Chicken Behavior Analysis Using Computer Vision Techniques

View at publisher Summarise with VIRAT
Authors: Md Roman Bhuiyan, Philipp Wree
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

A growing population of people is reliant on the essential sector of chicken breeding for subsistence. However, the chickens well-being is a major worry since bad living conditions create erratic behavior, which impairs the flock’s health and output. To monitor and improve the animals well-being, a surveillance system that records chicken behavior and alerts farmers to potential concerns must be in place. Farmers have long depended on experts to identify and diagnose ailments in chickens. Farmers lose a lot of farmed chickens as a result of delayed diagnosis or a lack of reliable specialists. In this paper, we have solved a problem for chicken behaviour analysis from high-crowd chicken images with resolutions of 3840 × 2160 using computer vision techniques. One chicken image has around 400 chickens with high resolutions images and it is really difficult to identify the chicken behaviors. Our proposed method employs a Mask-RCNN based approach which is the state of the art for behavior analysis used for monitoring in chicken farms. This study aims to address a computer vision-based system for identifying and tracking chicken behaviour in poultry farms. Chicken, chicken_eating and chicken_sitting are the three behaviors that the system classifies. To address this issue, this research aims to develop a dataset based on chicken behavior analysis from the chicken farm. The proposed method for chicken behaviour analysis achieved an accuracy of mAP (means average precision) = 96.75% with the three classes respectively. The proposed chicken behavior analysis results outperformed the state-of-the-art model in most cases.

Fields of Study & Subjects

Smart Agriculture and AI Water Quality Monitoring Technologies Advanced Neural Network Applications

Publication details

Journal / source
CRC Press eBooks
Pages
32–45
Publication date
27 March 2026
Language
EN
Document type
Book chapter