一种基于高光谱图像的糯稻种子淀粉含量测定方法及装置
By employing techniques such as closed-loop adaptive iterative preprocessing, causal inference engine, and dynamic spatial-spectral fusion, the problems of cumbersome process, high destructiveness, and insufficient model generalization ability in starch content detection of glutinous rice seeds have been solved, achieving efficient and accurate starch content determination, which is suitable for large-scale breeding and storage quality inspection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- RICE RES ISTITUTE ANHUI ACAD OF AGRI SCI
- Filing Date
- 2026-05-13
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for detecting starch content in glutinous rice seeds suffer from problems such as cumbersome detection procedures, high destructiveness, high cost, insufficient model generalization ability, crude feature fusion, and disconnect between preprocessing and modeling, making it difficult to meet the needs of high-throughput and accurate detection.
By employing closed-loop adaptive iterative preprocessing, causal inference engine, dynamic spatial-spectral fusion, and spectral-morphological heterogeneous graph inference, and through adaptive feature purification, causal relationship analysis, dynamic feature fusion, and cross-modal inference, a hyperspectral image detection model is constructed to achieve non-destructive, rapid, accurate, and high-throughput determination of starch content in glutinous rice seeds.
It achieves efficient, accurate, and high-throughput determination of starch content in glutinous rice seeds, improves the model's generalization ability and robustness, enhances the interpretability of the test results, and is suitable for large-scale breeding and storage quality inspection scenarios.
Smart Images
Figure CN122223713B_ABST