THE ROLE OF EDGE INFORMATION TO ESTIMATE THE PERCEIVED UTILITY OF NATURAL IMAGES

In the quality assessment task, observers evaluate a natura l image based on its perceptual resemblance to a reference. For the u tility assessment task, observers evaluate the usefulness of a nat ur l image as a surrogate for a reference. Humans are willing to use t he information captured by an imaging system and tolerate dist ort ons as long as the underlying task is performed reliably. Conven tional notions of perceived quality cannot generally predict the p erceived utility of a natural image. Many estimators have been design ed to estimate perceived quality scores, and a recent estimator, refe red to as the natural image contour evaluation (NICE), has been dev eloped to estimate perceived utility scores. An analysis of edge in formation in natural images drives object recognition mechanisms in t he human visual system, so this paper examines the role of an edgebased analysis in both popular objective quality estimators and N ICE for estimating the perceived quality or utility of distorted na tur l images. Among the estimators evaluated, results show that est imators that emphasize an edge-based analysis provide the most accu rate estimates of perceived utility scores. Estimators that aug ment the edge-based analysis with an energy-based analysis provide the most accurate estimates of perceived quality scores.

[1]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[2]  Alan C. Bovik,et al.  Image information and visual quality , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[3]  Alan C. Bovik,et al.  Image information and visual quality , 2006, IEEE Trans. Image Process..

[4]  Ronald G. Driggers,et al.  Current infrared target acquisition approach for military sensor design and wargaming , 2006, SPIE Defense + Commercial Sensing.

[5]  Bernd Girod,et al.  What's wrong with mean-squared error? , 1993 .

[6]  Patrick Le Callet,et al.  Image utility assessment and a relationship with image quality assessment , 2009, Electronic Imaging.

[7]  Sheila S. Hemami,et al.  Natural image utility assessment using image contours , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[8]  Carolyn G. Ford,et al.  Subjective video quality assessment methods for recognition tasks , 2009, Electronic Imaging.

[9]  P. O. Bishop,et al.  Spatial vision. , 1971, Annual review of psychology.

[10]  Refractor.  Vision , 2000, The Lancet.

[11]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.