{"tags":[{"tag":"textarea","content":"- histogram of oriented gradients\n- Divides image into localized regions\n- Generates histogram for each localized region\n- Histograms are created using gradients and orientation of pixel intensities\n- In this way, different shape accounts for different histogram\n- How it works:\n 1. Convert image to grayscale\n 2. calculate magnitude and angle of gradients (you may use sobel)\n 3. Divide images into grid-cells\n 4. For each cell, calculate the histogram of gradient direction\n 5. Block-normalize cells to reduce imabalanced highlights \/ shadows \n 6. Use this combination as feature vector","code_language":"whatever"}]}