Hybrid Multi-Core Algorithm Based Image Segmentation for Plant Disease Identification using Mobile Application

Abstract

Agriculture is the backbone of Indian economy. Right around 70% individuals rely on it and shares major part of the Gross Domestic Product (GDP). Diseases in yields, is generally identified by the symptoms on the leaves which affects both quality and quantity of farming items. Impression of human eye is less grounded, to watch minute variations in the tainted portion of leaf. In this paper, we propose an Android application for naturally recognizing and ordering plant leaf infections. The project enables the users to identify plant leaf infections based on photographs of the plant leaves taken with a mobile phone. This mode of approach is expected to upgrade the efficiency of harvests. It incorporates a few stages viz. image acquisition through mobile camera, image pre-processing to improve the quality of picture using smoothing filter, segmentation of image using hybrid multi-core algorithm, feature extraction using colour co-occurrence and classification of diseases using Baye's classifier. The research results indicated a better accuracy rate of 97.42% in detecting and classifying diseases.

Linear Interpolation Algorithms and their Architectures for Image Scaling – A Survey

“…The Bandelet transform reduces the effect of jagged edges, and improves the visual effect [30]. Bandelet transform is more complex and computationally intensive.…”