A Deep Learning Based RefineNet Model for Chilli Plant Disease Classification
Main Article Content
Abstract
Diseases of the chili leaf are a challenge in chili crop production and quality, hence the reduced quality and quantity. Early detection and control of diseases in the chili leaf are important to ensure the quality and quantity of the produce. However, early detection of diseases in the natural image of the chili leaf is challenging, especially when the leaves have irregularly shaped yellow spots and curls. Current models problem, this paper presents the use of the RefineNet model, which is specifically tailored to ensure effective early detection of diseases in the chili leaves. The proposed method involves the use of the RefineNet model, whose composition includes four blocks: the encoder/decoder block, the residual convolutional block (RCU), the multi-scale fusion block (MRF), and the chained resolution pooling (CRP) block. While the encoder block produces low-resolution maps, the other block produces high-resolution maps. However, the use of residual connections in the RCU block helps reduce vanishing problems, hence facilitating information transfer from different levels. In the MRF block, the feature map is processed at different scale levels. Dilated convolution is used in this block to enhance the receptive field, hence facilitating effective extraction of contextual information. Additionally, the leaky ReLU activation function replaces the standard ReLU activation function, enhancing the feature extraction ability for diseased leaves with irregularly shaped yellow spots and curls. Moreover, the auxiliary classifier is simplified, reducing the model’s complexity without compromising the recognition rate. Experimental results demonstrate that compared to conventional models and advanced multi-scale models, RefineNet exhibits superior classification performance for chili leaf diseases, achieving an average accuracy of 99.78%. This evaluation confirms that RefineNet is robust, stable, and highly accurate in practical applications for detecting chili leaf diseases. Furthermore, the evaluation reveals that RefineNet demonstrates a good recall rate, F1 score, and confusion matrix.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
References
Sharma K, Kaur M, Singh G (2022) Adoption status of various rice residue management technologies in sri muktsar sahib district of punjab. Journal of Community Mobilization and Sustainable Development p 387.
Sujatha R, Chatterjee JM, Jhanjhi N, et al. (2021) Performance of deep learning vs machine learning in plant leaf disease detection. Microprocessors and Microsystems 80:103,615.
Vallabhajosyula S, Sistla V, Kolli VKK (2022) Transfer learning-based deep ensemble neural network for plant leaf disease detection. Journal of Plant Diseases and Protection 129(3):545–558.
Tiwari V, Joshi RC, Dutta MK (2021) Dense convolutional neural networks based multiclass plant disease detection and classification using leaf images. Ecological Informatics 63:101,289.
Kanaparthi KR, Ilango SS (2023) A survey on training issues in chili leaf diseases identification using deep learning techniques. Procedia Computer Science 218:2123–2132.
Naik BN, Malmathanraj R, Palanisamy P (2022) Detection and classification of chilli leaf disease using a squeeze-and-excitation-based cnn model. Ecological Informatics 69:101,663.
Ahmad N, Asif HMS, Saleem G, et al. (2021) Leaf image-based plant disease identification using color and texture features. Wireless Personal Communications 121(2):1139–1168.
Wahab AHBA, Zahari R, Lim TH (2019) Detecting diseases in chilli plants using k-means segmented support vector machine. In: 2019 3rd international conference on imaging, signal processing and communication (ICISPC), IEEE, pp 57–61.
Kurmi Y, Gangwar S (2022) A leaf image localization based algorithm for different crops disease classification. Information processing in agriculture 9(3):456–474.
Patil A, Lad K (2021) Chili plant leaf disease detection using svm and knn classification. In: Rising Threats in Expert Applications and Solutions: Proceedings of FICR-TEAS 2020, Springer, pp 223–231.
Hossain E, Hossain MF, Rahaman MA (2019) A color and texture based approach for the detection and classification of plant leaf disease using knn classifier. In: 2019 international conference on electrical, computer and communication engineering (ECCE), IEEE, pp 1–6.
Sajitha P, Andrushia AD, Anand N, et al. (2024) A review on machine learning and deep learning image-based plant disease classification for industrial farming systems. Journal of Industrial Information Integration p 100572.
Yang G, He Y, Yang Y, et al. (2020) Fine-grained image classification for crop disease based on attention mechanism. Frontiers in Plant Science 11:600,854.
Ferentinos KP (2018) Deep learning models for plant disease detection and diagnosis. Computers and electronics in agriculture 145:311–318.
Tolstikhin IO, Houlsby N, Kolesnikov A, et al. (2021) Mlp-mixer: An all-mlp architecture for vision. Advances in neural information processing systems 34:24,261–24,272.
Malik A, Vaidya G, Jagota V, et al. (2022) Design and evaluation of a hybrid technique for detecting sunflower leaf disease using deep learning approach. Journal of Food Quality 2022(1):9211,700.
Liu H, Wang S, Jing G, et al. (2023) Combined cnn and rnn neural networks for gpr detection of railway subgrade diseases. Sensors 23(12):5383.
Thai HT, Le KH, Nguyen NLT (2023) Towards sustainable agriculture: A lightweight hybrid model and cloud-based collection of datasets for efficient leaf disease detection. Future Generation Computer Systems 148:488–500.
Sharma V, Kannan S, Tanya S, et al. (2024) Detecting plant diseases at scale: A distributed cnn approach with pyspark and hadoop. Procedia Computer Science 235:1044–1057.
Zhang G, Xu T, Tian Y (2022) Hyperspectral imaging-based classification of rice leaf blast severity over multiple growth stages. Plant Methods 18(1):123.
Yang W, Liu C, Zeng S, et al. (2024) Rapid identification of moldy peanuts based on three-dimensional hyperspectral object detection. Journal of Food Composition and Analysis p 106400.
Karthik R, Alfred JJ, Kennedy JJ (2023) Inception-based global context attention network for the classification of coffee leaf diseases. Ecological Informatics 77:102,213.
Karim MJ, Goni MOF, Nahiduzzaman M, et al. (2024) Enhancing agriculture through real-time grape leaf disease classification via an edge device with a lightweight cnn architecture and grad-cam. Scientific Reports 14(1):16,022.
Ramcharan A, Baranowski K, McCloskey P, et al. (2017) Deep learning for image-based cassava disease detection. Frontiers in plant science 8:1852.
Xu D, Wu Y (2020) Improved yolo-v3 with densenet for multi-scale remote sensing target detection. Sensors 20(15):4276.
Saleem MH, Khanchi S, Potgieter J, et al. (2020) Image-based plant disease identification by deep learning meta-architectures. Plants 9(11):1451.
Alirezazadeh P, Schirrmann M, Stolzenburg F (2023) Improving deep learningbased plant disease classification with attention mechanism. Gesunde Pflanzen 75(1):49–59.
Azour F (2022) Breast abnormality diagnosis using transfer and ensemble learning. PhD thesis, Université d’Ottawa/University of Ottawa.
Khanna M, Singh LK, Thawkar S, et al. (2024) Planet: a robust deep convolutional neural network model for plant leaves disease recognition. Multimedia Tools and Applications 83(2):4465–4517.
Salai B, Ramapuram C (2024) Deep learning approaches for disease detection based on plant leaf image: A review. Intelligent Systems Design and Applications: Deep Learning, Volume 2 p 323.
TugasAkhir (2023) chili disease detection1 dataset. URL https://universe.roboflow.com/tugasakhir-zdajz/chili-disease-detection1, visited on 2023-11-26.
Ullah I, Carrion D, Escalera S, et al. (2022) Meta-album: Multi-domain metadataset for few-shot image classification. In: Thirty-sixth Conference on Neural Information Processing Systems Datasets and Benchmarks Track.