Image De-Noising With Machine Learning: A Review

Abstract

Images are susceptible to various kinds of noises, which corrupt the pictorial information stored in the images. Image de-noising has become an integral part of the image processing workflow. It is used to attenuate the noises and accentuate the specific image information stored within. Machine learning is an important tool in the image-de-noising workflow in terms of its robustness, accuracy, and time requirement. This paper explores the numerous state-of-the-art machine-learning-based image de-noisers like dictionary learning models, convolutional neural networks and generative adversarial networks for a range of noises like Gaussian, Impulse, Poisson, Mixed and Real-World noises. The motivation, algorithm and framework of different machine learning de-noisers are analyzed. These de-noisers are compared using PSNR as quality assessment metric on some benchmark datasets. The best de-noising results for different noise type is discussed along with future prospects. Among various Gaussian noise de-noisers, GCBD, BRDNet and PReLU network prove to be promising. CNN+LSTM, and MC2RNet are most suitable CNNbased Poisson de-noisers. For impulse noise removal, Blind CNN, and CNN+PSO perform well. For mixed noise removal, WDL, EM-CNN, CNN, SDL, and Mixed CNN are prominent. De-noisers like GRDN and DDFN show accurate results in the domain of real-world de-noising.

Impact of noise to edge detection on images of different complexity

“…It is statistical and additive in nature that follows the normal distribution with zero mean and σ standard deviation and affects all pixels in the image. The cause of its appearance are fluctuations in sensor temperature and variations in ambient lighting [3]. PDF of Gaussian noise that the following equation [2,3]: (1) where z is the value of gray intensity, σ is the standard deviation and μ is the mean.…”

“…The cause of its appearance are fluctuations in sensor temperature and variations in ambient lighting [3]. PDF of Gaussian noise that the following equation [2,3]: (1) where z is the value of gray intensity, σ is the standard deviation and μ is the mean. The mathematical model of Gaussian noise represents an exact approximation of a scenario in the real world.…”

“…Thus, Salt and Pepper noise affects only certain pixels and replaces their values with dark pixels if that pixel was bright, that is, of higher intensity, and vice versa. Salt and Pepper noise is given [1,3]…”

“…It is an additive or multiplicative (Speckle) noise according to nature, that is, noise pixels are added or multiplied with the reference image. According to the classification of sources, this is often called quantization noise or photon noise [3].…”

A Neighborhood Impact Driven K-Medoid Clustering and Fuzzy Logic Blended Approach for High Density Impulse Noise Detection and Removal

“…In the field of image processing, eliminating impulse noise [1] is a challenging task. Noise may have a significant impact on digital images throughout the collection, transmission, and atmospheric turbulence processes.…”