基于模糊C均值聚类的医学图像压缩算法  

The Medical Image Compression Method Based on Fuzzy C-mean Clustering

作　　者：穆克[1] 李文娜[1] 

机构地区：[1]辽宁石油化工大学信息科学与控制工程学院,辽宁抚顺113001

出　　处：《控制工程》2016年第5期706-710,共5页Control Engineering of China

基　　金：辽宁省自然科学基金(2014020106)

摘　　要：作为图像存储、传输系统和远程医疗的关键技术,图像压缩应该以无损低压缩率的方法提供好的视觉效果以保证诊疗质量。随着医学图像的尺寸和分辨率的提高,亟需更高性能的压缩方法。提出一个基于模糊C均值分割和矩形分裂合并的医学图像压缩编码方法。首先通过模糊C均值分割方法将图像分为几部分,然后通过我们的方法获得差值图像。第1个数据流标识分割后留下数据的位置,这些信息通过基于二值图像的矩形分裂合并算法进行编码;第2个数据流包含差值图像,采用无损压缩的方法对其进行压缩编码。实验结果表明此算法能获得高压缩率,好的诊疗质量和改进的参数性能。The medical image compression is the key to study picture archiving, communication systems（PACS） and telemedicine. The image should keep the diagnostic validity with better quality but lower compression ratio using lossless compression algorithm. Higher performance compression algorithms are needed with the increase of resolution and size of medical images. This paper presents a new lossless compression algorithm based on FCM-segment and rectangular split and merge coding for medical images. Firstly, image data is segmented into several parts using fuzzy c-means algorithm and then difference image data can be obtained in our method. The first stream identifies the positions of the remaining data after segmentation, and the information is compressed by binary image coding based on rectangular split and merge coding. The second stream losslessly compressed with the proposed algorithm contains the exact difference image data values. The experimental results show that the method can obtain higher compression rates, better diagnostic image quality and improved performance parameters.

关 键 词：医学图像编码 模糊C均值 图像分割 矩形分裂合并编码 

分 类 号：TP391.58[自动化与计算机技术—计算机应用技术]