Performance Evaluation of Optimal Parameters for Pest Image Segmentation using FCM and ACO

In gift agricultural subject, pest identity causes enormous discount in both fine and quantity of crop cultivation. If you want to growth the production price of crop, the presence of tiny pests such as aphids, whiteflies, and spider mites which purpose leaf deformation is the main hassle. Consequently early pest detection is a primary project in agricultural field. This research specializes in photograph processing, sensible set of rules and computer technology to expand a new pest detection device that’s vital and imperious to crop cultivation. In existing system, the pest detection using fuzzy c-manner clustering address the problem of overlapping of gadgets can purpose for hiding the pest. As cluster length will increase the rims are brittle and compactness of the clusters gets altered. Therefore the prevailing FCM segmentation now not able to cope with the constituent parts of the picture implicitly. The proposed technique known as ant colony optimization method resolves this hassle, and offers the implicit pest segmentation. The segmentation approach uses the swarm intelligence approach based totally at the behavior of the ant colonies. Ant colony optimization became used to extract the area of the insect pest and also to attain the most suitable constant parameters of recognized insect pest. In the beginning the captured pix are processed for pre-processing. Then photo segmentation is done based totally on ACO to get pest target location. Later constant parameters are measured for pest image segmentation the use of ACO which include structural content, top signal to noise ratio, normalized correlation coefficient, and average difference and normalized absolute mistakes. The matlab simulation experiments demonstrate that this proposed ACO technique is extra powerful than fuzzy c-way clustering consequently it can phase the pest photograph higher.