Research on Non-destructive Identification of Chick Embryo Gender Based on Deep Learning

DOI: https://doi.org/10.62517/jbdc.202401112

Author(s)

Xinghui Jiao1,#, Ling Wang1,#,*, Xiaojuan Liu2,#, Yizhong Zhang1, Yihua Zheng1, Shuo Chen1, Shuyang Shi1, Pan Ding1,#,*

Affiliation(s)

1Henan agricultural University, Zhengzhou, China 2Zhengzhou Normal University, Zhengzhou, China #These authors contributed equally to this work *Corresponding Author.

Abstract

In the chick hatching industry, a common practice is to directly eliminate male chicks after hatching. However, this practice results in significant resource wastage. Timely detection of embryo gender and selection of male embryos are of great significance for reducing resource wastage and improving economic benefits. To address the serious lack of gender identification technology during chick hatching, this paper proposes a non-destructive identification method for chick embryos based on deep learning. We use the PyTorch framework to build a deep learning model and divide the dataset into 80% training set and 20% validation set for model training and validation. Experimental results show that our proposed model achieves an accuracy of 72.5% on the validation set. This study not only solves key technical problems for non-destructive identification of chick embryo gender but also provides new research ideas for precise gender identification of other oviparous species, promoting the intelligent development of production and breeding industries.

Keywords

Embryo; Gender Detection; Deep Learning; Machine Vision

References

[1] Krautwald-Junghanns ME, Cramer K, Fischer B, Förster A, Galli R, et al. Current approaches to avoid the culling of day-old male chicks in the layer industry, with special reference to spectroscopic methods. Poultry Sci, 2018, 97: 749-757. [2] De Haas EN, Oliemans E, van Gerwen MAAM. The need for an alternative to culling day-old male layer chicks: A survey on awarene b css, alternatives, and the willingness to pay for alternatives in a selected population of Dutch citizens. Front. Vet. Sci, 2021, 8: 662197. [3] Zhang J, Gao C .Hatching Egg Image Segmentation Based on Dense Blocks and the Hierarchical Sampling of Pixels. American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS), 2020, 74(1):189-199. [4] Alin K, Fujitani S, Kashimori A. Non-invasive broiler chick embryo sexing based on opacity value of incubated eggs. Computers and electronics in agriculture, 2019, 158: 30-35. [5] Huang L, He A, Zhai M, et al. A Multi-Feature Fusion Based on Transfer Learning for Chicken Embryo Eggs Classification. Symmetry, 2019, 11(5):606. [6] Bamelis FR, Tona K, Baerdemaeker JGD, Decuypere EM. Detection of early embryonic development in chicken eggs using visible light transmission. British Poultry Science, 2002, 43(2): 204–212. [7] Haghighat M, Abdel-Mottaleb M, Alhalabi W. Discriminant Correlation Analysis: Real-Time Feature Level Fusion for Multimodal Biometric Recognition. IEEE Transactions on Information Forensics and Security, 2016, 11(9): 1984–1996. [8] Rahman A, Syduzzaman M, Khaliduzzaman A, Fujitani S, Kashimori A, Suzuki T, Kondo N. Nondestructive Sex-Specific Monitoring of Early Embryonic Development Rate in White Layer Chicken Eggs using Visible Light Transmission. British Poultry Science, 2020, 60: 209-216. [9] Wang YL, Jin GF, Ma MH, Xiang XL. Sex differences in serum steroid hormonelevels during embryonic development in hen eggs. Poultry Science, 2019, 98(11):6053-6062. [10] Vapnik V, Golowich SE, Smola A. Support vector method for function approximation, regression estimation and signal processing. Advances in Neural Information Processing Systems, 1996, 9.