Dangerous items recognition and face recognition system based on K210 and YOLO
Pengna Jia, Zhengkang Zhou, Qingyu Wang · 2022
With the rapid development of artificial intelligence technology, dangerous items detection and face detection industry has gradually become the mainstream application technology, if you are using a general-purpose processor face detection for image data processing which is complete, it will produce a certain time, and there will be a single function, low computing power, low efficiency and so on. In this paper, a KPU neural network accelerator based on K210 chip is designed. The neural network model calculation of dangerous items detection and face detection is completed by YOLO network, and the functions of dangerous items detection and face detection are quickly realized. Experimental results show that the detection speed of the system is fast and the recognition result is more accurate.