Classification

Once you have extracted the features for all the samples in the dataset, it is time to start the classification process. The target of this classification process is to learn how to make accurate predictions automatically based on the training examples. There are many approaches to this problem. In this section, we will talk about machine learning algorithms in OpenCV, including neural networks, support vector machines, and k-nearest neighbors.

Classification process

Classification is considered supervised learning. In a classification problem, a correctly labelled training set is necessary. A model is produced during the training stage which makes predictions and is corrected when predictions are wrong. Then, the model is used for predicting in other applications. The model needs to be trained every time you have more training data. The following figure shows an overview of the classification process:

Overview of the classification process

The choice of learning algorithm to...