Saliency-induced reduced-reference quality index for natural scene and screen content images
Introduction
The quick advancements of transmission technologies have boosted various remote applications such as telecommuting and cloud computing, which bring massive computer-generated content called “screen content”. The so-called screen content has some distinctive characteristics different from natural scene because of the contained computer generated content, e.g., texts, icons, tables, graphics, etc. Those distinctive characteristics which sometimes violate natural scene statistics (NSS) cause some failures in traditional natural scene image (NSI) based applications. Hence some specialized technologies for screen content image (SCI) have been proposed, such as screen content video compression [1].
The booming of screen content also calls for SCI-specific image quality measures. Limited work has been done concerning SCI quality assessment (QA). In [2], the authors constructed a screen image quality assessment database (SIQAD), which shows that state-of-the-art image quality assessment (IQA) measures do not work efficiently for SCIs. It is reasonable since current IQA measures are implicitly designed for NSIs and somehow rely on NSS. Wang et al. [3] also constructed a database called quality assessment of compressed SCI (QACS). In [4], the authors proposed a full-reference (FR) saliency-guided quality measure named SQMS for SCI. SQMS exploits gradient magnitude similarity as the quality map, which is then weighted by a specific saliency map. Gu et al. [5], [6] learned blind quality evaluation engines for SCI from a huge group of SCIs and corresponding objective quality scores calculated by FR measures.
Although dozens of NSI quality estimators [7], [8], [9], [10], [11], [12], [13], [14], [15], [16] and several limited SCI quality measures [2], [3], [4], [5], [6] are proposed, they are either implicitly designed for NSIs or specifically developed for SCIs. Only very few quality measures can work for NSIs and SCIs simultaneously. Min et al. [17] proposed a blind blockiness measure which works for JPEG compressed NSIs and SCIs. Xu et al. [18] developed a measure for NSIs and SCIs. In [19], Min et al. constructed a cross-content-type database, and proposed a unified content-type adaptive blind IQA measure for compressed natural, graphic and screen content images. In practical multimedia communication systems, we may encounter both types of images, and sometimes we do not have any prior knowledge about the image types. Efficient general quality measures ignoring image types are highly needed in such circumstances. In this paper, we extract quality features efficient for both types of images and develop a general reduced-reference (RR) quality measure without any explicit image type classification.
The proposed method is based on visual saliency detection. Visual saliency detection is an important research topic in areas of psychology, image processing and computer vision [20]. Visual attention and quality assessment are two closely related research topics [7], [9], [10], [12], [13], [21], [22], [23]. Quality degradation can influence visual attention [21]. Contrarily, visually salient positions should be more carefully processed since subjects judge the image quality according to the observations of some limited positions, and a typical use of visual attention model is to optimize resource allocation and improve the perceptual quality under the constraints of bandwidth [24], [25], [26], [27], [28].
Motivated by the interaction between visual attention and quality assessment, some researchers used visual attention map as a weighting map during the quality pooling stage of IQA [9], [10], [13], [22], [23]. Min et al. [13] collected some visual attention data for main-stream IQA databases. Zhang et al. [22], [23] studied the use of saliency model in objective quality assessment models. Liu et al. [9] used the saliency map to highlight the visually salient areas. Besides highlighting the salient regions, Saha and Wu [10] used the dissimilarity between saliency maps of the reference and distorted images to highlight the more distorted image content. Besides visual attention maps, some measures utilize other kinds of weighting maps such as phase congruency map [7] and gradient magnitude map [12]. Although without explicit visual attention prediction or visual saliency detection processes, such kinds of weighting maps have also highlighted the visually salient positions, which can be also deemed as one kind of visual saliency.
Instead of as a weighting map, visual saliency can be also used as a quality feature since quality degradation can strongly affect saliency detection. Zhang et al. [8] proposed a FR IQA method named VSI by measuring the similarity between the reference image’s and the distorted image’s visual saliency. VSI is a FR measure since it utilizes not only saliency, but also the gradient magnitude and chrominance. All extracted feature maps have the same resolution as the reference image. Actually, deriving a gray scale saliency map from a color image is an operation of dimension reduction, which motivates us to develop a saliency-induced reduced-reference (SIRR) IQA measure.
SIRR detects saliency map of the reference image as the reference data, and then measure the similarity between the reference and distorted images’ saliency maps. We try to reduce the reference data from two aspects. First, we down-sample the reference image to a coarser scale to detect saliency, whose resolution is only one over sixty-four of the original resolution. We take full advantage of such down-sampling operation to reduce the reference data. Second, we exploit a binary image descriptor called “image signature” [29] to detect image saliency. The image saliency is represented by the binary image signature, which also significantly reduces the reference data. The final quality is described by the similarity between two images’ saliency maps. In this work, the similarity is evaluated by the classical image fidelity measure SSIM [30].
We perform extensive experiments to test the proposed SIRR in both NSIs and SCIs. Five large-scale NSI QA databases and two recent SCI QA databases are used. Among the NSI databases, LIVE [31], TID2008 [32] and CSIQ [33] are general-purpose IQA databases, whereas LIVEMD [34] focuses on multiply distorted images and CID2013 [35] consists of contrast changed images. As it to the SCI databases, SIQAD [2] is a general-purpose one and QACS [3] concentrates on compressed SCIs. The all seven databases can give an overall description of both NSIs and SCIs. As will be presented in the experiments part, the proposed SIRR is efficient for both types of images. SIRR can be comparable to or outperform state-of-the-art FR and RR IQA measures on all seven IQA databases.
The remainder of this paper is organized as follows. Section 2 describes the proposed saliency-induced reduced-reference quality measure. Experimental results are given in Section 3. We compare the proposed method with state-of-the-art FR and RR quality measures in this section. Section 4 concludes this paper.
Section snippets
Saliency-induced reduced-reference quality measure
As described in Section 1, visual saliency has been widely used in IQA, but it is generally used as a weighting map during the final pooling. Few work has considered saliency as a quality feature. Most bottom-up saliency models highly rely on the low-level features, which are sensitive to quality degradation. Fig. 1 illustrates the influence of quality degradation on image saliency. From this figure, we can observe that perceptible quality degradation can cause perceptible change of image
Validation of SIRR
The proposed SIRR measure is validated on both NSIs and SCIs. The details are as follows.
Conclusion
In this paper, we propose a saliency-induced reduced-reference (SIRR) IQA measure for the two most common but quite different types of images encountered in realistic multimedia communication systems, i.e., NSI and SCI. We develop SIRR based on the observations that quality degradation can significantly affect saliency detection, and that saliency detection is in fact an operation of dimension and data reduction. SIRR evaluates quality by measuring the similarity between two images’ saliency
Acknowledgments
This work was supported in part by National Natural Science Foundation of China under grants 61422112, 61371146, 61521062, and 61527804.
References (50)
- et al.
Learning a blind quality evaluation engine of screen content images
Neurocomputing
(2016) - et al.
Stereoscopic image quality assessment method based on binocular combination saliency model
Signal Process.
(2016) - et al.
Full-reference image quality assessment by combining global and local distortion measures
Signal Process.
(2016) - et al.
Biologically inspired image quality assessment
Signal Process.
(2016) - et al.
Visual attention analysis and prediction on human faces
Inf. Sci.
(2017) - et al.
Perceptual visual quality metrics: a survey
J. Vis. Commun. Image Represent.
(2011) - et al.
Reduced-reference quality assessment of screen content images
IEEE Trans. Circuits Syst. Video Technol.
(2016) - et al.
Overview of the emerging HEVC screen content coding extension
IEEE Trans. Circuits Syst. Video Technol.
(2016) - et al.
Perceptual quality assessment of screen content images
IEEE Trans. Image Process.
(2015) - et al.
Subjective and objective quality assessment of compressed screen content images
IEEE J. Emerging Sel. Top. Circuits Syst.
(2016)
Saliency-guided quality assessment of screen content images
IEEE Trans. Multimedia
No-reference quality assessment of screen content pictures
IEEE Trans. Image Process.
FSIM: a feature similarity index for image quality assessment
IEEE Trans. Image Process.
VSI: a visual saliency-induced index for perceptual image quality assessment
IEEE Trans. Image Process.
A fast reliable image quality predictor by fusing micro-and macro-structures
IEEE Trans. Ind. Electron.
Visual attention data for image quality assessment databases
Proceedings of the IEEE International Symposium on Circuits and Systems
A psychovisual quality metric in free-energy principle
IEEE Trans. Image Process.
Cross-dimensional perceptual quality assessment for low bit-rate videos
IEEE Trans. Multimedia
Model-based referenceless quality metric of 3d synthesized images using local image description
IEEE Trans. Image Process.
Blind quality assessment of compressed images via pseudo structural similarity
Proceedings of the IEEE International Conference on Multimedia and Expo
Blind image quality assessment based on high order statistics aggregation
IEEE Trans. Image Process.
Unified blind quality assessment of compressed natural, graphic, and screen content images
IEEE Trans. Image Process.
State-of-the-art in visual attention modeling
IEEE Trans. Pattern Anal. Mach. Intell.
Influence of compression artifacts on visual attention
Proceedings of the IEEE International Conference on Multimedia and Expo
The application of visual saliency models in objective image quality assessment: a statistical evaluation
IEEE Trans. Neural Netw. Learn. Syst.
Cited by (73)
PGF-BIQA: Blind image quality assessment via probability multi-grained cascade forest
2023, Computer Vision and Image UnderstandingAutomatic No-Reference kidney tissue whole slide image quality assessment based on composite fusion models
2023, Biomedical Signal Processing and ControlCitation Excerpt :The objective image quality assessment (IQA) approaches that can be embedded in the intelligent system are divided into three groups involving full-reference (FR), reduced-reference (RR), and no-reference (NR) [6], according to whether distortion-free images are used as reference. FR-IQA [7] and RR-IQA models [8] evaluated the visual quality by calculating the difference between the reference image and the distorted image. They are reliable, but the application scenarios are extremely limited due to the reference information is difficult to obtain.
A brain-inspired computational model for extremely few reference image quality assessment
2023, DisplaysCitation Excerpt :Generally speaking, brain-like intelligence is regarded to be the ultimate goal of artificial intelligence, and has become the focus of attention from countries around the world. Researchers on brain science have obtained lots of important findings, including free energy principle [1,2] and attention perception mechanism [3,4], causing various types of applications in the fields of image and video processing [5–12], vision perception and understanding [13–23], environment monitoring and protection [24–29], etc. Among them, those above two findings play a particularly significant role in the field of image quality assessment (IQA).
Blind light field image quality assessment by analyzing angular-spatial characteristics
2021, Digital Signal Processing: A Review JournalCitation Excerpt :Saad et al. [21] developed an IQA method that the generalized Gaussian density function was modeled by block Discrete Cosine Transform (DCT) coefficients in BLind Image Integrity Notator using DCT Statistics (BLIINDS-II). Min et al. [22] designed a Pseudo-Reference Image (PRI) and integrated the PRI-based distortion-specific metrics into a general-purpose blind IQA method named Blind PRI-based (BPRI) metric. Moreover, they utilized Multiple Pseudo Reference Images (MPRIs) by further degrading the distorted image in several ways and to certain degrees, and then compared the similarities between the distorted image and the MPRIs [23].