Center-Emphasized Visual Saliency and a Contrast-Based Full Reference Image Quality Index

Layek, Md. Abu; Uddin, Azher; Le, Tuyen P.; Chung, T. J.; Huh, Eui‐Nam

doi:10.3390/sym11030296

Cited by 13 publications

(3 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This bias is widely used in SRD, but is rare in image-quality estimation. Layek et al [66] propose an image quality metric that uses visual saliency and contrast, and extra attention is paid to the center by increasing the sensitivity of the similarity maps between the reference image and the distorted image. In their study, an image is splitted into 3 × 3 blocks, as shown in Figure 2D, and, to enhance the prediction performance, not only the center region but also its whole image are quantified from saliency similarity and contrast similiarty.…”

Section: Discussionmentioning

confidence: 99%

Salient Region Guided Blind Image Sharpness Assessment

Liu

Zhao

et al. 2021

Sensors

View full text Add to dashboard Cite

Salient regions provide important cues for scene understanding to the human vision system. However, whether the detected salient regions are helpful in image blur estimation is unknown. In this study, a salient region guided blind image sharpness assessment (BISA) framework is proposed, and the effect of the detected salient regions on the BISA performance is investigated. Specifically, three salient region detection (SRD) methods and ten BISA models are jointly explored, during which the output saliency maps from SRD methods are re-organized as the input of BISA models. Consequently, the change in BISA metric values can be quantified and then directly related to the difference in BISA model inputs. Finally, experiments are conducted on three Gaussian blurring image databases, and the BISA prediction performance is evaluated. The comparison results indicate that salient region input can help achieve a close and sometimes superior performance to a BISA model over the whole image input. When using the center region input as the baseline, the detected salient regions from the saliency optimization from robust background detection (SORBD) method lead to consistently better score prediction, regardless of the BISA model. Based on the proposed hybrid framework, this study reveals that saliency detection benefits image blur estimation, while how to properly incorporate SRD methods and BISA models to improve the score prediction will be explored in our future work.

show abstract

Section: Discussionmentioning

confidence: 99%

Salient Region Guided Blind Image Sharpness Assessment

Liu

Zhao

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Our proposed algorithm was compared to several state-of-the-art FR-IQA metrics, including SSIM [12], MS-SSIM [14], MAD [49], GSM [56], HaarPSI [20], MDSI [57], CSV [58], GMSD [19], DSS [59], VSI [60], PerSIM [61], BLeSS-SR-SIM [62], BLeSS-FSIM [62], BLeSS-FSIMc [62], LCSIM1 [40], ReSIFT [63], IQ(L T ) [28], MS-UNIQUE [64], RVSIM [65], 2stepQA [66], SUMMER [67], CEQI [68], CEQIc [68], VCGS [69], and DISTS [70], whose original source code are available online. Moreover, we reimplemented SSIM CNN [41] in MATLAB R2019a (Available : https://github.com/Skythianos/ SSIM-CNN).…”

Section: Comparison To the State-of-the-artmentioning

confidence: 99%

A Combined Full-Reference Image Quality Assessment Method Based on Convolutional Activation Maps

Varga

2020

Algorithms

View full text Add to dashboard Cite

The goal of full-reference image quality assessment (FR-IQA) is to predict the perceptual quality of an image as perceived by human observers using its pristine (distortion free) reference counterpart. In this study, we explore a novel, combined approach which predicts the perceptual quality of a distorted image by compiling a feature vector from convolutional activation maps. More specifically, a reference-distorted image pair is run through a pretrained convolutional neural network and the activation maps are compared with a traditional image similarity metric. Subsequently, the resulting feature vector is mapped onto perceptual quality scores with the help of a trained support vector regressor. A detailed parameter study is also presented in which the design choices of the proposed method is explained. Furthermore, we study the relationship between the amount of training images and the prediction performance. Specifically, it is demonstrated that the proposed method can be trained with a small amount of data to reach high prediction performance. Our best proposal—called ActMapFeat—is compared to the state-of-the-art on six publicly available benchmark IQA databases, such as KADID-10k, TID2013, TID2008, MDID, CSIQ, and VCL-FER. Specifically, our method is able to significantly outperform the state-of-the-art on these benchmark databases.

show abstract

“…Our proposed algorithm was compared to several state-of-the-art FR-IQA metrics, including SSIM [12], MS-SSIM [14], MAD [46], GSM [53], HaarPSI [20], MDSI [54], CSV [55], GMSD [19], DSS [56], VSI [57], PerSIM [58], BLeSS-SR-SIM [59], BLeSS-FSIM [59], BLeSS-FSIMc [59], LCSIM1 [39], ReSIFT [60], MS-UNIQUE [61], RVSIM [62], 2stepQA [63], SUMMER [64], CEQI [65], CEQIc [65], VCGS [66], and DISTS [67], whose original source code are available online. Moreover, we reimplemented SSIM CNN [40] in MATLAB R2019a 1 .…”

Section: Comparison To the State-of-the-artmentioning

confidence: 99%

A combined full-reference image quality assessment approach based on convolutional activation maps

Varga

2020

Preprint

View full text Add to dashboard Cite

The goal of full-reference image quality assessment (FR-IQA) is to predict the quality of an image as perceived by human observers with using its pristine, reference counterpart. In this study, we explore a novel, combined approach which predicts the perceptual quality of a distorted image by compiling a feature vector from convolutional activation maps. More specifically, a reference-distorted image pair is run through a pretrained convolutional neural network and the activation maps are compared with a traditional image similarity metric. Subsequently, the resulted feature vector is mapped onto perceptual quality scores with the help of a trained support vector regressor. A detailed parameter study is also presented in which the design choices of the proposed method is reasoned. Furthermore, we study the relationship between the amount of training images and the prediction performance. Specifically, it is demonstrated that the proposed method can be trained with few amount of data to reach high prediction performance. Our best proposal -ActMapFeat -is compared to the state-of-the-art on six publicly available benchmark IQA databases, such as KADID-10k, TID2013, TID2008, MDID, CSIQ, and VCL-FER. Specifically, our method is able to significantly outperform the state-of-the-art on these benchmark databases. This paper is accompanied by the source code of the proposed method: https://github.com/Skythianos/FRIQA-ActMapFeat.

show abstract

Center-Emphasized Visual Saliency and a Contrast-Based Full Reference Image Quality Index

Cited by 13 publications

References 39 publications

Salient Region Guided Blind Image Sharpness Assessment

Salient Region Guided Blind Image Sharpness Assessment

A Combined Full-Reference Image Quality Assessment Method Based on Convolutional Activation Maps

A combined full-reference image quality assessment approach based on convolutional activation maps

Contact Info

Product

Resources

About