局部自交干扰的全变分图像自适应滤噪算法

针对当前图像滤噪算法虽然能够在一定程度上抑制噪声;但无法保证图像质量,导致图像细节丢失、图像变模糊的弊端,提出一种新的局部自交干扰的全变分图像自适应滤噪算法。定义一个图像局部功率,求解自适应全变分算法的能量函数最小化问题。通过拉格朗日算子获取图像局部功率的非约束最小化全变分形式,计算规整化可信度参数和噪声分布。通过全变分模型将轮廓尺度图像从含有噪声图像中分离出来,对含有噪声图像和轮廓尺度图像进行差运算,获取含有噪声的残差纹理细节图像。将获取的规整化可信度参数代入全变分模型,对含有噪声图像进行处理获取最终的滤噪图像。实验结果表明,经所提算法滤噪处理后,图像质量高,滤噪效果好。

Adaptive image denoising algorithm based on total variation of local self interference

in view of the current image denoising algorithm can suppress the noise to a certain extent, but can not guarantee the quality of the image, resulting in loss of image detail, image blur defects, proposed adaptive image denoising algorithm with a new local self interference total variation, the definition of a local image power, energy minimization problem solving adaptive total variation algorithm. By using the Lagrange operator to obtain the local minimization of the local power of the image, the regularization parameter and the noise distribution are calculated. The contour scale image from noisy images separated by total variation model, difference operation is performed on the noisy images and contour scale image, obtaining residual texture details of the image with noise, the regularization parameters to obtain the reliability of total variation model, a noise image denoising image acquisition of the final. The experimental results show that the proposed algorithm has high image quality and good filtering effect.