No abstract

Background: Suppression of noise from color images is essential for many image-processing applications. Various median filter variants have been introduced to suppress impulse noise but fails to retain the image detail information in one way or the other, especially for color images. Method: Recently, denoising convolutional neural networks (DnCNNs) have been used for suppression of noise with outstanding results. Result: In this article, we suggest a frequency median filter method combined with DnCNN for color images polluted by Salt and Pepper (SnP) noise wherein the restored value for the center pixel in a noisy image is evaluated by the frequency median and the pretrained DnCNN is then hired for suppressing the remaining noise to attain the output image finally with refined quality. Conclusion: Simulation results on the taken color images show that the proposed method surpasses state-of-the-art de-noising methods visually and in terms of quality metrics such as PSNR, SSIM, NMSE, Entropy, IEF, NCC, PCC, and Running Time.

Elimination of impulse noise in image snap shots with side renovation is one of the complex duties in digital image processing. In this paper, the removal of random impulse noise is done in two important levels. In first level, the detection of the impulse noise is done on the premise of a double threshold selecting strategy after which in the another level, elimination of impulse noise is done by the usage of median filter and directional weighted median filter relying upon the noise map (Nmap) construction of corrupted pixels detected within the first level. The proposed method makes use of the statistical characteristics of noisy image graphs and the brink obtained is adaptable to one of a kind of snap shots and noise conditions. Comparative evaluation with different widespread de-noising techniques shows that the proposed method outperforms in terms of PSNR, SSIM, NMSE and Computation Time (CT) of the distinct trying out test images, with exclusive noise levels.

Impulse noise generally occurs because of bit errors in progression of image acquisition and transmission. It is well known that median filtering method is an impulse noise removal method. Lots of modified median filters have been proposed in the last decades to improve the methods for noise suppression and detail preservation, which have their own deficiencies while identifying and restoring noise pixels. In this article, after deeply analyzing the reasons, such as decreased noise detection and noise removal accuracy that forms the basis of the deficiencies, this article proposes a modified weighted median filter method for color images corrupted by salt-and-pepper noise. In this method, a pixel is classified into either “noise free pixel” or “noise pixel” by checking the center pixel in the current filtering window with the extreme values (0 or 255) for an 8-bit image using noise detection step. Directional differences and the number of “good” pixels in the current filtering window modify the detected noise pixels. Simulation effects on considered test images reveal the proposed method to be improved over state-of-the-art de-noising methods in terms of PSNR and SSIM with pictorial comparative analysis.

With the explosion in the number of color digital images taken every day, the demand for more accurate and visually pleasing images is increasing. Images that have only one component in each pixel are called scalar images. Correspondingly, when each pixel consists of three separate components from three different signal channels, these are called color images. Image denoising, which aims to reconstruct a high-quality image from its degraded observation, is a classical yet still very active topic in the area of low-level computer vision. Impulse noise is one of the most severe noises which usually affect the images during signal acquisition stage or due to the bit error in the transmission. The use of color images is increasing in many color image processing applications. Restoration of images corrupted by noises is a very common problem in color image processing. Therefore, work is required to reduce noise without losing the color image features.