A SURVEY ON, ANALYSIS AND DETECTION OF DISEASE ON COTTON LEAVES

Computer Engineering

Diseases on plants lead to an increased loss in quality and quantity of agricultural crops around the world. So method to identify the disease plays an important role in disease management. Detecting disease from image is more efficient than visiting the farm. But image contains more complex information which human eye can extract and understand after years of training. This article introduces one approach to detect diseases on cotton leaf using cotton leaf image. In this article, we are focusing on cotton leaf instead of focusing on whole cotton plant. There are different types of diseases on cotton leaf with different pattern of color in each disease. Comparing image directly with another image is not an efficient way of disease detection. So the proposed system includes three main phases that are Pre-processing, Feature Extraction and Classification. In pre-processing phase image is blurred and converted into Gray color scale from RGB color scale. Next, in feature extraction phase, color feature and texture features are extracted. Since color feature and texture feature are the most important features that describe the cotton images, the proposed system uses Color Coherent Vectors (CCV) for color feature extraction and Entropy, Local Range, Standard Deviation for texture feature extraction. Using k-means clustering algorithm different clusters are formed. Finally Support Vector Machine (SVM) classifier is used to train dataset during training phase and to identify disease by comparing extracted features with trained dataset during testing phase. After detection of disease, solution for the disease is provided to the user.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.