A Performance evaluation of ESP32 Camera Face Recognition for various projects

Abstract

Images Recognition and Face Recognition are a form of artificial intelligence developed on Robots and Smart CCTV, such as technology currently developing, such as YOLO Real-time Object Detection and Images Recognition using Python language. In this research, the tool trying to be used is the ESP32 camera and his approach to several types of study, such as smart door locks.  To activate this ESP32 camera, use the ATmega 328 mini Microcontroller, Arduino board, and FTDI Programmer.  The thing that can be analyzed is the delay or inference time (ms) on the ESP32 camera.  Face recognition technology in this research is devoted to using the ES32 camera, which is expected to be an inexpensive image recognition solution.  This research using the ESP32 camera was made as easy as possible to be applied by everyone, but by not leaving the analysis side, one of which is on the time side of the camera's detection of the object in front of it, how far and fast (ms), this object can be explicitly detected.  Speed is seen from the resolution specifications used for face recognition and image recognition, such as CIF and QVGA, and throughput (bytes) analysis of ESP32-Cam and Server using Wireshark.