Original article | Journal of Biometry Studies 2022, Vol. 2(2) 48-56
Automated identification of plant disease using deep learning
Kazi Riad UDDIN, Md Habib KHAN, Mohammad Mizanur RAHMAN, Arafath Al FAHIM
pp. 48 - 56 | DOI: https://doi.org/10.29329/JofBS.2022.445.02 | Manu. Number: MANU-2209-19-0002.R2
Published online: December 29, 2022 | Number of Views: 48 | Number of Download: 278
Abstract
The preliminary identification of plant diseases plays a predominant role in preventing loss of production. The laboratory identification process of plant diseases is time-consuming and could not be conducted in the countryside, where experiment facilities are rarely found. This paper shows a deep learning approach to confine the infection area and identify the diseases by using images of their leaves. Deep learning works well with large amounts of data. So we can increase the accuracy and reduce the loss by engrossing a plethora of data. However, it will not increase the efficiency of the models. In this paper, we use several cutting-edge deep learning models, such as MobileNet, ResNet, and EfficientNet, along with Faster R-CNN and SSD on a small dataset. The dataset contains 2366 images of 27 types. The dataset was taken in a real environment. The data augmentation technique cannot be used with a small dataset. All state-of-the-art deep learning model are trained as a baseline to work on the efficiency of the models. We experiment with the best performer for computation cost. So, to increase the efficiency of the model we implement cyclic learning rate which performs 53.81% map@.50 on best performer EfficientDet. It also lessens the variance, which suggests that cyclic learning not only works as a learning rate but also functions as a data augmentation. In the future, we will apply this learning rate to a dataset containing a large number of plant disease collection images, where different types of data augmentation can be used to not only increase the images but also decrease the generalization loss. Farmers can predict plant diseases more accurately using this system.
Keywords: PlantDoc, Deep learning, Convolution neural network, Cyclic learning rate
| How to Cite this Article? |
|---|
|
APA 6th edition Harvard Chicago 16th edition |
| References |
|---|
|
Amara, J., Bouaziz, B., & Algergawy, A. (2017). A Deep Learning-based Approach for Banana Leaf Diseases Classification. Lecture Notes in Informatics, 266, 79-88. The Global Economy (2019). Bangladesh: Employment in agriculture. https://www.theglobaleconomy.com/Bangladesh/Employment_in_agriculture/ The Global Economy (2021). Bangladesh: GDP share of agriculture. https://www.theglobaleconomy.com/Bangladesh/Share_of_agriculture/ Cañizares, M. C., Rosas-Díaz, T., Rodríguez-Negrete, E., Hogenhout, S. A., Bedford, I. D., Bejarano, E. R., Navas-Castillo, J., & Moriones, E. (2015). Arabidopsis thaliana, An Experimental Host for Tomato Yellow Leaf Curl Disease-associated Begomoviruses by Agroinoculation and Whitefly Transmission. Plant Pathology, 64(2), 265-271. https://doi.org/10.1111/ppa.12270 He, K., Zhang, X., Ren, S., & Sun, J. (2015). Deep residual learning for image recognition. http://image-net.org/challenges/LSVRC/2015/ Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., & Andreetto, M. (2017). MobileNets: Efficient convolutional neural networks for mobile vision applications. https://arxiv.org/abs/1704.04861 Jiang, P., Chen, Y., Liu, B., He, D., & Liang, C. (2019). Real-Time Detection of Apple Leaf Diseases Using Deep Learning Approach Based on Improved Convolutional Neural Networks. IEEE Access, 7, 59069-59080. https://doi.org/10.1109/ACCESS.2019.2914929 Lu, J., Hu, J., Zhao, G., Mei, F., & Zhang, C. (2017). An In-field Automatic Wheat Disease Diagnosis System. Computers and Electronics in Agriculture, 142(1), 369-379. https://doi.org/10.1016/j.compag.2017.09.012 Ramcharan, A., Baranowski, K., McCloskey, P., Ahmed, B., Legg, J., & Hughes, D. P. (2017). Deep Learning for Image-based Cassava Disease Detection. Frontiers in Plant Science, 8, 1852. https://doi.org/10.3389/fpls.2017.01852 Rangarajan, A. K., Purushothaman, R., & Ramesh, A. (2018). Tomato Crop Disease Classification Using Pre-trained Deep Learning Algorithm. Procedia Computer Science, 133, 1040-1047. https://doi.org/10.1016/j.procs.2018.07.070 Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., Berg, A. C., & Fei-Fei, L. (2015). ImageNet Large Scale Visual Recognition Challenge. International Journal of Computer Vision, 115(3), 211-252. https://doi.org/10.1007/S11263-015-0816-Y Sibiya, M., & Sumbwanyambe, M. (2019). A Computational Procedure for the Recognition and Classification of Maize Leaf Diseases Out of Healthy Leaves Using Convolutional Neural Networks. AgriEngineering, 1(1), 119-131. https://doi.org/10.3390/agriengineering1010009 Singh, D., Jain, N., Jain, P., Kayal, P., Kumawat, S., & Batra, N. (2020). PlantDoc: A dataset for visual plant disease detection [Oral presentation]. International Conference on Data Science and Management of Data, Hyderabad, India. Smith, L. N. (2015). Cyclical learning rates for training neural networks [Oral presentation]. IEEE Winter Conference on Applications of Computer Vision, Santa Rosa, USA. Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking model scaling for convolutional neural networks [Oral presentation]. 36th International Conference on Machine Learning, Long Beach, USA. Tan, M., Pang, R., & Le, Q. V. (2020). EfficientDet: Scalable and efficient object detection [Oral presentation]. IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, USA. Too, E. C., Yujian, L., Njuki, S., & Yingchun, L. (2019). A Comparative Study of Fine-tuning Deep Learning Models for Plant Disease Identification. Computers and Electronics in Agriculture, 161, 272-279. https://doi.org/10.1016/j.compag.2018.03.032 Türkoğlu, M., & Hanbay, D. (2019). Plant Disease and Pest Detection Using Deep Learning-based Features. Turkish Journal of Electrical Engineering and Computer Sciences, 27(3), 1636-1651. https://doi.org/10.3906/elk-1809-181 |
