|
|
|
|
|
|
| Using DANN to Classify the Mango Varieties With NIR Spectroscopy |
| LI Tong-le1, CHEN Xiao2, CHEN Xiao-jing1, CHEN Xi1, YUAN Lei-ming1, SHI Wen1, HUANG Guang-zao1* |
1. College of Electrical and Electronic Engineering, Wenzhou University, Wenzhou 325000, China
2. Wenzhou Institute of Industry & Science, Wenzhou 325000, China
|
|
|
|
|
Abstract Different cultivars of mangoes can not only represent different qualities but also produce different economic benefits. Traditional mango variety identification methods often rely more on the experience of practitioners and are time-consuming and laborious. Therefore, how to quickly classify mango cultivars is an emerging problem that needs to be solved. Near-infrared (NIR) spectroscopy technology is a fast and non-destructive approach. Users can often identify different mangoes by combining machine learning methods with near-infrared spectroscopy data. However, the NIR spectral information of the same variety of mangoes can vary due to variations in different instruments, seasons, and years. These differences result in a different distribution between the previously measured sample data (source domain) and the newly measured sample data (target domain). Consequently, the present classification model cannot correctly classify new mango samples. Domain adaptation methods can solve this problem of model inapplicability caused by different data distributions. This article focuses on the distribution differences of mango near-infrared spectroscopy data caused by factors such as working temperature and season. The domain adaptation methods can solve the problem of model unsuitability caused by different data distributions. This article used a deep domain adaptive neural network (DANN) model to solve this problem. The DANN model effectively achieves cross-domain sample classification models by aligning features between two domains through adversarial learning. This article compared DANN with unsupervised dynamic orthogonal projection (uDOP) and joint distribution adaptation (JDA), two traditional domain adaptation methods based on statistical learning. The experimental results of applying these three methods in this article showed that the DANN model could achieve a classification accuracy of 94% for the test set in the binary classification task of mango varieties. In the multi-classification task of mango varieties, the classification accuracy of the DANN model was over 10% higher than that of uDOP and JDA. The results indicated that the DANN model could effectively solve the problem of mango variety recognition caused by the different distribution of near-infrared spectral data between two fields.
|
|
Received: 2024-06-24
Accepted: 2024-09-14
|
|
|
|
Corresponding Authors:
HUANG Guang-zao
E-mail: guangzh@wzu.edu.cn
|
|
[1] Aly Farag El Sheikha, Isabelle Métayer, Didier Montet. Food Biotechnology, 2011, 25: 115.
[2] Nicolaï B M, Beullens K, Bobelyn E, et al. Postharvest Biol. Postharvest Biology and Technology, 2007, 46(2): 99.
[3] Rasool Khodabakhshian, Bagher Emadi, Mehdi Khojastehpour, et al. International Journal of Food Properties, 2017, 20(1): 41.
[4] Rong D, Wang H, Ying Y, et al. Electron. Computers and Electronics in Agriculture, 2020, 175: 105553.
[5] Jerome J, Workman J. Applied Spectroscopy, 2018, 72(3): 340.
[6] Wang Yongdong, Veltkamp D J, Kowalski B R. Analytical Chemistry, 1991, 63(23): 2750.
[7] Bouveresse E, Massart D L. Chemometrics and Intelligent Laboratory Systems, 1996, 32(2): 201.
[8] Du W, Chen Z P, Zhong L J. Analytica Chimica Acta, 2011, 690(1): 64.
[9] Nikzad Langerodi R, Zellinger W, Lughofer E, et al. Analytical Chemistry, 2018, 90(11): 6693.
[10] Huang G, Chen X, Li L. Chemometrics and Intelligent Laboratory Systems，2020, 201：103986.
[11] Zeaiter M, Roger J M, Bellon-Maurel V. Chemometrics and Intelligent Laboratory Systems, 2006, 80(2)：227.
[12] Diaz V F, Roger J M, Saeys W. Analytica Chimica Acta, 2022, 1225：340154.
[13] Pan S J, Tsang I W, Kwok J T, et al. IEEE Transactions on Neural Networks, 2011, 22(2)：199.
[14] Mishra P, Roger J M, Rutledge D N, et al. Postharvest Biology and Technology, 2020, 170: 111326.
[15] Zhao X, Liu S, Que H, et al. Sensors, 2023, 23(19).
[16] Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, et al. Journal of Machine Learning Research, 2016, 17(59): 1.
[17] Rodriguez-Vazquez J, Fernandez-Cortizas M, Perez-Saura D, et al. Remote Sensing，2023, 15(6): 1700.
[18] Anderson N T, Walsh K B, Subedi P P, et al. Postharvest Biology and Technology, 2020, 168: 111202.
|
| [1] |
ZHAO Yu-xia1, ZHANG Ming-jin1, 3*, WANG Ru1, ZHANG Shi-zhi2, YIN Bo1, 3. Discriminative Study on Huzhu Qingke Liquor by Back Propagation
Neural Network Combined With Ultraviolet-Near Infrared Fusion
Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(05): 1290-1299. |
| [2] |
ZHU Yong-bing1, CAI Yu-qin1, JIANG Li-yao1, LEI Chun1, TENG Long1, WANG De-wang3, TAO Zhi2*. Study on Quantitative Analysis Method of TDLAS Intravenous Drug
Concentration Based on ECA-1D-CNN[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(05): 1341-1347. |
| [3] |
YAO Cheng-shuo1, 2, WANG Chang-kun1, 2*, LIU Jie1, 2, GUO Zhi-ying1, 2, MA Hai-yi1, 2, YUAN Zi-ran1, 2, WANG Xiao-pan1, 3, PAN Xian-zhang1, 2. Spectral Prediction of Soil Fertility Attributes in Typical Croplands of Sanjiang Plain Based on Band Selection[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(05): 1422-1431. |
| [4] |
LI Jia-qi1, 2, 3, TIAN Xi2, 3, WANG Qing-yan2, 3, HE Xin2, 3, HUANG Wen-qian2, 3*. Research on the Method of Online Detection of Hollow Watermelons Based on Full-Transmission Near-Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(05): 1440-1447. |
| [5] |
CHEN Zhuo-ting1, WANG Qiao-hua1, 2*, WANG Dong-qiao1, CHEN Yan-bin1, LI Shi-jun1, 2. Non-Destructive Detection of Pre-Incubation Breeding Duck Egg Fertilization Information Based on Visible/Near Infrared Spectroscopy and Joint Optimization Strategy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(05): 1469-1475. |
| [6] |
WEN Yu-kuan1, DONG Gui-mei1, LI Liu-an2, YU Xiao-xue2, YU Ya-ping1*. Nondestructive Identification of Egg Yolk Color Based on Near Infrared Spectrum and Multivariate Data Processing[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(04): 1015-1021. |
| [7] |
CHEN Xin-gang1, 2, ZHANG Wen-xuan1, MA Zhi-peng1*, ZHANG Zhi-xian1, WAN Fu3, AO Yi1, ZENG Hui-min1. Improved Convolutional Neural Network Quantification of Mixed Fault Characterization Gases in Transformers Based on Raman Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(04): 932-940. |
| [8] |
TIAN Fu-chao1, 2, 3, ZHANG Hai-long1, 2, 3, SU Jia-hao1, 2, 3*, LIANG Yun-tao1, 2, 3, WANG Lin1, 2, 3, WANG Ze-wen1, 2, 3. Pressure Compensation of Industrial Ambient Gases and Their Prediction Based on Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(04): 994-1007. |
| [9] |
CUI Jia-cheng1, SONG Wei-ran1, YAO Wei-li2, JI Jian-xun1, HOU Zong-yu1, 3, WANG Zhe1, 3*. Improving LIBS Quantification by Combining Domain Factors and Multilayer Perceptron Method[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(04): 1022-1027. |
| [10] |
SHANG Yan-xia, HOU Ming-yu*, CUI Shun-li, LIU Ying-ru, LIU Li-feng, LI Xiu-kun*. Construction of Near-Infrared Detection Models for Peanut Protein and Their Components With Different Seed Coat Colors[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(04): 1129-1136. |
| [11] |
ZHANG Yi-ting1, 2, LU Dong-hua1, 2*, WU Ding1, 2, GAO Yan1, 2. Extraction of Impervious Surfaces in Towns Based on UAV Hyperspectral Imagery[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(04): 1150-1158. |
| [12] |
CHEN Jin-ni, TIAN Gu-feng*, LI Yun-hong, ZHU Yao-lin, CHEN Xin, MEN Yu-le, WEI Xiao-shuang. Near-Infrared Spectral Prediction Model for Cashmere and Wool Based on Two-Way Multiscale Convolution[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(03): 678-684. |
| [13] |
LIAO Juan1, CAO Jia-wen1, TIAN Ze-feng1, LIU Xiao-li1, YANG Yu-qing1, ZOU Yu2, WANG Yu-wei1, ZHU De-quan1*. Rapid Discrimination Method for Empty Grain Content Grade of Rice Seed Based on Near-Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(03): 692-699. |
| [14] |
ZHAO Xin1, 4, SHI Yu-na1, LIU Yi-tong1, JIANG Hong-zhe2, CHU Xuan3, ZHAO Zhi-lei1, 4, WANG Bao-jun1, 4*, CHEN Han1. Key Feature Analysis in Identification and Authenticity of Ziziphi Spinosae Semen by Using Hyperspectral Images Based on 1DCNN and PLSDA[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(03): 869-877. |
| [15] |
YIN Xiong1, CUI Hong-shuai1, LIU Xue-jing1, MA Shi-yi1, ZHOU Yan1*, CHONG Dao-tong1, XIONG Bing2, LI Kun2. Study on the Detection of Wear Particle Content in Lubricating Oil Based on Near-Infrared Spectroscopy Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(03): 816-825. |
|
|
|
|
