Postal envelope address block location by fractal-based approach

Lacunarity as a Texture Measure for Address Block Segmentation

“…The time complexity (O(n) ) of the regiongrowing algorithm is the same as in other approaches. But, simplicity of this approach gives a time performance gain (6 times faster), compared with [2], which performs an iterative process (k-means algorithm) and uses large box sizes for computing the fractal dimension.…”

“…The solution appears to work fast for well-constrained envelopes, whereby a large separation exists between the image regions since they mentioned a large drawback in the figures if the envelopes have more than one stamp for example. Eiterer et al [2] present a segmentation method based on calculation of fractal dimension from 2D variation procedure and k-means clustering. The authors have also computed the influence of box size r used in each image pixel.…”

Similarity Based Cross-Section Segmentation in Rough Log End Images

“…State-of-the-art region-based segmentation approaches are based on graph cuts ( [11], [12]), active contours ( [13], [14]) or mean shift ( [15]) algorithms. In the past two decades manifold publications showed up high level segmentation algorithms combined with well matured texture features like Markov-random-eld statistics ( [16]), Wavelet features ( [17], [18]), Gabor lters ( [19]) and Fractal features ( [20]). Furthermore, there exist many more textural spectral-temporal features (literature review see [21]) and various new texture features and extensions were published (e.g.…”

Texture Based Segmentation using Statistical Properties for Mammographic Images

“…Texture image segmentation identifies image regions that have homogeneous with respect to a selected texture measure. Recent approaches to texture based segmentation are based on linear transforms and multiresolution feature extraction [1], Markov random filed models [2,3], Wavelets [4 -6] and fractal dimension [7]. Although unsupervised texture-based image segmentation is not a novel approach, these have limited adoption due to their high computational complexity.…”