MATLAB环境中的人脑医学图像融合与识别方法研究  

Study on fusion and recognition method of human brain medical image under MATLAB environment

作　　者：李同英[1,2] 朱洪波[2] 

机构地区：[1]南通理工学院计算机与信息工程学院,江苏南通226002 [2]南京邮电大学宽带无线通信与传感网技术教育部重点实验室,南京210003

出　　处：《中国科技论文》2016年第20期2368-2374,共7页China Sciencepaper

基　　金：国家高技术研究发展计划(863计划)资助项目(2014AA01A705);国家自然科学基金资助项目(61271237)

摘　　要：运用MATLAB平台开发了人脑医学图像融合系统,引入小波变换方法,提出了基于坐标映射的小波最佳分解级数选择准则。对使用不同高低频选择规则和融合算法时的融合结果都进行了主客观分析。实验表明,基于小波变换的融合方法比其他方法更好。由于已经融合的灰度图像不能包括两幅源图像的所有细节信息,而且灰度图像在细节上不如彩色图像容易辨别,所以对已融合图像进行了伪彩色变换。实验表明,伪彩色图像增强了融合图像某些细节信息的可视性。Human brain medical image fusion system was developed on MATLAB platform. The fusion scheme based on wavelet transform was introduced and the rule of wavelet optimal decomposition level based on coordinate mapping was proposed. The subjective and objective analysis was made on the fusion results of different high and low frequency selection rules and fusion algo-rithms. The experiments results show that the fusion scheme based on wavelet transform presents the better performance than other methods. However, it is impossible for the fused gray-level image to contain all detail in the two source images. At the same time, it is more difficult to discern the details in gray level image than the color image. So it is necessary for a pseudo color image based on grayscale transform. The result shows that the pseudo color image enhances the visibility of certain details.

关 键 词：图像融合系统 融合效果评价 医学图像融合 小波变换 伪彩色 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]