Front vehicle parameter identification method based on multitask convolution nerve network

Abstract

The invention discloses a vehicle multiparameter identification method based on a multitask convolution nerve network. The method comprises the following steps of designing and training of a convolution nerve network structure; and vehicle parameter identification based on the convolution nerve network. In the invention, through the convolution nerve network, original data is converted into an abstract high-level expression through a simple and nonlinear model. Therefore, in the convolution nerve network, a recessive character reflecting an essence of an object to be identified can be learned from a lot of training samples. Compared to a shallow learning classifier, by using the method, high expandability is possessed, identification of various kinds of objects in a traffic environment is satisfied and identification precision is high. When being applied to the complex traffic environment, the method has a high anti-environment interference capability. In the invention, the convolution nerve network is applied to multiparameter identification of a vehicle; the trained convolution nerve network is used to identify a type characteristic, pose information and a vehicle light state of the vehicle in an image so that predictable performance of a potential vehicle behavior is enhanced.

Description

technical field [0001] The invention belongs to the field of vehicle intelligence, and in particular relates to a method for identifying parameters of a vehicle in front. Background technique [0002] Vehicle recognition in traffic scenes belongs to the category of vehicle intelligence. Accurate and effective identification of vehicle parameter information is a key factor for improving the intelligence of intelligent vehicles and driver assistance systems (ADAS) and realizing collision avoidance between vehicles, and it is also a key prerequisite for judging and preventing collisions. [0003] The identification of vehicle parameters refers to the process of identifying vehicle targets in traffic scene images and obtaining information that can reflect the possible impact of the preceding vehicle on the vehicle, so that the driver can predict the information and prevent collisions. At present, the recognition method of vehicle parameters in front usually only recognizes a ce...

AI Technical Summary

Problems solved by technology

The limited hierarchical depth of the shallow learning model makes it limited to solving binary classification problems, and it is difficult to deal with the problem of target multi-parameter identification, which has limitations that are not easy to expand

[0004] Most of the existing technologies described above only identify at the level of whether the target vehicle exists, and there is a problem of simplification of vehicle parameter identification, so that it is difficult to achieve simultaneous acquisition of multiple parameters

Images