The Real Time Human Detection

Abstract

For many computer vision applications, including visual surveillance, picture retrieval, and driver assistance systems, human detection is crucial. The Project Real Time Human Detection addresses the issues of face recognition, face detection, and human detection. In a given photograph or video, the project is capable of identifying a human and its face. A crucial component of video surveillance is person detection, a technique that predetermines the human forms in images and videos and ignores everything else which plays an irreplaceable role in video Surveillance. The Histogram of Oriented Gradient (HOG) based Descriptor significantly perform better than any other gradient based descriptor. This field is one of the most popular fields of study for many researchers nowadays as it has been already seen its implementation in self-driving car and for security purpose. The use of machine learning and image processing, using Convolution Neural Networks (CNN) and OpenCV, an open-source computer vision library, is the foundation of the entire effort.