Agricultural product quality detection method and system
By using multimodal data acquisition and cross-modal feature fusion networks, combined with the Light-Transformer model, we have achieved rapid and accurate assessment of agricultural product quality, solving the problems of time-consuming and labor-intensive traditional detection methods and improving detection efficiency and accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Filing Date
- 2026-04-22
- Publication Date
- 2026-07-17
AI Technical Summary
Traditional agricultural product testing methods are difficult to comprehensively and accurately assess quality, and the step-by-step testing methods are time-consuming and complex, affecting market supply and economic benefits.
A multimodal data acquisition module is used to simultaneously collect spectral data, image data and environmental parameters. Features are extracted through a cross-modal feature fusion network with an attention mechanism, and the Light-Transformer model is used for multi-task prediction to output appearance quality, internal nutritional components and harmful residue indicators.
It enables rapid and accurate multi-faceted quality assessment, improves testing efficiency and accuracy, reduces data processing complexity, and ensures the quality and safety of agricultural products and market supply.
Smart Images
Figure CN122416418A_ABST