Besides a high distinctiveness, robustness (or invariance) to image degradations is very desirable for texture feature\nextraction methods in real-world applications. In this paper, focus is on making arbitrary texture descriptors invariant\nto blur which is often prevalent in real image data. From previous work, we know that most state-of-the-art texture\nfeature extraction methods are unable to cope even with minor blur degradations if the classifier�s training stage is\nbased on idealistic data. However, if the training set suffers similarly from the degradations, the obtained accuracies\nare significantly higher. Exploiting that knowledge, in this approach the level of blur of each image is increased to a\ncertain threshold, based on the estimation of a blur measure. Experiments with synthetically degraded data show that\nthe method is able to generate a high degree of blur invariance without loosing too much distinctiveness. Finally, we\nshow that our method is not limited to ideal Gaussian blur.
Loading....