AUTHOR=Li Xiaolong , He Zhenni , Liu Fei , Chen Rongqin TITLE=Fast Identification of Soybean Seed Varieties Using Laser-Induced Breakdown Spectroscopy Combined With Convolutional Neural Network JOURNAL=Frontiers in Plant Science VOLUME=Volume 12 - 2021 YEAR=2021 URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2021.714557 DOI=10.3389/fpls.2021.714557 ISSN=1664-462X ABSTRACT=Soybean seed purity is a critical factor influencing the agricultural production, standardization of seed quality and food processing. In this study, laser-induced breakdown spectroscopy (LIBS) as a fast and micro-damage technology was successfully used to identify ten varieties of soybean seeds. We improved the traditional sample preparation scheme for LIBS by pressing soybean seeds into rubber sand in a culture plate using a ruler instead of grinding and squashing to ensure a relatively uniform surface height, greatly reducing the time cost. In our experimental scheme, three LIBS spectra were finally collected on the surface of one soybean seed. A majority vote based on three spectra was applied as the final decision on the variety of a single soybean seed. The results showed that support vector machine (SVM) obtained the optimal identification accuracy of 90% in the prediction set. In addition, PCA-ResNet (propagation coefficient adaptive ResNet) and PCSA-ResNet (propagation coefficient synchronous adaptive ResNet) were designed based on common ResNet structure through changing the way of self-adaption of propagation coefficients. Combined with a new form of input data called spectral matrix, PCSA-ResNet obtained the optimal performance with the discriminate accuracy of 91.75% in the prediction set. T-SNE was used to visualize the clustering process of the extracted features from the PCSA-ResNet. For the interpretation of the good performance of PCSA-ResNet coupled with spectral matrix, saliency maps were applied to visually show the pixel positions of spectral matrix that had a high influence on the discrimination results, indicating that the content and proportion of elements in soybean seeds could reflect the variety differences.