| dc.description.abstract | High Dynamic Range (HDR) technology transforms video material by increasing the overall visual experience by delivering a greater range of luminance and color. However, because to the complexity of their content, measuring the quality of HDR videos is a difficult process. This master's thesis addresses the critical need for an objective quality assessment metric built exclusively for HDR video.
The major goal of this thesis is to propose and evaluate HDR-FP, a feature-pooled objective metric for judging HDR video quality. HDR-FP is based on considerable research and employs advanced image processing algorithms to extract important defects such as sharpness, compression faults, color distortion, banding, dynamic range, increased noise, blurring, and ghosting. These distortions are critical elements that have a considerable impact on the perceptual quality of HDR videos.
An extensive dataset of HDR films with a varied variety of material is used to evaluate the performance of HDR-FP. The suggested metric is extensively compared to two state-of-the-art metrics, HDR-VQM and HDR-VDP-2, and shows superior accuracy and consistency. The Spearman Rank Order Correlation Coefficient (SROCC), Pearson Linear Correlation Coefficient (PLCC), and Kendall Rank Order Correlation Coefficient (KROCC) are used to assess the correlations between the objective quality scores derived from HDR-FP and the subjective Mean Opinion Scores (MOS). The high correlation values found demonstrate HDR-FP's ability to align with human visual judgments.
This study makes a substantial contribution to the field of HDR video quality assessment by developing a strong and dependable objective metric for precisely quantifying the quality of HDR films. The availability of such a metric is critical for a variety of applications, including content development, broadcasting, streaming, and quality control, allowing content providers and service providers to provide viewers with immersive HDR experiences. | |
