GENETIC ALGORITHMS FOR CLUSTERING IN MACHINE VISION

The paper presents a genetic algorithm for clustering objects in image s based on their visual features. In particular, a novel solution code (named Boolean Matching Code) and a correspondent reproduction operat or (the Single Gene Crossover) are defined specifically for clustering and are compared with other standard genetic approaches. The paper de scribes the clustering algorithm in detail, in order to show the suita bility of the genetic paradigm and underline the importance of effecti ve tuning of algorithm parameters to the application. The algorithm is evaluated on some test sets and an example of its application in auto mated visual inspection is presented.