As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
The aim of this study is to design, implement and evaluate an image-based software solution for automatic detection of abnormal on paddy fields that damages by insects, diseases. The methodology of the proposed is composed of third phases. This content is a combined approach for the segmentation of images. In the first phase, watershed segmentation algorithm, second phases, using the K-means clustering algorithm and final in the third phases reducing noise in image gradients. We have taken paddy filed images as a case study and evaluated the proposed approach using abnormal area paddy field images for the pest prediction. In experimental results, indicate that the proposed approach can automatic detection and of abnormal area on paddy fields for the pest prediction. In conclusion, the proposed methods show that detect abnormal region efficiently.