RESEARCH ON SOLUTIONS TO DETECT DISEASES ON ROSE LEAVES USING IMAGE PROCESSING TECHNIQUES

Abstract

Nowadays, Image processing techniques are widely applied in agricultural production, especially on agricultural products. In this paper, we propose a solution to detect diseases on rose leaves by an image processing algorithm based on leaf features. This algorithm helps detect diseases on rose leaves through collected images. The method is solved through stages such as: collecting images, formatting the size of images, separating the leaf image area from the background and then detecting and zoning the candidate areas that are likely to be diseases by the SimpleBlobDetector algorithm. Experiments on a dataset of 500 images showed that the model had an accuracy of 93.30%. This is a new method for automatic disease detection on rose leaves. The source images and collected images were taken by us and processed by our algorithm. The results show that the proposed method is able to apply for real applications in the future.