Safe pre-warning device for anti-collision at rear of vehicle based on binocular stereo vision

Abstract

The invention relates to a safe pre-warning device for anti-collision at the rear of a vehicle based on binocular stereo vision. The device consists of five parts, namely a vehicle speed sensor, a vehicle steering signal switch, an image acquisition module, a pre-warning module and a controller module, wherein the vehicle speed sensor is an electromagnetic vehicle speed sensor; the image acquisition module comprises a lens R, a lens L, a COMS (Complementary Metal Oxide Semiconductor) image acquisition sensor R, a COMS image acquisition sensor L and a bracket, and is arranged at the back of a vehicle roof in a vehicle compartment; the pre-warning module comprises a pre-warning signal lamp driving module, an anti-rear end collision pre-warning signal lamp and a road merging pre-warning signal lamp; the controller module comprises a real time digital signal processor, a data memory, a program memory, a power supply and a peripheral communication circuit; the vehicle speed sensor transmits detected signals Vf and T to the controller module; the image acquisition module transmits acquired signal images IR and IL to the controller module; and the controller module processes the signals Vf, T, IR and IL to generate a pre-warning lamp control signal I, and respectively transmits anti-collision pre-warning information to a driver of the vehicle and a driver of a following vehicle through the anti-rear end collision pre-warning signal lamp and the road merging pre-warning signal lamp.

Description

【Technical field】 [0001] The present invention relates to a vehicle driving safety early warning system, more particularly, refers to a vehicle rear collision safety early warning device based on binocular stereo vision. 【Background technique】 [0002] Improving traffic safety and efficiency has always been a hot topic in the automotive industry. Preventing vehicle rear-end collision and merging collision accidents has always been the key task of the vehicle rear collision safety early warning device. At present, such devices generally use radar, laser or infrared detection as the rear collision avoidance sensor of the vehicle, and provide early warning of the danger of rear-end collision and merging by detecting the relative speed, acceleration and distance of the vehicle. [0003] However, the vehicle rear collision avoidance safety warning device using this type of active sensor cannot distinguish the vehicle from other obstacles (roadblocks, road railings, green belts, ...

AI Technical Summary

Problems solved by technology

[0003] However, the vehicle rear collision avoidance safety warning device using this type of active sensor cannot distinguish the vehicle from other obstacles (roadblocks, road railings, green belts, etc.), so the false alarm rate is high; When the width of the road section changes or the vehicle is not driving in the middle, it is often impossible to effectively perform merge safety and rear-end collision prevention warnings; this type of device generally cannot monitor multiple target vehicles behind the vehicle in real time at the same time, so the anti-collision warning capability is limited; When the device is in the safety warning of merging lanes, it generally only regards whether there are cars in a certain preset range of merging lanes as the only basis for alarming, and does not consider the relative speed, acceleration and position relationship between vehicles. Therefore, it is obtained at the expense of traffic efficiency. Road safety early warning, and can not effectively reduce the driver's driving intensity; this type of device only judges the alarm based on the relative distance between vehicles, acceleration or speed threshold values ​​in the anti-rear collision warning, and does not calculate the best for different driving conditions. The safe distance between vehicles, so the traffic efficiency is low, and the false alarm rate is high, which obviously needs to be improved

Images