Region-based approach for discriminant snakes

Abstract

This paper proposes a statistic framework for segmenting textured areas over real images by discriminant snakes. Our active contour model has the ability to learn different texture prototypes and generate a global statistical model from a multi-valued function. This function is generated by means of filter responses over the texture regions. Linear discriminant analysis is performed to obtain a statistical classifier embodied into the snake scheme. Given an input image composed by different texture types, a likelihood map is built and the discriminant snake deforms on it to delineate regions with similar texture descriptions according to the learned texture patterns. Our method is tested on two different image applications: aerial images and medical (ultrasound) images, and the results are very encouraging.

Geometric Methods for Vessel Visualization and Quantification — A Survey

“…Hu [38] proposes a snake with variable stiffness parameters for vessel boundary extraction to allow to adapt to strong and smooth or missing edge features. Different kinds of discrete statistical snakes have been applied to segment cross-sections of vessels in intravascular ultrasound images [75,84] and angiographies [98]. McInerney and Terzopolous [64] demonstrated the power of 2D T-snakes in application to vessel trees in angiographies.…”

GIST: an interactive, GPU-based level set segmentation tool for 3D medical images

“…These methods (and many others) focus on image edges, but the the literature documents several other strategies for fitting level sets to image data. For instance, several authors have propose using the statistics of the greyscale interior of the model to control the motion [12,13]. Alternatively, the motion of the level set can depend on a variational formulation that positions the interface to create discontinuities that best model the discontinuities in the input data [14][15][16].…”

Video object segmentation and tracking using region-based statistics

“…That is, if a portion of the snake is inside the face region, it is pushed outside, and if it is outside the face region, it is pulled inside. A method for image segmentation that uses the texture information as a region descriptor has been presented in [18]. The method computes a set of derivative features for multiple spatial orientations and scales to capture the texture information.…”