Monitoring equipment-based target object speed calculation method

Abstract

The invention relates to a monitoring equipment-based target object speed calculation method, and aims at calculating speeds of target objects according to video images shot by monitoring equipment mounted on running vehicles. The method comprises the following steps of: judging whether a coordinate, under an image physical coordinate system, of an original video image is known or not, if the judging result is positive, removing distortion of the video image through coordinate transformation, and if the judging result is negative, removing the distortion of the video image through an image correction tool so as to obtain a corrected video image; establishing a relative two-dimensional coordinate system, and calculating a transformation parameter between the relative two-dimensional coordinate system and an image pixel coordinate system; calculating a coordinate, under the relative two-dimensional coordinate system, of a grounding point of a target object according to the normalized andnonlinearly optimized transformation parameter, so as to obtain relative positions of the target object and a running vehicle; and calculating a running speed of the target object according to changeconditions of the relative positions of the target object and the running vehicle. Compared with the prior art, the method has the advantages of removing image distortion and being small in result error and strong in operability.

Description

technical field [0001] The invention relates to the technical field of intelligent transportation, in particular to a method for calculating the speed of an object based on monitoring equipment. Background technique [0002] At present, with the widespread use of monitoring equipment (such as driving recorders, surveillance cameras, etc.) in vehicles and road traffic, we can easily understand the driving picture of the vehicle and even the occurrence of road traffic accidents from the recorded video images. For example, the driving recorder can record the real-time speed, acceleration and other information of the vehicle during driving, but we cannot directly know the relative position and speed information of the front target (such as vehicles, pedestrians, etc.) and the vehicle from the recorded picture; The driving recorder usually uses a monocular, wide-angle visual sensor, and there is a nonlinear relationship in its imaging system-distortion. Due to the existence of di...

AI Technical Summary

Problems solved by technology

[0003] At present, there are some deficiencies in the research method of using video to calculate vehicle speed, which are: first, most of the research fields involved in existing patents or professional papers are only aimed at the speed of participants in surveillance cameras without image distortion. Solving; second, it is limited to knowing the geometric parameters of the target, such as wheelbase, vehicle length, etc.; third, the result error is large. When the target and the video screen have a large angle, the obtained speed will be wrong. very big

Images