Saveetha Engineering College, Chennai, India.
SMK Fomra Institute of Technology, Chennai, India.
J Med Syst. 2019 Jul 12;43(8):280. doi: 10.1007/s10916-019-1376-4.
A new denoising algorithm using Fast Guided Filter and Discrete Wavelet Transform is proposed to remove Gaussian noise in an image. The Fast Guided Filter removes some part of the details in addition to noise. These details are estimated accurately and combined with the filtered image to get back the final denoised image. The proposed algorithm is compared with other existing filtering techniques such as Wiener filter, Non Local means filter and bilateral filter and it is observed that the performance of this algorithm is superior compared to the above mentioned Gaussian noise removal techniques. The resultant image obtained from this method is very good both from subjective and objective point of view. This algorithm has less computational complexity and preserves edges and other detail information in an image.
提出了一种新的去噪算法，该算法使用快速导向滤波器和离散小波变换来去除图像中的高斯噪声。快速导向滤波器在去除噪声的同时还会去除一部分细节。这些细节被准确估计，并与滤波后的图像结合，以得到最终的去噪图像。将所提出的算法与其他现有的滤波技术（如维纳滤波器、非局部均值滤波器和双边滤波器）进行了比较，结果表明，该算法的性能优于上述高斯噪声去除技术。从主观和客观的角度来看，这种方法得到的结果图像都非常好。该算法的计算复杂度较低，可以保留图像中的边缘和其他细节信息。
