Complex surface shape detection method based on binocular vision structured light

Abstract

The invention designs a complex surface shape detection method based on binocular vision structured light. The method is used for detecting shape features of steps, grooves and other objects on the surface of a complex workpiece. The binocular vision structured light image acquisition system comprises a controller, a left CCD camera, a right CCD camera and a structured light generator. The detection method comprises system calibration, image acquisition, algorithm processing and three-dimensional model reconstruction, and any CCD camera in the system can realize three-dimensional reconstruction under the assistance of structured light. That is to say, when a certain camera cannot capture structural light spots normally, three-dimensional reconstruction is achieved through the other camera,the recognition rate of a binocular system on the surface of a complex workpiece can be effectively increased, and the detection precision is effectively improved. For a step with a slope angle smaller than a right angle, the method can accurately reconstruct a real three-dimensional model, an undetected area does not need to be corrected by using algorithms such as interpolation and the like, and the application range of binocular vision detection is widened.

Description

technical field [0001] The invention relates to a complex surface shape detection method based on binocular vision structured light, and belongs to the technical field of binocular vision three-dimensional reconstruction and detection in the welding process. Background technique [0002] In the field of laser welding technology, welding defects such as welding cracks, steps, and grooves are prone to occur during automatic melting welding of metal materials. To solve this problem, the existing monocular or binocular vision systems can effectively detect these defects. does it exist. However, when the depth jump of cracks, steps, and grooves is too large, or the distribution of these welding defects is complex and diverse, due to the influence of factors such as the inclination angle between the camera and the measured object, the existing vision system cannot normally capture the image of a specific position. Therefore, parameters such as the depth of cracks, the slope of st...

AI Technical Summary

Problems solved by technology

However, when the depth jump of cracks, steps, and grooves is too large, or the distribution of these welding defects is complex and diverse, due to the influence of factors such as the inclination angle between the camera and the measured object, the existing vision system cannot normally capture the image of a specific position. Therefore, parameters such as the depth of cracks, the slope of steps and grooves often cannot be accurately detected

If it is necessary to further detect and analyze these parameters, and detect problems such as welding quality online in a timely manner, the traditional monocular or binocular vision system can no longer meet the requirements

[0003] The Chinese invention patent "Surface Shape Acquisition and Sensing System of Welding Pool Based on Binocular Vision" (201010566206.2) proposes a sensing system suitable for detecting the surface shape of the welding pool in the arc welding process, through which the shape of the molten pool can be proposed Surface shape, and used for welding modeling and intelligent control; utility model patent "welding stud detection system based on stereo vision principle" (201820683462.1) discloses a system for detecting screw welding position and angle; utility model patent "aluminum Alloy automatic welding binocular vision molten pool monitoring sensor" (200920142300.8) discloses a sensor device that guides the welding torch to automatically track the weld seam and monitor the state of the molten pool during the welding process. The disadvantages of the above three patents are: None of the disclosed technologies propose methods and solutions for analyzing and judging parameters such as the depth of cracks at welds, the slopes of steps and grooves, etc.

The Chinese invention patent "A Method for Acquiring the Three-Dimensional Coordinates of the Center of Light Strips Based on Binocular Structured Light in Welding Inspection" (CN201711327867) proposes that the method of using binocular structured light, combined with the principle of monocular structured light three-dimensional measurement, and the use of noise Different from the position of the structured light stripe after the image is mapped to the plane in the space, most of the noise in the welding can be effectively removed through image processing, and its center line can be obtained, which makes the welding visual inspection more accurate, but for the welding process For some steps and grooves that appear, if a camera cannot capture structured light normally, the parameters of welding steps or grooves cannot be accurately obtained

Images