How to Train a Object Detection Engine with HOG in OpenCV
In the previous post, you saw that OpenCV can extract features from an image using a technique called the Histogram of Oriented Gradients (HOG). In short, this is to convert a “patch” of an image into a numerical vector. This vector, if set up appropriately, can identify key features within that patch. While you can use HOG to compare images for similarity, one practical application is to make it the input to a classifier so you can detect objects in an image.
In this post, you will learn how to create a classifier with HOG. Specifically, you will learn:
- How to prepare input data for classifier training
- How to run the training and save the model for reuse