Collision time calculation method and device, readable storage medium and computer equipment

Abstract

The embodiment of the invention provides a collision time calculation method and device, a readable storage medium and computer equipment. In the technical scheme provided by the embodiment of the invention, a target image is calculated through a straight line detection algorithm and a perspective transformation algorithm to generate an aerial view, the aerial view is calculated through a horizontal integration algorithm to obtain the safety line of the vehicle, and the target object around the vehicle is identified through a target detection algorithm. A vertical coordinate pixel distance between the safety line and the target object is calculated through the collision time model to obtain the collision time between the target object and the vehicle, auxiliary and manual calibration through a GPS is not needed, online calibration can be conducted when the pose of the camera changes greatly, time and labor are saved, operation is easy, and cost is low.

Description

【Technical field】 [0001] The invention relates to the field of data processing, in particular to a collision time calculation method, device, readable storage medium and computer equipment. 【Background technique】 [0002] Studies have found that driver inattention, fatigue driving and immature driving behavior are the main factors causing traffic accidents. Under this background, collision warning technology has been developed rapidly. [0003] In related technologies, there are mainly two methods for calculating the collision time: a method based on millimeter wave radar and a method based on vision. Among them, the method based on millimeter-wave radar has high calculation accuracy, but the cost is relatively expensive. The cost of the vision-based method is relatively low, and generally consists of vehicle distance detection and speed detection to calculate the collision time. Speed ​​detection is generally provided by a Global Positioning System (GPS). There are two m...

AI Technical Summary

Problems solved by technology

[0004] However, the above two vehicle distance detection methods require GPS assistance and manual calibration, which is time-consuming and laborious. Once the camera pose changes, the actual camera pose or mapping table needs to be re-calibrated manually, and the calibration cost is relatively high.

Therefore, the collision time calculation method in the related art requires GPS for assistance and manual calibration, which is time-consuming, labor-intensive, complicated to operate, and high in cost

Images