论文部分内容阅读
【目的】遥感图像重建容易引入噪声或边缘出现不规则性,而它又在遥感图像的处理中能起到非常重要的作用,提出新的算法来得到更好的处理结果。【方法】通过对遥感图像进行分块,利用K-SVD算法对遥感图像自身进行字典学习,获得能够稀疏表示高分辨率遥感图像的字典,然后进行特征提取、独立成分分析降维、高分辨率遥感图像的重建等操作。【结果】实现了对遥感图像超分辨率的重建。【结论】该方法提高了图像的峰值信噪比,通过实验验证了算法高效性。
【Objective】 Remote sensing image reconstruction is easy to introduce noise or edge irregularity, and it plays a very important role in the processing of remote sensing images. A new algorithm is proposed to get better processing results. 【Method】 The remote sensing image was divided into blocks and the remote sensing images were dictionary-learned by using K-SVD algorithm to obtain a dictionary which can sparsely represent high-resolution remote sensing images. Then, feature extraction, independent component analysis, dimension reduction and high resolution Remote sensing image reconstruction and other operations. 【Result】 The reconstruction of the super-resolution of the remote sensing image was realized. 【Conclusion】 This method improves the peak signal-to-noise ratio of the image and validates the algorithm’s efficiency through experiments.