Abstract
In this paper, we propose an algorithm for assessing image quality using defocus blur estimation. The algorithm automatically detects regions of interest that contain features of edges, then uses a histogram-matching technique to estimate the extent of blurriness. Real defocus blurred regions and synthesized defocus blurred regions are compared, and the optimal extent of blurriness is obtained by iteratively changing the convolution parameters. Finally, the image quality is estimated by computing the average extent of blurriness of all regions. Experimental results demonstrate that our approach enables a highly accurate assessment of image quality.
Acknowledgment
The authors would like to thank the National Science Council of Taiwan for financially supporting this research under Contract NSC-99-2221-E-194-005-MY3.