A method of detecting a bearing
By driving the inner ring of the bearing to rotate, which in turn causes the rollers to rotate, and combining a high frame rate camera and image processing algorithms, the problems of low detection efficiency and incompleteness in existing technologies are solved, and efficient and high-precision full-circumference detection of bearing rollers is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUANGSHI BANGKE TECH CO LTD
- Filing Date
- 2026-01-15
- Publication Date
- 2026-05-29
AI Technical Summary
Existing bearing roller inspection technologies suffer from low inspection efficiency and incomplete inspection. Individual inspection modes are cumbersome, and post-assembly inspection is limited by structural constraints, resulting in the omission of defects and failing to meet the demands of high-precision and high-efficiency production.
By driving the inner ring of the bearing without an outer ring to rotate, the roller can simultaneously revolve and rotate. Combined with high frame rate shooting by a camera and image preprocessing, the YOLO series algorithm is used for defect detection to ensure that the entire circumference of the roller is captured and identified.
It achieves high efficiency and high precision in full-circumference inspection of bearing rollers, reduces defect omissions, and improves production efficiency and product quality.
Smart Images

Figure CN122108587A_ABST