Predicting Banana Leaf Diseases: Feature Extraction with BL-FEOT and Enhanced Classification using the BAT-KNN Hybrid Algorithm

International Journal of Electronics and Communication Engineering

© 2024 by SSRG - IJECE Journal

Volume 11 Issue 5

Year of Publication : 2024

Authors : Ravi Kumar Tirandasu, Prasanth Yalla

10.14445/23488549/IJECE-V11I5P119

How to Cite?

Ravi Kumar Tirandasu, Prasanth Yalla, "Predicting Banana Leaf Diseases: Feature Extraction with BL-FEOT and Enhanced Classification using the BAT-KNN Hybrid Algorithm," SSRG International Journal of Electronics and Communication Engineering, vol. 11,  no. 5, pp. 195-206, 2024. Crossref, https://doi.org/10.14445/23488549/IJECE-V11I5P119

Abstract:

Banana cultivation, fundamental to many rural economies, confronts persistent threats from many foliar diseases. Rapid and precise disease identification is critical for effective management and containment. This research introduces a pioneering method for detecting and classifying diseases on banana leaves, specifically targeting Cordana, Pestalotiopsis, and Sigatoka. Our process coined the Banana Leaf Feature Extraction and Optimization Technique (BL-FEOT), is a systematic approach encompassing manual background elimination, green color removal to emphasize the disease manifestations, and contour detection to mark out the compromised zones. Distinctive features are extracted for each disease type, culminating in a comprehensive dataset tailored for disease classification. Incorporating an extensive suite of feature extraction techniques, our methodology ensures the maximal retrieval of pivotal information from the infected sites. We leverage the Bat Optimization Technique to fine-tune the extracted features, especially within the RGB color space. This streamlines the feature dimensions and zeroes in on the most relevant components, amplifying the efficiency of the ensuing classification phase. The seminal contribution lies in integrating the Bat Optimization Technique with the K-Nearest Neighbors (KNN) algorithm, resulting in the novel hybrid algorithm, BAT+KNN. This algorithm is applied to the refined feature set for classification purposes. To substantiate the efficacy of BL-FEOT, its performance metrics are juxtaposed against prevailing algorithms using the same feature dataset. A thorough evaluation, encapsulating metrics such as precision, recall, accuracy, F1 score, ROC curve, and error rate, is presented. The experimental results, derived from available datasets, underscore the superior capabilities of our hybrid BAT+KNN algorithm in banana leaf disease identification. This research asserts that the BL-FEOT, powered by the BAT+KNN hybrid algorithm, offers a ground-breaking avenue for the automated and precise detection of banana leaf diseases. Its potential integration into real-time monitoring systems could revolutionize early disease detection and intervention in banana plantations.

Keywords:

Banana, Feature extraction, Banana leaf feature extraction and optimization technique, Bat optimization technique, BAT- K-Nearest Neighbors (KNN) algorithm.

References:

[1] C. Rajamanickam, “Foliar Application of Arka Banana Special as Micronutrients Increase Yield of Banana,” Journal of Krishi Vigyan, vol. 10, no. 1, pp. 101-104, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[2] J.C. Jayasudha and S. Lalithakumari, “Feature Extraction and Classification for Weld Flaw Detection Using Phased Array Ultrasonic Image,” Journal of Advanced Research in Dynamical and Control Systems, vol. 12, no. 7, pp. 1934-1943, 2020.
[CrossRef] [Publisher Link]
[3] Namita M. Butale, and Dattatraya.V.Kodavade, “Survey Paper on Detection of Unhealthy Region of Plant Leaves Using Image Processing and Soft Computing Techniques,” International Journal of Computer Engineering in Research Trends, vol. 5, no. 12, pp. 232-235, 2018.
[CrossRef] [Publisher Link]
[4] Anamika Sharma, and Parul Malhotra, “LDA Based Tea Leaf Classification on the Basis of Shape, Color and Texture,” International Journal of Computer Engineering In Research Trends, vol. 4, no. 12, pp. 543-546, 2017.
[Google Scholar] [Publisher Link]
[5] Silvânio Rodrigues dos Santos, Marcos Koiti Kondo, and M. Sai Kiran, “Multimodal Fusion for Robust Banana Disease Classification and Prediction: Integrating Image Data with Sensor Networks,” Frontiers in Collaborative Research, vol. 1, no. 2, pp. 22-31, 2023.
[Publisher Link]
[6] Qazi Umar Farooq et al., “AgriAqua Intelligence: Advancing Smart Farming with Ecosystem-Based Water Management Solutions,” International Journal of Computer Engineering in Research Trends, vol. 11, no. 1, pp. 51-60, 2024.
[Publisher Link]
[7] Epsita Medhi, and Nabamita Deb, “PSFD-Musa: A Dataset of Banana Plant, Stem, Fruit, Leaf, and Disease,” Data in Brief, vol. 43, pp. 1-7, 2022.
[CrossRef] [Google Scholar] [Publisher Link]  
[8] Shifat E. Arman et al., “BananaLSD: A Banana Leaf Images Dataset for Classification of Banana Leaf Diseases Using Machine Learning,” Data in Brief, vol. 50, pp. 1-8, 2023.
[CrossRef] [Google Scholar] [Publisher Link] 
[9] Praveen Kumar et al., “A Comparative Analysis of Collaborative Filtering Similarity Measurements for Recommendation Systems,” International Journal on Recent and Innovation Trends in Computing and Communication, vol. 11, no. 3s, pp. 184-192, 2023.
[CrossRef] [Google Scholar] [Publisher Link]   
[10] Walter Ocimati, Sivalingam Elayabalan, and Nancy Safari, “Leveraging Deep Learning for Early and Accurate Prediction of Banana Crop Diseases: A Classification and Risk Assessment Framework,” International Journal of Computer Engineering in Research Trends, vol. 11, no. 4, pp. 46-57, 2024.
[Publisher Link]
[11] Farah Mohammad, and Saad Al Ahmadi, “Alzheimer’s Disease Prediction Using Deep Feature Extraction and Optimization,” Mathematics, vol. 11, no. 17, pp. 1-17, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Mohamed Hamada, and A. Al-Fayadh, “Wavelet-Aided Selective Encoding for Enhanced Lossless Image Compression,” Frontiers in Collaborative Research, vol. 1, no. 2, pp. 1-9, 2023.
[Publisher Link]
[13] Rachna Mehta, and Navneet Agarwal, “Splicing Detection for Combined DCT, DWT and Spatial Markov-Features Using Ensemble Classifier,” Procedia Computer Science, vol. 132, pp. 1695-1705, 2018.
[CrossRef] [Google Scholar] [Publisher Link]  
[14] R.S. Loomis, J. Rockström, and M. Bhavsingh, “Synergistic Approaches in Aquatic and Agricultural Modeling for Sustainable Farming,” Synthesis: A Multidisciplinary Research Journal, vol. 1, no. 1, pp. 32-41, 2023.
[Publisher Link] 
[15] Vinuthna Papana, Devireddy Sritha Reddy, and Kistipati Priyatham Reddy, “Transformative Approaches in Integrating Data Science for Disease Outbreak Prediction: A Comprehensive Survey in Epidemiology,” International Journal of Computer Engineering in Research Trends, vol. 10, no. 11, pp. 55-65, 2023.
[CrossRef] [Publisher Link] 
[16] Robbi Rahim, and Abdul Wahid, “Advancements in Plant Disease Detection: Integrating Machine Learning, Image Processing, and Precision Agriculture,” International Journal of Computer Engineering in Research Trends, vol. 10, no. 8, pp. 19-25, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Jing Liu et al., “Hyperspectral Remote Sensing Images Deep Feature Extraction Based on Mixed Feature and Convolutional Neural Networks,” Remote Sensing, vol. 13, no. 13, pp. 1-17, 2021.
[CrossRef] [Google Scholar] [Publisher Link] 
[18] J. Rockstroma, J. Barron, and Addepalli Lavanya, “Aquatic-Based Optimization Techniques for Sustainable Agricultural Development,” Frontiers in Collaborative Research, vol. 1, no. 1, pp. 12-21, 2023.
[Publisher Link]
[19] Songtao Liu, Di Huang, and Yunhong Wang, “Pay Attention to Them: Deep Reinforcement Learning-Based Cascade Object Detection,” IEEE Transactions on Neural Networks and Learning Systems, vol. 31, no. 7, pp. 2544-2556, 2020.
[CrossRef] [Google Scholar] [Publisher Link] 
[20] Dongyeong Choi, and Seong-Won Lee, “Texture Region Based Hybrid Stereo Matching,” IEIE Transactions on Smart Processing & Computing, vol. 7, no. 2, pp. 89-96, 2018.
[CrossRef] [Google Scholar] [Publisher Link]  
[21] M. Chitra Devi, “Skin Cancer Classification Using Dermoscopic Images Based on Ranklet Transform, Co-occurrence Features and Random Forest Classifier,” Medico-Legal Update, vol. 20, no. 3, pp. 344-350, 2020.
[CrossRef] [Google Scholar] [Publisher Link] 
[22] S. Harini Chandana, and R. Senthil Kumar, “A Deep Learning Model to Identify Twins and Look Alike Identification Using Convolutional Neural Network (CNN) and to Compare the Accuracy with SVM Approach,” Electrochemical Society Transactions, vol. 107, no. 1, pp. 14109-14121, 2022.
[CrossRef] [Google Scholar] [Publisher Link]  
[23] Kasula Kedhari Priya et al., “Towards a Greener Tomorrow: The Role of Data Science in Shaping Sustainable Farming Practices,” International Journal of Computer Engineering in Research Trends, vol. 11, no. 4, pp. 12-19, 2024.
[Publisher Link]  
[24] Ch.G.V.N. Prasad et al., “Edge Computing and Blockchain in Smart Agriculture Systems,” International Journal on Recent and Innovation Trends in Computing and Communication, vol. 10, no. 1s, pp. 265-274, 2022.
[CrossRef] [Google Scholar] [Publisher Link]