|
|
|
|
|
|
A Model Construction Method of Spectral Nondestructive Detection for Apple Quality Based on Unsupervised Active Learning |
ZHAO Xiao-kang, ZHAO Xin, ZHU Qi-bing*, HUANG Min |
Key Laboratory of Advanced Process Control for Light Industry (Ministry of Education), Jiangnan University, Wuxi 214122, China |
|
|
Abstract The essence of using near-infrared spectroscopy to realize non-destructive detection of agricultural products and food quality is to establish a machine learning model between sample spectral information and sample quality parameters. In order to obtain a machine learning model with good generalization performance, a large number of labeled samples are usually required. However, it is relatively easy to obtain spectral information of samples, but labeling samples quality parameters often involves a large amount of time and economic costs and is destructive. Active learning is a method to reduce the number of labeled samples in training set by selecting the most valuable samples for labeling instead of random selection. Therefore, active learning can control which samples are added to the training set, and the model no longer passively accepts samples for modeling. There have been many active learning algorithms in classification tasks. There are relatively few researches in regression tasks. Moreover, most of the existing active learning algorithms for regression tasks are supervised. That is, a small number of labeled samples are needed to train the initial model. In this paper, a training sample selection strategy based on unsupervised active learning is proposed. Firstly, the method divides the diversity of unlabeled (standard value) spectral datasets through hierarchical agglomerative clustering to obtain different clustering clusters. Then, the locally linear reconstruction method selects the most representative samples in each clustering cluster to form a training sample set and establish the partial least squares regression model based on the training set to predict the unlabeled samples. In this paper, partial least squares prediction models for soluble solids content and firmness prediction were constructed to evaluate the proposed method’s performance, using the near infrared spectrum data of three varieties of apples from two years. The experimental results show that the method proposed in this paper is superior to the existing sample selection strategy, which can effectively improve the model accuracy and reduce destructive physical and chemical experiments in model training. Meanwhile, compared with random sampling (RS), traditional Kennard-Stone (KS) and joint x-y distances (SPXY), the proposed method achieved the optimal performance. The root mean square error of the soluble solid content prediction models based on the unsupervised active learning algorithm proposed in this paper, which selects 200 samples as the training set, is reduced by 2.0%~13.2% compared with the other three algorithms, and the root means square error of the firmness prediction models is reduced by 1.2%~15.7%.
|
Received: 2020-12-18
Accepted: 2021-03-22
|
|
Corresponding Authors:
ZHU Qi-bing
E-mail: zhuqib@163.com
|
|
[1] Li X N, Huang J C, Xiong Y J, et al. Computers and Electronics in Agriculture, 2018, 155: 23.
[2] MAO Bo-hui, SUN Hong, LIU Hao-jie, et al(毛博慧, 孙 红, 刘豪杰, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2017, 48(S1): 160.
[3] TANG Jin-ya, HUANG Min, ZHU Qi-bing(唐金亚, 黄 敏, 朱启兵). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2015, 35(8): 2136.
[4] GUO Wen-chuan, ZHU De-kuan, ZHANG Qian, et al(郭文川, 朱德宽, 张 乾, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2020, 51(9): 350.
[5] MA Wen-qiang, ZHANG Man, LI Yuan, et al(马文强, 张 漫, 李 源, 等). Chinese Journal of Analytical Chemistry(分析化学), 2020, 48(12): 1737.
[6] WANG Li-guo, SHANG Hui, SHI Yao(王立国, 商 卉, 石 瑶). Journal of Harbin Engineering University(哈尔滨工程大学学报), 2020, 41(5): 731.
[7] DAI Xiang, HUANG Xi-feng, TANG Rui, et al(代 翔, 黄细凤, 唐 瑞, 等). Journal of South China University of Technology·Natural Science Edition(华南理工大学学报·自然科学版), 2019, 47(8): 84.
[8] Zhang L J, Chen C, Bu J J, et al. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(10): 2026.
[9] Mendoza F, Lu R F, Cen H Y. Postharvest Biology and Technology, 2012, 73: 89.
[10] Li H D, Liang Y Z, Xu Q S, et al. Analytica Chimica Acta, 2009, 648(1): 77.
[11] LIU Zi-ang, JIANG Xue, WU Dong-rui(刘子昂, 蒋 雪, 伍冬睿). Aata Automatica Sinica(自动化学报), https://doi.org/10.16383/j.aas.c200071.
[12] YAN Yue, ZHANG Hong-guang, LU Jian-gang, et al(鄢 悦, 张红光, 卢建刚, 等). Computers and Applied Chemistry(计算机与应用化学), 2017, 34(5): 351. |
[1] |
ZHENG Hong-quan, DAI Jing-min*. Research Development of the Application of Photoacoustic Spectroscopy in Measurement of Trace Gas Concentration[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 1-14. |
[2] |
CHENG Jia-wei1, 2,LIU Xin-xing1, 2*,ZHANG Juan1, 2. Application of Infrared Spectroscopy in Exploration of Mineral Deposits: A Review[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 15-21. |
[3] |
FAN Ping-ping,LI Xue-ying,QIU Hui-min,HOU Guang-li,LIU Yan*. Spectral Analysis of Organic Carbon in Sediments of the Yellow Sea and Bohai Sea by Different Spectrometers[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 52-55. |
[4] |
LI Jie, ZHOU Qu*, JIA Lu-fen, CUI Xiao-sen. Comparative Study on Detection Methods of Furfural in Transformer Oil Based on IR and Raman Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 125-133. |
[5] |
BAI Xi-lin1, 2, PENG Yue1, 2, ZHANG Xue-dong1, 2, GE Jing1, 2*. Ultrafast Dynamics of CdSe/ZnS Quantum Dots and Quantum
Dot-Acceptor Molecular Complexes[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 56-61. |
[6] |
XU Tian1, 2, LI Jing1, 2, LIU Zhen-hua1, 2*. Remote Sensing Inversion of Soil Manganese in Nanchuan District, Chongqing[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 69-75. |
[7] |
WANG Fang-yuan1, 2, HAN Sen1, 2, YE Song1, 2, YIN Shan1, 2, LI Shu1, 2, WANG Xin-qiang1, 2*. A DFT Method to Study the Structure and Raman Spectra of Lignin
Monomer and Dimer[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 76-81. |
[8] |
LIU Zhen1*, LIU Li2*, FAN Shuo2, ZHAO An-ran2, LIU Si-lu2. Training Sample Selection for Spectral Reconstruction Based on Improved K-Means Clustering[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 29-35. |
[9] |
YANG Chao-pu1, 2, FANG Wen-qing3*, WU Qing-feng3, LI Chun1, LI Xiao-long1. Study on Changes of Blue Light Hazard and Circadian Effect of AMOLED With Age Based on Spectral Analysis[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 36-43. |
[10] |
GAO Feng1, 2, XING Ya-ge3, 4, LUO Hua-ping1, 2, ZHANG Yuan-hua3, 4, GUO Ling3, 4*. Nondestructive Identification of Apricot Varieties Based on Visible/Near Infrared Spectroscopy and Chemometrics Methods[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 44-51. |
[11] |
ZHENG Pei-chao, YIN Yi-tong, WANG Jin-mei*, ZHOU Chun-yan, ZHANG Li, ZENG Jin-rui, LÜ Qiang. Study on the Method of Detecting Phosphate Ions in Water Based on
Ultraviolet Absorption Spectrum Combined With SPA-ELM Algorithm[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 82-87. |
[12] |
XU Qiu-yi1, 3, 4, ZHU Wen-yue3, 4, CHEN Jie2, 3, 4, LIU Qiang3, 4 *, ZHENG Jian-jie3, 4, YANG Tao2, 3, 4, YANG Teng-fei2, 3, 4. Calibration Method of Aerosol Absorption Coefficient Based on
Photoacoustic Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 88-94. |
[13] |
LI Xin-ting, ZHANG Feng, FENG Jie*. Convolutional Neural Network Combined With Improved Spectral
Processing Method for Potato Disease Detection[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 215-224. |
[14] |
XING Hai-bo1, ZHENG Bo-wen1, LI Xin-yue1, HUANG Bo-tao2, XIANG Xiao2, HU Xiao-jun1*. Colorimetric and SERS Dual-Channel Sensing Detection of Pyrene in
Water[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 95-102. |
[15] |
LEI Hong-jun1, YANG Guang1, PAN Hong-wei1*, WANG Yi-fei1, YI Jun2, WANG Ke-ke2, WANG Guo-hao2, TONG Wen-bin1, SHI Li-li1. Influence of Hydrochemical Ions on Three-Dimensional Fluorescence
Spectrum of Dissolved Organic Matter in the Water Environment
and the Proposed Classification Pretreatment Method[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 134-140. |
|
|
|
|