Red Soil Organic Matter Content Prediction Model Based on Dilated
Convolutional Neural Network
DENG Yun1, 2, WU Wei1, 2, SHI Yuan-yuan3, CHEN Shou-xue1, 2*
1. Guangxi Key Laboratory of Embedded Technology and Intelligent System, Guilin 541006, China
2. School of Information Science and Engineering, Guilin University of Technology, Guilin 541006, China
3. Guangxi Zhuang Autonomous Region Forestry Research Institute, Nanning 530002, China
Abstract：Soil Organic Matter (SOM) content is one of the important indicators used to measure soil fertility, and it is of great significance in accurately predicting SOM content from hyperspectral remote sensing images. Traditional machine learning methods require complex feature engineering. Still, they are not highly accurate, while deep learning methods represented by Convolutional Neural Networks (CNNs) are less studied in soil hyperspectral, and the modeling accuracy of small sample data is poor. The spatial feature extraction of spectral data is insufficient. This paper proposes a one-dimensional convolutional network model using a channel attention mechanism (SE Dilated Convolutional Neural Network, SE-DCNN). Taking 207 soil samples collected from Guangxi State-owned Huangmian Forest Farm and State-owned Yachang Forest Farm as research objects, this paper compares and analyzes the modeling effects of 3 machine learning and 4 deep learning methods under different spectral preprocessing. The results show that the SE-DCNN model, because of the use of dilated convolution and channel attention mechanism, expands the receptive field, extracts multi-scale features, and has good modeling accuracy and generalization fitting ability. The best prediction model in this paper is the SE-DCNN model established based on the spectral preprocessing method of Savitaky-Golay denoising (SGD) and first-order derivative (DR), the determination coefficient (R2) of the validation set is 0.971, the root mean square error (RMSE) is 2.042 g·kg-1, and the relative analysis error (RPD) is 5.273. Therefore, SE-DCNN can accurately predict the organic matter content of red soil in Guangxi forest land.
[1] WANG Qing-kui, WANG Si-long, FENG Zong-wei, et al(王清奎, 汪思龙, 冯宗炜, 等). Chinese Journal of Applied Ecology(应用生态学报), 2004, (10): 1947.
[2] ZHOU Wen-ting, HUANG Yi-bin, WANG Yi-xiang, et al(周文婷, 黄毅斌, 王义祥, 等). Pratacultural Science(草业科学), 2013, 30(11): 1725.
[3] WU Jian-hu, ZHANG Xiu-li, CONG Xiang-an(吴建虎, 张秀丽, 丛祥安). Gansu Agriculture(甘肃农业), 2006, 20(11): 382.
[4] HE Jun-liang, JIANG Jian-jun, ZHOU Sheng-lu, et al(贺军亮, 蒋建军, 周生路, 等). Scientia Agricultura Sinica(中国农业科学), 2007, 40(3): 638.
[5] BAO Qing-ling, DING Jian-li, WANG Jing-zhe, et al(包青岭, 丁建丽, 王敬哲, 等). Arid Land Geography(干旱区地理), 2019, 42(6): 1404.
[6] Shen L, Gao M, Yan J, et al. Remote Sensing, 2020, 12(7): 1206.
[7] Xu X, Chen S, Xu Z, etal. Remote Sensing, 2020, 12(22): 3765.
[8] SHANG Tian-hao, MAO Hong-xin, ZHANG Jun-hua, et al(尚天浩, 毛鸿欣, 张俊华, 等). Chinese Journal of Ecology(生态学杂志), 2021, 40(12): 4128.
[9] Carvalho J K, Moura-Bueno J M, Ramon R, et al. Geoderma Regional, 2022, 29: e00530.
[10] Hu W, Huang Y, Wei L, et al. Journal of Sensors, 2015, 2015: 258619.
[11] LIU Huan-jun, BAO Yi-lin, MENG Xiang-tian, et al(刘焕军, 鲍依临, 孟祥添, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2020, 36(12): 90.
[12] Donoho D L, Johnstone I M. Biometrika, 1994, 81(3): 425.
[13] Galvao R K H, Araujo M C U, José G E, et al. Talanta, 2005, 67(4): 736.
[14] Yu F, Koltun V. Multi-Scale Context Aggregation by Delated Convolutions, CoRR, 2015, abs/1511.07122.
[15] Khalfaoui-Hassani I, Pellegrini T, Masquelier T. Dilated Covolution with Learnable Spacings, ICLR 2023, arXivpreprintarXiv: 2112. 03740, 2021.
[16] PAN Ning, WANG Shuai, LIU Yan-xu, et al(潘 宁, 王 帅, 刘焱序, 等). Acta Ecologica Sinica(生态学报), 2019, 39(13): 4615.
[17] FAN Rong, MENG Da-zhi, XU Da-shun(樊 嵘，孟大志，徐大舜). Mathematical Modeling and Its Applications(数学建模及应用)，2014，3(1)：12.
[18] ZHANG Sen, LU Xia, NIE Ge-ge, et al(张 森, 卢 霞, 聂格格, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2020, 40(2): 556.
[19] Qiao X, Wang C, Feng M, etal. Spectroscopy Letters, 2017, 50(3): 156.
