Medical Image Segmentation Algorithm Based on Quantum Clonal Evolution and Two-Dimensional Tsallis Entropy

The paper proposes an improved clone quantum evolutionary algorithm in view of the shortcomings existed in the image segmentation by evolutionary algorithm such as slow convergence and easy prematurity.The improved algorithm is able to overcome the weaknesses like the singe variation of quantum gate, the fault of fixed-size, and avoid the prematurity in small parts by using diverse population information in quantum space and imposing different chaotic perturbation on each unit which depends its fitness in quantum variation.The algorithm also can partially improve optimization ability, increase the converging speed and transmit the information of optimal unit to next generation by using clonal operator.The algorithm will be applied to seek the two dimension optimal Tsallis entropy, divide the picture eventually.It is showed that the algorithm can not only effectively solve the problems of slow converging and liability to local extremum, but considerably increase both dividing speed and precision, it meets the request of rebuilding in three dimension.