization ...