3D detection method and device based on three-camera cooperative laser

Abstract

The invention discloses a 3D detection method and device based on three-camera cooperation laser. The method comprises steps that grid laser is adopted as an active light source to irradiate the surface of an object; a laser line image is collected through a trinocular camera; the three-dimensional information of the laser ray image is obtained through image processing and calculation so that point cloud of the contour of the target object is obtained; if denser point clouds need to be obtained, the translation laser can be controlled through the stepping motor, so the target object is scanned, three-dimensional point cloud data of a complete outline of the surface of the target object is obtained, point cloud data with higher precision can be obtained, and three-dimensional information such as the position and the posture of the target object can be further obtained. The method solves problems of difficult feature point matching, low point cloud precision, high requirements for illumination conditions and the like of the existing visual measurement scheme, does not need to paste markers on the surface of the target object, and is wide in application range and high in operability.

Description

technical field [0001] The invention relates to the technical field of optical and electronic measurement, in particular to a 3D detection method and device based on a three-camera cooperative laser. Background technique [0002] 3D detection has important demands in industrial production, aerospace, automatic driving and other fields. Common 3D detection methods include visual measurement (including monocular, binocular, and multi-eye), lidar measurement, and multi-sensor fusion measurement. Among them, the measurement method based on binocular stereo vision is the most widely used, has the advantages of simple structure, low power consumption, etc., but has certain limitations, high requirements for lighting conditions, and feature point matching for the surface texture and material of the target object. higher. In recent years, lidar has developed rapidly, but lidar has disadvantages such as high power consumption, high cost, and complex structure, and the resolution of...

AI Technical Summary

Problems solved by technology

In recent years, lidar has developed rapidly, but lidar has disadvantages such as high power consumption, high cost, and complex structure, and the resolution of point clouds is also limited.

The solution of binocular camera and line laser scanning can better solve the problem of difficult matching of feature points, but the line laser scanning cycle is long and the dynamic performance of the system is poor

Images