论文部分内容阅读
图像融合是解决多源遥感图像综合的最有效技术手段,针对不同数据源选择最佳的融合方法是提高图像融合质量的关键。在分析了B ROVEY融合和小波融合的理论、算法和融合过程的基础上,对Q U ICKB IRD的全色波段图像和多光谱波段图像数据进行融合实验,然后从定性和定量两个方面对融合效果进行了分析与评价。定性分析是从色调、纹理和清晰度等方面进行分析,而定量分析是根据熵、平均梯度和光谱真实性等指标进行分析,实验结果表明:在处理Q U ICKB IRD遥感图像时,采用小波融合的图像既保持了较高的空间分辨率,又具有较好的光谱特性;而采用B ROVEY融合的图像虽然图像空间分辨率也较高,但光谱信息丢失较大,因此B ROVEY融合方法并不适用于处理Q U ICKB IRD遥感图像。
Image fusion is the most effective technique to solve multi-source remote sensing image synthesis. The key to improve image fusion quality is to select the best fusion method for different data sources. Based on the analysis of the theory, algorithm and fusion process of B ROVEY fusion and wavelet fusion, the fusion experiment of QU ICKB IRD panchromatic and multispectral band image data is carried out, and then the fusion effect is analyzed qualitatively and quantitatively Carried out analysis and evaluation. Qualitative analysis is based on the aspects of hue, texture and sharpness. The quantitative analysis is based on entropy, average gradient and spectral authenticity. The experimental results show that when QUICKB IRD remote sensing images are processed, wavelet fusion The image maintains high spatial resolution and good spectral characteristics. However, the B ROVEY fusion method does not apply to the B ROVEY fusion although the spatial resolution of the image is high, but the spectral information loss is large For processing QU ICKB IRD remote sensing images.