Traditional Chinese medicine particle detection method, device and equipment, storage medium and program product

By improving the YOLOv8 model and the multi-core attention module to enhance feature extraction capabilities, the problem of low accuracy in particle detection of traditional Chinese medicine powders was solved, and efficient and accurate particle detection and quality control were achieved.

CN120656166AActive Publication Date: 2025-09-16JIANGSU KANION PHARMA CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202510692020.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-05-27
Publication Date
2025-09-16
Estimated Expiration
2045-05-27

Smart Images

  • Figure CN120656166A_ABST
    Figure CN120656166A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of target detection, and discloses a traditional Chinese medicine particle detection method, device and equipment, a storage medium and a program product. Inputting the to-be-detected traditional Chinese medicine particle image into a pre-constructed improved YOLOv8 traditional Chinese medicine particle detection model to obtain a traditional Chinese medicine particle detection result; wherein the improved YOLOv8 traditional Chinese medicine particle detection model is obtained by replacing the backbone network of the original YOLOv8 model and other stepping convolution layers except the first stepping convolution layer of the backbone network in the neck network with a cross-sub-block multi-kernel attention space depth module and then training the cross-sub-block multi-kernel attention space depth module. According to the method, the YOLOv8 model is combined with the cross-subblock multi-kernel attention space depth module, so that the feature learning capability of the model on small-particle-size particles is enhanced, the overall detection precision of the particles is remarkably improved, and rapid detection and accurate positioning of the traditional Chinese medicine particles are effectively realized.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Wood defect detection method and system based on deep neural network

    CN117876663A

  • Target detection method based on fine-grained features

    CN118887378A

  • Lightweight bolt detection method based on DSCCC-YOLO

    CN119624859A

  • Synthetic aperture radar (SAR) image target detection method

    US20230169623A1