Applications of self-organization networks spatially isomorphic to patterns

Shanmukh, K.; Murthy, C.N.S. Ganesh; Venkatesh, Y. V.

doi:10.1016/s0020-0255(98)10067-1

Cited by 3 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They used a modified neural networks proposed by Ganesh Murthy and Venkatesh [2] and used by Shanmukh et al [3], that applies isomorphic self-organizing networks (SON) to the patterns. Venkatesh and Rishikesh explored the simplicity and elegance of this model to adapt to the problem of contour extraction.…”

mentioning

confidence: 99%

Analysis of mammogram using self-organizing neural networks based on spatial isomorphism

Ferreira

Nascimento

Tsang

et al. 2007

2007 International Joint Conference on Neural Networks

View full text Add to dashboard Cite

Abstract-The correct segmentation and measurement of mammography images is of fundamental importance for the development of automatic or computer-aided cancer detection systems. In this paper we propose a method to segment mammogram image using a self-organizing neural network based on spatial isomorphism. The method used is a modified version of the algorithm proposed by Venkatesh and Rishikesh [1] to extract object boundaries in an image. This model explores the principle of spatial isomorphism and selforganization in order to create flexible contours that characterize shapes in images. We modified the original algorithm to overcame problems of local minimum, poor performance for image object with large concavity and imprecise results when simple or far from object border contour are chosen. A comparison of both algorithm and original segmentation used by the MIAS database [9] is presented.

show abstract

mentioning

confidence: 99%

Analysis of mammogram using self-organizing neural networks based on spatial isomorphism

Ferreira

Nascimento

Tsang

et al. 2007

2007 International Joint Conference on Neural Networks

View full text Add to dashboard Cite

show abstract

“…In other words, human recognition could be considered to be guided by the amount of mental deformation the exemplar has to undergo to match the given unknown pattern. This idea has been exploited in object and character recognition by [9], [12]. In these references, the authors have employed a network of neurons for each exemplar, with the neurons in each network arranged in exactly the same way as the corresponding exemplar.…”

Section: Network Models Based On Human Perceptionmentioning

confidence: 99%

“…This approach has already been shown to be efficient for character and object recognition [12]. Application to face recognition and tracking in a video sequence is under study.…”

Section: Open Contoursmentioning

confidence: 99%

Modelling active contours using neural networks isomorphic to boundaries

Venkatesh

Rishikesh²

Proceedings of International Conference on Neural Networks (ICNN'97)

Self Cite

View full text Add to dashboard Cite

W e propose a new technique, basedIn using deformable templates, the a priori shape (or on self-organization, for localizing salient contours boundary) information is #assumed to be available in the in an image, with applications to, for instance, ob-form of an inexact model of the object as a sketch or a ject and character recognition, stereopsis and motion binary template. The problem of object recognition is tracking. A neuronal network which is isomorphic then formulated as the pralcess of matching a deformable to the template/initial contour is created. This net-contour to the object boundary in an input image. For work acts as an active contour, which, using self-an early result on the representation of the deformation organization, undergoes deformation in an attempt as a probabilistic transforimation on the prototype temto cling on to the nearest salient contour in the test plate, see Grenander [l]. image. The application area3 Of the n~d e lProposed Snakes, proposed by K t s s et al.[2], are an example are similar to 'snake' [2]' but distinct from it both of a technique of contour pxtraction by meof energy in the underlying mathematics and impb"tation. nlininization. In this reference , the authors choose en-The new technique w ergy functionals which attract the snakes to salient fea-Indexing tfx1-1~3 :Active DefOmable tures (lines, edges, and terminations) in images. Elom a given starting point, the snake deforms itself into wn- Mapping OfNeural network, Self-formity with the nearest s,dient contour. In effect, their OryurLizuLz'orL, Snukw.model is a controlled-continuity spline under the influence of image forces and external constraint forces. The internal spline forces serve to impose a smoothness con-1669

show abstract

Imposing tree-based topologies onto self organizing maps

Astudillo

Oommen

2011

Information Sciences

View full text Add to dashboard Cite

Summary. The beauty of the Kohonen map is that it has the property of organizing the codebook vectors, which represent the data points, both with respect to the underlying distribution and topologically. This topology is traditionally linear, even though the underlying lattice could be a grid, and this has been used in a variety of applications [19,25,28]. The most prominent efforts to render the topology to be structured involves the Evolving Tree (ET) due to Pakkanen et al. [26], and the Self-Organizing Tree Maps (SOTM) due to Guan et al. [15], among others. In this paper we propose a strategy, the Tree-based Topology-Oriented SOM (TTO-SOM) by which we can impose an arbitrary, user-defined, tree-like topology onto the codebooks. Such an imposition enforces a neighborhood phenomenon which is based on the user-defined tree, and consequently renders the so-called bubble of activity to be drastically different from the ones defined in the prior literature. The map learnt as a consequence of training with the TTO-SOM is able to infer both the distribution of the data and its structured topology interpreted via the perspective of the user-defined tree. The TTO-SOM also reveals multi-resolution capabilities, which are helpful for representing the original data set with different numbers of points, and this can be obtained without the necessity of recomputing the whole tree. The ability to extract an skeleton, which is a "stick-like" representation of the image in a lower dimensional space, is discussed as well. These properties has been confirmed by our experimental results on a variety of data sets.

show abstract

Applications of self-organization networks spatially isomorphic to patterns

Cited by 3 publications

References 15 publications

Analysis of mammogram using self-organizing neural networks based on spatial isomorphism

Analysis of mammogram using self-organizing neural networks based on spatial isomorphism

Modelling active contours using neural networks isomorphic to boundaries

Imposing tree-based topologies onto self organizing maps

Contact Info

Product

Resources

About