Performance metrics for image contrast

In this paper, contrast level of the images are quantified by the two proposed metrics. These metrics are Histogram Flatness Measure (HFM) and Histogram Spread (HS). Computation of these metrics is based on the shape of the histogram. Extensive simulation results reveal that HS is more meaningful than HFM. Low contrast images have low HS value, while high contrast images have higher value of HS. Thus HS metric can be used to distinguish between the images having different contrast level. Accuracy of the metric is also verified for natural and medical images. This metric has broad applications in image retrieval, image database management, visualization, rendering and image classification.

[1]  Oscar C. Au,et al.  Optimal contrast enhancement for tone-mapped low dynamic range images based on high dynamic range images , 2009, 2009 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing.

[2]  Hanseok Ko,et al.  Enhancement of image degraded by fog using cost function based on human visual model , 2008, 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems.

[3]  Jean-Philippe Tarel,et al.  Fast visibility restoration from a single color or gray level image , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[4]  Steven Kay,et al.  Modern Spectral Estimation: Theory and Application , 1988 .