Non-Destructive Testing of Fertilization Information of Pre-Incubation Duck Eggs Based on Convolutional Neural Network and Spectral Features
LI Qing-xu1, WANG Qiao-hua1, 2*, GU Wei1, GAO Sheng1, MA Mei-hu3
1. College of Engineering, Huazhong Agricultural University, Wuhan 430070, China
2. Key Laboratory of Agricultural Equipment in Mid-Lower Yangtze River, Ministry of Agriculture and Rural Agriculture, Wuhan 430070, China
3. National Egg Research and Development Center, Wuhan 430070, China
Abstract:China is a big country for the production and consumption of duck eggs and duck meat. For this reason, a large number of ducklings need to be hatched each year to meet production needs. Since the Infertile egg cannot hatch the duckling during the hatching process, it can be avoided by eliminating it as early as possible. At present, it is necessary to carry out artificial photographing eggs in the country for about 7 days after hatching, so that the infertile eggs can be removed. For the low-efficiency and elimination of artificial eggs, there is no edible value problem. In this paper, the duck eggs before hatching were taken as research objects, and the visible/near-infrared transmission spectroscopy combined with the convolutional neural network was used to realize the duck eggs before hatching Non-destructive identification of fertilization information. The method preprocesses the acquired 400~1 000 nm raw spectral information (removing the noise band and Savitzky-Golay convolution smoothing) to eliminate noise and extraneous information, and applies the competitive adaptive re-weighting algorithm (CARS) and continuous projection algorithm (SPA) to select the characteristic wavelengths, and converts the selected characteristic wavelengths into spectral two-dimensional information matrix. The spectral two-dimensional information matrix can not only represent the effective information of the characteristic spectrum, but also can transmit the spectral information to the neural network for training. For the characteristics of spectral data, the network is too deep to cause over-fitting of the model, and the shallow network will cause under-fitting of the model. A four-layer convolutional neural network (CNN) is built to train the spectral information, including three convolutional layers and one fully-connected layer. The convolutional layer is used to extract the spectral two-dimensional information matrix automatically. Effective information, the fully connected layer is integrated for the output layer by integrating the local features extracted by the convolutional layer. In addition, the introduction of local corresponding normalization layer, pooling layer and dropout in the convolutional neural network can accelerate the convergence speed of the network and prevent model overfitting. The accuracy of the model training set established by the SPA extracted feature wavelength is 97.71%, the test set accuracy rate is 97.41%, and the verification set accuracy rate is 98.29%. The model training set accuracy rate established by CARS extraction is 97.42%, the test set accuracy rate 97.41%, the verification set accuracy rate is 97.44%,the traditional machine learning model test set established using the characteristic wavelengths extracted by SPA and CARS has a precision of only 87.39%. The results show that the combination of deep learning and spectroscopy can realize the non-destructive identification of the fertilization information of duck eggs before hatching, which can provide efficient, non-destructive and rapid model support for the subsequent development of dynamic online detection equipment.
李庆旭,王巧华,顾 伟,高 升,马美湖. 基于卷积神经网络和光谱特征的孵前种鸭蛋受精信息无损检测[J]. 光谱学与光谱分析, 2020, 40(12): 3847-3853.
LI Qing-xu, WANG Qiao-hua, GU Wei, GAO Sheng, MA Mei-hu. Non-Destructive Testing of Fertilization Information of Pre-Incubation Duck Eggs Based on Convolutional Neural Network and Spectral Features. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2020, 40(12): 3847-3853.
[1] Ghaderi M, Banakar A, Masoudia A. Measurement, 2018, (114): 191.
[2] Hashemzadeh M, Farajzadeh N. International Journal of Computational Intelligence Systems, 2016, 9(5): 850.
[3] ZHU Zhi-hui, LIU Ting, MA Mei-hu(祝志慧,刘 婷, 马美湖). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2015, 31(15): 285.
[4] ZHANG Wei, TU Kang, LIU Peng, et al(张 伟, 屠 康, 刘 鹏, 等). Transactions of the Chinese Society of Agricultural Machinery(农业机械学报), 2012, 43(2): 140.
[5] Dong Jun, Dong Xiaoguang, Li Yanlei, et al. Computers and Electronics in Agriculture, 2019, 157:471.
[6] ZHU Zhi-hui, XIE De-jun, LI Wan-qing, et al(祝志慧, 谢德君, 李婉清,等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2015, 31(2): 312.
[7] YUAN Pei-sen, LI Wei, REN Shou-gang, et al(袁培森, 黎 薇, 任守纲,等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2018, 34(5): 152.
[8] FU Dan-dan, WANG Qiao-hua(付丹丹, 王巧华). Food Science(食品科学), 2016, 37(22): 173.
[9] GAO Sheng, WANG Qiao-hua, FU Dan-dan, et al(高 升, 王巧华, 付丹丹,等). Acta Optica Sinica(光学学报), 2019,10(39): 1.
[10] ZHU Zhi-hui, HONG Qi, WU Lin-feng, et al(祝志慧,洪 琪,吴林峰,等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2019, 39(9): 2780.