An object recognition system uses an image classifier to classify the images into known categories. Extremely Random Forests (ERFs) are very popular in the field of machine learning because of their speed and accuracy. This algorithm is based on decision trees. Their differences compared to classical decision trees are in the choice of the points of division of the tree. The best division to separate the samples of a node into two groups is done by creating random subdivisions for each of the randomly selected features and choosing the best division between those.
Training an image classifier using Extremely Random Forests
Getting ready
In this recipe, we will use ERFs to train our image classifier. We basically construct...