Gender and Age based Census System for Metropolitan Cities

This paper presents a smart computer-vision based system which helps ensure disciplined entry of people in places by ascertaining the allowable age and gender of the person, as well as storing the data for census use. The system involves deep learning based facial image classifiers, which predict the gender and age from the image of face captured by the camera in real-time. The actuating response is generated according to the prediction of the computer vision algorithm, which further keeps the record of encountered faces with the respective counts of persons with corresponding facial traits. The log of the entire surveillance process can be monitored online through the integrated IOT capability, thereby can be used as primary data source. The successful experimental runs demonstrate the system’s usability in the real world applications.