Xiuhua Jiang

Objective Perceptual Video Quality Measurement using a Foveation-Based Reduced Reference Algorithm

The properties and limitations of human visual system (HVS) are usually incorporated into objective video quality techniques for better correlation with subjective methods. In this paper, we propose a new objective assessment algorithm based on reduced reference video quality metrics (RR-VQM). The key point of our method is to design multilayer Spatial-Temporal (ST) regions based on humans fovea region, which can utilize the nonuniform resolution properties of HVS for high definition (HD) videos. Quality features are then extracted from ST regions for computing the video quality. Experimental results shows that this kind of multi-layer ST regions can improve the performance of objective methods, not only the results of evaluation, but can also accelerate the evaluation speed and reduce the amount of quality parameters to a large extent.

A QOE-driven approach to rate adaptation for dynamic adaptive streaming over http

In this paper, we focus on the quality of experience (QoE) optimization for dynamic adaptive streaming over HTTP (DASH) services. First, we propose a QoE evaluation model based on playback continuity, segment media quality and perceptual quality fluctuations caused by bitrate switching. Plenty of subjective mean-opinion-score (MOS) tests have shown a good correlation between this model and subjective perception. Second, based on our QoE evaluation model, we formulate the rate adaptation for DASH as a constrained convex optimization problem. The design objective is to maximize the overall QoE for the end users while keep the receiver buffer from underflow. Numerous experiment results demonstrate that our proposed rate adaptation scheme achieves near optimal tradeoff between overall QoE and playback continuity.

Objective Image Quality Assessment Based on Image Complexity and Color Similarity

This paper proposes a new objective assessment method for image quality based on content complexity and color similarity. Considering that human visual perception is sensitive to color distortion in compressed images due to different content complexity, we analyze the effects of image complexity and color similarity to images quality. Based on above analysis, we propose a method to establish the relationship between image quality and those two factors. The method includes the following steps: carrying out image color space conversion, calculating the image complexity and the color similarity, adjusting color similarity according to the different image complexity, and then obtaining the evaluation results. Experimental results show that our method can obtain a good consistency with the subjective results.

An improved image quality objective assessment method using non-linear calibration algorithm

This paper firstly presents an image quality objective assessment method on the base of color moments. Then by analyzing the statistical characteristics of the subjective assessment results, this paper proposes a non-linear calibration algorithm using the worst image quality video sequence as the reference to separate the score interval of objective assessment. Combined with the similarity metric algorithm based on the color moment, an improved objective assessment algorithm of the image quality is established. The results show that the subjective evaluation results and objective evaluation values gained by this algorithm are with high consistency.

Visual perception features of color damage in High Definition video compression

High Definition video color information research and analysis, mainly considering compression process brought about a decline in the quality of visual perception. Four subjective evaluation is designed and experiments has been done: The diversity of the human eye perception of time; rate selection; and two subjective evaluation experiments of the video with DSCQS as the standard. The High Definition video main characteristics of the human eye perception between color and compression rate injury. Before the experiment, statistical methods were used to calculate the actual sample size; according to R-REC-BT series of standards for subjective evaluation results of the statistical data after the experiment. Finally, analyzed the data results combined with color visual characteristics of the human eye.

The New Improved Data Processing Methods in Video Quality Subjective Assessment

Data screening used in video quality subjective assessment contribute to eliminate abnormal observers and improve the precision and effectiveness of the results. The paper firstly introduces a kind of standard and mainstream data-processing method, which is recommended by ITU-R BT.500. Then the paper has made the improvement to the standard method by using a method called STRESS in the field of Psychophysics and proposes an improved data-processing method which is used to choose observers whose score trends have low correlation with the overall trend and then eliminate observers whose STRESS order number is small. Finally we analyze the applicability of the two data-processing methods and verify the effectiveness of the improved method.

Assessment of subjective video quality on 4K Ultra High Definition videos

With the wide spread use of digital video, video quality assessment has become a hot research. In this paper, a subjective video quality assessment on 4K Ultra-High Definition (UHD) videos was presented in detail. We described the source sequences which were chosen and the preparation of the impaired test sequences. The test environment and test equipment was also presented and we chose DSCQS for 4K-UHD subjective quality assessment methodology. 21 human subjects participated in the test and the differential mean opinion scores (DMOS) of each test sequence was recorded. Results of the subjective quality assessment show that at the bitrate of 23Mbps achieved an obviously good video quality with the average DMOS of 12.7.

A No-reference video quality assessment based on two-pass edge analysis and background subtraction

A novel no-reference video quality assessment is proposed in this paper. The proposed metric is based on the combination of two-pass edge analysis and background subtraction. The two-pass edge analysis is a blur metric we proposed in our previous work, which has gain an efficient performance in image blur assessment and other image processing applications. We apply the two-pass edge analysis to the video quality assessment in order to calculate the quality of each frame in the video. On the other hand, considering that one will pay his most attention on the foreground of the video. So we also apply the background subtraction to get the foreground of the video and only use these foreground to calculate the video quality. Experiments in the last section shows that our proposed no-reference metric has a good performance in the test sequences.

A method on digital watermarking image against geometric distortion

This paper describes a method against geometric distortion of digital watermarking image by means of Hausdorff distance measuring, which matches the feature points, and calculates similarity between two point sets (maximum distance). In this paper, Harris feature points are firstly extracted into feature point set according to the response of the Harris detector and the field size, and then secondly the Hausdorff distance is used to eliminate the image geometric deformation, thirdly, DWT(Discrete Wavelet Transform) transformation is conducted with the Harris feature points as centers, and also DCT(Discrete Cosine Transform) transformation is conducted in low-frequency LL domain; at last, watermark could be embedded and extracted in the DCT coefficient. The experimental results demonstrate that the embedded watermark image is well extracted and has high similarity after such geometric attacks as translation, rotation, scaling and shearing, so this method is a blind watermarking detection algorithm with strong robustness.

Subjective quality assessment of compressed 3D video

Since 3D video has the potential to provide stronger immersive perception, evaluating the quality of 3D video becomes an important subject. This paper mainly studies the subjective quality assessment of 3D video of two views compressed by the state-of-the-art standard H.264/MVC in 3DTV system. Our goal is to obtain available opinion scores which can be applied in the research of objective quality assessment in the future. The subjective evaluation is conducted under the guidance of ITU recommendations, including test materials, test environment, test equipment, assessment factors, subjective methodologies and analysis of assessment results. Experiments showed that Mean Opinion Scores (MOS) of overall quality increases with bitrate and different 3D sequences with different complexity have different quality trends.