A Gabor Feature-Based Quality Assessment Model for the Screen Content Images.

Area of Science:

• Computer Vision
• Image Processing
• Human Visual System Modeling

Background:

• Objective image quality assessment (IQA) is crucial for evaluating screen content images (SCIs).
• The human visual system (HVS) is sensitive to edge information, a key feature in image perception.
• Gabor filters mimic HVS response and are effective for edge detection.

Purpose of the Study:

• To propose an accurate and efficient full-reference IQA model for SCIs.
• To leverage Gabor features for objective image quality evaluation.
• To develop a model with high consistency with human perception and low computational cost.

Main Methods:

• Extraction of Gabor features from luminance components of reference and distorted SCIs using the imaginary part of Gabor filters.
• Measurement of local similarities between Gabor features and two chrominance components in the LMN color space.
• Application of a Gabor-feature pooling strategy to combine measurements and generate a final quality score.

Main Results:

• The proposed Gabor feature-based model (GFM) demonstrates high consistency with human perception in SCI assessment.
• GFM exhibits lower computational complexity compared to classical and state-of-the-art IQA models.
• Experimental results on large SCI databases validate the model's effectiveness.

Conclusions:

• The GFM provides an accurate and efficient objective IQA solution for SCIs.
• The model's reliance on Gabor features effectively captures HVS-sensitive information.
• GFM offers a promising alternative for image quality evaluation with improved performance and efficiency.