一种目标区域识别方法、图像采集系统及分选机
By acquiring transmission and color sorting images during ore transportation, and based on contour mapping and grayscale correction, the problem of inaccurate image registration in ore identification and sorting is solved, achieving high-precision ore identification and sorting.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING HONEST TECHNOLOGY CO LTD
- Filing Date
- 2025-12-30
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, the rolling and flipping of ore during transportation makes it difficult to accurately register X-ray images and visible light images, resulting in low accuracy in ore identification and sorting.
By acquiring transmission and color sorting images, the first contour is determined based on the transmission image and mapped to the color sorting image. The foreground and background regions are divided, candidate sub-regions are introduced and merged to determine the target region. Gray-scale binarization and centroid position are used to correct the ore contour offset, thus establishing an accurate target region identification method.
It achieves high-precision matching between transmission images and color sorting images under ore pose changes, ensuring the integrity of ore contour and color information, improving the accuracy of ore identification and sorting, and reducing the probability of misjudgment and missed judgment.
Smart Images
Figure CN121504969B_ABST