We developed a method to identify, count, and classify chickens and eggs inside nesting boxes of a chicken coop. Utilizing an IoT AWS Deep Lens Camera for data capture and inferences, we trained and deployed a custom single-shot multibox (SSD) object detection and classification model. This allows us to monitor a complex environment with multiple chickens and eggs moving and appearing simultaneously within the video frames. The models can label video frames with classifications for 8 breeds of chickens and/or 4 colors of eggs, with 98% accuracy on chickens or eggs alone and 82.5% accuracy while detecting both types of objects. With the ability to directly inferred and store classifications on the camera, this setup works in a low/no internet bandwidth setting. Having these classifications, provides the necessary base data required for accurately measuring the individual egg production of every chicken in the flock and supports additional flock production analysis.
Lubich, Jeremy; Thomas, Kyle; and Engels, Daniel W.
"Identification and Classification of Poultry Eggs: A Case Study Utilizing Computer Vision and Machine Learning,"
SMU Data Science Review: Vol. 2
, Article 20.
Available at: https://scholar.smu.edu/datasciencereview/vol2/iss1/20
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License