Intelligent detection and sorting system for fragile fruits
The intelligent detection and sorting system for fragile fruits utilizes multidimensional image streams and cross-domain coherence compensation technology to solve the problem of misjudgment caused by dirty carriers, achieving high-precision non-destructive sorting of fragile fruits and ensuring the adaptive identification and robustness of the sorting equipment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANDONG QINGQINGDADI FRUIT & VEGETABLE CO LTD
- Filing Date
- 2026-04-08
- Publication Date
- 2026-07-10
AI Technical Summary
Existing sorting technologies are easily affected by dirt and obstruction of the carrier, leading to visual misjudgment at the bottom and distortion due to mechanical shock wave attenuation, making it difficult to achieve high-precision non-destructive sorting of fragile fruits.
An intelligent detection and sorting system for fragile fruits is adopted, including a first extraction module, a second extraction module, a cross-domain coherence compensation module, a decision module, and an execution control module. By acquiring multi-dimensional image streams, calculating the spatiotemporal attenuation index of local luminous flux and the spatial micron-level amplitude response spectrum, and combining non-rigid elastic registration and optical flow field offset alignment techniques, a hidden interface rheological feature matrix is generated to achieve accurate detection and sorting of the true viscoelastic state of the bottom of the fruit.
It effectively removes shock wave dissipation artifacts caused by dirt, provides a pure, high signal-to-noise ratio data source, ensures high-precision diversion and adaptive identification capabilities, reduces misjudgments, and achieves equipment robustness and operational reliability for continuous high-throughput operations.
Smart Images

Figure CN122367709A_ABST