A method and related device for biometric identification of driver's identity

Abstract

The present application discloses a method and a related device for biometric identification of a driver's identity, so as to avoid loopholes such as a false login system and false login with an identity medium. The application includes: obtaining an infrared picture and a depth picture of the driver's face; identifying the infrared picture, and obtaining a face area; extracting a face feature point according to the face area; according to the face feature point and the depth Whether the picture is a living body is detected by the support vector machine algorithm, and if so, the identity authentication is performed on the driver's face picture. The support vector machine algorithm is a generalized linear classifier for binary classification of data.

Description

technical field [0001] The embodiments of the present application relate to the technical field of intelligent driving, and in particular, to a method and a related device for biometric identification of a driver's identity. Background technique [0002] In recent years, with the rapid increase in the number of motor vehicles and drivers in our country, road traffic accidents have become a serious social problem. According to statistics, more than 90% of road traffic accidents and road traffic congestion are caused by drivers' poor driving behavior. Therefore, it is particularly important to train drivers to develop good driving habits. Driving habits, in addition to paying attention to the way of training, the length of practice is also very important. [0003] The traditional driving training timing method is to install timing equipment on the training vehicle. When students and coaches get on the bus and start training, they log in to the timing system through identifica...

AI Technical Summary

Problems solved by technology

[0004] However, this method is prone to loopholes such as false login system and fraudulent use of identity media for false login. However, this method still has the cheating loopholes of misappropriation and illegal imitation of biometric features, and the continuous video recording method not only has high storage costs, but also has a huge workload in screening and detecting fraud. At the same time, the face recognition method based on vision Susceptible to light interference, resulting in poor recognition

Images