Wyniki wyszukiwania - BazTech

1

A connected component labelling algorithm for greyscale mammography image processing as a pre-processing tool

Yapa R. D., Harada K.

Machine Graphics and Vision

|

2007

|

Vol. 16, No. 3/4

305-327

EN

A new algorithm for connected component-labelling is presented in this paper. The proposed algorithm requires only one scan through an image for labelling connected components. Once this algorithm encounters a starting pixel of a component, it traces in full all the contour pixels and all internal pixels of that particular component. The algorithm recognizes components of the image one at a time while scanning in the raster order. This property will be useful in areas such as image matching, image registration, content-based information retrieval and image segmentation. It is also capable of extracting the contour pixels of an image and storing them in a clock-wise directional order, which will provide useful information in many applications. The algorithm assigns consecutive label numbers to different components, and therefore requires a minimum number of labels. We have used the algorithm in mammography image processing as a pre-processing tool, and have demonstrated the possibility of using it for breast tissue segmentation and for detecting regions of interest in breast tissue. Another important advantage of the algorithm is that it can be used as a content-based image retrieval tool for retrieving images based on the visual contents of a given image. This would be very useful in retrieving related images from large scale medical databases.

2

Solving point-feature labeling placement problem by parallel Hopfield neural network on GPU graphics card

He Z., Harada K.

Machine Graphics and Vision

|

2006

|

Vol. 15, No. 1

99-120

EN

This paper discusses the application of parallel Hopfield neural networks in solving the point-feature labeling placement (PFLP) problem by using programmable graphics hardware found in a commodity PC. In this paper, we focus on two aspects. The first aspect concerns mapping the PFLP onto parallel Hopfield neural network. The second aspect is the detailed method of implementing the parallel Hopfield neural network on graphics hardware. We demonstrate the effectiveness of implementing the parallel Hopfield network by solving the PFLP problem. Moreover, our proposal makes use of the advantages of the parallel Hopfield network on low-cost platforms.

3

Mesh construction based on differential surface property

Miyazaki R., Harada K.

Machine Graphics and Vision

|

2006

|

Vol. 15, No. 2

197-210

EN

In this paper, we propose a novel method for extraction of a feature from range data to create of a simplified triangular mesh. The feature extracted by our method is based on the surface property. In the past, vertices on a highly curved surface were extracted from range data, and the extracted vertices were triangulated to construct a simplified triangular mesh. However, the vertices extracted with such a method were not enough to create a simplified triangular mesh in some cases. For example, when the shape shows less change in one direction, like a parabolic shape, too many vertices are extracted by the traditional method. We introduce a surface property defined by a combination of the mean curvature and the gaussian curvature. Vertices are classified into three categories according to the surface property. Then, the structures consisting of vertices and edges are extracted suitably for constructing a simplified triangular mesh. Following that, we can create the desirable trianguiation through constrained Delaunay trianguiation using the extracted structures.

4

A new topology-based watermarking method for layered 3D triangular mesh models

Rahman M., Kaneda K., Harada K.

Machine Graphics and Vision

|

2005

|

Vol. 14, No. 4

427-439

EN

A new topology-based watermarking method is proposed to embed information in objects with layered 3D triangular meshes such as those reconstructed from CT or MRI data. The main idea of the method is to compare the heights of the vertices of a triangle lying in the same layer. A watermark message is converted into a binary bit sequence, and then embedded into the modelin such a way that the first vertex of a triangle in the upper level carries information 1, and the first vertex of a triangle in the lower level carries information 0. For experimental purposes, a watermark message is embedded in a mouse embryo model. It is robust against translation, rotation, re-sectioning, local deformation and scaling. It left some artifacts after re-arrangement of local or global numbering. It is useful for shape sensitive 3D geometric models.

5

Vertex correspondence between polygons in different applications

Dharmaratne A., Harada K.

Machine Graphics and Vision

|

2003

|

Vol. 12, No. 3

311-333

EN

This paper describes some applications in computer graphics and computational geometry where establishing a vertex correspondence between a pair of objects is necessary. We present an algorithm for establishing a correspondence between the vertices of a pair of polygons by inserting new vertices, regardless of the polygons' locations, orientations, sizes, shapes, or numbers of vertices. The new vertices can be inserted either at the existing vertices or along the corresponding edges of the polygons. After establishing a successful correspondence between the vertices, its role in compatible convex decompositions, skeleton construction and polygon morphing is discussed.

6

Sequential projective reconstruction with factorization

Xu X., Harada K.

Machine Graphics and Vision

|

2003

|

Vol. 12, No. 4

477-487

EN

The objective of the present study is to propose a new sequential projective factorization approach for recovering the structure and motion. In contrast to the previous sequential factorization methods or nonlinear filter methods, it starts with the first two images and the accurate result can be obtained even for the first image pair without considering provision of a reliable initial estimation. In each recursive step, the infinite sequences can be effectively handled while keeping the size of the factorization matrix unchanged. The experimental data in the paper show that sufficient accuracy can be attained with absence of noise.

7

A complement to the 8 point algorith in matching uncalibrated image sequences of a scene

Tefera D. A., Harada K.

Machine Graphics and Vision

|

2001

|

Vol. 10, No. 1

15-28

EN

This paper presents an algorithm which generates an initial set of matching corner points between two uncalibrated images of the same scene. Matching different images of a single scene remains one of the bottlenecks in computer vision. The major part of matching algorithms is to recover the epipolar geometry, which heavily depends on the accuracy of the set of eigh corresponding points that are used to calculate the fundamental matrix between the sets of images under consideration. Thus, an algorithm to locate a reliable initial guess for matching corner points is proposed. The algorithm has been tested and shown reliable with different types of images indicating its ability to localize reliable sets of matching corner points. With the help of non-maximum response suppression technique our approach yields better results than other methods.

8

Compatible convex decompositions of simple polygons

Dharmaratne A., Harada K.

Machine Graphics and Vision

|

2000

|

Vol. 9, No. 4

851-866

EN

In this paper, we presnt an algorithm for constructing compatible convex decompositions of two simple polygons. Given two simple polygons with an equal number of vertices, convex decompositions of these polygons are to be compatible if there exist a one-to-one mapping between them such that the two polygons are decoposed into an equal number of convex sub-polygons and the coresponding sub-polygons are defined by the corresponding sequence of vertices. In general, it may not be possible to decompose two polygons compatibly if additional vertices inside the polygon (commonly called Steiner points) are not allowed. Our algorithm calculates a compatible convex decomposition of two polugons with/without introducing Steiner points.