论文部分内容阅读
当卫星遥感图像的对比度较低和暗原色干扰较大时,图像的增强效果不好。提出基于改进Retinex算法的卫星遥感图像增强技术,采用小波多尺度分解方法进行卫星遥感图像的降噪预处理,在特征模板中提取卫星遥感图像的多尺度Retinex颜色特征分量,在Retinex颜色特征基础上融合卫星遥感图像的边缘轮廓特征点,通过模板自适应匹配实现卫星遥感图像的增强。仿真结果表明,采用该算法进行卫星遥感图像增强处理,能有效减弱噪声的干扰,抑制不感兴趣的特征,提高对图像的输出峰值信噪比,进而提高对卫星遥感图像的识别能力。
When the contrast of the satellite remote sensing image is low and the interference of the dark primary color is large, the image enhancement effect is not good. The improved remote sensing image enhancement technology based on improved Retinex algorithm is proposed. The wavelet multiscale decomposition method is used to reduce noise in the satellite remote sensing image. The multi-scale Retinex color feature components of the satellite remote sensing image are extracted from the feature template. Based on the Retinex color features The edge contour feature points of the satellite remote sensing image are fused, and the satellite remote sensing image enhancement is achieved through adaptive template matching. The simulation results show that this algorithm can effectively reduce the interference of noise, restrain the non-interested features, increase the peak signal-to-noise ratio of the image and improve the ability of recognizing satellite remote sensing images.