A MACHINE LEARNING-BASED FRAMEWORK FOR CROP DISEASES DETECTION USING IMAGES
A MACHINE LEARNING-BASED FRAMEWORK FOR CROP DISEASES DETECTION USING IMAGES
Authors:
Jatin Kumar Singh
Department of Computer Engineering MIT ADT University Loni Kalbhor, Pune, Maharashtra, India
ABSTRACT
Agricultural productivity and food security are majorly affected by crop diseases, especially in developing countries like India, where diagnosing crop diseases early can be difficult for small and marginal farmers. Traditional methods of diagnosing crop diseases are based on visual inspection and expert judgement; they take too much time to complete; the process is subjective and not easily accessible for the rural community. Although new methods using machine learning and deep learning to detect crop diseases from images have been successful in controlled environments, the methods will not perform as well in a real-world agricultural environment. The purpose of this article is to present a framework for detecting crop diseases based on leaf images using machine learning and image processing methods, which provide for a combination of robust methods that allow for an earlier diagnosis of crop disease and less crop damage, and promote more sustainable practices of agriculture to reduce crop losses.
KEYWORDS:
Index Terms—Crop disease detection, deep learning, leaf image analysis, feature extraction, automated diagnosis.