Design and Practice Analysis of Video Desensitization for Intelligent Connected Vehicles Desensitization design and practice analysis
Tonghong Chong, Chao Ma, Yanan Zhang, Tianyu Liu · 2022 International Conference on Image Processing, Computer Vision and Machine Learning (ICICML) · 2022
There are many application scenarios and large data volume of intelligent networked vehicle data, and the data security problem of the Internet of Vehicles, especially the illegal collection of external faces and license plates, has become increasingly prominent. In order to solve the common problem of deploying desensitization algorithm at vehicle end, the lightweight design of desensitization algorithm for vehicle end deployment is completed by selecting NanoDet puls vehicle model. Combine the license plate and face into a model, and successfully deploy it in Raspberry Pie. It can achieve face and license plate detection rate $\gt 96$% and false detection rate$\lt 5$%.