Multiscale Analysis for Improving Texture Classification

Information from an image occurs over multiple and distinct spatial scales. Image pyramid multiresolution representations are a useful data structure for analysis manipulation spectrum of This paper employs the Gaussian–Laplacian to separately treat different frequency bands texture. First, we generate three images corresponding levels input capture intrinsic details. Then, aggregate features extracted gray color texture using bioinspired descriptors, information-theoretic measures, gray-level co-occurrence matrix feature Haralick statistical descriptors into single vector. Such aggregation aims at producing that characterize textures their maximum extent, unlike employing each descriptor separately, which may lose some relevant textural information reduce classification performance. The experimental results on histopathologic datasets have shown advantages proposed method compared state-of-the-art approaches. findings emphasize importance multiscale corroborate mentioned above complementary.