Naturalness index for a tone-mapped high dynamic range image.

Area of Science:

• Computer Vision
• Image Processing
• Perception

Background:

• High dynamic range (HDR) imaging requires tone-mapping for compatibility with low dynamic range (LDR) systems.
• Quality assessment (QA) of tone-mapped HDR images is crucial yet challenging for HDR imaging research.

Purpose of the Study:

• To propose a novel naturalness index for predicting the quality of tone-mapped HDR images.
• To develop a no-reference QA method for tone-mapped images.

Main Methods:

• Extracting statistical features of luminance values to assess brightness naturalness.
• Utilizing perceptive color, image contrast, and detail information to predict content naturalness.
• Combining four naturalness components for an overall quality score.

Main Results:

• The proposed naturalness index effectively evaluates image brightness naturalness without reference data.
• Experimental results show superior performance compared to traditional LDR QA and existing tone-mapped QA methods.
• Validation on a public database confirms the method's effectiveness.

Conclusions:

• The developed naturalness index provides a robust and accurate method for tone-mapped HDR image quality assessment.
• This no-reference approach simplifies the QA process for tone-mapped HDR images.
• The method demonstrates significant improvements in predicting the perceptual quality of tone-mapped images.