No-reference image quality assessment in contourlet domain

We present a no-reference (NR) image quality assessment (IQA) algorithm that is inspired by the representation of visual scenes in the primary visual cortex of the human visual system. Specifically, we use the sparse coding model of the area V1 to construct an overcomplete dictionary for sparsely representing pristine (undistorted) natural images. First, we empirically demonstrate that the distribution of the sparse representation coefficients of natural images have sharp peaks and heavy tails, and can therefore be modeled using a Univariate Generalized Gaussian Distribution (UGGD). We then show that the UGGD model parameters form good features for distortion estimation and formulate our no-reference IQA algorithm based on this observation. Subsequently, we find UGGD model parameters that are representative of the class of pristine natural images. This is achieved using a training set of undistorted natural images. The perceptual quality of a test image is then defined to be the likelihood of its sparse coefficients being generated from the pristine UGGD model. We show that the proposed algorithm correlates well with subjective evaluation over several standard image databases. Further, the proposed method allows us to construct a distortion map that has several useful applications like distortion localization, adaptive rate allocation etc. Finally and importantly, the proposed NR-IQA algorithm does not make use of any distortion information or subjective scores during the training process.

Being able to automatically predict digital picture quality, as perceived by human observers, has become important in many applications where humans are the ultimate consumers of displayed visual information. Standard dynamic range (SDR) images provide 8 bits/color/pixel. High dynamic range (HDR) images which are usually created from multiple exposures of the same scene, can provide 16 or 32 bits/color/pixel, but must be tonemapped to SDR for display on standard monitors. Multi-exposure fusion (MEF) techniques bypass HDR creation, by fusing the exposure stack directly to SDR format while aiming for aesthetically pleasing luminance and color distributions. Here we describe a new no-reference image quality assessment (NR IQA) model for HDR pictures that is based on standard measurements of the bandpass and on newly-conceived differential natural scene statistics (NSS) of HDR pictures. We derive an algorithm from the model which we call the Gradient Image Quality Assessment algorithm (G-IQA). NSS models have previously been used to devise NR IQA models that effectively predict the subjective quality of SDR images, but they perform significantly worse on tonemapped HDR content. Towards ameliorating this we make here the following contributions: (1) We design a HDR picture NR IQA model and algorithm using both standard space-domain NSS features as well as novel HDR-specific gradient based features that significantly elevate prediction performance, (2) We validate the proposed models on a large-scale crowdsourced HDR image database, and (3) We demonstrate that the proposed model also perform well on legacy natural SDR images. The software is available at: http://signal.ece.utexas.edu/%7Edebarati/higradeRelease.zip.

No-reference image quality assessment based on spatial and spectral entropies

Measuring visual quality, as perceived by human observers, is becoming increasingly important in many applications where humans are the ultimate consumers of visual information. Significant progress has been made for assessing the subjective quality of natural images, such as those taken by optical cameras. Natural Scene Statistics (NSS) is an important tool for no-reference visual quality assessment of natural images, where the reference image is not needed for comparison. In this paper, we take an important step towards using NSS to automate visual quality assessment of photorealistic synthetic scenes typically found in video games and animated movies. Our primary contributions are (1) conducting subjective tests on our publicly available ESPL Synthetic Image Database containing 500 distorted images (20 distorted images for each of the 25 original images) in 1920 × 1080 format, and (2) evaluating the performance of 17 no-reference image quality assessment (IQA) algorithms using synthetic scene statistics. We find that similar to natural scenes, synthetic scene statistics can be successfully used for IQA and certain statistical features are good for certain image distortions.

An optimized CNN-based quality assessment model for screen content image

The human visual system is sensitive to structural information in images, and modeling this information has been regarded as useful for predicting their perceptual quality. In this study, we propose a no-reference (NR) image quality assessment (IQA) method based on a sparse representation of the distribution of structural information. The grayscale fluctuation map of an image is first calculated and divided into patches of fixed size that are rearranged into column vectors, which are regarded as structural elements of the image. Following this, using sparse coding, these structural elements can be represented by sparse representation coefficients and a trained dictionary. By using the former, a probability vector for observing different elements in the trained dictionary can then be obtained. Finally, the prediction model is trained using support vector regression. The results of experiments to test the proposed method show that it can accurately predict humans’ perception of image quality and is competitive in comparison with prevalent NR–IQA methods.

In the past years, deep learning evolution has helped the development of computer vision systems. However, the quality of images plays a significant role in the effectiveness of these systems and it would be useful to know the quality of the images that are imported into our systems. No-reference image quality assessment is a challenging procedure, which tries to predict the quality of an image without using any reference image. In this paper, we evaluate the performance of widely used deep learning models in no-reference image quality assessment. To that end, we used transfer learning on 8 pre-trained models which we fit into 3 datasets related to image quality assessment. The performance of these models was studied in terms of mean absolute percentage error (MAPE). Although most of the models performed reasonably well in a MAPE range of 15% to 40% depending on the dataset used, the best performing one in a single dataset was DenseNet201 with a MAPE of 9.8%, while the overall best performing model was ResNet50.

No-reference image quality assessment (NR-IQA), which aims to predict image quality without accessing to reference images, is a fundamental and challenging problem in the field of image processing. Nevertheless, there are few researches about contrast-distorted images and results of the existing NR-IQA methods which cannot be in accordance with the subjective assessment results further. Therefore, an effective NR-IQA method for contrast-distorted image is proposed in this paper. Firstly, the proposed method extracts five features of all images from the database based on the natural scene statistics (NSS) model. Then the curve fitting method is subsequently represented to calculate values of natural image features. Finally, an improved Support Vector Regression (SVR) learning method based on grid search is proposed to establish the mapping between image feature values and the quality score of a test image. Experiments proved that the proposed method is effective when compared with other related state-of-the-art image quality assessment (IQA) methods based on three standard databases.

In this work, we propose a supervised no-reference (NR) Image Quality Assessment (IQA) model for the objective evaluation of the perceptual quality of 3D virtual reality (VR) images. To achieve such practical algorithm, we first study the scene statistics of saliency maps of the individual left and right views of VR images, and empirically model these statistics with Univariate Generalized Gaussian Distribution (UGGD). We compute the UGGD model parameters at multi-scale and multi-orient steerable subband decomposition, and introduce these features as distortion discriminables. This is followed by the computation of the entropy and normalized root mean square scores of each subband and then these values are utilized as weights to pool the individual view features. We apply the popular 2D supervised BRISQUE model on the individual views to estimate the overall spatial quality of VR images. As the last step of the algorithm, the predicted saliency score and the spatial BRISQUE score are pooled to derive the final quality score of VR images. The performance of the proposed model is evaluated on the popular LIVE 3D VR IQA dataset. The results indicate robust and competitive performance against the off-the-shelf 2D full-reference and NR supervised algorithms.

Image Quality Assessment (IQA) is essential for the accuracy of systems for automatic recognition of tree species for wood samples. In this study, a No-Reference IQA (NR-IQA), wood NR-IQA (WNR-IQA) metric was proposed to assess the quality of wood images. Support Vector Regression (SVR) was trained using Generalized Gaussian Distribution (GGD) and Asymmetric Generalized Gaussian Distribution (AGGD) features, which were measured for wood images. Meanwhile, the Mean Opinion Score (MOS) was obtained from the subjective evaluation. This was followed by a comparison between the proposed IQA metric, WNR-IQA, and three established NR-IQA metrics, namely Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE), deepIQA, Deep Bilinear Convolutional Neural Networks (DB-CNN), and five Full Reference-IQA (FR-IQA) metrics known as MSSIM, SSIM, FSIM, IWSSIM, and GMSD. The proposed WNR-IQA metric, BRISQUE, deepIQA, DB-CNN, and FR-IQAs were then compared with MOS values to evaluate the performance of the automatic IQA metrics. As a result, the WNR-IQA metric exhibited a higher performance compared to BRISQUE, deepIQA, DB-CNN, and FR-IQA metrics. Highest quality images may not be routinely available due to logistic factors, such as dust, poor illumination, and hot environment present in the timber industry. Moreover, motion blur could occur due to the relative motion between the camera and the wood slice. Therefore, the advantage of WNR-IQA could be seen from its independency from a “perfect” reference image for the image quality evaluation.

In recent years, deep learning has achieved significant progress in many respects. However, unlike other research fields with millions of labeled data such as image recognition, only several thousand labeled images are available in image quality assessment (IQA) field for deep learning, which heavily hinders the development and application for IQA. To tackle this problem, in this paper, we proposed an error self-learning semi-supervised method for no-reference (NR) IQA (ESSIQA), which is based on deep learning. We employed an advanced full reference (FR) IQA method to expand databases and supervise the training of network. In addition, the network outputs of expanding images were used as proxy labels replacing errors between subjective scores and objective scores to achieve error self-learning. Two weights of error back propagation were designed to reduce the impact of inaccurate outputs. The experimental results show that the proposed method yielded comparative effect.

No-reference image quality assessment for photographic images based on robust statistics

Our approach to no-reference image quality assessment (IQA) is grounded on the assumption that natural scenes have certain statistical properties which are altered in the presence of distortion. Therefore, by measuring this statistic one can recognize the distortion affecting the image and provide a no-reference (NR) IQA. In this paper, we extract local entropy and standard deviation from non-overlapping 8 × 8 blocks in the image and its gradient version from 2 scales. Extracting features from gradient images can help to better distinguishing blur and noise effects on images. Having extracted local measures, we aggregate them by finding the parameters of a Gaussian mixture model fitted on these local measures. Finally, our quality metric is achieved via a Neural Network machine learning tool, and its performance is evaluated on a standard LIVE database. The experimental results point out that our proposed features not only can identify the type of distortion but also are consistent with the DMOS in subjective assessment in terms of the correlation coefficient.

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

In practice, images available to consumers usually undergo several stages of processing including acquisition, compression, transmission, and presentation, and each stage may introduce certain type of distortion. It is common that images are simultaneously distorted by multiple types of distortions. Most existing objective image quality assessment (IQA) methods have been designed to estimate perceived quality of images corrupted by a single image processing stage. In this letter, we propose a no-reference (NR) IQA method to predict the visual quality of multiply-distorted images based on structural degradation. In the proposed method, a novel structural feature is extracted as the gradient-weighted histogram of local binary pattern (LBP) calculated on the gradient map (GWH-GLBP), which is effective to describe the complex degradation pattern introduced by multiple distortions. Extensive experiments conducted on two public multiply-distorted image databases have demonstrated that the proposed GWH-GLBP metric compares favorably with existing full-reference and NR IQA methods in terms of high accordance with human subjective ratings.