基于图像质量评价参数的FDST域图像融合

为了提升多源图像融合精度,提出了一种基于图像 质量评价参数的有限离散剪切波变换(FDST)域图像自适应融合方法。利用FDST对源图像进行 多尺度、多方向分解,低频子带图像采用结构相似度与空间频率两种图像评价参数作为系数 权值,高频子带图像应用区域空间频率取大的融合策略。应用有限离散剪切波逆变换(FDSIT )重 构图像。采用多组多源图像进行融合实验,并对融合结果进行了客观评价。实验结果表明, 本文提出的融合方法在主观和客观评价上均优于其他多尺度融合方法,具有更好的融合效果 。

Adaptive image fusion based on image quality assessment parameter in FDST domain

To enhance multi-source images fusion accuracy,an adaptive fusion method based on image quality assessment parameters in finite discrete shearlet transform (FDST) domain is proposed.Source images a re decomposed to subband images with multi-scale and multi-direction by FDST.The low-frequency subband fusion rule is based on structural similarity index with spatial frequency as co efficient weights and the fusion rule of regional window spatial frequency selection max is adopted in the high frequency subbands.Fused low and high frequency coefficients are reconstructed to image by finite discrete shearlet in verse transform.Corresponding fusion experiments are done by several sets of different modality images and objective performance assessments are conducted for fusion results.The experi mental results indicate that the proposed algorithm performs better in subjective and objective assessments than a few existing multi-scale fusion tec hniques in the literature and can obtain better fusion performance.