结合灰度共生矩阵的多分辨率图像融合策略

多分辨率图像的融合策略是提升融合图像质量的关键。目前常用的融合策略都没有直接考虑融合图像的纹理信息,根据灰度共生矩阵(GLCM)能较好地反映纹理粗糙度和方向的特点,提出了一种结合灰度共生矩阵的多分辨率图像融合策略。首先计算多分辨率图像的低频和高频部分的灰度共生矩阵,然后将低频图像GLCM的能量和高频图像GLCM的对比度对应作为低频和高频系数选取的活性度量进行图像融合。实验表明,所提出的融合策略与目前常见的融合策略相比,在主观感官和客观评价上都能获得较好的融合效果。

Fusion strategy of multi-resolution image combined with gray level co-occurrence matrix

Fusion strategy of multi-resolution image is the key to enhance the quality of integration image. The most commonlyused strategies are not directly consider the image texture information. According to GLCM can reflect the characteristics ofgrain roughness and direction accurately,this paper proposed a fusion strategy of multi-resolution image which combined withGLCM. First, it calculated the GLCM of the low and high frequency part of the multi-resolution image. Then the energy of thelow-frequency GLCM and the contrast of the high-frequency GLCM corresponded as activity measure for image fusion to selectethe low and high frequency coefficients. Experiments show that compared to the current common fusion strategy, the proposedmethod can make fusion effect better through subjective sensory and objective evaluation.