A method for measuring the distance of monocular vision plane without image control

Abstract

The invention discloses a monocular vision plane distance measuring method free of photo-control. The method comprises the following steps of a measurement system integration stage: fixedly connectinga camera and a laser range finder to form a direct oriented vision measurement system; a measurement system calibration stage: performing integrated calibration on the measurement system to determinecamera internal parameters, distortion aberrations and an eccentric angle between the camera and the laser range finder, and correcting the nonlinear distortion, measured elevation and attitude of the camera; and a plane distance measurement stage: establishing an "image-distance" transformation relationship between the image plane and the object plane based on a central projection imaging principle of the collinearity of the image point, optic center and object point of the camera in an oblique angle of view, and finally realizing the monocular vision plane distance measurement free of photo-control. The method provided by the invention has low hardware cost and operational complexity, greatly reduces the workload in the field by completing the measuring point layout in a few minutes with no need to arrange photo-control points on the plane to be tested, and has universality without depending on the prior knowledge in the scene.

Description

technical field [0001] The invention relates to the field of plane distance measurement, in particular to an image control-free monocular vision plane distance measurement method. Background technique [0002] Visual measurement is a non-contact measurement technology formed with the development of computer vision technology. Compared with traditional measurement methods, it has the advantages of large measurement range, more measurement information, less restriction by the working environment, and no interference with the measurement object. According to the number of cameras used, the visual measurement system can be divided into monocular vision measurement, binocular vision measurement, multi-eye vision measurement and structured light vision measurement. Among them, the technology of using a single image captured by a single camera to measure the geometric size or spatial position of an object is called monocular vision measurement in the field of machine vision. Due t...

AI Technical Summary

Problems solved by technology

However, due to the lack of depth information in the imaging process of a single image, the mapping relationship between the three-dimensional world coordinates and the two-dimensional image coordinates is irreversible. Usually, additional coplanar constraints are required to establish the image plane and object plane described by the homography matrix. Reversible coordinate transformation relationship between planes

The main method used is the direct linear transformation method based on point correspondence to solve the homography matrix, that is, use more than 4 image control points in the plane to be measured to solve the homography matrix, and then use the homography matrix to solve the homography matrix. The coordinates of the image point in the object plane, and then calculate the distance between two points; its measurement accuracy is sensitive to factors such as the accuracy, quantity and distribution conditions of the image control point, and the image control point in the complex measurement environment such as the water surface The layout is more difficult, which often limits the use of this method

Other methods use the known distance or regular geometric structure in the field of view as prior knowledge to solve the homography matrix. For example, Wang et al. proposed a homography matrix solution method based on straight line correspondence. Point fitting is obtained, so the measurement accuracy is higher than that of the point corresponding method; Ding Weili et al. proposed a method to obtain measurement information using a rectangular structure, which can be used for different The position is measured; Wang Meizhen et al. proposed a single image plane geometric information extraction method based on cross-ratio projective transformation invariants, using the geometric structure in the scene to calculate the vanishing point and vanishing line in the plane, and using the previously measured distance in the plane to be measured The intersection ratio of the line segment is used to calculate the physical distance of the line segment to be obtained, which avoids the cumbersome layout of image control points; however, these methods need to use the geometric information in the image, which is only applicable to specific scenes and lacks universality

Images