Real Time Human Detection and Counting

The goal of this project is to use OpenCv and Tensorflow to build and implement a human detector and counting system. The quantity of people in a given location at a given moment is measured using human detecting technologies. The accuracy of this measurement is totally determined by the sophistication of the equipment employed. This project investigates and reports benchmarks for detecting and enumerating humans through real time images, videos and camera. This is very useful in various image processing and performing computer vision tasks. This schemes have been implemented in Python programming language, and using various tech-stacks like OpenCv, Tensorflow, etc. To build and implement such a system. It was attempted to assess its viability, improve its correctness, and highlight the problems encountered in its execution. The experiment was declared successful since it was able to detect and count a certain number of persons in a given location using photographs, Videos and camera. There are various vision-based algorithms for counting people. For various indoor and outdoor circumstances, each algorithm performs differently in terms of efficiency, adaptability, and accuracy. For this reason, the most widely used Haar Cascade Classifier and Histogram of Oriented Gradient based approaches are reviewed in our study with regard to many aspects such as camera orientation, lighting, occlusion, and so on. The accuracy and speed with which these algorithms performed in various settings indicates the need for more accurate and faster people counting methods.