Shellfish sorting device

TWM686141UActive Publication Date: 2026-08-01DABAO CLAM AQUATIC PRODUCTS CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
TW · TW
Patent Type
Utility models
Current Assignee / Owner
DABAO CLAM AQUATIC PRODUCTS CO LTD
Filing Date
2026-05-22
Publication Date
2026-08-01

Smart Images

  • Figure TWG2TB001904705_001
    Figure TWG2TB001904705_001
  • Figure TWG2TB001904705_002
    Figure TWG2TB001904705_002
  • Figure TWG2TB001904705_003
    Figure TWG2TB001904705_003
Patent Text Reader

Abstract

This invention provides a shellfish sorting device, comprising: a conveying platform on which one or more shellfish to be inspected are placed; a rejection mechanism connected to the end of the conveying platform; an emitter that emits X-rays toward the shellfish to be inspected; a receiver that converts the X-rays penetrating the shellfish to be inspected into an X-ray image; and an object detection module that identifies the shellfish to be inspected based on the X-ray image to generate a classification result for each. When the classification result of any shellfish to be inspected is defective, the rejection mechanism rejects the defective shellfish. Thus, this invention can automatically reject defective shellfish.
Need to check novelty before this filing date? Find Prior Art

Claims

1. A shellfish sorting device, comprising: A conveying platform includes a conveyor belt on which one or more shellfish to be inspected are placed; a rejection mechanism connected to the end of the conveyor belt; a transmitter located on one side of the rejection mechanism and used to emit X-rays toward the shellfish to be inspected; a receiver located on one side of the rejection mechanism and used to convert the X-rays penetrating the shellfish to be inspected into an X-ray image; an object detection module used to identify the shellfish to be inspected based on the X-ray image to generate a classification result; and a processing module coupled to and controlling the operation of the conveyor belt, the rejection mechanism, the transmitter, the receiver, and the object detection module; wherein, when the classification result of any of the shellfish to be inspected is a defective product, the processing module controls the rejection mechanism to reject the defective shellfish to be inspected.

2. The shellfish sorting device as described in claim 1, wherein, The waste rejection mechanism is equipped with a drop outlet and a baffle. The baffle is located behind the drop outlet. When any of the shellfish to be tested is classified as defective, the processing module controls the baffle to extend upwards for a period of time to prevent the defective shellfish from continuing to move along a discharge direction and falling downwards through the drop outlet.

3. The shellfish sorting device as described in claim 2, wherein, The waste removal mechanism is divided into a front channel and a rear channel with the drop outlet as the boundary, and the front channel and the rear channel form a segment difference.

4. The shellfish sorting device as described in claim 2, wherein, The baffle is connected to a pneumatic cylinder via a piston rod, and the baffle extends upward or retracts downward as the pneumatic cylinder pushes the piston.

5. The shellfish sorting device as described in claim 2, wherein, When the classification results of the shellfish to be tested show no defective products, the processing module keeps the baffle retracted downwards, allowing the shellfish to be tested to fly over the drop outlet and continue to move along the discharge direction.

6. The shellfish sorting apparatus as described in claim 2, wherein, The rejection mechanism also includes several diversion channels, each of which is equipped with an independently controllable drop outlet and baffle. These baffles are coupled to the processing module. When the shellfish to be tested on any diversion channel is classified as defective, the processing module controls the baffle of the corresponding diversion channel to extend upward, while the baffles of the other diversion channels remain retracted downward.

7. The shellfish sorting apparatus as described in claim 1, wherein, The object detection module uses YOLO as the core architecture of the model and integrates a path aggregation network.

8. The shellfish sorting apparatus as described in claim 7, wherein, The model uses Focal Loss and CIoU Loss as loss functions, and then uses the Cosine LR Scheduler to control the learning rate during training.