An intelligent approach using boosted support vector machine based arithmetic optimization algorithm for accurate detection of plant leaf disease

• Published in: Pattern analysis and applications : PAA Vol. 26; no. 1; pp. 367 - 379
• Main Authors: Prabu, M., Chelliah, Balika J.
• Format: Journal Article
• Language: English
• Published: London Springer London 01.02.2023; Springer Nature B.V
• Subjects: Algorithms; Arithmetic; Computer Science; Disease; Feature extraction; Image segmentation; Optimization; Optimization algorithms; Pattern Recognition; Plant diseases; Short Paper; Support vector machines; Plant leaf disease; Arithmetic optimization algorithm; Gray-level co-occurrence matrix; Boosted support vector machine; Vector-valued active contour model
• ISSN: 1433-7541; 1433-755X
• DOI: 10.1007/s10044-022-01086-z

Leaf disease is considered a serious threat which affects agricultural productivity and ultimately reduces the GDP of the Indian economy. The precise detection and timely analysis of foliar diseases can mitigate the spread of the disease to other parties. However, certain complications such as low precision, high calculation cost, and low recognition speed are raised when detecting leaf diseases. Therefore, to overcome these limitations, we proposed a novel technique called Boosted support vector machine-based Arithmetic optimization algorithm (BSVM-AOA) for accurate detection of plant leaf disease. In this case, image segmentation is done using the vector value active contour model, and feature extraction is done using the greyscale co-occurrence matrix. Furthermore, the performance of the proposed approach is determined by performance parameters such as accuracy, accuracy, recall, specificity, and f-rating. Finally, the comparative analysis is conducted between the different existing techniques and the proposed technique. The comparative results showed that the proposed BSVM-AOA approach is about 98.6% more accurate than other existing techniques.