A Method and System for Instrument Reading Recognition Based on Image Segmentation
By using image segmentation technology, training a semantic segmentation model, and performing rotation correction and triangle fitting, the difficulties in reading recognition caused by instrument occlusion and type changes are solved, achieving high-precision and widely applicable instrument reading recognition.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG SYNTHESIS ELECTRONICS TECH
- Filing Date
- 2022-10-31
- Publication Date
- 2026-05-26
AI Technical Summary
Existing technologies cannot accurately identify instrument readings when pointer-type instruments are obstructed or of different types, and existing machine vision methods lack generalization ability.
An image segmentation-based method is adopted. By training a semantic segmentation model, the mask information of the pointer area, the center area of the dial, and the set triangular area in the instrument image is extracted, rotation correction and triangle fitting are performed, and key points are calculated to identify the instrument reading.
It achieves accurate identification under obstructed conditions and different types of instruments, improving the accuracy and generalization ability of instrument reading identification, with an error within 2%.
Smart Images

Figure CN115690419B_ABST