Haiqing Du

Blind Image Quality Assessment Based on High Order Statistics Aggregation

Blind image quality assessment (BIQA) research aims to develop a perceptual model to evaluate the quality of distorted images automatically and accurately without access to the non-distorted reference images. The state-of-the-art general purpose BIQA methods can be classified into two categories according to the types of features used. The first includes handcrafted features which rely on the statistical regularities of natural images. These, however, are not suitable for images containing text and artificial graphics. The second includes learning-based features which invariably require large codebook or supervised codebook updating procedures to obtain satisfactory performance. These are time-consuming and not applicable in practice. In this paper, we propose a novel general purpose BIQA method based on high order statistics aggregation (HOSA), requiring only a small codebook. HOSA consists of three steps. First, local normalized image patches are extracted as local features through a regular grid, and a codebook containing 100 codewords is constructed by K-means clustering. In addition to the mean of each cluster, the diagonal covariance and coskewness (i.e., dimension-wise variance and skewness) of clusters are also calculated. Second, each local feature is softly assigned to several nearest clusters and the differences of high order statistics (mean, variance and skewness) between local features and corresponding clusters are softly aggregated to build the global quality aware image representation. Finally, support vector regression is adopted to learn the mapping between perceptual features and subjective opinion scores. The proposed method has been extensively evaluated on ten image databases with both simulated and realistic image distortions, and shows highly competitive performance to the state-of-the-art BIQA methods.

Research on Adaptive FEC for Video Delivery over WLAN

This paper presents an algorithm to mend the Enhanced Adaptive FEC (EAFEC) mechanism for improving video delivery over WLAN. The algorithm is based on two factors, first of which is the queue length in the access point (AP), indicating network traffic load; the other is packet retransmission times, indicating wireless channel state. According to these factors, two different redundant FEC packets were calculated separately, through setting the redundant FEC packets different weight value, the final redundant FEC packets is acquired. The simulation test proved there is much improvement especially in sudden changing video scenes, when our algorithm is adopted, compared to EAFEC.

A no-reference perceptual blur metric based on complex edge analysis

This paper presents an efficient no-reference blur metric where image quality is quantified by perceptual-based edge analysis. Instead of statistical measurement of the whole image, it computes local statistics in the vicinity of detected edges at a largely reduced computational cost. Unlike existing objective no-reference blur metrics, the proposed metric is able to distinguish blurred edges in a cost-effective way. Evaluation of the proposed metric shows its high prediction accuracy when it is applied to Gaussian blurred images. Experiments using the LIVE blur database demonstrate that the proposed algorithm correlates well with subjective quality evaluations.

Local feature aggregation for blind image quality assessment

Previous feature learning based blind image quality assessment (BIQA) methods invariably require large codebook or codebook updating procedure to obtain satisfying performance. In this paper, we propose a novel general purpose BIQA method, local feature aggregation (LFA) model, which requires only a much smaller codebook without the need for codebook updating. The proposed model consists of three steps. Firstly, normalized local raw image patches are extracted as local features through a regular grid and a 100 codeword codebook is constructed by K-means clustering. Secondly, the soft weighted differences between local features and codewords are aggregated to the global quality aware representation. Finally, support vector regression (SVR) is utilized to learn the mapping between features and subjective opinion scores. The proposed method is evaluated on two large image databases and achieves comparable performance to state-of-the-art BIQA methods.

Blind image quality assessment on authentically distorted images with perceptual features

Almost all present successful general purpose blind image quality assessment (BIQA) models are designed only towards singly and synthetically distorted images. However, real world images generally contain multiply types of distortions other than single type of distortion. Even some state of the art BIQA methods cannot work well on authentically distorted images. In this paper we propose a novel effective BIQA method for authentically distorted images to remedy the shortage of the popular BIQA models. First, except for traditional natural scene statistics (NSS) based features, we introduce a series of quality aware features based on human visual system (HVS), such as image dynamic range, color information, blurriness. Then support vector regression (SVR) is utilized to learn the mapping between the combined effective features and human opinion scores. Experimental results demonstrate the promising performance of the proposed method.

Blind image quality assessment with complete pixel-level information

In this paper, we develop a novel method for blind image quality assessment (BIQA) based on image complete pixel level information. First, traditional rotation invariant uniform local binary pattern (LBP) histogram is extracted from grayscale image as perceptual quality aware feature. Second, except for the signs of local pixel differences, the magnitudes of local pixel differences in grayscale image are also encoded by LBP, and the joint histogram between the signs and magnitudes of local pixel differences is also calculated as part of the perceptual feature. Finally, the support vector regression (SVR) is utilized to learn the mapping between the combined perceptual feature and human opinion scores. Experimental results show that the proposed method is highly correlated with human opinion scores and achieves competitive performance with state-of-the-art methods for quality evaluation and distortion classification.

Statistical metric fusion for image quality assessment

In this paper, we propose two novel Statistical Metric Fusion (SMF) methods for Image Quality Assessment (IQA) metric enhancement. First, local quality map is constructed from existing state-of-the-art IQA algorithm. After that several statistical indices are extracted from local quality map. Finally, the extracted statistical indices are fused by Supervised Statistical Metric Fusion (SMF-S) based on Support Vector Regression (SVR) and Unsupervised Statistical Metric Fusion (SMF-U) based on Reciprocal Rank Fusion (RRF) to obtain the final quality score, respectively. Experimental results on the largest public IQA database TID2013 have demonstrated that the two proposed SMF methods can generally enhance the quality prediction performance of the fused IQA metric in terms of high correlation with human opinion scores.

A no-reference perceptual blur metric based on the blur ratio of detected edges

In this paper, we present an efficient no-reference image blur metric which is based on the analysis of the spread of edge and the study of human blur perception for varying contrast values. Our method calculates blur ratio of significant edges and global vertical edges respectively, and final score is a weighted average of the two ratios because giving different edges different corresponding weights will improve prediction accuracy. Evaluation of the proposed metric shows its high prediction accuracy when it is applied to Gaussian blurred images. Experiments using the LIVE and TID Gaussian blur dataset demonstrate that the proposed algorithm correlates well with subjective quality evaluations.