Related Experiment Video
Updated: Jul 7, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
A distortion measure for blocking artifacts in images based on human visual sensitivity.
S A Karunasekera1, N G Kingsbury
1Dept. of Eng., Cambridge Univ.
Summary
This study introduces a visual model to measure blocking artifacts in images. The model accurately predicts the visibility of blocking errors, correlating well with human perception.
Area of Science:
- Computer Vision
- Image Processing
- Human Visual Perception
Background:
- Blocking artifacts are common distortions in digital images.
- Existing methods may not accurately reflect human perception of these artifacts.
Purpose of the Study:
- To develop a visual model for quantifying blocking artifact visibility.
- To align image distortion measures with human visual sensitivity.
Main Methods:
- Deriving a model based on human visual sensitivity to edge artifacts.
- Conducting psychovisual experiments to measure sensitivity variations.
- Estimating model parameters using experimental data.
- Testing the model on real-world images.
Main Results:
- The model quantifies blocking error visibility numerically.
- Psychovisual experiments revealed sensitivity variations based on luminance, activity, edge length, and amplitude.
- The model's predictions showed strong correlation with subjective rankings of error visibility.
Conclusions:
- The developed visual model effectively measures blocking artifact visibility.
- The model's accuracy is validated by its correlation with human perception.
- This model can improve image quality assessment by incorporating visual perception.
