Quality assessment of images with multiple distortions based on phase congruency and gradient magnitude

Miao, Xin; Chu, Hairong; Liu, Hui; Yang, Yanxi; Li, Xiaolong

doi:10.1016/j.image.2019.08.013

Cited by 11 publications

(16 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Miao et. al [45] have proposed another method based on LBP in Phase Congruency domain to take into account low level features. It also employs gradient magnitude as a weighing factor.…”

Section: Related Workmentioning

confidence: 99%

Frequency-domain blind quality assessment of blurred and blocking-artefact images using Gaussian Process Regression model

Viqar

Moinuddin

Khan

et al. 2022

Signal Processing: Image Communication

View full text Add to dashboard Cite

“…Miao et. al [45] have proposed another method based on LBP in Phase Congruency domain to take into account low level features. It also employs gradient magnitude as a weighing factor.…”

Section: Related Workmentioning

confidence: 99%

Frequency-domain blind quality assessment of blurred and blocking-artefact images using Gaussian Process Regression model

Viqar

Moinuddin

Khan

et al. 2022

Signal Processing: Image Communication

View full text Add to dashboard Cite

“…𝑃𝐶 𝑚 = max(PC 1 , PC 2 ) is the original and augmented image's maximum local phase consistency matrix [28].…”

Section: 𝑘(𝑔 𝑔mentioning

confidence: 99%

Image Improvement via Fuzzy Inference Centered Contextual-Dissimilarity Histogram Equalization

2022

IJIES

View full text Add to dashboard Cite

To solve the drawbacks of existing picture enhancement methods which include the error that occur in the pixel intensity expression, a fuzzy inference-centered contextual-dissimilarity histogram equalization (FICDHE) approach is suggested. Proposed system has three components. The first module generates intensity membership functions based on predicted intensity intervals. Second, fuzzy inference techniques are constructed and contextualdissimilarity of every pixel is determined. In the last module, dissimilarity histograms are cut and leveled. The fuzzy system in this study not only addresses the uncertainty source of pixel gray level expression, but is also adaptable. In this approach, the fuzzy inference system membership function parameter is automatically selected depending on picture pixel gray. Its versatility creates an algorithm that is more accessible. Two typical medical photos from BrainWeb are used in experiments. Performance of the suggested technique was compared against a number of enhancement algorithms based on subjective evaluation and picture quality measurement indexes. Experiments show that the proposed method is superior to the other algorithms in terms of improving the contrast improvement index (CII) (03.603), the peak signal-to-noise ratio (PSNR) (22.877), the entropy (E) (06.781), the enhancement measures (EME and EMEE) (56.688 and 3344.63 respectively), the quality index based on local variance (QILV) (00.965), and the feature similarity index (FSIM Index) (00.905).

show abstract

“…The best filter parameters can be obtained in [2][3][4][5]31,35]. Finally, from the contrast threshold we reach the error sensitivity that can be expressed as follow:…”

Section: Conceiving and Configuring A Wavelet Foveal Filtermentioning

confidence: 99%

“…To attain this objective, one may find in the literature a series of "Image Fidelity Assessor" (IFA) [21][22][23][24]. Based on the "Visible Difference Predictor" VDP [25] and its wavelet-based variant WVDP [26][27][28][29][30][31][32][33][34][35], we suggest a psycho-visual assessor that integrates the "Human Visual System HVS" quality criterions [31][32][33]. Our foveal metric named thereafter FWVDP adapts the original wavelet coefficients, provides a foveal score FPS aiming to predict the probability of detecting the coder introduced errors.…”

Section: Introductionmentioning

confidence: 99%

A Psychovisual Optimization of Wavelet Foveation-Based Image Coding and Quality Assessment Based on Human Quality Criterions

Bajit¹,

Nahid²,

Tamtaoui³

et al. 2020

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

In the present article, we introduce a foveation-based optimized embedded and its optimized version image coders thereafter called VOEFIC/MOEFIC and its related foveation wavelet visible difference predictor FWVDP coding quality metric. It advances a visually advanced foveal weighting mask that regulates the wavelet-based image spectrum before its encoding by the SPIHT encoder. It intends to arrive at a destined compression rate with a significant quality improvement for a disposed of binary budget, witnessing separation, and a foveal locale that locates the object in the zone of concern ROI. The coder embodies a couple of masking achieves build on the human psycho-visual quality criteria. Hence, the coder administers the foveal model to weigh the source wavelets samples, reshapes its spectrum content, adapts its shape, discards or somewhat shrinks the redundant excess and finally enhances the visual quality. The foveal weighting mask is computed indoors wavelet subbands as come after. First, it administers the foveal wavelet-based filter depending on the intention point so that it removes or at least reduces the imperceptible frequencies around the zone of concern. Next, it augments the picture contrast according to wavelet JND thresholds to manage brightening and nice the contrast above the distortion just notable. Once refined, the weighted wavelet spectrum will be embedded coded using the standard SPIHT to reach a desired binary bit budget. The manuscript also advances a foveationbased objective quality evaluator that embodies a psycho-visual quality criterion identified with the visual cortex framework. This investigator furnishes a foveal score FPS having the power of detecting probable errors and measuring objectively the compression quality. Keep in mind that the foveal coder VOEFIC and its visually upgraded variant MOEFIC, have similar complexity as their reference SPIHT. In contrast, their gathered data highlight the visual coding advancement and the boost ratio purchased in its quality gain.

show abstract

Quality assessment of images with multiple distortions based on phase congruency and gradient magnitude

Cited by 11 publications

References 49 publications

Frequency-domain blind quality assessment of blurred and blocking-artefact images using Gaussian Process Regression model

Frequency-domain blind quality assessment of blurred and blocking-artefact images using Gaussian Process Regression model

Image Improvement via Fuzzy Inference Centered Contextual-Dissimilarity Histogram Equalization

A Psychovisual Optimization of Wavelet Foveation-Based Image Coding and Quality Assessment Based on Human Quality Criterions

Contact Info

Product

Resources

About