Vis-NIR Spectra Discriminant of Pesticide Residues on the Hami Melon Surface by GADF and Multi-Scale CNN
YU Guo-wei1, MA Ben-xue1,2*, CHEN Jin-cheng1,3, DANG Fu-min4,5, LI Xiao-zhan1, LI Cong1, WANG Gang1
1. College of Mechanical and Electrical Engineering, Shihezi University, Shihezi 832003, China
2. Key Laboratory of Northwest Agricultural Equipment, Ministry of Agriculture and Rural Affairs, Shihezi 832003, China
3. Mechanical Equipment Research Institute, Xinjiang Academy of Agricultural and Reclamation Science, Shihezi 832000, China
4. Food Quality Supervision and Testing Center (Shihezi), Ministry of Agriculture, Shihezi 832000, China
5. Analysis and Testing Center, Xinjiang Academy of Agricultural and Reclamation Science, Shihezi 832000, China
Abstract:Given the costly and destructive detection of pesticide residues on the Hami melon surface, the feasibility of visible/near-infrared (Vis/NIR) spectroscopy for the qualitative discriminant was assessed. In this study, Hami melon was taken as experimental samples. Two pesticides were taken as the research objects, including chlorothalonil and imidacloprid. Hami melon’s Vis/NIR spectra of Hami melon with no, chlorothalonil and imidacloprid residues were collected in the diffuse reflectance mode. Then the one-dimensional spectrum was transformed into a two-dimensional image by using gramian angular fields (GAF). The GAF image data set was constructed. A multi-scale convolutional neural network (CNN) architecture incorporatedan Inception module was developed, including aninput layer, three convolution layers, amerging layer, aflatten layer, two fully-connected layers, and an output layer. The confusion matrix result of the multi-scale CNN model suggested that the best method for expressing Vis/NIR spectral features was gramian angular difference fields (GADF) transformation. Moreover, two CNN models (AlexNet and VGG-16) and two machine learning models (support vector machine (SVM) and extreme learning machine (ELM)) were established toverify the proposed model performance. With higher average accuracy than SVM and ELM models, the CNN models had a better effect to identifying whether there were pesticide residues on the Hami melon surface. Compared with AlexNet and VGG-16 models, the proposed multi-scale CNN model had the best performance with the shortest training time of 14 s and the highest test accuracy of 98.33%.The multi-scale CNN structure can capture different level and scale features by using combinations of various small-size filters (1 1, 3 3 and 5 5) and stacking of parallel convolutions. The multi-scale deep feature fusion was carried out in the concatenation mode, which can improve the feature extraction ability of the CNN model. Compared with traditional CNN models with large depth, the model proposed in this study improved the discriminant accuracy while keeping the computational complexity constant. The overall research results reflected that GADF transformation combined with a multi-scale CNN model can effectively achieve the qualitative spectral data analysis. Vis/NIR spectroscopy can realize the qualitative discriminant of pesticide residues on the Hami melon surface. These findings can provide a reference for the rapid non-destructive detection of pesticide residues on the surface of other large melons and fruit.
喻国威,马本学,陈金成,党富民,李小占,李 聪,王 刚. 基于GADF变换和多尺度CNN的哈密瓜表面农药残留可见-近红外光谱判别方法[J]. 光谱学与光谱分析, 2021, 41(12): 3701-3707.
YU Guo-wei, MA Ben-xue, CHEN Jin-cheng, DANG Fu-min, LI Xiao-zhan, LI Cong, WANG Gang. Vis-NIR Spectra Discriminant of Pesticide Residues on the Hami Melon Surface by GADF and Multi-Scale CNN. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(12): 3701-3707.
[1] WEI Kai-li, ZHOU Xiao-long, YAN Qiao-li, et al(韦凯丽, 周晓龙, 闫巧俐, 等). Food & Machinery(食品与机械),2019, 35(8): 90.
[2] JIANG Xue-song, WANG Wei-qin, XU Lin-yun, et al(蒋雪松, 王维琴, 许林云, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报),2016, 32(20): 267.
[3] Zhang X, Zhang N N. Agricultural Biotechnology,2018, 7(5): 222.
[4] Sun J, Ge X, Wu X H, et al. Journal of Food Process Engineering,2018, 41(6): e12816.
[5] Yazici A, Tiryaki G Y, Ayvaz H. Journal of the Science of Food and Agriculture,2020, 100(5): 1980.
[6] Yang J, Xu J F, Zhang X L, et al. Analytica Chimica Acta, 2019, 1081: 6.
[7] Zhang X L, Lin T, Xu J F, et al. Analytica Chimica Acta, 2019, 1058: 48.
[8] Zhang X L, Xu J F, Yang J, et al. Analytica Chimica Acta, 2020, 1119: 41.
[9] Wu M M, Sun J, Lu B, et al. Journal of Food Process Engineering,2019, 42(3): e13005.
[10] National Health Commission of the People’s Republic of China, Ministry of Agriculture and Rural Affairs of the People’s Republic of China, State Administration for Market Regulation(中华人民共和国国家卫生健康委员会, 中华人民共和国农业农村部, 国家市场监督管理总局). GB 2763—2021 National Food Safety Standard-Maximum Residue Limits for Pesticides in Food(GB 2763—2021食品安全国家标准 食品中农药残留最大限量),2021.
[11] Zhang G, Si Y J, Wang D, et al. IEEE Access,2019, 7: 171570.
[12] Szegedy C, Liu W, Jia Y Q, et al. IEEE Computer Society,2015, 1.
[13] Sun J T, Ma B X, Dong J, et al. Journal of Food Process Engineering,2017, 40: e12496.
[14] Hu R, Zhang L X, Yu Z Y, et al. Infrared Physics and Technology,2019, 102: 102999.