- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Blind Image Quality Assessment Using A Deep Bilinear Convolutional Neural Network
摘要： We propose a deep bilinear model for blind image quality assessment (BIQA) that works for both synthetically and authentically distorted images. Our model constitutes two streams of deep convolutional neural networks (CNN), specializing in the two distortion scenarios separately. For synthetic distortions, we first pre-train a CNN to classify the distortion type and level of an input image, whose ground truth label is readily available at a large scale. For authentic distortions, we make use of a pre-train CNN (VGG-16) for the image classification task. The two feature sets are bilinearly pooled into one representation for a final quality prediction. We fine-tune the whole network on target databases using a variant of stochastic gradient descent. Extensive experimental results show that the proposed model achieves state-of-the-art performance on both synthetic and authentic IQA databases. Furthermore, we verify the generalizability of our method on the large-scale Waterloo Exploration Database, and demonstrate its competitiveness using the group maximum differentiation competition methodology.
关键词： Blind image quality assessment,convolutional neural networks,bilinear pooling,perceptual image processing,gMAD competition
更新于2025-09-09 09:28:46
-
Full-Reference Image Quality Assessment by Combining Features in Spatial and Frequency Domains
摘要： Objective employs mathematical and computational theory to objectively assess the quality of output images based on the human visual system (HVS). In this paper, a novel approach based on multifeature extraction in the spatial and frequency domains is proposed. We combine the gradient magnitude and phase congruency maps to generate a local structure (LS) map, which can perceive local structural distortions. The LS matches well with HVS and highlights differences with details. For complex visual information, such as texture and contrast sensitivity, we deploy the log-Gabor filter, and spatial frequency, respectively, to effectively capture their variations. Moreover, we employ the random forest (RF) to overcome the limitations of existing pooling methods. Compared with support vector regression, RF can obtain better prediction results. Extensive experimental results on the five benchmark databases indicate that the proposed method precedes all the state-of-the-art image quality assessment metrics in terms of prediction accuracy. In addition, the proposed method is in compliance with the subjective evaluations.
关键词： log-Gabor filter,random forest (RF),contrast sensitivity function (CSF),full-reference,Image quality assessment (IQA)
更新于2025-09-09 09:28:46
-
Blind Image Quality Assessment with Semantic Information
摘要： No-reference (NR) image quality assessment (IQA) aims to evaluate the quality of an image without reference image, which is greatly desired in the automatic visual signal processing system. Distortions degrade the visual contents and affect the semantics acquisition during the process of human perception. Although the existing methods evaluate the quality of images based on the structure, texture, or statistical characteristics, and deliver high quality prediction accuracy, they do not take the spatial semantics into account. From the perspective of human perception, distortions decrease the structural semantics that represent the structural information, and disturb the spatial semantics that describe the contents of images. Therefore, we attempt to measure the image quality by its degradation of semantics in an image. To extract the semantics of an image, a semantic network is proposed. The network contains convolutional neural networks (CNN) and Long Short-Term Memory (LSTM) that correspond to structural semantics and spatial semantics, respectively. CNN can be regarded as a coarse imitation of human visual mechanism to obtain the structural information, and LSTM can express the contents of an image. Then, by measuring the degradations of different semantics on images, a novel NR IQA is introduced. The proposed approach is evaluated on the databases of LIVE, CSIQ, TID2013, and LIVE multiply distorted database as well as LIVE in the wild image quality challenge database, and the results show superior performance to other state-of-the-art NR IQA methods. Furthermore, we explore the generalization capability of the proposed approach, and the experimental results indicate the proposed approach has a high robustness.
关键词： spatial semantics,No-reference image quality assessment,structural semantics,human perception,semantic network
更新于2025-09-09 09:28:46
-
Blind Image Quality Assessment Based on Joint Log-Contrast Statistics
摘要： During recent years, quality-aware features extracted from natural scene statistics (NSS) models have been used in development of blind image quality assessment (BIQA) algorithms. Generally, the univariate distributions of bandpass coefficients are used to fit a parametric probabilistic model and the model parameters serve as the quality-aware features. However, the inter-location, inter-direction and inter-scale correlations of natural images cannot be well exploited by such NSS models, as it is hard to capture such dependencies using univariate marginal distributions. In this paper, we build a novel NSS model of joint log-contrast distribution to take into account the across space and direction correlations of natural images (inter-scale correlation to be explored as the next step). Furthermore, we provide a new efficient approach to extract quality-aware features as the gradient of log-likelihood on the NSS model, instead of using model parameters directly. Finally, we develop an effective joint-NSS model based BIQA metric called BJLC (BIQA based on joint log-contrast statistics). Extensive experiments on four public large-scale image databases have validated that objective quality scores predicted by the proposed BIQA method are in higher accordance with subjective ratings generated by human observers compared with existing methods.
关键词： partial least square,Blind image quality assessment (BIQA),no-reference (NR),natural scene statistics
更新于2025-09-09 09:28:46
-
[IEEE 2018 IEEE International Symposium on Haptic, Audio and Visual Environments and Games (HAVE) - Dalian, China (2018.9.20-2018.9.21)] 2018 IEEE International Symposium on Haptic, Audio and Visual Environments and Games (HAVE) - No-Reference Quality Assessment for Stereoscopic 3D Images Based on Binocular Visual Perception
摘要： In this paper, we propose a blind/no-reference 3D image quality assessment scheme that utilizes binocular visual characteristics. The design of this scheme is motivated by studies on the perception of distorted stereoscopic images. Specifically, after the log-Gabor filter processing, the local amplitude, local phase and visual saliency are extracted from a stereopair and concatenated to form feature vectors. In addition, the binocular energy responses are also obtained as quality-predictive features. Experimental results show that the proposed scheme achieves superiority over other compared methods in terms of consistent alignment with human subjective judgments for stereoscopic images.
关键词： binocular visual perception,3D image quality assessment,no-reference
更新于2025-09-09 09:28:46
-
[IEEE 2018 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) - San Diego, CA, USA (2018.7.23-2018.7.27)] 2018 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) - Quality Assessment for Tone-Mapped HDR Images Using Multi-Scale and Multi-Layer Information
摘要： Tone mapping operators and multi-exposure fusion methods allow us to enjoy the informative contents of high dynamic range (HDR) images with standard dynamic range devices, but also introduce distortions into HDR contents. Therefore methods are needed to evaluate tone-mapped image quality. Due to the complexity of possible distortions in a tone-mapped image, information from different scales and different levels should be considered when predicting tone-mapped image quality. So we propose a new no-reference method of tone-mapped image quality assessment based on multi-scale and multi-layer features that are extracted from a pre-trained deep convolutional neural network model. After being aggregated, the extracted features are mapped to quality predictions by regression. The proposed method is tested on the largest public database for TMIQA and compared to existing no-reference methods. The experimental results show that the proposed method achieves better performance.
关键词： multi-scale and multi-layer,tone-mapped HDR images,no-reference image quality assessment
更新于2025-09-09 09:28:46
-
A Reduced-Reference Image Quality Assessment Model Based on Joint-Distribution of Neighboring LOG Signals
摘要： Previous work have validated that the output of retinal ganglion cells in human visual pathway, which can be modeled as an LOG (Laplacian of Gaussian) filtration, can whiten the power spectrum of not only the natural images, but also the distorted images, hence the first-order (average luminance) and the second-order (contrast) redundancies have been removed when applying the LOG filtration. Considering the fact that human vision system (HVS) always ignores the first-order and the second-order information when sensing image local structures, the LOG signals should be efficient features in IQA (image quality assessment) task and a lot of LOG based IQA models have been proposed. In this paper, we focus on an interesting question that has not been investigated carefully yet: what is an efficient way to represent image structure features that is perceptual quality aware based on relationship between the LOG signals. We examine the to represent neighboring LOG signals and propose relationship by computing the joint distribution of neighboring LOG signals, and thus propose a set of simple but efficient RR IQA feature and consequently yield an excellent RR IQA model. Experimental results on three large scale subjective IQA databases show that our proposed method works robustly across different databases and stay in the state-of-the-art RR IQA models.
关键词： Whiten,Reduced-Reference,Laplacian of Gaussian,Image Quality Assessment,Joint Distribution
更新于2025-09-09 09:28:46
-
Blind Image Quality Assessment Using Multiscale Local Binary Patterns
摘要： This article proposes a new no-reference image quality assessment method that is able to blindly predict the quality of an image. The method is based on a machine learning technique that uses texture descriptors. In the proposed method, texture features are computed by decomposing images into texture information using multiscale local binary pattern (MLBP) operators. In particular, the parameters of local binary pattern operators are varied, which generates MLBP operators. The features used for training the prediction algorithm are the histograms of these MLBP channels. The results show that, when compared with other state-of-the-art no-reference methods, the proposed method is competitive in terms of prediction precision and computational complexity.
关键词： MLBP,machine learning,multiscale local binary pattern,texture descriptors,no-reference image quality assessment
更新于2025-09-09 09:28:46
-
[IEEE 2018 Tenth International Conference on Quality of Multimedia Experience (QoMEX) - Cagliari (2018.5.29-2018.6.1)] 2018 Tenth International Conference on Quality of Multimedia Experience (QoMEX) - A Hybrid Quality Metric for Non-Integer Image Interpolation
摘要： A great need of High-Resolution (HR) images has boosted the development of interpolation techniques. However, it is still a challenging task to objectively evaluate the perceptual quality of interpolated images, especially when the interpolation factor is a non-integer. To address this issue, we propose a hybrid quality metric for non-integer image interpolation that combines both reduced-reference and no-reference philosophies. To validate the proposed metric, we construct a non-integer interpolated image database and conduct a subjective user study to collect subjective opinions for each image. Experiments on the new database show that the proposed metric outperforms previous methods by a large margin.
关键词： high-resolution images,perceptual image processing,Image quality assessment,image interpolation
更新于2025-09-04 15:30:14
-
Feature selection algorithm for no-reference image quality assessment using natural scene statistics
摘要： Images play an essential part in our daily lives and the performance of various imaging applications is dependent on the user’s quality of experience. No-reference image quality assessment (NR-IQA) has gained importance to assess the perceived quality, without using any prior information of the nondistorted version of the image. Different NR-IQA techniques that utilize natural scene statistics classify the distortion type based on groups of features and then these features are used for estimating the image quality score. However, every type of distortion has a different impact on certain sets of features. In this paper, a new feature selection algorithm is proposed for distortion identification based image verity and integration evaluation that selects distinct feature groups for each distortion type. The selection procedure is based on the contribution of each feature on the Spearman rank order correlation constant (SROCC) score. Only those feature groups are used in the prediction model that have majority features with SROCC score greater than mean SROCC score of all the features. The proposed feature selection algorithm for NR-IQA shows better performance in comparison to state-of-the-art NR-IQA techniques and other feature selection algorithms when evaluated on three commonly used databases.
关键词： feature selection,No-reference image quality assessment,classification,distortion identification based image verity and integration evaluation,support vector regression
更新于2025-09-04 15:30:14