Skip to main content
ARS Home » Southeast Area » Athens, Georgia » U.S. National Poultry Research Center » Egg and Poultry Production Safety Research Unit » Research » Publications at this Location » Publication #397940
Research Project: Reduction of Foodborne Pathogens and Antimicrobial Resistance in Poultry Production Environments

Location: Egg and Poultry Production Safety Research Unit

Title: Monitoring behaviors of broiler chickens at different ages with deep learning

Author
item GUO, YANGYANG - University Of Georgia
item AGGREY, SAMUEL - University Of Georgia
item WANG, PENG - College Of Biosystems Engineering And Food Science
item Oladeinde, Adelumola - Ade
item CHAI, LILONG - University Of Georgia

Submitted to: Animals
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/30/2022
Publication Date: 12/2/2022
Citation: Guo, Y., Aggrey, S.E., Wang, P., Oladeinde, A.A., Chai, L. 2022. Monitoring behaviors of broiler chickens at different ages with deep learning. Animals. https://doi.org/10.3390/ani12233390.
DOI: https://doi.org/10.3390/ani12233390

Interpretive Summary: We previously demonstrated that we could accurately count and determine the location of a broiler chickens in broiler houses using a machine vision-based method. In this study, we sought to use this method to predict the activities broiler chickens are engaged in e.g., feeding, drinking, standing, and resting. To do this, we tested different deep learning behavior posture recognition models and found that the DenseNet-264 network was the best method for predicting broiler behaviors when birds were 2, 9, 16 and 23 days old. This study demonstrated that it is possible to automatically monitor the behavior of broilers and it is a proof-of-concept for developing a machine vision-based method early detection of sick/diseased chickens.

Technical Abstract: Animal behavior monitoring allows the gathering of animal health information and living habits and is an important technical means in precision animal farming. To quickly and accurately identify the behavior of broilers at different days, we adopted different deep learning behavior recognition models. Firstly, the top-view images of broilers at 2, 9, 16 and 23 days of age old were obtained. In each stage, 300 images of each of the four broilers behaviors (i.e., feeding, drinking, standing, and resting) were segmented, totaling 4800 images. After image augmentation processing, 10,200 images were generated for each day including 8,000 training sets, 2,000 validation sets, and 200 testing sets. Finally, the performance of different convolutional neural network models (CNN) in broiler behavior recognition at different days was analyzed. The results show that the overall performance of the DenseNet-264 network was the best, with the accuracy rates of 88.5%, 97%, 94.5%, and 90% when birds were 2, 9, 16 and 23 days old, respectively. In addition, the efficient channel attention was introduced into the DenseNet-264 network (ECA-DenseNet-264), and the results (accuracy rates: 85%, 95%, 92%, 89.5%) confirmed that the DenseNet-264 network was still the best overall. The research results demonstrate that it is feasible to apply deep learning technology to monitor the behavior of broilers at different days.