摘要
由于光学遥感成像的距离较远和CCD物理尺寸的限制,获得的图像分辨率都比较低。为了在保持原始图像信息的情况下,提高图像的空间分辨率,改善图像的峰值信噪比,在深入分析前期研究中提出的小波双三次插值搜索算法的基础上,建立了小波双三次高倍插值搜索算法。该算法可以根据需要重建出原始遥感图像的高倍高分辨率遥感图像。实验结果表明,新算法重建出原始图像的2×2倍高分辨率图像的细节信息和视觉效果最好,3×3倍和4×4倍次之,5×5倍较差,更高倍数的高分辨率重建图像不宜在实际应用中采用。
Due to the far imaging distance of optical remote sensing and the limitation of the physical size of CCD, the resolution of the remote sensing image is usually low. In order to improve the resolution and PSNR of the obtained image without damaging its original information, the Wavelet Bicubic Interpolation Search Algorithm was proposed in the previous work. On the basis of deep analysis and investigation of the Wavelet Bicubic Interpolation Search Algorithm, the Multi-time Wavelet Bicubic Interpolation Search Algorithm is proposed. The new established algorithm can realize the multi-time and high-resolution image reconstruction of optical remote sensing images. The experiment results prove that the detail information and visual effect of the 2 × 2 times high-resolution image reconstructed by the proposed algorithm is better, the 3 ×3 times and 4 × 4 times is good, 5 × 5 times is bad and more reconstructed times is not suitable to be applied in practice.
出处
《光电工程》
EI
CAS
CSCD
北大核心
2006年第5期62-65,75,共5页
Opto-Electronic Engineering
基金
国家863-708项目资助
关键词
小波变换
图像重建
遥感图像
信噪比
算法
Wavelet transform
Image reconstruction
Remote sensing images
Signal-noise rate
Algorithm
