Class-Aware Fully Convolutional Gaussian and Poisson Denoising (original) (raw)

Image Denoising in the Deep Learning Era

Over the last decade, the number of digital images captured per day witnessed a massive explosion. Nevertheless, the visual quality of photographs is often degraded by noise during image acquisition or transmission. With the re-emergence of deep neural networks, the performance of image denoising techniques has been substantially improved in recent years. The objective of this paper is to provide a comprehensive survey of recent advances in image denoising techniques based on deep neural networks. In doing so, we commence with a thorough description of the fundamental preliminaries of the image denoising problem followed by highlighting the benchmark datasets and the widely used metrics for objective assessments. Subsequently, we study the existing deep denoisers in the supervised and unsupervised categories and review the technical specifics of some representative methods within each category. Last but not least, we conclude the analysis by remarking on trends and challenges in the...

Learning Pixel-Distribution Prior with Wider Convolution for Image Denoising

ArXiv, 2017

In this work, we explore an innovative strategy for image denoising by using convolutional neural networks (CNN) to learn pixel-distribution from noisy data. By increasing CNN's width with large reception fields and more channels in each layer, CNNs can reveal the ability to learn pixel-distribution, which is a prior existing in many different types of noise. The key to our approach is a discovery that wider CNNs tends to learn the pixel-distribution features, which provides the probability of that inference-mapping primarily relies on the priors instead of deeper CNNs with more stacked nonlinear layers. We evaluate our work: Wide inference Networks (WIN) on additive white Gaussian noise (AWGN) and demonstrate that by learning the pixel-distribution in images, WIN-based network consistently achieves significantly better performance than current state-of-the-art deep CNN-based methods in both quantitative and visual evaluations. \textit{Code and models are available at \url{this ...

State of the Art on: Deep Image Denoising

2020

Recent technological and methodological advances have allowed the employment of deep learning techniques, in particular deep artificial neural networks, in a large variety of fields. One of the fields that most is benefiting from the introduction of deep learning is image processing and computer vision, which mainly exploits convolutional neural networks (CNNs) for addressing visual understanding problems. For instance, the use of CNNs for image classification and object detection has led to outstanding results. In the last years, deep learning models have been successfully employed also for the tasks of image restoration. Starting from a corrupted image (e.g. noisy, blurred), the goal of image restoration is to recover the original image (i.e. the clean image). Depending on the type of corruption, image restoration tasks can be divided into deblurring, super-resolution, denoising, inpainting, text removal and many others. In particular, the main focus of our research work is image ...

Gated Texture CNN for Efficient and Configurable Image Denoising

Computer Vision – ECCV 2020 Workshops, 2020

Convolutional neural network (CNN)-based image denoising methods typically estimate the noise component contained in a noisy input image and restore a clean image by subtracting the estimated noise from the input. However, previous denoising methods tend to remove high-frequency information (e.g., textures) from the input. It caused by intermediate feature maps of CNN contains texture information. A straightforward approach to this problem is stacking numerous layers, which leads to a high computational cost. To achieve high performance and computational efficiency, we propose a gated texture CNN (GTCNN), which is designed to carefully exclude the texture information from each intermediate feature map of the CNN by incorporating gating mechanisms. Our GTCNN achieves state-of-the-art performance with 4.8 times fewer parameters than previous state-of-the-art methods. Furthermore, the GTCNN allows us to interactively control the texture strength in the output image without any additional modules, training, or computational costs.

Self-Supervised Poisson-Gaussian Denoising

2021 IEEE Winter Conference on Applications of Computer Vision (WACV), 2021

We extend the blindspot model for self-supervised denoising to handle Poisson-Gaussian noise and introduce an improved training scheme that avoids hyperparameters and adapts the denoiser to the test data. Self-supervised models for denoising learn to denoise from only noisy data and do not require corresponding clean images, which are difficult or impossible to acquire in some application areas of interest such as low-light microscopy. We introduce a new training strategy to handle Poisson-Gaussian noise which is the standard noise model for microscope images. Our new strategy eliminates hyperparameters from the loss function, which is important in a self-supervised regime where no ground truth data is available to guide hyperparameter tuning. We show how our denoiser can be adapted to the test data to improve performance. Our evaluations on microscope image denoising benchmarks validate our approach.

Dilated Residual Network for Image Denoising

arXiv preprint arXiv: 1708.05473, 2017

Variations of deep neural networks such as convolutional neural network (CNN) have been successfully applied to image denoising. The goal is to automatically learn a mapping from a noisy image to a clean image given training data consisting of pairs of noisy and clean image patches. Most existing CNN models for image denoising have many layers. In such cases, the models involve a large amount of parameters and are computationally expensive to train. In this paper, we develop a dilated residual CNN for Gaussian image denoising. Compared with the recently proposed residual denoiser, our method can achieve comparable performance with less computational cost. Specifically, we enlarge receptive field by adopting dilated convolution in residual network, and the dilation factor is set to a certain value. Appropriate zero padding is utilized to make the dimension of the output the same as the input. It has been proven that the expansion of receptive field can boost the CNN performance in image classification, and we further demonstrate that it can also lead to competitive performance for denoising problem. Moreover, we present a formula to calculate receptive field size when dilated convolution is incorporated. Thus, the change of receptive field can be interpreted mathematically. To validate the efficacy of our approach, we conduct extensive experiments for both gray and color image denoising with specific or randomized noise levels. Both of the quantitative measurements and the visual results of denoising are promising comparing with state-of-the-art baselines.

Fully Symmetric Convolutional Network for Effective Image Denoising

Applied Sciences

Neural-network-based image denoising is one of the promising approaches to deal with problems in image processing. In this work, a deep fully symmetric convolutional–deconvolutional neural network (FSCN) is proposed for image denoising. The proposed model comprises a novel architecture with a chain of successive symmetric convolutional–deconvolutional layers. This framework learns convolutional–deconvolutional mappings from corrupted images to the clean ones in an end-to-end fashion without using image priors. The convolutional layers act as feature extractor to encode primary components of the image contents while eliminating corruptions, and the deconvolutional layers then decode the image abstractions to recover the image content details. An adaptive moment optimizer is used to minimize the reconstruction loss as it is appropriate for large data and noisy images. Extensive experiments were conducted for image denoising to evaluate the FSCN model against the existing state-of-the-...

The Art of Image Restoration: Unmasking the Secrets of Denoising Techniques

mage denoising is a crucial preprocessing step in many computer vision and image processing applications, aimed at enhancing image quality by reducing noise while preserving significant details. Over the years, various techniques have been developed, ranging from traditional methods like spatial and frequency domain filtering to more advanced machine learning-based approaches. This paper presents a comprehensive review of existing image denoising techniques, categorizing them into classical methods (such as Gaussian filters, wavelet transforms, and non-local means) and modern approaches leveraging deep learning architectures (e.g., convolutional neural networks, autoencoders, and GANs). We analyze the strengths and weaknesses of each method, with a particular focus on their performance under different noise models, computational complexity, and generalization across diverse datasets. Finally, we discuss emerging trends in the field and propose directions for future research, particularly emphasizing the role of hybrid models and unsupervised learning in improving denoising performance.

Noise2Grad: Extract Image Noise to Denoise

2021

In many image denoising tasks, the difficulty of collecting noisy/clean image pairs limits the application of supervised CNNs. We consider such a case in which paired data and noise statistics are not accessible, but unpaired noisy and clean images are easy to collect. To form the necessary supervision, our strategy is to extract the noise from the noisy image to synthesize new data. To ease the interference of the image background, we use a noise removal module to aid noise extraction. The noise removal module first roughly removes noise from the noisy image, which is equivalent to excluding much background information. A noise approximation module can therefore easily extract a new noise map from the removed noise to match the gradient of the noisy input. This noise map is added to a random clean image to synthesize a new data pair, which is then fed back to the noise removal module to correct the noise removal process. These two modules cooperate to extract noise finely. After co...

ResDNN: deep residual learning for natural image denoising

IET Image Processing, 2020

Image denoising is a thoroughly studied research problem in the areas of image processing and computer vision. In this work, a deep convolution neural network with added benefits of residual learning for image denoising is proposed. The network is composed of convolution layers and ResNet blocks along with rectified linear unit activation functions. The network is capable of learning end-to-end mappings from noise distorted images to restored cleaner versions. The deeper networks tend to be challenging to train and often are posed with the problem of vanishing gradients. The residual learning and orthogonal kernel initialisation keep the gradients in check. The skip connections in the ResNet blocks pass on the learned abstractions further down the network in the forward pass, thus achieving better results. With a single model, one can tackle different levels of Gaussian noise efficiently. The experiments conducted on the benchmark datasets prove that the proposed model obtains a significant improvement in structural similarity index than the previously existing state-of-the-art techniques.