Revisiting Image Vignetting Correction by Constrained Minimization of Log-Intensity Entropy

Lopez-Fuentes, Laura; Oliver, Gabriel; Massanet, Sebastià

doi:10.1007/978-3-319-19222-2_38

Cited by 19 publications

(24 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…NUIC is typically applied to remote sensing images [10], optical microscope images [11], aerial images [12], and scanned images [13]. Also, there is a need to remove vignetting in optical photography [14,15]. According to the image enhancement model, the traditional NUIC algorithms are mainly included three categories: incident-reflection multiplicative model, mathematical-statistical model, and light additive model.…”

Section: Related Workmentioning

confidence: 99%

NUICNet: Non-Uniform Illumination Correction for Underwater Image Using Fully Convolutional Network

et al. 2020

View full text Add to dashboard Cite

Absorption and scattering in aqueous media would attenuate light and make imaging difficult. Therefore, an artificial light source is usually utilized to assist imaging in the deep ocean. However, the artificial light source typically alters the light conditions to a large extent, resulting in the non-uniform illumination of images. To solve this problem, we propose a non-uniform illumination correction algorithm based on a fully convolutional network for underwater images. The proposed algorithm model the original image as the addition of the ideal image and a non-uniform light layer. We replace the traditional pooling layer with dilated convolution to expand the receptive field and achieve higher accuracy in non-uniform illumination recognition. To improve the perception ability of the network effectively, the original image and parameters which pre-trained on the ImageNet are concentrated. The concentrated information is used as input to the network. Due to the color shift and blurred details of the underwater image, we design the novel loss function, which includes three parts of feature loss, smooth loss, and adversarial loss. Moreover, we built a dataset of the underwater image with non-uniform illumination. Experiments show that our method performs better in subjective assessment and objective assessment than some traditional methods.

show abstract

Section: Related Workmentioning

confidence: 99%

NUICNet: Non-Uniform Illumination Correction for Underwater Image Using Fully Convolutional Network

et al. 2020

View full text Add to dashboard Cite

show abstract

“…While various calibration methods have been proposed in the vignetting literature, such methods require that users follow complicated observation procedures in controlled environments, e.g., taking reference images with known camera configurations to estimate vignetting functions [2], [5]. Goldman [1] showed that vignetting functions can be estimated without special calibration objects or controlled lighting conditions; however, his method still requires multiple images with the same vignetting pattern.…”

Section: Related Workmentioning

confidence: 99%

Variational-Bayesian Single-Image Devignetting

Sonogashira

Iiyama

Minoh

2018

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Vignetting is a common type of image degradation that makes peripheral parts of an image darker than the central part. Singleimage devignetting aims to remove undesirable vignetting from an image without resorting to calibration, thereby providing high-quality images required for a wide range of applications. Previous studies into single-image devignetting have focused on the estimation of vignetting functions under the assumption that degradation other than vignetting is negligible. However, noise in real-world observations remains unremoved after inversion of vignetting, and prevents stable estimation of vignetting functions, thereby resulting in low quality of restored images. In this paper, we introduce a methodology of image restoration based on variational Bayes (VB) to devignetting, aiming at high-quality devignetting in the presence of noise. Through VB inference, we jointly estimate a vignetting function and a latent image free from both vignetting and noise, using a general image prior for noise removal. Compared with state-of-the-art methods, the proposed VB approach to single-image devignetting maintains effectiveness in the presence of noise, as we demonstrate experimentally.

show abstract

“…Amaral's algorithm employs background subtraction for illumination compensation, and subsequent histogram based thresholding with morphological filtering [9]. Compensation of Uneven illumination (vignetting) is suggested based on filtering and illumination modeling, but without segmentation based assessment in literature [16][17][18]. Our algorithm is different from others in the sense that it is segmentation centric, and usually, there is only one parameter in the threshold is to be adjusted iteratively with range [0,1].…”

Section: Introductionmentioning

confidence: 99%

Illumination Compensated Segmentation of Microscopic Images of Activated Sludge Flocs

Khan¹,

Nisar²,

Ng³

et al. 2015

2015 International Conference on Digital Image Computing: Techniques and Applications (DICTA)

View full text Add to dashboard Cite

Image processing and analysis is a useful tool for monitoring of activated sludge wastewater treatment plant. However its effectiveness is dependent on performance of the segmentation algorithms. The activated sludge plant is monitored by image processing and analysis of images acquired through trinocular microscope. The sample observed under microscope is collected from aeration tank of the plant. In this paper, a segmentation technique with integrated illumination compensation is proposed for the microscopic images of the activated sludge samples. The illumination noise was modeled and estimated as Gaussian distribution symmetric about a threshold value determined by global Otsu thresholding algorithm. The performance of the algorithm was evaluated using time required for segmentation, Rand index, accuracy and quantification of flocs. In order to compare with the state-of-theart algorithms, gold approximations of ground truth images were manually prepared. The performance was assessed by combining the evaluation metrics in an integrated perspective. The proposed algorithm exhibits better performance in terms of both integrated and non-integrated perspectives.

show abstract

Revisiting Image Vignetting Correction by Constrained Minimization of Log-Intensity Entropy

Cited by 19 publications

References 14 publications

NUICNet: Non-Uniform Illumination Correction for Underwater Image Using Fully Convolutional Network

NUICNet: Non-Uniform Illumination Correction for Underwater Image Using Fully Convolutional Network

Variational-Bayesian Single-Image Devignetting

Illumination Compensated Segmentation of Microscopic Images of Activated Sludge Flocs

Contact Info

Product

Resources

About