Human counter using deep learning

Artificial Intelligence

Counting people in visual inspection is a very difficult and complex problem. Automatic counting of people in public places is very important for security management. Many ideas and methods are planned in advance. This model focuses on creating a physical counter using deep learning (DL) technology. The aim is to introduce a system that can count the number of people in a venue using image recognition algorithms. Improve the efficiency and effectiveness of human search and calculation of specific medications for stores by creating a system with a graphical user interface and management functions. This method does not provide accuracy and high performance in serious situations. To provide solutions that are more accurate to people counting and self-identification methods, a method based on expectation extraction and expectation maximization (EM) is now proposed. This study also means social distance measures along with covid-19 safety measures. The Single Shot Detection algorithm (SSD) uses the camera's live stream and the Convolutional Neural Network (CNN) will identify the person and provide an identification number for calculation. Keywords: real-time, human detection, deep-sort, yolov3.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.