Orange & Orange leaves diseases detection using Computerized Techniques
Mayen Uddin Mojumdar, Narayan Ranjan Chakraborty
Abstract
Since agriculture employs 47% of the population and contributes about 19.9% of the GDP in Bangladesh, disease detection and management are critical for farmers in order to harvest a higher percentage of utilizable fruits that are fit for consumption. Fruit diseases are a major source of agricultural losses. Fruit monitoring by hand is unreliable since it is entirely dependent on the naked eye's interpretation, and it is also impractical to have experts in the remote areas where the fruits develop. As a result, an automated disease detection system for Orange has been suggested, which uses image processing techniques to determine the extent of the disease and monitor yield loss. K-means clustering was used to segment the images. Using a gray-level co-occurrence matrix, thirteen features were extracted from the segmented image (GLCM). For disease detection and classification, a multi-class support vector machine (SVM) is used. As compared to other current algorithms, the results are experimentally checked and classification Overall accuracy of up to 82.3% is achieved.