近红外光谱与图像融合的杂交水稻种子分类方法研究

doi:10.3964/j.issn.1000-0593(2023)09-2935-07

摘要
参考文献
相关文章 (15)

全文: PDF (11497 KB)
输出: BibTeX | EndNote (RIS)

摘要：随着杂交水稻育种技术的快速发展，杂交水稻品种日益繁多，品质与价格也千差万别，利用智能化手段对杂交水稻种子进行快速分类、分级和品质检测成为杂交水稻研究领域的热点。首先研究了不同预处理方式对基于近红外光谱的一维卷积神经网络分类模型对杂交水稻种子的分类效果的影响，研究结果表明利用Savitzky-Golay卷积平滑算法预处理后的一维卷积神经网络分类模型可获得最佳的分类效果，其验证集与测试集的分类精度为95.4%和92.9%。利用随机森林特征波长选择算法选取3个最重要的特征波长分别构建基于单波长灰度图像数据集和基于3波长重构的伪彩色图像数据集，研究了基于图像数据集的卷积神经网络VGG和残差网络ResNet的杂交水稻种子分类模型，其研究结果表明基于伪彩色图像数据集的VGG卷积神经网络模型能够获得最优的分类效果，其验证集与测试集的分类精度分别为92.8%和92.8%，相比基于伪彩色图像数据集的ResNet分类模型，其验证集提升3.6%，测试集提升4.9%。为了进一步提高分类的精度，提出了一种图像信息与光谱信息融合的杂交水稻种子分类方法，该方法利用1D-CNN网络分支提取种子的光谱特征，利用2D-CNN网络分支提取种子图像的空间维度特征，最终构建基于图谱融合的2Branch-CNN卷积神经网络分类模型，其验证集与测试集的分类精度都得到明显改善，分别达到98%和96.7%。并利用混淆矩阵评估了2Branch-CNN分类模型对于各个种类的杂交水稻种子的分类效果。研究结果表明通过图谱融合能有效提升卷积神经网络模型的分类精度，构建基于光谱与图像数据融合的二分支卷积神经网络模型将为杂交品种的种子快速筛选与分级提供新思路。

关键词：深度学习；近红外高光谱；杂交水稻种子；特征融合

Abstract：With the rapid development of hybrid rice breeding technology, hybrid rice varieties are becoming increasingly diverse, and their quality and price vary widely. The use of intelligent means for rapid classification, grading and quality detection of hybrid rice seeds has become a hot spot in hybrid rice research. In this paper, we first investigate the effect of different preprocessing methods on the accuracy of a 1D Convolutional Neural Network (1D-CNN) classification model constructed based on the near-infrared spectra of 10 hybrid rice seeds. The results show that the overall validation and prediction accuracy can be up to 95.4% and 92.9% respectively when the near-infrared spectra are preprocessed with the Savitzky-Golay convolution smoothing algorithm (SG). Secondly, the three most important feature wavelengths were selected by the random forest feature wavelength selection algorithm to build a single-wavelength grayscale image dataset and a 3-wavelength reconstructed pseudo-color image dataset, and the hybrid rice seed classification model based on the convolutional neural network VGG and the residual network ResNet of the image dataset was constructed and studied. The results show that the VGG model based on the pseudo-color image dataset can obtain the optimal classification effect, and the classification accuracies of its validation set and test set are 92.8% and 92.8%, respectively. Compared with the ResNet classification model based on the pseudo-color image dataset, an improved value of 3.6% is achievedin the validation set and 4.9% in the test set. In order to further improve the classification accuracy, a hybrid rice seed classification method based on the fusion of image information and spectral information is proposed. This methodextracts spectral features using the 1D-CNN network branch and extracts dimensionalspatial features using the 2D-CNN network branch. 2Branches-CNN model is then constructed based on the fusion of image and spectral features, and the classification accuracy reaches high values of 98% and 96.7% for the validation set and test set. The classification effect of the 2Branch-CNN model for each type of hybrid rice seeds is also evaluated by calculating the confusion matrix. The results of this paper show that the classification accuracy of the convolutional neural network model can be effectively improved by image-spectrum fusion, and the construction of a two-branch convolutional neural network model based on image-spectrum fusion will provide new ideas for rapid screening and classification of hybrid seed varieties.

Key words：Deep learning; Near-infrared hyperspectrum; Hybrid rice seeds; Feature fusion

收稿日期: 2022-05-04 修订日期: 2023-02-15

中图分类号:

TP391

基金资助: 岭南现代农业实验室科研项目（NT2021009），广东省重点领域研发计划项目（2019B020214 005），国家自然科学基金项目（12174120），高等学校学科创新引智计划项目(D18019)，广东省基础与应用基础研究基金项目（2022A151501 0411）资助

通讯作者: 邓海东，龙拥兵 E-mail: dhdong@scau.edu.cn；yongbinglong@126.com

作者简介: 叶文超，1996年生，华南农业大学电子工程学院（人工智能学院）硕士研究生 e-mail: 1299870652@qq.com

引用本文:

叶文超，罗水洋，李金豪，李钊荣，范志文，徐海涛，赵静，兰玉彬，邓海东，龙拥兵. 近红外光谱与图像融合的杂交水稻种子分类方法研究[J]. 光谱学与光谱分析, 2023, 43(09): 2935-2941.
YE Wen-chao, LUO Shui-yang, LI Jin-hao, LI Zhao-rong, FAN Zhi-wen, XU Hai-tao, ZHAO Jing, LAN Yu-bin, DENG Hai-dong, LONG Yong-bing. Research on Classification Method of Hybrid Rice Seeds Based on the Fusion of Near-Infrared Spectra and Images. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(09): 2935-2941.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2023)09-2935-07 或 https://www.gpxygpfx.com/CN/Y2023/V43/I09/2935

[1] XIE Fang-ming, PENG Shao-bing(谢放鸣，彭少兵). Science Bulletin (科学通报), 2016, 61(35): 3858.
[2] ZHU De-feng, ZHANG Yu-ping, CHEN Hui-zhe(朱德峰，张玉屏，陈惠哲, 等). Scientia Agricultura Sinica(中国业科学), 2015, 48(17): 3404.
[3] Yang G Y, Wang Q Y, Liu C, et al. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2018, 200: 186.
[4] LIU Xiao-dan, FENG Xu-ping, LIU Fei, et al(刘小丹，冯旭萍，刘飞, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2017, 33(22): 189.
[5] PENG Yan-kun, ZHAO Fang, LI Long, et al(彭彦昆，赵芳，李龙, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2018, 34(5): 159.
[6] YUE Jia-ming, CONG Xiao-xiang, LI Man-li(岳佳铭，丛晓翔，李曼莉). Seed(种子), 2021, 40(10): 129.
[7] Li H, Jia S, Le Z. Sensors, 2019, 19(20): 4355.
[8] Gutiérrez S, Wendel A, Underwood J. Computers and Electronics in Agriculture, 2019, 157: 126.
[9] Zhang X, Liu F, He Y, et al. Sensors (Basel, Switzerland), 2012, 12(12): 17234.
[10] YIN Wen-jun, RU Chen-lei, ZHENG Jie, et al(殷文俊，茹晨雷，郑洁，等). China Journal of Chinese Materia Medica(中国中药杂志), 2021, 46(4): 923.
[11] Li P, Zhang X, Li S, et al. Sensors, 2020, 20(6): 1586.
[12] WANG Cheng-kun, ZHAO Peng(王承琨，赵鹏). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2021, 41(6): 1713.
[13] He X T, Feng X P, Sun D W, et al. Molecules, 2019, 24(12): 2227.
[14] Baek I, Kim M S, Cho B K, et al. Applied Sciences, 2019, 9(5): 1027.
[15] WANG Liu-san, HUANG Zi-liang, WANG Ru-jing(汪六三，黄子良，王儒敬). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2021, 52(6): 361.
[16] Zhang L, Rao Z H, Ji H Y, et al. Sensors, 2019, 19(14): 3147.
[17] Genuer R, Poggi J M, Tuleau-Malot C. Pattern Recognition Letters, 2016, 31(14): 2225.
[18] Yao R, Li J, Hui M, et al. IEEE Access, 2020, 8: 159151.
[19] Zhu Y L, Duan J M, Li Y H, et al. Textile Research Journal, 2022, 92(7-8): 1012.
[20] LIU Shu, JIANG Qi-gang, MA Yue, et al(刘舒，姜琦刚，马玥, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2017, 48(1): 119.
[21] ZHANG Wen-jie, JIAO An-ran, TIAN Jing, et al(张文杰，焦安然，田静，等). Journal of Instrumental Analysis(分析测试学报), 2021, 40(7): 6.
[22] ZHAI Nai-qi, YUN Li-jun, YE Zhi-xia, et al(翟乃琦，云利军，叶志霞, 等). Computer Engineering & Science(计算机工程与科学), 2021, 43(10): 1833.
[23] Li Z, Li F, Zhu L, et al. International Journal of Computational Intelligence Systems, 2020, 13(1): 559.
[24] QI Chao, ZUO Yi, CHEN Zhe-qi, et al(戚超，左毅，陈哲琪, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2021, 52(5): 301.
[25] ZHANG Yao-xin, ZHU Rong-guang, MENG Ling-feng, et al(张垚鑫，朱荣光，孟令峰, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2021, 37(18): 331.
[26] Xie W J, Wei S, Zheng Z H, et al. Food and Bioprocess Technology, 2021, 14(7): 1361.