Classification of Citrus Diseases with Convolutional Neural Network Based Deep Learning Model

Year 2022, , 352 - 368, 24.03.2022

Halit Çetiner , İbrahim Çetiner

https://doi.org/10.17798/bitlisfen.1028323

Abstract

Plant diseases are vital for crop yield in agricultural production. It is difficult and tiring to detect diseases in plants at an early stage due to the similarity of features such as colour, shape and texture in plants. Detecting diseases in plants at an early stage and taking precautions is a necessary step to prevent damage to the crop. For this reason, a deep learning-based model has been developed within the scope of the study to classify leaf diseases that affect citrus imports and cause great financial losses to producers. In addition, leaf diseases were classified with three different models based on DenseNet121, MobileNetV2 and ResNet50 architectural models. Fine-tuned transfer learning technique was used in the creation of these models. With the 15-layer CNN model proposed within the scope of the study, 99% accuracy rates were achieved with the Adamax optimization method and 97% with the RMSProp optimization method. Blackspot (citrius black spot (CBS)), canker (citrius bacterial cancer (CBC)), greening (huanglongbing (HLB)) and (healthy), which are the most common citrus leaf diseases, are 100%, 100%, 98%, respectively, in Health classes. and 100% success rates have been reached.

Keywords

CNN, Citrus leaf disease, Classification, Transfer learning

Konvolüsyonel Sinir Ağı Tabanlı Derin Öğrenme Modeli ile Narenciye Hastalıklarının Sınıflandırılması

Abstract

Tarımsal üretimde mahsul verimi için bitki hastalıkları hayati öneme sahiptir. Bitkilerde bulunan renk, şekil, doku gibi özelliklerin birbirine benzemesinden dolayı bitkilerdeki hastalıkların erken aşamada tespiti zor ve yorucu olmaktadır. Bitkilerdeki hastalıkların erken aşamada tespit edilerek önlem alınması mahsule gelen zararın engellemesi için gerekli bir adımdır. Bu nedenle, narenciye ithalatını etkileyerek üreticileri maddi olarak büyük zararlara uğratan yaprak hastalıklarını sınıflandırmak için yapılan çalışma kapsamında derin öğrenme tabanlı bir model geliştirilmiştir. Buna ek olarak DenseNet121, MobileNetV2 ve ResNet50 mimari modellerini temel alan üç ayrı model ile de yaprak hastalıkları sınıflandırılmıştır. Bu modellerin oluşturulmasında ince ayarlı transfer öğrenme tekniği kullanılmıştır. Yapılan çalışma kapsamında önerilen 15 katmanlı CNN modeli ile Adamax optimizasyon yöntemi ile %99, RMSProp optimizasyon yöntemi ile de %97 doğruluk oranlarına ulaşılmıştır. En sık karşılaşılan narenciye yaprak hastalıklarından olan Blackspot (citrius siyah nokta (CBS)), canker (citrius bakteriyel kanseri (CBC)), greening (huanglongbing (HLB)) ile (sağlıklı) Health sınıflarında ise sırasıyla %100, %100, %98 ve %100 başarı oranlarına erişilmiştir.

Keywords

CNN, Narenciye yaprak hastalığı, Sınıflandırma, Transfer öğrenme

References

• [1] E. C. Too, L. Yujian, S. Njuki, and L. Yingchun, 2019. A comparative study of fine-tuning deep learning models for plant disease identification, Comput. Electron. Agric., vol. 161, pp. 272–279, doi: https://doi.org/10.1016/j.compag.2018.03.032.
• [2] S. Zhang, W. Huang, and C. Zhang, 2019. Three-channel convolutional neural networks for vegetable leaf disease recognition, Cogn. Syst. Res., vol. 53, pp. 31–41, doi: https://doi.org/10.1016/j.cogsys.2018.04.006.
• [3] G. Shrivastava, 2021. Review on Emerging Trends in Detection of Plant Diseases using Image Processing with Machine Learning, Int. J. Comput. Appl., vol. 174, doi: 10.5120/ijca2021920990.
• [4] A. Gargade and S. A. Khandekar, 2019. A Review: Custard Apple Leaf Parameter Analysis and Leaf Disease Detection using Digital Image Processing, in 2019 3rd International Conference on Computing Methodologies and Communication (ICCMC), pp. 267–271, doi: 10.1109/ICCMC.2019.8819867.
• [5] M. Bin Tahir et al., 2021. Recognition of Apple Leaf Diseases using Deep Learning and Variances-Controlled Features Reduction, Microprocess. Microsyst., p. 104027, doi: https://doi.org/10.1016/j.micpro.2021.104027.
• [6] P. Bansal, R. Kumar, and S. Kumar, 2021. Disease Detection in Apple Leaves Using Deep Convolutional Neural Network, Agriculture , vol. 11, no. 7, doi: 10.3390/agriculture11070617.
• [7] D. Tiwari, M. Ashish, N. Gangwar, A. Sharma, S. Patel, and S. Bhardwaj, 2020. Potato Leaf Diseases Detection Using Deep Learning.
• [8] R. Sujatha, J. M. Chatterjee, N. Z. Jhanjhi, and S. N. Brohi, 2021. Performance of deep learning vs machine learning in plant leaf disease detection, Microprocess. Microsyst., vol. 80, p. 103615, doi: https://doi.org/10.1016/j.micpro.2020.103615.
• [9] O. Russakovsky et al., 2015. ImageNet Large Scale Visual Recognition Challenge, Int. J. Comput. Vis., vol. 115, no. 3, pp. 211–252, doi: 10.1007/s11263-015-0816-y.
• [10] W. Xue, X. Hu, Z. Wei, X. Mei, X. Chen, and Y. Xu, 2019. A fast and easy method for predicting agricultural waste compost maturity by image-based deep learning, Bioresour. Technol., vol. 290, p. 121761, doi: https://doi.org/10.1016/j.biortech.2019.121761.
• [11] Z. Zhang, H. Liu, Z. Meng, and J. Chen, 2019. Deep learning-based automatic recognition network of agricultural machinery images, Comput. Electron. Agric., vol. 166, p. 104978, doi: https://doi.org/10.1016/j.compag.2019.104978.
• [12] N. Kussul, M. Lavreniuk, S. Skakun, and A. Shelestov, 2017. Deep learning classification of land cover and crop types using remote sensing data, IEEE Geosci. Remote Sens. Lett., vol. 14, no. 5, pp. 778–782.
• [13] A. Z. da Costa, H. E. H. Figueroa, and J. A. Fracarolli, 2020. Computer vision based detection of external defects on tomatoes using deep learning, Biosyst. Eng., vol. 190, pp. 131–144, doi: https://doi.org/10.1016/j.biosystemseng.2019.12.003.
• [14] A. Molotoks, P. Smith, and T. P. Dawson, 2021. Impacts of land use, population, and climate change on global food security, Food Energy Secur., vol. 10, no. 1, p. e261, doi: https://doi.org/10.1002/fes3.261.
• [15] M. van Dijk, T. Morley, M. L. Rau, and Y. Saghai, 2021. A meta-analysis of projected global food demand and population at risk of hunger for the period 2010–2050, Nat. Food, vol. 2, no. 7, pp. 494–501, doi: 10.1038/s43016-021-00322-9.
• [16] M. Dutt, C. El Mohtar, and N. Wang, 2020. Biotechnological Approaches for the Resistance to Citrus Diseases, pp. 245–257.
• [17] N. T. Tran et al., 2017. Sexual Reproduction in the Citrus Black Spot Pathogen, Phyllosticta citricarpa, Phytopathology®, vol. 107, no. 6, pp. 732–739, doi: 10.1094/PHYTO-11-16-0419-R.
• [18] E. National Academies of Sciences and Medicine, 2018. A Review of the Citrus Greening Research and Development Efforts Supported by the Citrus Research and Development Foundation. Washington, D.C.: National Academies Press.
• [19] S. F. Syed-Ab-Rahman, M. H. Hesamian, and M. Prasad, 2021. Citrus disease detection and classification using end-to-end anchor-based deep learning model, Appl. Intell., doi: 10.1007/s10489-021-02452-w.
• [20] P. M. M. Martins, M. de Oliveira Andrade, C. E. Benedetti, and A. A. de Souza, 2020. Xanthomonas citri subsp. citri: host interaction and control strategies, Trop. Plant Pathol., vol. 45, no. 3, pp. 213–236, doi: 10.1007/s40858-020-00376-3.
• [21] S. A. de Carvalho et al., 2014. Comparison of Resistance to Asiatic Citrus Canker Among Different Genotypes of Citrus in a Long-Term Canker-Resistance Field Screening Experiment in Brazil, Plant Dis., vol. 99, no. 2, pp. 207–218, doi: 10.1094/PDIS-04-14-0384-RE.
• [22] J. Martínez-Minaya, D. Conesa, A. López-Quílez, and A. Vicent, Climatic distribution of citrus black spot caused by Phyllosticta citricarpa. A historical analysis of disease spread in South Africa, Eur. J. Plant Pathol., vol. 143, no. 1, pp. 69–83, doi: 10.1007/s10658-015-0666-z.
• [23] G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, 2017. Densely connected convolutional networks, in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 4700–4708.
• [24] K. He, X. Zhang, S. Ren, and J. Sun, 2016. Identity mappings in deep residual networks,in European conference on computer vision, pp. 630–645.
• [25] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. B. Wojna, 2016. Rethinking the Inception Architecture for Computer Vision.
• [26] C. Szegedy, S. Ioffe, V. Vanhoucke, and A. A. Alemi, 2017. Inception-v4, inception-resnet and the impact of residual connections on learning.
• [27] M. Khanramaki, E. Askari Asli-Ardeh, and E. Kozegar, 2021. Citrus pests classification using an ensemble of deep learning models, Comput. Electron. Agric., vol. 186, p. 106192, doi: https://doi.org/10.1016/j.compag.2021.106192.
• [28] S. Sun, Z. Cao, H. Zhu, and J. Zhao, 2019. A survey of optimization methods from a machine learning perspective, IEEE Trans. Cybern., vol. 50, no. 8, pp. 3668–3681.
• [29] S. Ioffe and C. Szegedy, 2015. Batch normalization: Accelerating deep network training by reducing internal covariate shift,” in International conference on machine learning, pp. 448–456.
• [30] B. Liu, Y. Zhang, D. He, and Y. Li, 2018. Identification of Apple Leaf Diseases Based on Deep Convolutional Neural Networks, Symmetry (Basel)., vol. 10, no. 1, p. 11, doi: http://dx.doi.org/10.3390/sym10010011.
• [31] M. Turkoglu, D. Hanbay, and A. Sengur, 2019. Multi-model LSTM-based convolutional neural networks for detection of apple diseases and pests, J. Ambient Intell. Humaniz. Comput., doi: 10.1007/s12652-019-01591-w.
• [32] M. Khan, T. Akram, M. Sharif, K. Javed, M. Raza, and T. Saba, 2020. An Automated System for Cucumber Leaf Diseased Spot Detection and Classification using Improved Saliency Method and Deep Features Selection, Multimed. Tools Appl., vol. 79, doi: 10.1007/s11042-020-08726-8.
• [33] S. Rajora, D. kumar Vishwakarma, K. Singh, and M. Prasad, 2018. CSgI: A Deep Learning based approach for Marijuana Leaves Strain Classification, in 2018 IEEE 9th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON), pp. 209–214, doi: 10.1109/IEMCON.2018.8615011.
• [34] Z. Iqbal, M. A. Khan, M. Sharif, J. H. Shah, M. H. ur Rehman, and K. Javed, 2018. An automated detection and classification of citrus plant diseases using image processing techniques: A review, Comput. Electron. Agric., vol. 153, pp. 12–32, doi: https://doi.org/10.1016/j.compag.2018.07.032.
• [35] H. T. Rauf, B. A. Saleem, M. I. U. Lali, M. A. Khan, M. Sharif, and S. A. C. Bukhari, 2019. A citrus fruits and leaves dataset for detection and classification of citrus diseases through machine learning, Data Br., vol. 26, p. 104340, doi: https://doi.org/10.1016/j.dib.2019.104340.
• [36] S. Xing, M. Lee, and K. Lee, 2019. Citrus Pests and Diseases Recognition Model Using Weakly Dense Connected Convolution Network, Sensors , vol. 19, no. 14, doi: 10.3390/s19143195.
• [37] Q. Chen et al., 2019. Deep Convolutional Network for Citrus Leaf Diseases Recognition.
• [38] M. Sharif, M. A. Khan, Z. Iqbal, M. F. Azam, M. I. U. Lali, and M. Y. Javed, 2018. Detection and classification of citrus diseases in agriculture based on optimized weighted segmentation and feature selection, Comput. Electron. Agric., vol. 150, pp. 220–234, doi: https://doi.org/10.1016/j.compag.2018.04.023.
• [39] C. B. Wetterich, R. Felipe de Oliveira Neves, J. Belasque, and L. G. Marcassa, 2016. Detection of citrus canker and Huanglongbing using fluorescence imaging spectroscopy and support vector machine technique, Appl. Opt., vol. 55, no. 2, pp. 400–407, doi: 10.1364/AO.55.000400.
• [40] Kaggle, 2020. Citrus Leaves Prepared, https://www.kaggle.com/dtrilsbeek/citrus-leaves-prepared.
• [41] V. Chouhan et al., 2020. A Novel Transfer Learning Based Approach for Pneumonia Detection in Chest X-ray Images, Applied Sciences , vol. 10, no. 2., doi: 10.3390/app10020559.
• [42] A. G. Howard et al., 2017. Mobilenets: Efficient convolutional neural networks for mobile vision applications, arXiv Prepr. arXiv1704.04861.
• [43] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, 2018. Mobilenetv2: Inverted residuals and linear bottlenecks, in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 4510–4520.
• [44] S. Rajpal, N. Lakhyani, A. K. Singh, R. Kohli, and N. Kumar, 2021. Using handpicked features in conjunction with ResNet-50 for improved detection of COVID-19 from chest X-ray images, Chaos, Solitons & Fractals, vol. 145, p. 110749, doi: 10.1016/j.chaos.2021.110749.
• [45] S. Vallabhajosyula, V. Sistla, and V. K. K. Kolli, 2021. Transfer learning-based deep ensemble neural network for plant leaf disease detection, J. Plant Dis. Prot., doi: 10.1007/s41348-021-00465-8.
• [46] D. M. Bulanon, T. F. Burks, D. G. Kim, and M. A. Ritenour, 2013. Citrus black spot detection using hyperspectral image analysis, Agric. Eng. Int. CIGR J., vol. 15, no. 3, pp. 171–180.
• [47] D. Kim, T. F. Burks, M. A. Ritenour, and J. Qin, 2014. Citrus black spot detection using hyperspectral imaging, Int. J. Agric. Biol. Eng., vol. 7, no. 6, pp. 20–27, doi: 10.3965/j.ijabe.20140706.004.
• [48] M. Zhang and Q. Meng, 2011. Automatic citrus canker detection from leaf images captured in field, Pattern Recognit. Lett., vol. 32, no. 15, pp. 2036–2046, doi: https://doi.org/10.1016/j.patrec.2011.08.003.
• [49] D. Xiaoling, Y. Lan, X. Xiaqiong, M. Huilan, L. Jiakai, and H. Tiansheng, 2016. Detection of citrus Huanglongbing based on image feature extraction and two-stage BPNN modeling, Int. J. Agric. Biol. Eng., vol. 9, no. 6, pp. 20–26.
• [50] X. Deng et al., 2020. Detection of Citrus Huanglongbing Based on Multi-Input Neural Network Model of UAV Hyperspectral Remote Sensing, Remote Sensing , vol. 12, no. 17., doi: 10.3390/rs12172678.