基于改进E-DWT算法和深度学习模型的红小豆锈病诊断方法

doi:10.3964/j.issn.1000-0593(2025)09-2648-10

Abstract
References
Related Citation (6)

Download: PDF (16174 KB)
Export: BibTeX | EndNote (RIS)

Abstract Adzuki bean rust is a common fungal disease that significantly reduces crop yield by infecting leaves and impairing photosynthesis. This paper proposes a novel diagnostic method for adzuki bean rust based on an improved Empirical Mode Decomposition-Wavelet Transform (E-DWT) algorithm and a deep learning model. Using “Baoqinghong” adzuki beans as the experimental material, 960 leaf spectral datasets were collected over 10 days using a handheld visible/near-infrared spectrometer, obtaining reflectance data in the wavelength range of 326~1 075 nm. First, the improved E-DWT algorithm was applied for spectral denoising. This algorithm combines Empirical Mode Decomposition (EMD) and wavelet threshold denoising technology to retain effective signal information while removing noise. The optimal wavelet basis function (sym5) and the number of decomposition layers (4 layers) were determined by comparing the RMSE and SNR indicators. To further reduce redundancy in high-dimensional data, the Successive Projections Algorithm (SPA) was employed to select 12 representative wavelengths from the initial 750 features, resulting in a 98.4% reduction in feature wavelength count. Next, the Gramian Angular Field (GAF) method was employed to convert the one-dimensional wavelength sequence into a two-dimensional spectral image, thereby enhancing correlations between different bands for subsequent model training. The designed deep learning model combines a Convolutional Neural Network (CNN) with a Convolutional Block Attention Module (CBAM). The CBAM module effectively discriminates the weights of different feature wavelengths and time nodes through channel and spatial attention mechanisms, enabling the model to focus on key features influencing adzuki bean rust identification. Experimental results show that the CBAM-CNN model achieved 99.31% accuracy in the training set, 98.33% accuracy in the test set, and 98.89% recall, significantly outperforming traditional CNN models. Compared to existing methods, the proposed model exhibits superior performance in terms of recognition accuracy, stability, and training convergence speed. Additionally, the model structure is more concise, which optimizes parameter adjustment and improves operability in practical applications. In conclusion, the proposed diagnostic method based on the improved E-DWT algorithm and CBAM-CNN model not only achieves efficient and precise disease detection but also provides a theoretical foundation and technical support for constructing data-driven crop disease diagnosis systems in the future.

Key words：Adzuki bean rust; Spectral data processing; E-DWT algorithm; Deep learning model; Diagnostic model

Received: 2024-12-13 Accepted: 2025-04-19

ZTFLH:

TP391.4

Corresponding Authors: GUAN Hai-ou E-mail: gho123@163.com

Cite this article:

FU Qiang,GUAN Hai-ou,LI Jia-qi. A Diagnostic Method for Adzuki Bean Rust Based on an Improved E-DWT Algorithm and Deep Learning Model[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(09): 2648-2657.

URL:

https://www.gpxygpfx.com/EN/10.3964/j.issn.1000-0593(2025)09-2648-10 OR https://www.gpxygpfx.com/EN/Y2025/V45/I09/2648

[1] GONG Run-hua, YANG Xin-ting, GUO Xiao-hui, et al（龚润华，杨信廷，郭晓晖，等）. Journal of the Chinese Cereals and Oils Association（中国粮油学报），2024, 39(10): 190.
[2] MA Xiao-xuan, HAN Dong, KE Xi-wang, et al（麻晓萱，韩冬，柯希望，等）. China Plant Protection（中国植保导刊），2020, 40(9): 47.
[3] ZHI Ye, DUAN Can-xing, SUN Su-li, et al（支叶，段灿星，孙素丽，等）. Acta Phytopathologica Sinica（植物病理学报），2014, 44(6): 581.
[4] XIAO Ke-yu, JIA Rui-min, YANG Li-sa, et al（肖珂雨，贾瑞敏，杨利飒，等）. Journal of Plant Protection（植物保护学报），2024, 51(2): 314.
[5] Meng R, Lv Z, Yan J, et al. Remote Sensing，2020, 12(19): 3233.
[6] Barthel D, Cullinan C, Mejia-Aguilar A, et al. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy，2023, 303: 123246.
[7] SHEN Meng-jiao, BAO Hao, ZHANG Yan（沈梦姣，鲍浩，张艳）. Progress in Biochemistry and Biophysics（生物化学与生物物理进展）, 2025, 52(1): 233.
[8] Arun R A, Umamaheswari S. Journal of Intelligent & Fuzzy Systems, 2022, 43: 5185.
[9] Ghazi M M, Yanikoglu B, Aptoula E. Neurocomputing，2017, 235: 228.
[10] Wei B, Ma X, Guan H, et al. Ecological Informatics，2023, 75: 102070.
[11] LIU Jian-feng, LIU Meng-qi, DONG Qian-wen, et al（刘建锋，刘梦琪，董倩雯，等）. Journal of Electric Power Science and Technology（电力科学与技术学报），2023, 38(6): 55.
[12] WANG Yuan-zhe, LIANG Teng-fei, ZENG Yu-qiao, et al（王元喆，梁腾飞，曾宇乔，等）. Information and Control（信息与控制），2024, 53(3): 287.
[13] GUO Lin-ge, YIN Yong, YU Hui-chun, et al（郭林鸽，殷勇，于慧春，等）. Food Science（食品科学），2024, 45(7): 164.
[14] ZHAO Yang, GENG Li-min, HU Xun-quan, et al（赵扬，耿莉敏，胡循泉，等）. Journal of Xi'an Jiaotong University（西安交通大学学报），2024, 58(11): 27.
[15] SHENG Yang, JIAO Jun, TENG Yan, et al（圣阳，焦俊，滕燕，等）. Journal of Analytical Science（分析科学学报），2022, 38(5): 552.
[16] CHEN Qi, PAN Tian-hong, LI Yu-qiang, et al（陈琦，潘天红，李鱼强，等）. Spectroscopy and Spectral Analysis（光谱学与光谱分析），2021, 41(9): 2776.