Abstract

Crop growth is a core element of the field management. Suitable evaluation and diagnosis of crop disease in the field is very critical for the increased production. Agricultural crops in India are under constant threat of pests affecting their roots as well as leaves. Diseased plants can exhibit a variety of symptoms and making diagnosis was extremely difficult.

Common symptoms are includes abnormal leaf growth, color distortion, stunted growth, shriveled and damaged pods. Image acquisition devices are used to acquire images of plantations at regular intervals. These images are then subjected to pre-processing, transformation and clustering. The leaf images are segmented using clustering techniques. Clustering is the process of partitioning a group of data points into a small number of clusters. In this paper we present a clustering technique called Spatial FCM (SFCM) to identify the pest & the disease. Also the performance of proposed technique is compared with other clustering techniques such as K-means, Fuzzy CMeans (FCM), Kernel based FCM (KFCM) & Spatial FCM (SFCM).

Then the features such as color, texture are extracted from diseased leaf image & then compared with normal leaf image. The neural network method is used to classify the pest & Disease in crops. The evaluation parameters considered for comparison of spatial FCM with other clustering techniques are as follows: Specificity, Sensitivity, Accuracy, Area of affected leaf, Percentage of disease infection, etc. The neural network classification method is used to classify the type of disease and the pathogen or pest that causes that disease.