Classification of Hybrid and Peking Duck DOD Varieties Based on Feather Images Using CNN

Khoironi (Informatics Engineering, Department of Informatics and Computer Engineering, Surabaya State Polytechnic of Electronics, Surabaya, East Java, Indonesia)
I Wayan Rangga Pinastawa (Informatics, Computer Science Faculty, University Pembangunan Nasional Veteran Jakarta, DKI Jakarta, Indonesia)

Teknika
Vol. 14 No. 2 (2025): July 2025

Hybrid meat duck is a fast-growing variety with superior meat structure, the result of crossbreeding between Peking duck and laying duck to increase egg productivity. Day Old Duck (DOD) hybrid duck is highly sought after in the rearing industry, and this crossbreeding process produces hybrid and Peking varieties with different feather color patterns. Sorting DOD is important to maintain the consistency of hybrid varieties, but the process requires high expertise and a long time. The feather color of DOD meat duck can be used to distinguish between hybrid and Peking varieties. In this study, a deep learning approach with the Convolutional Neural Network (CNN) algorithm was applied to classify DOD varieties based on feather images. The study used 1702 images including hybrid and Peking ducklings. The model was tested with test data and evaluated using a confusion matrix to measure the performance of the developed model in identifying DOD varieties.