Improving Rice Leaf Disease Identification with Object Detection and Image Enhancement
Loading...
Files
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Department of Computer Science and Engineering(CSE), Islamic University of Technology(IUT), Board Bazar, Gazipur-1704, Bangladesh
Abstract
Highly accurate rice leaf disease identification using lighter models can ensure food
security globally and also less crop loss by maximizing the rate of disease identifica tion in proper time. Our thesis focus is on the process where the quality of the image
is enhanced and the diseased part is detected for predicting the disease in the output.
Image enhancement refers to the process where noisy, low-contrast, and low-quality
image is taken care of. The object detection mechanism detects which part of the im age contains disease. The identification process specifies the disease from the given
input. The currently available disease detection model seems to have limitations, such
as - high-quality training and testing images, requiring high computational power,
images with biased backgrounds, etc. Our proposed technique works on these limita tions by including an image enhancement technique, Lite SR-GAN, that gives a better
result even if we train our models on low-quality images. To make the computation
less complex we have used lightweight architectures in the different phases (image
enhancement, detection, disease identification) of the model. Next, to address the is sue of biased images, we have used an object detection technique, YOLO-s, to detect
the diseased part of the image. In the final stage, we introduce a lightweight model,
EfficientViT that, with the help of previous layers, works equally fine with pictures of
all qualities and identifies the disease using less computation power with an accuracy
of 97.615%, inference time 33.99ms per image, parameter count 4.56M, model size
6.95MB and flop count 33.56B.
Description
Supervised by
Mr. Tareque Mohmud Chowdhury,
Assistant Professor,
Mr. Njayou Youssouf,
Lecturer,
Ms. Sabrina Islam,
Lecturer,
Department of Computer Science and Engineering (CSE)
Islamic University of Technology (IUT)
Board Bazar, Gazipur, Bangladesh
This thesis is submitted in partial fulfillment of the requirement for the degree of Bachelor of Science in Computer Science and Engineering, 2024
Keywords
Disease_identification, lightweight_model, Lite_SRGAN, Yolov8n, EfficientViT
Citation
[1] G. Singh, A. Mittal, et al., “Various image enhancement techniques-a critical re view,” International Journal of Innovation and Scientific Research, vol. 10, no. 2, pp. 267–274, 2014. [2] D. Xiangyu, S. Wenhao, Z. Wenwei, et al., “A comprehensive overview of image enhancement techniques,” Archives of Computational Methods in Engineering, 2022. [Online]. Available: https://doi.org/10.1007/s11831-021-09587-6. [3] Z. Zou, K. Chen, Z. Shi, Y. Guo, and J. Ye, “Object detection in 20 years: A sur vey,” Proceedings of the IEEE, vol. 111, no. 3, pp. 257–276, 2023. [4] R. Szeliski,Computer vision: algorithms and applications. Springer Nature, 2022. [5] L. Oz, “17 interesting applications of object detection for businesses,” alwaysAI, 2022. [Online]. Available: https://alwaysai.co/blog/object-detection for-businesses. [6] A. Voulodimos, N. Doulamis, A. Doulamis, and E. Protopapadakis, “Deep learn ing for computer vision: A brief review,” Computational intelligence and neuro science, vol. 2018, no. 1, p. 7 068 349, 2018. [7] K. P. Ferentinos, “Deep learning models for plant disease detection and diag nosis,” Computers and electronics in agriculture, vol. 145, pp. 311–318, 2018. [8] B. R. R. Institute. “Rice in bangladesh.” (), [Online]. Available: https://www. knowledgebank-brri.org/riceinban.php. [9] C. Muppala and V. Guruviah, “Machine vision detection of pests, diseases, and weeds: A review,” Journal of Phytology, pp. 9–19, Apr. 2020. doi: 10.25081/jp. 2020.v12.6145. [10] C. R. Rahman, P. S. Arko, M. E. Ali, M. A. I. Khan, S. H. Apon, and F. Nowrin, “Identification and recognition of rice diseases and pests using convolutional neural networks,” Biosystems Engineering, vol. 194, no. 3, pp. 112–120, 2020. 70 [11] J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, “Imagenet: A large scale hierarchical image database,” in 2009 IEEE Conference on Computer Vi sion and Pattern Recognition, 2009, pp. 248–255. doi: 10 . 1109 / CVPR . 2009 . 5206848. [12] R. R. Atole and D. Park, “A multiclass deep convolutional neural network clas sifier for detection of common rice plant anomalies,” International Journal of Advanced Computer Science and Applications, vol. 9, no. 1, 2018. [13] GeeksforGeeks. “Vgg-16 cnn model.” Accessed: 2024-07-03. (2023), [Online]. Available: https://www.geeksforgeeks.org/vgg-16-cnn-model/. [14] A. Obinguar. “Simple implementation of inceptionv3 for image classification using tensorflow and keras.” Accessed: 2024-07-03. (2020), [Online]. Available: https : / / medium . com / @armielynobinguar / simple - implementation - of- inceptionv3- for- image- classification- using- tensorflow- and keras-6557feb9bf53. [15] “Squeezenet.” Accessed: 2024-07-03. (2023), [Online]. Available: https : / / paperswithcode.com/method/squeezenet. [16] B. R. R. Institute. “Brri’s online available dataset.” (), [Online]. Available: https: / / drive . google . com / drive / folders / 1ewBesJcguriVTX8sRJseCDbXAF _ T4akK. [17] M. P. Babu, B. S. Rao, et al., “Leaves recognition using back propagation neural network-advice for pest and disease control on crops,” IndiaKisan. Net: Expert Advisory System, pp. 607–626, 2007. [18] M. Jiang, C. Feng, X. Fang, Q. Huang, C. Zhang, and X. Shi, “Rice disease iden tification method based on attention mechanism and deep dense network,” Electronics, 2023. [Online]. Available: https://api.semanticscholar.org/ CorpusID:256148712. [19] J. Hu, L. Shen, S. Albanie, G. Sun, and E. Wu, “Squeeze-and-excitation net works,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 42, no. 8, pp. 2011–2023, 2020. doi: 10.1109/TPAMI.2019.2913372. [20] J. Pan, T. Wang, and Q. Wu, “Ricenet: A two stage machine learning method for rice disease identification,” Biosystems Engineering, vol. 225, pp. 25–40, Jan. 2023. doi: 10.1016/j.biosystemseng.2022.11.007. [21] D. Argüeso, A. Picon, U. Irusta, et al., “Few-shot learning approach for plant disease classification using images taken in the field,” Computers and Electron ics in Agriculture, vol. 175, p. 105 542, 2020, issn: 0168-1699. doi: https : / / 71 doi . org / 10 . 1016 / j . compag . 2020 . 105542. [Online]. Available: https : //www.sciencedirect.com/science/article/pii/S0168169920302544. [22] H. Sherif, H. El-Saadawy, M. Al-berry, and M. Tolba, “Lite-srgan and lite-unet: Towards fast and accurate image super-resolution, segmentation, and localiza tion for plant leaf diseases,” IEEE Access, vol. PP, Jun. 2023. doi: 10 . 1109 / access.2023.3289750. [23] A. Alidev. “Plantvillage dataset.” (), [Online]. Available: https://www.kaggle. com/datasets/abdallahalidev/plantvillage-dataset. [24] M. M. Hasan, T. Rahman, A. F. M. S. Uddin, et al., “Enhancing rice crop man agement: Disease classification using convolutional neural networks and mo bile application integration,” Agriculture, vol. 13, no. 8, 2023, issn: 2077-0472. doi: 10 . 3390 / agriculture13081549. [Online]. Available: https : / / www . mdpi.com/2077-0472/13/8/1549. [25] G. Developers. “Softmax.” Accessed: 2024-07-03. (2023), [Online]. Available: https : / / developers . google . com / machine - learning / crash - course / multi-class-neural-networks/softmax. [26] M. S. I. Sobuj, M. I. Hossen, and M. F. Mahmud, “Leveraging pre-trained cnns for efficient feature extraction in rice leaf disease classification,” Journal of Agri cultural Informatics, vol. 8, no. 3, pp. 1–18, 2021. [27] B. In. “Histogram of oriented gradients (hog).” Accessed: 2024-07-03. (2023), [Online]. Available: https : / / builtin . com / articles / histogram - of - oriented-gradients. [28] N. Takano and G. Alaghband, Srgan: Training dataset matters, 2019. arXiv: 1903. 09922 [cs.CV]. [Online]. Available: https://arxiv.org/abs/1903.09922. [29] Z. Ge, S. Liu, F. Wang, Z. Li, and J. Sun, “Yolox: Exceeding yolo series in 2021,” arXiv preprint arXiv:2107.08430, 2021. [30] L. Hu and Y. Li, “Micro-yolo: Exploring efficient methods to compress cnn based object detection model.,” in ICAART (2), 2021, pp. 151–158. [31] N. Ketkar, J. Moolayil, N. Ketkar, and J. Moolayil, “Convolutional neural net works,” Deep learning with Python: learn best practices of deep learning models with PyTorch, pp. 197–242, 2021. [32] B. Koonce and B. Koonce, “Mobilenetv3,” Convolutional Neural Networks with Swift for Tensorflow: Image Recognition and Dataset Categorization, pp. 125–144, 2021. 72 [33] K. Dong, C. Zhou, Y. Ruan, and Y. Li, “Mobilenetv2 model for image classifi cation,” in 2020 2nd International Conference on Information Technology and Computer Application (ITCA), IEEE, 2020, pp. 476–480. [34] A. Vedaldi and A. Zisserman, “Vgg convolutional neural networks practical,” Department of Engineering Science, University of Oxford, vol. 66, 2016. [35] B. Koonce and B. Koonce, “Efficientnet,” Convolutional neural networks with swift for Tensorflow: image recognition and dataset categorization, pp. 109–123, 2021. [36] B. Lyu, L.-Y. Shen, and C.-M. Yuan, “Mixnet: Mix different networks for learn ing 3d implicit representations,” Graphical Models, vol. 129, p. 101 190, 2023. [37] F. Chollet, “Xception: Deep learning with depthwise separable convolutions,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 1251–1258. [38] C. Szegedy, W. Liu, Y. Jia, et al., “Going deeper with convolutions,” in Proceed ings of the IEEE conference on computer vision and pattern recognition, 2015, pp. 1–9. [39] “Paddy doctor-paddy disease classification.” (), [Online]. Available: https:// www.kaggle.com/c/paddy-disease-classification/data. [40] “Dhan-shomadhan: A dataset of rice leaf disease classification for bangladeshi local rice.” (), [Online]. Available: https://data.mendeley.com/datasets/ znsxdctwtt/1. [41] “Rice leaf diseases dataset.” (), [Online]. Available: https : / / www . kaggle . com/datasets/vbookshelf/rice-leaf-diseases/data. [42] X. Liu, H. Peng, N. Zheng, Y. Yang, H. Hu, and Y. Yuan, “Efficientvit: Memory efficient vision transformer with cascaded group attention,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 14 420–14 430. [43] K. Wu, J. Zhang, H. Peng, et al., “Tinyvit: Fast pretraining distillation for small vision transformers,” in European conference on computer vision, Springer, 2022, pp. 68–85.