Related Experiment Video
Updated: Jun 28, 2025
Analysis of Neural Crest Migration and Differentiation by Cross-species Transplantation
Published on: February 7, 2012
An enhancement algorithm for head characteristics of caged chickens detection based on cyclic consistent migration
Zhenwei Yu1, Liqing Wan1, Khurram Yousaf2
1College of Mechanical and Electronic Engineering, Shandong Agricultural University, Tai'an 271018, China; Shandong Provincial Engineering Laboratory of Agricultural Equipment Intelligence, Shandong Provincial Key Laboratory of Horticultural Machineries and Equipment, Tai'an 271018, China.
This study introduces a novel algorithm to remove cage-gate interference in poultry farm images, significantly improving the detection accuracy of chicken combs and eyes. This advancement aids in better health monitoring for caged chickens.
Area of Science:
- Agricultural Engineering
- Computer Vision
- Animal Science
Background:
- Industrial caged chicken breeding relies on enclosed multistory housing.
- Accurate detection of chicken health indicators like combs and eyes is crucial for farm management.
- Image detection accuracy is often compromised by enclosure entrances, reducing precision.
Purpose of the Study:
- To propose a deep learning-based algorithm for removing cage-gate obstructions in poultry images.
- To enhance the accuracy of detecting caged chicken combs and eyes for health assessment.
- To validate the effectiveness of the proposed algorithm using various YOLOv8 models.
Main Methods:
- Development of a cage-gate removal algorithm using a Cyclic Consistent Migration Neural Network (CCMNN).
- Application of the CCMNN network for automatic elimination and restoration of image information.
- Evaluation of image quality using Structural Similarity Index Measure (SSIM) and Peak Signal-to-Noise Ratio (PSNR).
- Comparative analysis of target detection algorithm performance (YOLOv8 variants) before and after CCMNN processing.
Main Results:
- The CCMNN algorithm achieved an SSIM of 91.14% and a PSNR of 25.34dB for image recovery.
- Comb detection precision improved by 10.2–12.8% across different YOLOv8 models after CCMNN processing.
- Eye detection precision improved by 2.4–10.2% across different YOLOv8 models after CCMNN processing.
Conclusions:
- The proposed CCMNN algorithm effectively removes cage-gate interference, yielding more complete chicken images.
- Significant enhancements in comb and eye detection precision contribute to improved health monitoring in caged chickens.
- This research provides a foundation for developing advanced detection equipment for poultry production and farm condition assessment.
