小生境演化算法下的WDCT图像压缩方法

WDCT Image Compression Based on Niching Evolutionary Algorithm

针对传统WDCT图像压缩编码算法频率卷曲参数的难选取问题,提出了小生境演化算法下的WDCT图像压缩方法(NEAWDCT).利用小生境演化算法全局寻优的特点,与WDCT图像压缩编码方法相结合,达到自适应选取最优频率卷曲参数的目的.针对频率卷曲参数特征,设计染色体编码方式及演化算子,以加快收敛速度.由于图像能量多集中于低频部分,选择特定范围内的数值初始化种群,不仅能加快算法收敛速度,还保留了种群的多样性.实验结果表明,利用新的NEAWDCT算法生成的WDCT矩阵能有效提高峰值信噪比.

In order to select the frequency warping parameter in WDCT image compression ,this paper presents a new algo-rithm-WDCT image compression algorithm based on niching evolutionary algorithm （NEAWDCT） .With the global optimization of niching evolutionary algorithm ,it is combined with WDCT image compression algorithm to select the optimal parameter adaptively . To improve the convergence speed ,the coding methods and evolutionary operators are designed according to the parameter charac-teristics .The energy of image signal ordinarily is distributed in the part with lower frequency ,therefore population initialized by some numbers in certain interval .The initial population can improve the convergence speed and maintain the population diversity . The experiments show that the new algorithm （NEAWDCT） can improve the peak signal to noise ratio effectively .