A Camera-Based Overtaking Scene Data Acquisition, Recognition and Extraction System

Abstract

The present invention provides a camera-based overtaking scene data collection, recognition and extraction system, including a natural driving data collection device, a target scene recognition subsystem and a target scene mining system, the natural driving data collection device collects target car data, And transmit the target vehicle data to the target scene recognition subsystem and the target scene mining system, the target scene recognition subsystem will identify whether the current scene is the target scene through the system input parameters, and send the recognition result after the recognition operation to The mining system of the target scene, the mining system of the target scene decides whether to perform the scene extraction operation according to the recognition result. The camera-based overtaking scene data collection, recognition and extraction system of the present invention has the functions of automatic scene recognition and extraction. Compared with manual scene recognition and extraction, it can greatly reduce manpower consumption and has higher recognition efficiency.

Description

technical field [0001] The invention belongs to the fields of natural driving data collection and scene recognition, and in particular relates to a camera-based overtaking scene data collection, recognition and extraction system. Background technique [0002] In recent years, with the rapid development of sensor technology and the decline of hardware equipment prices, the commercialization of ADAS technology has gradually matured. At the same time, the more advanced L3~L4 level driver assistance technology has also become the key research direction of major OEMs. The natural driving scene refers to the real traffic conditions encountered by the driver on the actual road. The research on the natural driving scene has important practical significance for the development of driver assistance technology and automatic driving technology, and the compilation of test cases. The extraction of typical natural driving scenes is the basis of scene research. At present, relevant researc...

AI Technical Summary

Problems solved by technology

For the special overtaking scene where the self-vehicle overtakes a large vehicle, due to the large width of the large vehicle and the large fluctuation of the lateral position, when the self-vehicle overtakes a low-speed large vehicle, the meeting distance between the self-vehicle and the large vehicle will be relatively large during the overtaking process. Small, there is a risk of side collision

The negative pressure area at the rear of a large vehicle due to turbulence will give the vehicle a lateral force towards the large vehicle when the vehicle overtakes the large vehicle, which further increases the risk of side collisions and is easy to cause damage to the driver. People bring bad driving experience

Images