Shinji Koka

Ridge Detection with the Steepest Ascent Method

We propose a new method to detect ridges from digital elevation map (DEM) data. We call it “the steepest ascent method” which is based on steepest ascent lines obtained by selecting the maximum inclined direction from eight neighbors. Then ridge lines are extracted by their steepest ascent lines on a surface. In this paper it is shown that our method is able to detect ridge lines sharply. Besides we claim that our algorithm is simple and suitable for huge area such as the entire area of Japan. Finally, we give the ridge detected terrain map of the entire area of Japan and Korea by applying “the steepest ascent method”.

A resolution reduction method for multi-resolution terrain maps

Raster images such as raster terrain maps are commonly used in computer graphics. For rapid processing such as rendering and rapid feature extraction, rapid resolution reduction methods are required that keep the quality of huge images. This study deals with the resolution reduction methods.

8k-ary Grid graph models of tabular forms

Tabular forms are commonly used in software. Those tabular forms are represented as rectangular dissections. In rectangular dissections, ruled line oriented operations such as cell merge, line and column operations are often used. With respect to ruled line oriented operations, 8k-ary grid graphs have been introduced as models of rectangular dissections that provide fast algorithms. This paper surveys octal and hexa-decimal grid graph models of rectangular dissections. First, octal grids, called octgrids, for single layer rectangular dissections and related algorithms are introduced. Next, hexa-decimal grid graphs for multiple layer rectangular dissections, called hexadeci-grids, and related algorithms are introduced. Furthermore, tetraicosa-grid graphs for rectangular solid dissections for CG applications, called tetraicosa-grids and related algorithms are introduced.

Tabular form editing with a hexadecimal grid graph model

We discuss tabular forms. It is known in the previous studies that the “octgrid” is a data structure for single layer tabular forms with ruled line preserving algorithms. And the “hexadeci-grid” is known as a data structure for multiply layered tabular forms which is a generalization of the “octgrid”. Our purpose of this paper is to propose a new algorithm based on the “hexadeci-grid” model for editing tabular forms.

A data structure for triangular dissection of multi-resolution images

In this work, the heterogeneous rectangular dissections that represent multi-resolution images of raster data are considered. Specifically, heterogeneous rectangular dissections are changed to triangular dissections in order to provide more effective feature extraction. We propose a method of generating triangular dissections that maintains “octgrid” properties and have developed a list structure suitable for extracting image features (ridges, valleys, etc.) from terrain maps. We propose a detailed list structure called “H12Code” and present examples of feature extraction using H12Code lists.

Reduction of resolution for binary images by an octal grid graph representation model

We deal with a data structure which is suitable for expression of the binary images. In general, binary images are expressed in raster data or rectangular dissections. In this paper, we describe that it is possible to reduce effectively the number of nodes in raster data represented by “octgrid” which is a type of octal grid graphs to handle heterogeneous rectangular dissections. Precisely, we propose algorithms to reduce the resolution of binary images represented by “octgrid”. Furthermore, we compare by examples our method with the Hilbert scanning, and show that our method reduces the rectangular number by less than 50% than the Hilbert scanning does, in several examples.

A Method of Ridge Detection in Triangular Dissections Generated by Homogeneous Rectangular Dissections

In order to display 3D terrain map effectively , detections of features on maps are very important. In this paper, we discuss a method for detection of ridges. It is known in the previous work that the steepest ascent method is effective for a ridge detection on terrain map represented by rectangular dissections. We will introduce the steepest ascent method in triangular dissections generated by homogeneous rectangular dissections.

A Comparison of Ridge Detection Methods for DEM Data

We deal with ridge detection methods from digital elevation map (DEM) data. As ridge detection methods, the O (N2) -time steepest ascent method and the O (N) -time discrete Lap lace transform (D.L.T.) method are known, where N is the number of cells. However, the D.L.T. method is too blurry to form ridge lines. In this paper, we introduce a 12 neighbor D.L.T. method which is a modification of the 4 neighbor D.L.T. method. And we also introduce another ridge detection method by the classification of local shapes around each cell. We can consider 32 patterns for ridges or valleys. Furthermore, we compare and evaluate their ridge detection methods in a certain area. We note that our two methods provide blurry terrain maps, but it require only O (N) -time for N cells, in comparison with the steepest ascent method.

Row manipulation in the heterogeneous tabular forms with a hexadecimal grid graph model

We consider hexadeci-grids as a model of multiply layered heterogeneous tabular forms, which are generalization of the octgrid model. We introduce column deletion and multiple row deletion algorithms on the hexadeci-grid model.

Row manipulation in the heterogenous tabular forms with an octal grid model

We propose combined algorithms such as a multiple row deletion algorithm based on octgrid graph model for a heterogeneous table.