Qualitative Modeling Method of Mango Species in Near Infrared Based on Attention Mechanism Residual Neural Network
WANG Shu-tao1, WAN Jin-cong1*, LIU Shi-yu2, ZHANG Jin-qing1, WANG Yu-tian1
1. Department of Instrument Science and Engineering, School of Electrical Engineering，Yanshan University，Qinhuangdao 066004，China
2. College of Quality and Technical Supervision, Hebei University, Baoding 071002, China
Abstract：In recent years, the rapid development of modern spectral detection technology is closely related to deep learning. As an end-to-end model, the deep neural network can get more information from the spectra, thus improving the robustness of the model. A one-dimensional residual neural network (1D-AD-ResNet-18) model based on a convolutional block attention module was proposed to explore the feasibility of qualitative prediction of mango species by near-infrared spectroscopy combined with deep learning. Firstly, to reduce the interference of redundant information in the spectra, the CBAM convolution attention module is added to the traditional one-dimensional residual neural network, which can focus on the local useful information of the spectra. Secondly, to avoid the disappearance of gradient and the occurrence of overfitting, ResNet-18 is used to solve the problem of network “degradation”. For 186 mango samples, 70% of the samples were trained, and 30% were tested. Accuracy, Precision, Recall, F1-score, Macro-average, and weighted average were used as evaluation indexes of the model. Three comparison models were established, including traditional one-dimensional ResNet-18, SNV-SVM, and PCA-KNN. Compared with the above three methods, the established 1D-AD-ResNet-18 model obtained the optimal prediction results, and the accuracy of the four qualitative analysis models was 96.42%，80.35%，76.78% and 67.85%. The experimental results show that the 1D-AD-ResNet-18 model can accurately identify and classify mango species, which provides a new idea for the qualitative analysis of mango species by NIR spectroscopy.
王书涛，万金丛，刘诗瑜，张金清，王玉田. 基于注意力机制残差神经网络的近红外芒果种类定性建模方法[J]. 光谱学与光谱分析, 2024, 44(08): 2262-2267.
WANG Shu-tao, WAN Jin-cong, LIU Shi-yu, ZHANG Jin-qing, WANG Yu-tian. Qualitative Modeling Method of Mango Species in Near Infrared Based on Attention Mechanism Residual Neural Network. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(08): 2262-2267.
[1] Yahia E M, Ornelas-Paz J D J, Brecht J K, et al. Arabian Journal of Chemistry, 2023, 16(7): 104860.
[2] CHU Xiao-li, CHEN Pu, LI Jing-yan, et al(褚小立, 陈 瀑, 李敬岩, 等). Journal of Instrumental Analysis(分析测试学报), 2020, 39(10): 1181.
[3] NI Chao, LI Zhen-ye, ZHANG Xiong, et al(倪 超, 李振业, 张 雄, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2019, 50(12): 170.
[4] Rong D, Wang H Y, Ying Y B, et al. Computers and Electronics in Agriculture, 2020, 175: 9.
[5] Yang J, Ma X D, Guan H O, et al. Infrared Physics & Technology, 2023, 128: 10.
[6] Guan H O, Yu M, Ma X D, et al. Infrared Physics & Technology, 2022, 127: 14.
[7] Chen X H, Cheng G Y, Liu S H, et al. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2022, 279: 9.
[8] Wang Y, Zhang Z, Feng L, et al. Computers and Electronics in Agriculture, 2021, 184: 106090.
[9] Munawar A A, Kusumiya T I, Wahyun I D. Data Brief, 2019, 27: 104789.
[10] Yu H Z, Li Z Z, Guo W B, et al. Industrial Crops and Products, 2023, 196: 116455.
[11] Chen L J, Yao H D, Fu J Y, et al. Engineering Structures, 2023, 275: 16.
[12] SUN Pan, SHI Xiu-dong, HE Ying-jie(孙 攀, 石秀东, 何英杰). Manufacturing Automation(制造业自动化), 2023, 45(4): 34.
[13] ZENG Jing-xiang, ZHANG Jin-xi, CAO Dan-dan, et al(曾靖翔, 张金喜, 曹丹丹, 等). Journal of South China University of Technology(Natural Science Edition)[华南理工大学学报（自然科学版）], 2022, 50(3): 50.
[14] CHEN Jia-yue, LI Fei(陈嘉跃, 李 飞). Electronic Measurement Technology(电子测量技术), 2022, 45(1): 104.
