| 摘要 |
A method for estimating if an image (14) is either blurred or sharp includes the steps of (i) determining an image gradient histogram distribution (410) of at least a portion of the image (14), and (ii) comparing at least a portion of the image gradient histogram distribution (410) to a Gaussian model gradient histogram distribution (414) of at least a portion of the image (14) to estimate if the image (14) is either blurred or sharp. In most cases, a sharp image (14) has a relatively large tailed distribution when compared to the Gaussian model, while a blurred image has a relatively small tailed distribution when compared to the Gaussian model. |
