Monocular-camera-based distance measurement method and distance measurement system

Abstract

The invention provides a monocular-camera-based distance measurement method and distance measurement system. The method comprises the steps: acquiring information, including position information of atarget, of the target object through a camera unit including a vehicle-mounted camera; acquiring lane information according to the position information of the target object, and acquiring a vertical distance between the vehicle-mounted camera and the target object according to the lane information and an internal reference proportionality coefficient of the vehicle-mounted camera; acquiring an offset angle of the target object relative to a vehicle-mounted camera coordinate system according to position information of the vehicle-mounted camera and the position information of the target object;and on the basis of the offset angle and the vertical distance between the vehicle-mounted camera and the target object, acquiring a horizontal distance between the vehicle-mounted camera and the target object to obtain vehicle distance information including the vertical distance between the vehicle and the target object and the horizontal distance between the vehicle and the target object. According to the invention, the distance measurement method has advantages of simple algorithm, high accuracy and efficiency, and low cost and is suitable for popularization and application.

Description

technical field [0001] The present invention relates to the technical field of ranging methods, in particular to a ranging method and ranging system based on a monocular camera. Background technique [0002] With the development of autonomous driving technology, autonomous vehicles have gradually entered the public's field of vision. In the driving scene of autonomous vehicles, the identification and detection of obstacles is an essential link. In order to realize the identification and detection of obstacles, self-driving vehicles generally use sensors such as lidar and cameras configured in the vehicle. Among them, if the laser radar is used, the cost is high. For example, only a 64-line laser radar costs tens of thousands of yuan, and the laser radar has the disadvantage of not being able to identify the object type. Tracking and Motion Law Estimation. In addition, because the camera lacks depth information, it generally requires binocular cameras to extract features to...

AI Technical Summary

Problems solved by technology

Among them, if the laser radar is used, the cost is high. For example, only a 64-line laser radar costs tens of thousands of yuan, and the laser radar has the disadvantage of not being able to identify the object type. Tracking and Motion Law Estimation

In addition, because the camera lacks depth information, it generally needs a binocular camera to extract features to estimate the depth, which cannot be used efficiently in outdoor scenes.

[0003] Patent document CN106708084B discloses a UAV automatic obstacle detection and obstacle avoidance method in a complex environment, including: using the binocular camera set on the UAV to perform real-time image acquisition, based on the collected left and right two images matching Calculate and generate the disparity map and perform preprocessing; based on the information of the disparity map, cluster the contour and gray value of the disparity map to obtain a region block with a clear structure, and remove the noise interference of the region block to obtain potential obstacles area; compare the corresponding potential obstacle areas in the front and back two frames of disparity images, and combine the magnification characteristics of the obstacles to obtain the final obstacle area based on the distance between the UAV and the potential obstacle area and the area of ​​the potential obstacle area; based on the Get the final obstacle area, and output the action command of UAV obstacle avoidance; but this method has a large amount of calculation

Images