Quantum-Inspired Evolutionary Clustering Algorithm

The 20th century is a resplendent age with an all-time development of technology andcivilization. Matter, energy and communication is three key topics. Computers centered Modern Information Processing and digital information characterized Modern Information Transmission are in process of their close combination. As an activeresearch direction of intelligent information processing, Computational Intelligence (CI) has attracted many scientists'attention. In recent years, CI is generally considered as a new computational method based on the full development of its three branches-Neural Network (NN), Fuzzy System (FS) and Evolutionary Computation (EC). In fact, CI includes abundant implications. Over a long period, the worldwide researchers are going in different directions and using various methods to approach the essence of CI. Naturally, CI is an abstract subject spanning Physics, Mathematics, Computer Science, Communication, Physiology, Evolution and Psychology. Accordingly, using the extracted knowledge of these subjects can make a deeper investigation into CI and give a basis for the optimization, also help to build up a more uniformly intelligent method of system design. In this paper, a method named Quantum-Inspired Evolutionary Algorithm is proposed, and which can be successfully used for data clustering and image processing. At the same time, several other techniques were used in the paper for making comparison, such as Genetic Algorithm (GA), Immune evolutionary Algorithm (IEA) and Fuzzy C-Means (FCM). The main contributions can be listed as follows:1) The first one is Quantum-Inspired Evolutionary Clustering Algorithm Based on Manifold Distance (QEAM). This algorithm require the manifold distance as distance measurement function, with which can reflect global consistency, and can break the concept traditional Euclidean distance that"A straight line is the shortest distance between two points". The experiments showed that the algorithm can accelerates the speed of convergence and raises the probability of converging to the best results.2) The second one is an image segmentation algorithm based on Quantum-Inspired Evolutionary Clustering Algorithm (QEAC). In this algorithm, we introduced Quantum-Inspired Evolutionary Clustering Algorithm to the image segmentation. In order to enhance the convergency speed and make the segmentation result more accurate,.we take use of the spatial information among pixels in images and make the image texture features as clustering data sets This algorithm performed well both in the speed of convergence and the quality of segmentation. 3) The third one is an image segmentation algorithm based on watershed and Quantum-Inspired Evolutionary Clustering Algorithm (QWEA).This method divided the image into blocks. First, the image was segmented by watershed algorithm, and then Quantum-Inspired evolutionary algorithm is used for post-processing...