NoiseNet: Signal-Dependent Noise Variance Estimation with Convolutional Neural Network

Uss, Mikhail; Vozel, Benôıt; Chehdi, Kacem

doi:10.1007/978-3-030-01449-0_35

Cited by 11 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…DnCNN [88] model has been trained to handle Gaussian denoising with unknown noise level (i.e., random noise, blind Gaussian denoising). While for the rest of the models [104], [105] the level of noise was estimated, using techniques as in [131], [132]. Then the respective model with the adjusted weights was used to handle this specific perturbation.…”

Section: Methodsmentioning

confidence: 99%

Countering Adversarial Attacks on Autonomous Vehicles Using Denoising Techniques: A Review

Kloukiniotis

Papandreou

Lalos

et al. 2022

IEEE Open J. Intell. Transp. Syst.

View full text Add to dashboard Cite

The evolution of automotive technology will eventually permit the automated driving system on the vehicle to handle all circumstances. Human occupants will be just passengers. This poses security issues that need to be addressed. This paper has two aims. The first one investigates strategies for robustifying scene analysis of adversarial road scenes. A taxonomy of the defense mechanisms for countering adversarial perturbations is initially presented, classifying those mechanisms in three major categories: those that modify the data, those that propose adding extra models, and those that focus on modifying the models deployed for scene analysis. Motivated by the limited number of surveys in the first category, we further analyze the approaches that utilize input transformation operations as countermeasures, further classifying them in supervised and unsupervised methods and highlighting both their strengths and weaknesses. The second aim of this paper is to publish CarlaScenes dataset produced using the CARLA simulator. An extensive evaluation study, on CarlaScenes, is performed testing the supervised deep learning approaches that have been either proposed for image restoration or adversarial noise removal. The study presents insights on the robustness of the aforementioned approaches in mitigating adversarial attacks in scene analysis operations.INDEX TERMS Countering adversarial images, robust road scene analysis, deep learning.

show abstract

Section: Methodsmentioning

confidence: 99%

Countering Adversarial Attacks on Autonomous Vehicles Using Denoising Techniques: A Review

Kloukiniotis

Papandreou

Lalos

et al. 2022

IEEE Open J. Intell. Transp. Syst.

View full text Add to dashboard Cite

show abstract

“…Потенційно, ефективність методів третьої групи може бути підвищена за допомогою згорткових нейронних мереж. Такі дослідження були вже проведені і показали досить успішні результати [20]. Проте, цей напрям знаходиться тільки на початковому етапі свого розвитку.…”

Section: вступunclassified

“…Метою даної роботи є побудування точної системи оцінювання характеристик завад на зображеннях, отриманих за допомогою цифрових камер мобільних пристроїв, та розробка мобільного додатку, що реалізує роботу цієї системи на платформі Android. Основу системи оцінювання характеристик завад складає нещодавно розроблена згорткова нейрона мережа (ЗНН) NoiseNet [20].…”

Section: вступunclassified

“…Як було зазначено вище, в якості основи розроблюваної системи автоматичного оцінювання характеристик завад на зображеннях, отриманих за допомогою мобільних пристроїв, передбачається використовувати згорткову нейронну мережу NoiseNet [20]. Цю нейронну мережу було розроблено для оцінювання характеристик просторовонекорельованих завад, що описуються моделлю (1) Після цього, було проаналізовано отримані оцінки СКВ шуму за допомогою NoiseNet, використовуючи всі зображення з бази даних Tampere17.…”

Section: згорткова нейронна мережа Noisenetunclassified

See 1 more Smart Citation

Оцінка Характеристик Шуму На Реальних Зображеннях З Використанням Згорткової Нейронної Мережі На Мобільній Платформі

Цехмистро¹,

Абрамова²,

Rubel³

et al. 2019

reks

View full text Add to dashboard Cite

The subject of the study is the noise characteristics in real images obtained by mobile devices. The goal is to create a demo mobile application in Android platform, which realizes real-time estimation of noise characteristics in such images. Tasks: to investigate the accuracy of noise characteristics estimation by NoiseNet neural network on test images from the Tampere17 database; to conduct a preliminary study of the type, intensity and correlation characteristics of the noise in images obtained by mobile devices; to investigate the possibility of using NoiseNet to assess the noise characteristics in these images. The following results were obtained. Analyzing the noise characteristics in test images from the Tampere17 database, distorted by white Gaussian noise, it was shown that in general, the NoiseNet neural network demonstrates a rather high estimation accuracy (the relative error of evaluation does not exceed 0.2). However, for some images, in particular, highly textured, the value of relative error can be several times higher. The noise characteristics of images taken in various conditions by cameras embedded in mobile devices from various manufacturers were studied. It is shown that the noise in such images is signal-dependent and is often characterized by a high degree of spatial correlation. At the same time, the degree of spatial correlation of noise largely depends on lighting conditions of photo taking and is higher for images obtained in dim light. Since the NoiseNet neural network is not designed to work with spatially correlated noise, for its applying the images were preprocessed to eliminate the spatial correlation of noise. The ready-to-use NoiseNet neural network and the Android demo application for testing are available on the GitHub resource: https://github.com/radiuss/NoiseNet.

show abstract

“…In the case when an image is corrupted by a noise which distribution differs from distributions used for the design of the noise estimation algorithms, these algorithms will produce erroneous estimates. This problem is most contentious for noise level estimators which are based on deep convolutional neural networks (DCNNs) [11][12][13]. Such CNNs, pre-trained under specific noise (e.g., AWGN), may produce wrong results of prediction in real situations.…”

Section: Introductionmentioning

confidence: 99%

Blind estimation of noise level based on pixels values prediction

Ponomarenko¹,

Oleksandr²,

Vladimir³

et al. 2022

Self Cite

View full text Add to dashboard Cite

Noise parameters estimation is required in various stages of digital image processing. Many efficient algorithms of noise estimation were proposed during last two decades. However, most of these algorithms are efficient only for a specific type of noise for which they are designed. For example, methods of variance estimation of additive white Gaussian noise (AWGN) will not work in the case of additive colored Gaussian noise (ACGN) or, in general, in the case of a noise with non AWGN distribution. In this paper, a totally blind method of noise level estimation is proposed. For a given image, a distorted image with a discarded portion of pixels (around 10%) is generated. Then an inpainting (or impulse noise removal) method is applied to recover those discarded pixels values. The difference between the true and recovered pixel values is used to robustly estimate image noise level. The algorithm is applied for different image scales to estimate a noise spectrum. In this paper, we propose a convolutional neural network called PIXPNet for effective prediction of values of missing pixels. A comparative analysis confirms that the proposed PIXPNet provides smallest error of recovered pixel values among all existing methods. A good efficiency of application of the proposed method in both AWGN and spatially correlated noise suppression is demonstrated.

show abstract

NoiseNet: Signal-Dependent Noise Variance Estimation with Convolutional Neural Network

Cited by 11 publications

References 12 publications

Countering Adversarial Attacks on Autonomous Vehicles Using Denoising Techniques: A Review

Countering Adversarial Attacks on Autonomous Vehicles Using Denoising Techniques: A Review

Оцінка Характеристик Шуму На Реальних Зображеннях З Використанням Згорткової Нейронної Мережі На Мобільній Платформі

Blind estimation of noise level based on pixels values prediction

Contact Info

Product

Resources

About