Steel ball surface defect detecting device and method based on double-line-array camera and ring light

Abstract

The invention relates to a steel ball surface defect detecting device and method based on a double-line-array camera and ring light, and belongs to the field of industrial nondestructive detection. The defect detecting device comprises a camera and a steel ball expanding plate. Multiple steel balls to be measured are placed on the steel ball expanding plate. The defect detecting device further comprises a ring light source. The steel ball expanding plate is rotated in a certain angle under the control of a motor until the steel balls to be measured reach the lower part of the ring light source. A friction plate is installed below the steel ball expanding plate. The friction plate is driven to be moved along an XY direction by an XY workbench. The corresponding steel balls to be measured are driven to be rotated in the XY direction through the friction. The steel ball surface defect detecting device is capable of irradiating the steel balls to be measured by using the ring light, and collecting steel ball images by using the line-array camera. The steel balls can be rolled in the orthogonal direction under the action of the XY workbench and the friction plate, namely the full-surface expansion of the steel balls to be measured can be realized. The image matching and the image stitching do not need to be considered, and the collected images are space-time images. The comparison with the background is obvious, an image processing algorithm is simple, and the detection speed is greatly improved.

Description

technical field [0001] The invention relates to a steel ball surface defect detection device and method based on a dual line array camera and a ring light, in particular to realizing the full surface unfolding of a steel ball and defect detection through space-time images, belonging to the field of industrial non-destructive testing. Background technique [0002] Bearings are the most important basic parts in rotary machinery, and steel balls are the key basic components of bearings, and are the key to the key of rotary machinery. The surface quality of steel balls directly affects the performance and life of bearings. Due to the problems of processing technology and materials, during the production process of steel balls, surface defects such as wear, cracking, and pits will occur. These defects will affect the accuracy and performance of the bearing, and generate noise and vibration. In severe cases, it will cause the bearing to fail. . Therefore, it is of great significa...

AI Technical Summary

Problems solved by technology

[0003] Domestic steel ball factories now mainly use human eyes to spot-check the surface defects of steel balls manually under strong light. This method is limited by the limitations of manual inspection, which will cause missed inspections and wrong picks, and cannot achieve all inspections. Low automation, which limits the detection of steel balls for high-precision bearings and the development of the industry

[0004] The current automatic steel ball surface defects mainly include contact method and non-contact method. The contact method will cause wear to the detection steel ball, which is a destructive detection method and can only be used for random inspection, not for all detection. Need to be replaced; non-contact methods mainly include photoelectric detection technology and visual detection technology

The existing photoelectric detection technology mainly uses point scanning to detect the surface defects of the steel ball through meridian or helical expansion. This method needs to realize the full surface expansion of the steel ball. The expansion mechanism is relatively complicated and the movement of the steel ball is accurate. Control is a difficult point in the design of the mechanical structure, and it is prone to slipping and missed inspections; while the visual inspection technology mainly uses area scanning and area array image sensors to shoot a certain ball crown surface on the surface of the steel ball, the steel ball rolls, and the area array image sensor The sensor continues to shoot the surface of the steel ball, so it is necessary to find a control method so that the captured spherical crown covers the entire surface of the steel ball. This method needs to control the shooting time of the camera, and the rolling of the steel ball cannot slip; Subsequent image matching and image stitching are required, and the image processing algorithm is complex

Images