Application of Artificial Intelligence Algorithms for Image Processing

This article describes the approach to machine learning to identify objects that are capable of handling images extremely quickly and reaches high detection rates. This paper describes three main areas for research on image processing. First, it is the introduction of a new image called "Integral image", which allows you to quickly calculate the functions that our detector uses. The second is the learning algorithm based on AdaBoost, which selects a small number of critical visual functions from a larger set and provides extremely efficient classifiers [6]. The third installment is a method of combining increasingly complex classifiers in the "cascade", which allows background areas to quickly reject the image, spending more calculations on promising object-like areas. The cascade can be considered as a specific mechanism of focusing attention. In the face detection system, the system displays detection rates comparable to the best previous systems. This paper describes the process of facerecognition using OpenCV library in Python.