DISEASE DETECTION AND RECOGNITION OF ROSE PIANTS USING IMAGE PROCESSING AND MACHINE LEARNING

Abstract

The main motivation behind doing this research is to help the agriculture sector, especially the farmers to predict, detect and prevent plant diseases. In this paper we will work on black spot disease in leaves of rose plant. The current methodology which is being used is outdated and highly dependent upon human resources, it is practically impossible for the plant pathologist to help every farmer and it is also not possible for farmers to afford plant pathologists. The main objective of this research is not only to detect plant diseases using image processing techniques but also to prevent diseases by studying the hidden pattern. We'll be using the following algorithms – Random Forest from machine learning and Convolutional Neural Networks from deep learning. This paper includes various phases of implementation namely data creation, feature extraction, training the classifier and classification.

Keywords: Diseased and Healthy leaf, Feature extraction, Training, Classification, CNN, Random Forest