Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
December 8, 2025AgriEngineeringOpen Access

Automated Detection of Kinky Back in Broiler Chickens Using Optimized Deep Learning Techniques

View Full Paper
Ask AI
Bookmark
Share

Authors

RBRamesh Bahadur BistCPChaitanya Pallerla

Discussion

Loading...

Member takes

Overview

Automated techniques reveal skeletal disorders in broilers, suggesting improved welfare management in poultry production.

Key Points

  • This research aims to develop an optimized deep learning system for detecting Kinky Back in broiler chickens.
  • Utilized image data from overhead cameras to monitor Cobb 500 broilers over 7 weeks.
  • Evaluated various optimizers, image sizes, and data augmentation techniques for optimal performance.
  • Tested multiple YOLO model architectures to assess detection accuracy using precision, recall, and F1-score metrics.
  • SGD optimizer achieved the highest precision at 100% and mAP of 74.7%.
  • Image size of 960 × 960 pixels yielded a precision of 99.0% and recall of 99.4%.
  • Optimized YOLOv9 model's combination with data augmentation reached a precision of 99.1% and recall of 100%.

Cite This Study

Bist et al. (2025) studied this question.

synapsesocial.com/papers/693624ce4fa91c937236cdcdhttps://doi.org/10.3390/agriengineering7120415
View Full Paper
Ask AI
Bookmark
Share