Naturalness Preserved Nonuniform Illumination Estimation for Image Enhancement Based on Retinex

Gao, Yuanyuan; Hu, Hai‐Miao; Li, Bo; Guo, Qiang

doi:10.1109/tmm.2017.2740025

Cited by 88 publications

(21 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the paper [52] is proposed improved adaptive contrast enhancement method based on histogram compacting transform. In the papers [53,54] are presented low-lightning image enhancement based on Retinex model. The different Retinex models are proposed in order to enhance illumination and reflectance.…”

Section: Latest Image Enhancement Techniquesmentioning

confidence: 99%

A Review on Image Enhancement Techniques

2019

View full text Add to dashboard Cite

Image Enhancement is one of the most important and complex techniques in image processing technology. The main aim of image enhancement is to improve the visual appearance on an image and to offer a better representation of the image for Computer Vision Algorithms. In this paper,we is covered a few application fields of image enhancement with various images like grayscale, color, infrared and even with videos. The main objective of this paper is to highlight the drawbacks of the state of the art image enhancement techniques.

show abstract

Section: Latest Image Enhancement Techniquesmentioning

confidence: 99%

A Review on Image Enhancement Techniques

2019

View full text Add to dashboard Cite

show abstract

“…So local spatial enhancement is needed with a transformation function devised on neighborhood of pixel. The local transformation function may be called a filter, mask, kernel, template, or window, etc terminology [28,29,30,31]. The construction of the mask is used to convolved every pixel in image to enhance local visual details, including sharpening edges, denoising, deblurring and filter.…”

Section: Intensity Image Enhancementmentioning

confidence: 99%

A Comprehensive and Practical Image Enhancement Method

2019

KSII TIIS

View full text Add to dashboard Cite

Image enhancement is a challenging problem in the field of image processing, especially low-light color images enhancement. This paper proposed a robust and comprehensive enhancement method based several points. First, the idea of bright channel is introduced to estimate the illumination map which is used to attain the enhancing result with Retinex model, and the color constancy is keep as well. Second, in order eliminate the illumination offsets wrongly estimated, morphological closing operation is used to modify the initial estimating illumination. Furthermore, in order to avoid fabricating edges, enlarged noises and over-smoothed visual features appearing in enhancing result, a multi-scale closing operation is used. At last, in order to avoiding the haloes and artifacts presented in enhancing result caused by gradient information lost in previous step, guided filtering is introduced to deal with previous result with guided image is initial bright channel. The proposed method can get good illumination map, and attain very effective enhancing results, including dark area is enhanced with more visual features, color natural and constancy, avoiding artifacts and over-enhanced, and eliminating Incorrect light offsets.

show abstract

“…In contrast, illuminationinvariant representation methods try to estimate and remove unwanted illumination. Retinex-based image enhancement methods, which can recall the visual content of dark regions as well as keep the visual realism, are the mainstream methods (Gao, 2018;Shin, 2015;Chen, 2006). According to the assumption that illumination corresponds to low frequency information, Retinex-based methods estimate illumination information using low-pass Gaussian filter (Shin, 2015;Jobson, 1997) or total variation (TV) normalization model (Chen, 2006;Ng, 2011).…”

Section: Introductionmentioning

confidence: 99%

Adversarial Reconstruction CNN for Illumination-Robust Frontal Face Image Recovery and Recognition

Yang

2021

International Journal of Cognitive Informatics and Natural Intelligence

View full text Add to dashboard Cite

This article proposes an adversarial reconstruction convolution neural network (ARCNN) for non-uniform illumination frontal face image recovery and recognition. The proposed ARCNN includes a reconstruction network and a discriminative network. The authors employ GAN framework to learn the reconstruction network in an adversarial manner. This article integrates gradient loss and perceptual loss terms, which are able to preserve the detailed and spatial structure image information, into the overall reconstruction loss function to constraint the reconstruction procedure. Experiments are conducted on the typical illumination-sensitive dataset, extended YaleB dataset. The reconstructed results demonstrate that the proposed ARCNN approach can remove the illumination and shadow information and recover natural uniform illuminated face image from non-uniform illuminated ones. Face recognition results on the extended YaleB dataset demonstrate that the proposed ARCNN reconstruction procedure can also preserve the discriminative information of face image for classification task.

show abstract

Naturalness Preserved Nonuniform Illumination Estimation for Image Enhancement Based on Retinex

Cited by 88 publications

References 27 publications

A Review on Image Enhancement Techniques

A Review on Image Enhancement Techniques

A Comprehensive and Practical Image Enhancement Method

Adversarial Reconstruction CNN for Illumination-Robust Frontal Face Image Recovery and Recognition

Contact Info

Product

Resources

About