Social insect based Intelligent Optimization Techniques for Squish Piston Image Analysis in CI Engine

Abstract

Social insect are the various species of insects that survive in colonies and its three characteristics: overlap of generations, division of labour and group integration. Social insects are best exemplified by all termites, ants, by various bees and wasps. In this paper, a Swarm intelligence algorithm such as Artificial Bee Colony algorithm (ABC) is implemented to identify the defected region of the squish piston image in CI engine. The asymmetry approach is also implemented for investigating the squish piston image. In bilateral subtraction, the asymmetries between corresponding original squish piston image and used squish piston images are considered for defect identification. The squish piston image border and the center position are used as reference points for alignment of images. The artificial bee colony algorithm is used to identify the border and center position of the image. Then the bilateral subtraction method is used to segment the defected region of the image. Feature extraction method is used to extract the features from the segmented image of the squish piston and feature selection method is used to select the best features of the segmented image. The BPN Classifier is used to classify the defected percentage of the squish piston. Then the performance analysis method is validating the above method.