Detection of cotton waterlogging stress based on hyperspectral images and convolutional neural network

Jing Zhao, Fangjiang Pan, Zhiming Li, Yubin Lan, Liqun Lu, Dongjian Yang, Yuting Wen

Abstract


Waterlogging in the early stage of cotton will reduce the number of bolls and do harm to yield. Early detection of waterlogging will help farmers to adjust cotton management and save the loss. To evaluate the application of deep learning for the detection of early waterlogging, this study applied a convolutional neural network (CNN) to classify different durations of waterlogging stress (0, 2, 4, 6, 8, 10 d) based on hyperspectral images (HSIs) of cotton leaves. An experiment was designed to simulate the situation of cotton under waterlogging stress and collect HSIs of visible and near-infrared (VNIR 450-950 nm) spectra with 126 bands 66 d after cotton sowing (66 DAS). It was found the spectral curve reflectance of waterlogging cotton was higher than that of non-waterlogging cotton. Especially near 550 nm and 750 nm, and the spectral curve increased with durations of waterlogging stress and there were ‘blue shift’ phenomena for the position of the red edge of the spectra. The first principal components of HSIs after band randomly discarding and principal component analysis (PCA) were used to build a dataset. GoogLeNet Inception-v3 (GLNI-v3) and VGG-16 models were selected to detect cotton waterlogging stress with the dataset. The results showed that the average time for a round training for GLNI-v3 was 13.337 s, with a classification accuracy of 96.95% and a loss value of 0.09431. The average time for a round training for VGG-16 was 21.470 s, with a classification accuracy of 97.00% and a loss value of 0.17912. Though these two models had similar classification accuracy and loss value, GLNI-v3 achieved a high accuracy with fewer training iterations. The durations of waterlogging stress of cotton in a short-term can be detected by HSIs of cotton leaves and CNN models are suitable for the classification of HSIs, and this method can provide support for cotton yield estimation and loss assessment after waterlogging.
Keywords: cotton, waterlogging, hyperspectral image, convolutional neural network
DOI: 10.25165/j.ijabe.20211402.6023

Citation: Zhao J, Pan F J, Li Z M, Lan Y B, Lu L Q, Yang D J, et al. Detection of cotton waterlogging stress based on hyperspectral images and convolutional neural network. Int J Agric & Biol Eng, 2021; 14(2): 167–174.

Keywords


cotton, waterlogging, hyperspectral image, convolutional neural network

Full Text:

PDF

References


Hodgson A S, Chan K Y. The effect of short-term waterlogging during furrow irrigation of cotton in a cracking grey clay. Aust J Agric Res, 1982; 33(1): 109–116.

Reicosky D C, Meyer W S, Schaefer N L, Sides R D. Cotton response to short-term waterlogging imposed with a water-table gradient facility. Agricultural Water Management, 1985; 10(2): 127–143.

Hocking P J, Reicosky D C, Meyer W S. Effects of intermittent waterlogging on the mineral nutrition of cotton. Plant Soil, 1987; 101: 211–221.

Bange M P, Milroy S P, Thongbai P. Growth and yield of cotton in response to waterlogging. Field Crop Res, 2004; 88(2-3): 129–142.

Zhang Y J, Chen Y Z, Lu H Q, Kong X Q, Dai J L, Li Z H, et al. Growth, lint yield and changes in physiological attributes of cotton under temporal waterlogging. Field Crop Res, 2016; 194: 83–93.

Milroy P S, Bange P M. Reduction in radiation use efficiency of cotton (Gossypium hirsutum L.) under repeated transient waterlogging in the field. Field Crop Res, 2013; 140: 51–58.

Kuai J, Zhou Z G, Wang Y H, Meng Y L, Chen B L, Zhao W Q. The effects of short-term waterlogging on the lint yield and yield components of cotton with respect to boll position. Europ. J. Agronomy, 2015; 67: 61–74.

Zhang Y J, Song X Z, Yang G Z, Li Z H, Lu H Q, Kong X Q, et al. Physiological and molecular adjustment of cotton to waterlogging at peak-flowering in relation to growth and yield. Field Crop Res, 2015; 179: 164–172.

Milroy P S, Bange P M, Thongbai P. Cotton leaf nutrient concentrations in response to waterlogging under field conditions. Field Crop Res, 2009; 34: 246–255.

Wang H M, Chen Y L, Hu W, Snider J L, Zhou Z G. Short-term soil-waterlogging contributes to cotton cross tolerance to chronic elevated temperature by regulating ROS metabolism in the subtending leaf. Plant Physiol Bioch, 2019; 139: 333–341.

Najeeb U, Bange P M, Atwell B J, Tan D K Y. Understanding of the interactive effect of waterlogging and shade on cotton (Gossypium hirsutum L.) growth and yield. Procedia Environmental Sciences, 2015; 29: 85–86.

Baranowski P, Jedryczka M, Mazurek W, Babula-Skowronska D, Siedliska A, Kaczmarek J. Hyperspectral and thermal imaging of oilseed rape (Brassica napus) response to fungal species of the genus alternaria. Plos One, 2015; 10(3): e0122913.

Xia J A, Cao H X, Yang Y W, Zhang W X, Wan Q, Xu L, et al. Detection of waterlogging stress based on hyperspectral images of oilseed rape leaves (Brassica napus L.). Comput Electron Agr, 2019; 159: 59–68.

Moshou D, Pantazi X E, Kateris D, Gravalos I. Water stress detection based on optical multisensor fusion with a least squares support vector machine classifier. Biosyst Eng, 2014; 117: 15–22.

Mahlein A K, Steiner U, Hillnhütter C, Dehne W H, Oerke E C. Hyperspectral imaging for small-scale analysis of symptoms caused by different sugar beet diseases. Plant Methods, 2012; 8(1): 3.

Gu Q, Sheng L, Zhang T H, Lu Y W, Zhang Z J, Zheng K F, et al. Early detection of tomato spotted wilt virus infection in tobacco using the hyperspectral imaging technique and machine learning algorithms. Comput Electron Agr, 2019; 167: 105066.

Xie C Q, Shao Y N, Li X L, He Y. Detection of early blight and late blight diseases on tomato leaves using hyperspectral imaging. Scientific Reports, 2015; 5: 16564.

Khan I H, Liu H Y, Cheng T, Tian Y C, Cao Q, Zhu Y, et al. Detection of wheat powdery mildew based on hyperspectral reflectance through SPA and PLS-LDA. Int J Precis Agric Aviat, 2020; 3(1): 13–24.

Zhang M Y, Li C Y, Yang F Z. Classification of foreign matter embedded inside cotton lint using short wave infrared (SWIR) hyperspectral transmittance imaging. Comput Electron Agr, 2017; 139: 75–90.

Zhang R Y, Li C Y, Zhang M Y, Rodgers J. Shortwave infrared hyperspectral reflectance imaging for cotton foreign matter classification. Comput Electron Agr, 2016; 127: 260–270.

Prabhakar M, Prasad Y G, Thirupathi M, Sreedevi G, Dharajothi B, Venkateswarlu B. Use of ground based hyperspectral remote sensing for detection of stress in cotton caused by leafhopper (Hemiptera: Cicadellidae). Comput Electron Agr, 2011; 79: 189–198.

Yang C H, Everitt J H, Fernandez C J. Comparison of airborne multispectral and hyperspectral imagery for mapping cotton root rot. Biosyst Eng, 2010; 107(2): 131–139.

Yi Q X, Wang F M, Bao A M, Jiapaer G. Leaf and canopy water content estimation in cotton using hyperspectral indices and radiative transfer models. Int J Appl Earth Obs, 2014; 33: 67–75.

Yi Q X, Bao A M, Wang Q, Zhao J. Estimation of leaf water content in cotton by means of hyperspectral indices. Comput Electron Agr, 2013; 90: 144–151.

Han X, Yu J Y, Lan Y B, Kong F X, Yi L L. Determination of application parameters for cotton defoliants in the Yellow River Basin. Int J Precis Agric Aviat, 2019; 2(1): 1–5.

Schafer R W. What is a Savitzky-Golay filter?.[lecture notes] IEEE Signal Proc Mag, 2011; 28(4): 111–117.

Chen J, Jonsson P, Tamura M, Gu Z H, Matsushita B, Eklundh L. A simple method for reconstructing a high quality NDVI time-series data set based on the Savitzky-Golay filter. Remote Sens Environ, 2004; 91(3-4): 332–344.

Canny J. A computational approach to edge detection. IEEE T Pattern Anal, 1986; 8(6): 679–698.

Liu Z Y, Wu F H, Huang J F. Application of neural networks to discriminate fungal infection levels in rice panicles using hyperspectral reflectance and principal components analysis. Comput Electron Agr, 2010; 72(2): 99–106.

Golzarian M R, Frick R A. Classification of images of wheat, ryegrass and brome grass species at early growth stages using principal component analysis. Plant Methods, 2011; 7(28): 1–11.

Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z. Rethinking the inception architecture for computer vision. arXiv: 1512.00567, 2016.

LeCun Y, Bengio Y, Hinton G. Deep learning. Nature, 2015; 521: 436–444.

Schmidhuber J. Deep learning in neural networks: an overview. Neural Networks, 2015; 61(1): 85–117.

Abadi M, Agarwal A, Barham P, Brevdo E, Zhifeng C, Citro C, et al. TensorFlow: Large-scale machine learning on heterogeneous distributed systems. 2016. Available: https://blog.csdn.net/ynsshzwxhzyx/ article/details/79448631. Accessed on [2020-06-26].

Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arXiv: 1409.1556, 2014.

Akash A, Jinha J, Chang A J, Sungchan O, Murilo M, Juan L. A comparative study of RGB and multispectral sensor-based cotton canopy cover modelling using multi-temporal UAS data. Remote Sensing, 2019; 11(23): 2757. doi : 10.3390/rs11232757.

Pedro H A M, Fabio H R B, Túlio H D M, João V P F F, Larissa P R T, Carlos A S J, et al. Estimating spray application rates in cotton using multispectral vegetation indices obtained using an unmanned aerial vehicle. Crop Protection, 2021; 40: 105407.

Sungchan O, Chang A J, Akash A, Jinha J, Nothabo D, Murilo M, et al. Plant counting of cotton from UAS imagery using deep learning-based object detection framework. Remote Sensing, 2020; 12(18): 2981. doi : 10.3390/rs12182981.

Feng A J, Zhou J F, Vories E, Sudduth K A. Evaluation of cotton emergence using UAV-based imagery and deep learning. Computers and Electronics in Agriculture, 2020; 177: 105711. doi: 10.1016/j.compag.2020.105711

Adão N A, Witenberg S R S, Díbio L B. Cotton pests classification in field-based images using deep residual networks. Computers and Electronics in Agriculture, 2020; 174: 105488. doi: 10.1016/j.compag.2020.105488.




Copyright (c) 2021 International Journal of Agricultural and Biological Engineering

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

2023-2026 Copyright IJABE Editing and Publishing Office