Two-part Segmentation of Text Documents

We consider the problem of segmenting text documents that have a
two-part structure such as a problem part and a solution part. Documents
of this genre include incident reports that typically involve
description of events relating to a problem followed by those pertaining
to the solution that was tried. Segmenting such documents
into the component two parts would render them usable in knowledge
reuse frameworks such as Case-Based Reasoning. This segmentation
problem presents a hard case for traditional text segmentation
due to the lexical inter-relatedness of the segments. We develop
a two-part segmentation technique that can harness a corpus
of similar documents to model the behavior of the two segments
and their inter-relatedness using language models and translation
models respectively. In particular, we use separate language models
for the problem and solution segment types, whereas the interrelatedness
between segment types is modeled using an IBM Model
1 translation model. We model documents as being generated starting
from the problem part that comprises of words sampled from
the problem language model, followed by the solution part whose
words are sampled either from the solution language model or from
a translation model conditioned on the words already chosen in the
problem part. We show, through an extensive set of experiments on
real-world data, that our approach outperforms the state-of-the-art
text segmentation algorithms in the accuracy of segmentation, and
that such improved accuracy translates well to improved usability
in Case-based Reasoning systems. We also analyze the robustness
of our technique to varying amounts and types of noise and empirically
illustrate that our technique is quite noise tolerant, and
degrades gracefully with increasing amounts of noise

Segmentation of optic disc in retinal images using an improved gradient vector flow algorithm

Image segmentation plays an important role in the analysis of retinal images as the extraction of the optic disk provides important cues for accurate diagnosis of various retinopathic diseases. In recent years, gradient vector flow (GVF) based algorithms have been used successfully to successfully segment a variety of medical imagery. However, due to the compromise of internal and external energy forces within the resulting partial differential equations, these methods can lead to less accurate segmentation results in certain cases. In this paper, we propose the use of a new mean shift-based GVF segmentation algorithm that drives the internal/external energies towards the correct direction. The proposed method incorporates a mean shift operation within the standard GVF cost function to arrive at a more accurate segmentation. Experimental results on a large dataset of retinal images demonstrate that the presented method optimally detects the border of the optic disc.

Mean shift based gradient vector flow for image segmentation

In recent years, gradient vector flow (GVF) based algorithms have been successfully used to segment a variety of 2-D and 3-D imagery. However, due to the compromise of internal and external energy forces within the resulting partial differential equations, these methods may lead to biased segmentation results. In this paper, we propose MSGVF, a mean shift based GVF segmentation algorithm that can successfully locate the correct borders. MSGVF is developed so that when the contour reaches equilibrium, the various forces resulting from the different energy terms are balanced. In addition, the smoothness constraint of image pixels is kept so that over- or under-segmentation can be reduced. Experimental results on publicly accessible datasets of dermoscopic and optic disc images demonstrate that the proposed method effectively detects the borders of the objects of interest.

What motion distributions yield global transparency and spatial segmentation?

We have examined the ability of observers to parse bimodal local-motion distributions into two global motion surfaces, either overlapping (yielding transparent motion) or spatially segregated (yielding a motion boundary). The stimuli were random dot kinematograms in which the direction of motion of each dot was drawn from one of two rectangular probability distributions. A wide range of direction distribution widths and separations was tested. The ability to discriminate the direction of motion of one of the two motion surfaces from the direction of a comparison stimulus was used as an objective test of the perception of two discrete surfaces. Performance for both transparent and spatially segregated motion was remarkably good, being only slightly inferior to that achieved with a single global motion surface. Performance was consistently better for segregated motion than for transparency. Whereas transparent motion was only perceived with direction distributions which were separated by a significant gap, segregated motion could be seen with abutting or even partially overlapping direction distributions. For transparency, the critical gap increased with the range of directions in the distribution. This result does not support models in which transparency depends on detection of a minimum size of gap defining a bimodal direction distribution. We suggest, instead, that the operations which detect bimodality are scaled (in the direction domain) with the overall range of distributions. This yields a flexible, adaptive system that determines whether a gap in the direction distribution serves as a segmentation cue or is smoothed as part of a unitary computation of global motion.

Continuity - based and discontinuity - based segmentation in transparent and spatially segregated global motion

The mechanisms underlying the parsing of a spatial distribution of velocity vectors into two adjacent (spatially segregated) or overlapping (transparent) motion surfaces were examined using random dot kinematograms. Parsing might occur using either of two principles. Surfaces might be defined on the basis of similarity of motion vectors and then sharp perceptual boundaries drawn between different surfaces (continuity-based segmentation). Alternatively, detection of a high gradient of direction or speed separating the motion surfaces might drive the process (discontinuity-based segmentation). To establish which method is used, we examined the effect of blurring the motion direction gradient. In the case of a sharp direction gradient, each dot had one of two directions differing by 135°. With a shallow gradient, most dots had one of two directions but the directions of the remainder spanned the range between one motion-defined surface and the other. In the spatial segregation case the gradient defined a central boundary separating two regions. In the transparent version the dots were randomly positioned. In both cases all dots moved with the same speed and existed for only two frames before being randomly replaced. The ability of observers to parse the motion distribution was measured in terms of their ability to discriminate the direction of one of the two surfaces. Performance was hardly affected by spreading the gradient over at least 25% of the dots (corresponding to a 1° strip in the segregation case). We conclude that detection of sharp velocity gradients is not necessary for distinguishing different motion surfaces.

Application-Specific Instruction Set Processor for SoC implementation of Modern Signal Processing Algorithms

A novel application-specific instruction set processor (ASIP) for use in the construction of modern signal processing systems is presented. This is a flexible device that can be used in the construction of array processor systems for the real-time implementation of functions such as singular-value decomposition (SVD) and QR decomposition (QRD), as well as other important matrix computations. It uses a coordinate rotation digital computer (CORDIC) module to perform arithmetic operations and several approaches are adopted to achieve high performance including pipelining of the micro-rotations, the use of parallel instructions and a dual-bus architecture. In addition, a novel method for scale factor correction is presented which only needs to be applied once at the end of the computation. This also reduces computation time and enhances performance. Methods are described which allow this processor to be used in reduced dimension (i.e., folded) array processor structures that allow tradeoffs between hardware and performance. The net result is a flexible matrix computational processing element (PE) whose functionality can be changed under program control for use in a wider range of scenarios than previous work. Details are presented of the results of a design study, which considers the application of this decomposition PE architecture in a combined SVD/QRD system and demonstrates that a combination of high performance and efficient silicon implementation are achievable. © 2005 IEEE.

Fuzzy Coding of Genetic Algorithms

A well-cited paper suggesting fuzzy coding as an alternative to the conventional binary, grey and floating-point representations used in genetic algorithms.