CAJ | 학술논문

　　针对多传感器图像融合问题，提出一种基于非下采样Shearlet变换域与人眼视觉特性的图像融合方法。采用非下采样Shearlet变换对源图像进行多尺度、多方向稀疏分解，得到低频子带图像和一系列不同尺度、不同方向的高频子带图像；提出一种视觉敏感度系数作为各子带图像融合的考量标准完成对源图像各对应子带图像的融合，设计了基于非下采样Shearlet变换与人眼视觉特性的图像融合算法，并采用非下采样Shearlet逆变换获得最终融合图像。仿真结果表明：该方法不仅拥有更理想的融合效果，还具有较高的运行效率。
　　침대다전감기도상융합문제，제출일충기우비하채양Shearlet변환역여인안시각특성적도상융합방법。채용비하채양Shearlet변환대원도상진행다척도、다방향희소분해，득도저빈자대도상화일계렬불동척도、불동방향적고빈자대도상；제출일충시각민감도계수작위각자대도상융합적고량표준완성대원도상각대응자대도상적융합，설계료기우비하채양Shearlet변환여인안시각특성적도상융합산법，병채용비하채양Shearlet역변환획득최종융합도상。방진결과표명：해방법불부옹유경이상적융합효과，환구유교고적운행효솔。
A novel technique for image fusion based on the non?subsampled shearlet transform (NSST) domain and human visual characteristic (HVC) is proposed to resolve the problem of the multi?sensor image fusion. Multi?scale and multi?directional sparse decompositions of source images are performed by NSST, so that the low?frequency sub?images and a series of high?frequency ones with diverse scales and directions can be obtained. Then, as the e?valuation norm of sub?images fusion, the definition of visual sensitivity coefficient is presented to complete the fu?sion process of sub?images from each corresponding source image, respectively. Meanwhile, the algorithm for image fusion based on NSST and HVC is devised. The final fused image is achieved by utilizing inverse NSST to all fused sub?images. Experimental results show that the technique proposed has better performance, and higher running effi?ciency.