New Mean-Variance Gamma Method for Automatic Gamma Correction

Mahamdioua, Meriama; Bahaj, Mohamed

doi:10.5815/ijigsp.2017.03.05

Cited by 10 publications

(12 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this phase, we used the same steps described in [8]: Gaussian filter, gamma estimation, and the TT method [35]. However, we used the MV-gamma method proposed in [33] for gamma estimation for its calculation efficiency:…”

Section: Pre-processing Methodmentioning

confidence: 99%

“…In the literature, we were able to find only a few methods that automatically estimate inverse gamma values based on image quality. In this study, we used our method mean‐variance gamma (MV‐Gamma) which we proposed in [33], for its calculation efficiency. The value gamma =

γ^{- 1}

is then used to apply gamma correction to the compared images.…”

Section: Proposed Methodsmentioning

confidence: 99%

“…In this phase, we used the same steps described in [8]: Gaussian filter, gamma estimation, and the TT method [35]. However, we used the MV‐gamma method proposed in [33] for gamma estimation for its calculation efficiency: Gaussian smoothing: First, a Gaussian smoothing filter is applied to the compared images to reduce their sensitivity to noise. …”

Section: Proposed Methodsmentioning

confidence: 99%

“…The remainder of the pre‐processing steps are based on the TT method [35]: Correction of gamma by automatic estimation of gamma value : In the TT method proposed in [35], the gamma correction step used a fixed value of gamma, but this value is related to image deformation. In our method, we use the MV‐gamma method proposed in [33] to automatically estimate the gamma value according to image illumination quality. DoG filter: The operation

I false\leftarrow I * DoG

is applied to remove illumination variations. Normalisation : The images resulting from these pre‐processing stages contain a small mixture of extreme values produced by reflections, useless pixels at the edges of the images, and small dark areas, such as in the nostrils. For this reason, it is necessary to apply a normalisation operation to standardise a robust measurement of overall contrast or intensity variation and reduce the influence of extreme values on the subsequent stages of processing [35].…”

Section: Proposed Methodsmentioning

confidence: 99%

See 3 more Smart Citations

Automatic adaptation of SIFT for robust facial recognition in uncontrolled lighting conditions

Mahamdioua

Bahaj²

2018

IET Computer Vision

Self Cite

View full text Add to dashboard Cite

The scale invariant feature transform (SIFT), which was proposed by David Lowe, is a powerful method that extracts and describes local features called keypoints from images. These keypoints are invariant to scale, translation, and rotation, and partially invariant to image illumination variation. Despite their robustness against these variations, strong lighting variation is a difficult challenge for SIFT-based facial recognition systems, where significant degradation of performance has been reported. To develop a robust system under these conditions, variation in lighting must be first eliminated. Additionally, SIFT parameter default values that remove unstable keypoints and inadequately matched keypoints are not well-suited to images with illumination variation. SIFT keypoints can also be incorrectly matched when using the original SIFT matching method. To overcome this issue, the authors propose propose a method for removing the illumination variation in images and correctly setting SIFT's main parameter values (contrast threshold, curvature threshold, and match threshold) to enhance SIFT feature extraction and matching. The proposed method is based on an estimation of comparative image lighting quality, which is evaluated through an automatic estimation of gamma correction value. Through facial recognition experiments, the authors find significant results that clearly illustrate the importance of the proposed robust recognition system.

show abstract

Section: Pre-processing Methodmentioning

confidence: 99%

γ^{- 1}

is then used to apply gamma correction to the compared images.…”

Section: Proposed Methodsmentioning

confidence: 99%

Section: Proposed Methodsmentioning

confidence: 99%

I false\leftarrow I * DoG

Section: Proposed Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Automatic adaptation of SIFT for robust facial recognition in uncontrolled lighting conditions

Mahamdioua

Bahaj²

2018

IET Computer Vision

Self Cite

View full text Add to dashboard Cite

show abstract

“…The low-contrast effect reduces the visual quality of an image and thus, it should be handled properly to provide acceptable quality for digital images [9]. Hence, it is desired to redistribute the intensities of a given image to the entire dynamic range in order to improve its contrast and provide an adequate representation for its information [10].…”

Section: Introductionmentioning

confidence: 99%

A Low-Complexity Algorithm for Contrast Enhancement of Digital Images

Al-Ameen¹,

Hasan²

2018

IJIGSP

View full text Add to dashboard Cite

As known, the contrast is a highly important feature by which the visual quality of digital images can be judged as adequate or poor. Hence, many methods exist for contrast enhancement, where the complexity of those methods usually varies due to the utilization of different concepts. In this article, a simple yet efficient algorithm is introduced for contrast enhancement of digital images. The proposed algorithm consists of four distinct stages: In the first stage, the hyperbolic sine function is applied to provide a simple contrast modification. In the second stage, a modified power-law function is utilized to control the amount of contrast adjustment. In the third stage, the standard sigmoid function is used to remap the image pixels into an "S" shape, which can provide further contrast enhancement. In the final stage, a contrast stretching function is applied to remap the image pixels into their natural dynamic range. The performed computer experiments on different low-contrast images demonstrated the efficiency of the proposed algorithm in processing synthetic and real degraded images, as it provided better and clearer results when compared to several existing contrast enhancement algorithms. To end with, the proposed algorithm can be used as a contrast processing step in many image-related applications.

show abstract