A single-parameter text-line extraction algorithm is described along with an efJicient technique for estimating the optimal value for the parameter for individual images without need for ground truth. The algorithm is based on three simple tree operations, cut, glue and jlip. An XYtree representing the segmentation is incrementally transformed to reflect a change in the parameter while intrinsi...

We present an algorithm for image segmentation with irregular pyramids. Instead of starting with the original pixel grid, we rst apply an adaptive Voronoi tessellation to the image. For irregular pyramid construction we present a Hoppeld neural network which controls the decimation process. The validity of our approach is demonstrated by several examples in image segmentation.

Segmentation of images based in texture is a fundamental task in many computer vision environments. There are many objects in the real word whose main characteristic is not their mean gray level but other features related with their texture (like grain size, orientation, etc.) The existing methods for texture segmentation differ in the definition of the texture descriptors, and they range from ...

Overlapping a series of adaptive simple mathematical models can be used for image segmentation or data clustering. This paper presents model based evolutionary optimisation segmentation algorithms that incrementally include additional features to the model. Several artificially created images and real images are used to demonstrate the ability of the proposed algorithms.

This paper presents an integrated method for adaptive segmentation of brain tissues in three-dimensional (3-D) MRI (Magnetic Resonance Imaging) images. The method intends to do the volume segmentation in a slice-by-slice manner. Firstly, some slices in the volume are segmented using an automatic algorithm composed of watershed, fuzzy clustering (Fuzzy C-Means) and resegmentation. Then their adj...

In this paper we propose an object-driven online segmentation system for mobile robots. Among existing tracking and segmentation methods, the methods themselves are highly emphasized while properties of objects are usually not exploited enough. We propose an adaptive selection mechanism based on the properties of the objects to automatically choose an optimal tracking algorithm. For textured ob...

The adaptive distance preserving level set (ADPLS) method is fast and not dependent on the initial contour for the segmentation of images with intensity inhomogeneity, but it often leads to segmentation with compromised accuracy. And the local binary fitting model (LBF) method can achieve segmentation with higher accuracy but with low speed and sensitivity to initial contour placements. In this...

In the paper, we proposed a model called Adaptive Threshold Level Set Without Edge that is applied on the segmentation of GM in brain MR images. Threshold to find the boundary of GM is automatically obtained by fuzzy c mean algorithm. A similarity index (SI) is used for quantitative evaluation of the segmentation results. By testing 134 planar MR brain images and comparing to the gold standard ...

The major aim of this paper consists of a comprehensive quantitative evaluation of adaptive texture descriptors when integrated into an unsupervised image segmentation framework. The techniques involved in this evaluation are: the standard and rotation invariant Local Binary Pattern (LBP) operators, multichannel texture decomposition based on Gabor filters and a recently proposed technique that...

This paper compares the correlation dimension (D2) and Higuchi fractal dimension (HFD) approaches in estimating BIS index based on of electroencephalogram (EEG). The single-channel EEG data was captured in both ICU and operating room and different anesthetic drugs, including propofol and isoflurane were used. For better analysis, application of adaptive segmentation on EEG signal for estimating...