论文部分内容阅读
针对经典分形编码算法编码时间过长和基于K-均值聚类等快速分形编码算法依赖数据分布等问题,提出了一种基于免疫粒子群优化(IPSO)和核模糊聚类的快速分形图像编码算法.提出基于IPSO的核模糊聚类算法,将IPSO算法应用于聚类中心的求解中,并将其应用于分形图像编码,分别对子块和父块进行核模糊聚类,以更加合理的分类搜索取代全局搜索,减少编码时间.实验结果表明,新算法的编码时间约为经典分形编码算法的1/6,其峰值信噪比只略微下降;与基于K-均值聚类和基于粒子群优化聚类等快速分形图像编码算法相比,新算法能以更少的编码时间获得更高的峰值信噪比.
Aiming at the problems of the traditional fractal coding algorithm coding time-consuming and the fast fractal coding algorithm based on K-means clustering, which depend on the data distribution, a fast fractal image coding algorithm based on immune particle swarm optimization (IPSO) and kernel-based fuzzy clustering The IPSO based kernel fuzzy clustering algorithm is proposed in this paper. The IPSO algorithm is applied to solve the clustering center, which is applied to the fractal image coding. The sub - block and the parent block are respectively subjected to nuclear fuzzy clustering, Search to replace the global search and reduce the coding time.The experimental results show that the coding time of the new algorithm is about 1/6 of that of the classical fractal coding algorithm and the peak signal to noise ratio only slightly decreases.Compared with K-means clustering and Particle Swarm Optimization Compared with the fast fractal image coding algorithm such as clustering, the new algorithm can achieve higher peak signal-to-noise ratio with less coding time.