High-precision speed measurement method based on binocular stereoscopic vision

Abstract

The invention discloses a high-precision speed measurement method based on binocular stereoscopic vision. The method comprises the following steps of (a) setting a longitudinal speed measurement binocular camera and a transverse speed measurement binocular camera; (b) calibrating the longitudinal speed measurement binocular camera and the transverse speed measurement binocular camera to obtain a corresponding relation between the camera pixel points and the actual size; (c) acquiring an image of the moving object by using a longitudinal speed measurement binocular camera and a transverse speedmeasurement binocular camera, and acquiring feature points of the moving object through SIFT corner detection; (d) respectively calculating vehicle distances directly corresponding to two adjacent frames of images by utilizing a binocular parallax algorithm, and calculating longitudinal speed components and transverse speed components of the moving object according to the frame rate; and (e) properly synthesizing the longitudinal velocity component and the transverse velocity component to obtain the velocity of the moving object, so that the problems that the distance estimation can be carried out only by recognizing a measured object in advance in existing bullet screen distance measurement, the transverse movement speed of part of vehicles is ignored, and the measurement precision is not high enough, are solved.

Description

technical field [0001] The invention belongs to the field of computer vision, and in particular relates to a high-precision speed measurement method based on binocular stereo vision. Background technique [0002] If we can measure the speed of passing vehicles in the vehicle monitoring area in real time, and submit the obtained speed information to the management department in time, and seriously investigate and deal with speeding vehicles, it will greatly improve the road safety situation in our country. [0003] In recent years, computer vision, which began in the 1950s, has developed rapidly. It integrates image processing, pattern recognition, computer technology, computer graphics, neurophysiology and other disciplines, aiming to simulate the function of biological vision, that is, to recognize objects Carry out fine classification and determine the position and shape of objects. Due to the guidance of visual framework theory and the introduction of computer graphics, ...

AI Technical Summary

Problems solved by technology

[0005] The invention provides a high-precision speed measurement method based on binocular stereo vision, which solves the problem that the existing barrage distance measurement needs to identify the object to be measured before performing distance estimation, and ignores the speed of some vehicles moving laterally and the measurement accuracy is not high enough. question

Images