通过直方图中轴化提高融合图像光谱保真度

Increasing fusion image spectral fidelity by using midway histogram equalization

针对遥感图像融合过程中光谱失真问题,提出一种基于直方图中轴化策略的图像融合算法。首先,将多光谱图像进行IHS变换;然后,采用直方图中轴化策略调整多光谱图像强度分量图像和全色图像的像素直方图,使之趋于一致;最后,进行IHS反变换获得高质量的彩色图像。理论分析和实验结果表明,该算法不仅可以较好地抑制融合图像光谱失真,同时也能有效保留融合图像的空间分辨率,算法步骤简单、容易实现;与四种传统融合算法(IHS变换、主成分分析(PCA)法、小波变换(WT)法、Brovey)相比,该算法生成的融合图像具有良好的视觉效果,特别是在峰值信噪比(PSNR)、光谱扭曲度和信息熵等客观评价指标中明显优于对比算法。基于直方图中轴化策略融合的遥感图像光谱失真度小、空间信息保持度高。

To solve the problem of spectral distortion in remote sensing image fusion, an improved transform method based on midway image equalization was proposed. First, the muhispectral image was decomposed by IHS （Intensity, Hue, Saturation） transform; then by using the midway image equalization, the cumulative histogram of panchromatic image and the spectral intensity component of multispectral image were adjusted to be the same; finally, the inverse transform of IHS was implemented and a high quality fusion image was obtained. The theoretical analysis and experimental results show that the proposed algorithm can not only suppress the spectral distortion of the fusion image, but also preserve the spatial resolution of the fusion image effectively, and it is simple and easy to implement. Compared with the traditional fusion algorithms such as IHS transform, Principal Component Analysis （PCA）, Wavelet Transform （WT） and Brovey, the fusion images generated by the proposed algorithm have good visual effects; in addition, the proposed algorithm has better performance in terms of Peak- Signal-to-Noise Ratio （PSNR）, spectral distortion and information entropy. Fusion images obtained by midway image equalization maintain spatial information well and have little spectral distortion.