Farm chicken detection method, device, equipment and medium

By updating the network structure and processing methods in the YOLO series algorithm, the detection accuracy and target tracking stability problems in dense chicken environments are solved, and higher detection accuracy and more stable target tracking are achieved, which is suitable for complex breeding farm environments.

CN120014672AActive Publication Date: 2025-05-16SOUTH CHINA AGRICULTURAL UNIVERSITY
View PDF 3 Cites 3 Cited by

Patent Information

Application Number
CN202510148162.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-11
Publication Date
2025-05-16
Estimated Expiration
2045-02-11

AI Technical Summary

Technical Problem

In the prior art, the YOLO series algorithm has limited detection effect in dense environments of chickens, and is prone to missed detection and false detection, resulting in a decrease in detection accuracy and unstable target tracking.

Method used

By updating the C2f module in the backbone network of the first chicken detection model as the C2f_Conv3XC module, the path aggregation network in the neck network is a multi-scale adaptive feature pyramid network, and a double allocation method is used in the detection head network to build the second chicken detection model to improve detection accuracy and stability.

Benefits of technology

It significantly improves the accuracy of chicken detection, reduces missed and missed detection, improves the stability of target tracking, is suitable for complex farm environments, and improves the efficiency and accuracy of breeding management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120014672A_ABST
    Figure CN120014672A_ABST
Patent Text Reader

Abstract

The invention relates to a chicken detection method, device and equipment in a farm and a medium, and the method comprises the steps: obtaining a to-be-detected farm image frame containing a plurality of chickens in response to an instruction for carrying out chicken detection on the farm; updating a C2f module in a backbone network of the first chicken detection model to a C2fConv3XC module, updating a path aggregation network in a neck network to a multi-scale adaptive feature pyramid network, and constructing a second chicken detection model in a detection head network by adopting a dual-allocation method; and inputting the to-be-detected farm image frame into a second chicken detection model which is trained to be in a convergence state to determine the number of the chickens so as to complete the detection of the chickens in the farm. According to the method, the chicken detection precision can be remarkably improved, missing detection and false detection are reduced, and the target tracking stability can be improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Cage rearing broiler chicken state detection and counting system based on deep learning

    CN116665129A

  • Deep learning-based chicken and egg target detection method for chicken farm

    CN118537802A

  • Chicken farm detection method and system based on YOLO-CV model

    CN118537852A

Cited By

  • Network model and method for identifying egg type, and storage medium

    CN121121727A

  • An egg type recognition network model and method, and a storage medium

    CN121121727B

  • Dual-mode information detection system and method based on automatic driving scene

    CN121999460A