一种基于工艺模型的装配质量监测方法
By constructing a structured inspection process model and a deep learning network, the problems of unstructured data and low efficiency in manual assembly inspection mode are solved, realizing intelligent and efficient assembly quality inspection and adapting to the needs of multi-model co-line production.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG UNIV
- Filing Date
- 2026-05-18
- Publication Date
- 2026-07-17
AI Technical Summary
The existing manual assembly and inspection model suffers from unstructured data, low efficiency, and high cost. It cannot meet the needs of multi-model co-production and is difficult to inspect small parts in a narrow and deep space.
A structured inspection process model is constructed, and deep learning networks such as the YOLO detection network are used for image processing. Combined with image enhancement and preprocessing, automated and intelligent assembly quality inspection is achieved. The model is optimized through transfer learning, high-resolution image segmentation and detection are performed, and the detection results are output and iteratively optimized.
It enables fully structured storage and rapid traceability of assembly quality inspection data, significantly improving inspection accuracy and efficiency, reducing manpower input, adapting to the needs of multi-model co-production, and promoting the intelligent upgrade of assembly inspection mode.
Smart Images
Figure CN122223020B_ABST