A driving behavior identification method based on a Kinect camera

Abstract

The invention provides a driving behavior identification method based on a Kinect camera, which comprises the following steps of carrying out skeleton tracking on a foreground target driver by utilizing a video frame collected by the Kinect camera, and positioning skeleton data of the driver; obtaining a plurality of joint point coordinate values of the driver in each frame of the video frame according to the skeleton data of the driver, and performing normalization processing on the joint point coordinate values; performing feature extraction on the joint point coordinate values to generate aplurality of single-frame bone feature vectors; and finally, sending the plurality of single-frame skeleton feature vectors into a preset deep learning model for recognition, and judging whether thebehavior is a standard behavior or a violation behavior. Compared with the prior art, the method can achieve the recognition of the operation behavior of the driver according to the coordinates of a plurality of joint points of the skeleton of the driver and the corresponding skeleton feature vectors, achieves the prevention effect of alarming or early warning, is not influenced by light when skeleton information is extracted based on a Kinect camera, is higher in precision, and is more reliable.

Description

technical field [0001] The invention relates to the fields of computer vision and video image processing, in particular to a driving behavior recognition method based on a Kinect camera. Background technique [0002] At present, the violations of port drivers often directly lead to major port accidents. For example, a violation of the driver may often lead to the shutdown of the entire production line, and in severe cases, it will cause equipment damage and casualties. Through the investigation, it is found that the driver's violations are mainly concentrated in the operator's inattention, and illegal driving behaviors such as playing with mobile phones, answering calls, or taking hands off the joystick for a long time during driving. In view of this, real-time identification and early warning of the driver's current driving behavior has practical value and an important role in improving the safety and reliability of port operations. [0003] In the prior art, although ther...

AI Technical Summary

Problems solved by technology

The traditional recognition method is to extract features from pictures for recognition, such as gradient histogram (HOG), optical flow histogram (HOF), scale invariant feature change (SIFT), but it is often affected by complex environments, such as lighting Interference, etc.

From the above, it can be seen that there are two main problems in the existing behavior recognition: one is how to select powerful features to describe human behavior, and the other is how to model the dynamic time without losing effective information while maximizing use information to identify

Images