基于压缩感知的无人机图像组合去噪方法

摘要

针对无人机图像中存在的多种类型噪声干扰的情况,提出了一种基于压缩感知的无人机图像混合去噪方法.利用移动窗口平滑处理含噪图像中脉冲噪声,并去除该类噪声对图像稀疏性的破坏;对粗去噪图像进行稀疏表示,利用高斯观测矩阵对其测量,通过正交匹配追踪算法重构得到去噪后图像.实验结果表明,该算法改善了基于压缩感知图像去噪方法对含有脉冲噪声去噪效果差的问题,提高了去噪图像的峰值信噪比和视觉效果.%Aiming at the multiple types of noise jamming in unmanned aerial vehicles (UAV) images,a hybrid image denoising method based on compression sensing is proposed. Firstly,the moving window is used to smooth the impulse noise in the noisy image,and the sparseness of the image is removed. Then,the initial denoising image is sparsely expressed and measured by Gaussian observation matrix,then reconstructed by orthogonal matching pursuit algorithm. Experimental results show that the proposed algorithm improves the denoising effect of Impulse noise and improves the PSNR of the denoised image and the visual effect.