• 한국의류산업학회지
• /
• 제2권3호
• /
• pp.246-252
• /
• 2000

The purpose of this study was to analyze the characteristics of the factor structure in Clothing Silhouette Image and the differences of Image Evaluation related to Grouping of Line and Grouping of Body Emphasis using Silhouette Image Factor and Life Space as variables. Clothing Silhouette Image was estimated by the photographs of 9 Life Spaces ${\times}$ 23 Clothing Silhouettes with 15 semantic differentiated by-polar scales. The major finding were : The factor structure of Clothing Silhouette Image were found to include 4 factor dimensions-Capability Evaluation, Function, Activity. It was clear that Silhouette Image were affected by Life Spaces. There were the differences of Image Evaluation related to Grouping of Line and Grouping of Body Emphasis. There were the differences of Life Space Evaluation related to Grouping of Line and Grouping of Body Emphasis. Therefore it is important to considerate Life Space for selecting Clothing Silhouette.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2002년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.59-64
• /
• 2002

This paper presents a method for line segment extraction for 3-d building reconstruction. Building roofs are described as a set of planar polygonal patches, each of which is extracted by watershed-based image segmentation, line segment matching and coplanar grouping. Coplanar grouping and polygonal patch formation are performed per region by selecting 3-d line segments that are matched using epipolar geometry and flight information. The algorithm has been applied to high resolution aerial images and the results show accurate 3-d building reconstruction.

• 전기전자학회논문지
• /
• 제16권2호
• /
• pp.95-101
• /
• 2012

위성 영상에서 건물형태를 발생하기위한 새로운 접근방식이 본 논문에서 제안되었다. 제안된 알고리즘은 낮은 수준의 선소들을 연결하고 유사한 개체들을 군집화하기 위해 선소 측정함수가 적용된 신경망이다. 제안된 신경망은 윤곽선 영상에서 추출된 윤곽선들을 군집화 목적으로 사용된다. 본 논문에서는 3차원 선소의 오류에 의한 군집화 결과의 비현실적 건물모델의 발생을 근원적으로 차단하기 위하여, 높이 정보를 이용한 계층적 군집화를 제안하였다. 제안된 새로운 거리척도의 신경망과 군집화를 통해 성공적인 건물모델의 재구성을 실험으로 보여주었다.

• 한국의류학회지
• /
• 제26권11호
• /
• pp.1558-1569
• /
• 2002

Two components of brand image are brand association related to brand properties and brand persona which is measured by descriptive words. The purpose of this study is to analyze the brand image of apparel by qualitative approach including natural grouping method suggested by Aaker. For this, face-to-face interviews were carried out in March and April 200l.11 interviewees were respectively asked to classify' pre-selected several tens of apparel brands based on their image differentiation, and then to explain the reason of grouping and to describe resultant brand groups. In this process, many brand image associations and brand persona-descriptive words were collected. 9 types of brand association were identified, and these were summarized as three factors suggested by Keller -attributes, benefits, and attitudes/evaluations. And 3 words which used to refer brand image frequently -dokteukhan (unique), simple, and yeosungseureowoon (feminine)- were interpreted in their meaning. Brand persona-descriptive words implied diverse meaning which were dependent on context.

• 전자공학회논문지S
• /
• 제34S권3호
• /
• pp.63-73
• /
• 1997

For the practical applications of the EIT technology, it is essential to reconstruct sttic images iwth a higher spatial resolution in a reasonalble amount of processing time. Using the conventional EIT static image reconstruction algorithms, however, the processing time increases exponential with poor convergence characteristics as we try to get a higher spatial resolution. In order to overcome this problem, we developed a recursive mesh grouping method based on the Fuzzy-GA like algorithm. Computational simulation using the well-known improve dewton-raphson method with the proposed recursive mesh grouping algorithm shows a promising result that we can significantly reduce the processing time in the reconstruction of EIT static images of a higher spatial resolution.

• 스마트미디어저널
• /
• 제1권2호
• /
• pp.23-33
• /
• 2012

본 논문에서는 영상처리 및 컴퓨터비전 분야에서 다양하게 응용되고 있는 텐서보팅(tensor voting)의 특징을 고찰하고자 한다. 일반적으로 텐서보팅은 점 (points), 곡선 성분 (curve elements), 그리고 표면 조각 성분 (surface patch elements)로 주어진 n-차원의 데이터들로부터 교차점 (junction), 곡선 (curve), 영역 (region), 그리고 표면 (surface)과 같은 구조 특징을 추론할 수 있다. 영상 및 장면에서 구조적 추론을 이용한 지각기반 그룹핑 (perceptual grouping) 방법들이 다양한 분야에서 연구되어 응용되고 있다. 텐서보링은 잡음에 강건한 특징을 제공하며 다양한 응용을 통하여 효율성을 보여주고 있다.

• International Journal of Contents
• /
• 제1권1호
• /
• pp.1-5
• /
• 2005

This paper presents the development and application of electrical resistance imaging techniques for the visualization of two-phase flow fields. Two algorithms, the so-called the mesh grouping and the boundary estimation, are described for potential applications of electrical resistance tomography (ERT) and results from extensive numerical simulations are also presented. In the electrical resistance imaging for two-phase flows, numerical meshes fairly belonging to each phase can be grouped to improve the reconstruction performance. In many cases, the detection of phase boundary is a key subject and a mathematical model to estimate phase boundary can be formulated in a different manner. Our results indicated that the mesh grouping algorithm is effective to enhance computational performance and image quality, and boundary estimation algorithm to determine the phase boundary directly.

• IEIE Transactions on Smart Processing and Computing
• /
• 제3권6호
• /
• pp.358-365
• /
• 2014

This paper proposes a method that reduces the flickering effect caused by A-GLG (Adaptive Gray-Level Grouping) during video contrast enhancement. Of the GLG series, A-GLG shows the best contrast enhancement performance. The GLG series is based on histogram grouping. Histogram grouping is calculated differently between the continuous frames with a similar histogram and causes a subtle change in the transformation function. This is the reason for flickering effect when the video contrast is enhanced by A-GLG. To reduce the flickering effect caused by A-GLG, the proposed method calculates a modified transformation function. The modified transformation function is calculated using a previous and current transformation function applied with a weight separately. The proposed method was compared with A-GLG for flickering effect reduction and video contrast enhancement. Through the experimental results, the proposed method showed not only a reduced flickering effect, but also video contrast enhancement.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2002년도 춘계종합학술대회
• /
• pp.177-180
• /
• 2002

본 논문에서는 이진영상 데이터의 저작권 보호를 위한 디지털 워터마킹 방법을 제안하였다. 제안 방법은 먼저 이진영상을 기하학적 특징이 포함된 특징영역과 그 외의 일반영역으로 분리한다. 그리고 기하학적 특징이 포함된 이진영상의 특징영역을 보존하면서 인증을 위한 워터마크를 일반영역에만 삽입, 검출하는 효과적인 워터마킹 방법을 연구하였다. 특징영역은 특수런을 사용한 런길이부호화를 이용하여 일반영역과 분리하였다. 워터마크의 비가시성을 위해 워터마크는 일반영역 중에서 화소변화에 대한 민감도를 고려하여 삽입하였다. 제안 방법을 문사, 서명, 도장, 지문영상에 적용하여 성능을 평가하였다. 실험 결과 제안 방법은 원 영상의 중요한 특징영역을 보존하고, 또한 워터마킹된 영상의 비가시성이 높음을 알 수 있었다.

• 융합신호처리학회논문지
• /
• 제15권4호
• /
• pp.128-134
• /
• 2014

본 연구에서는 시공간 영상 해석을 이용한 미소혈관내 백혈구 운동의 검출 방법을 제안한다. 혈관벽에 부착하는 백혈구 운동은 영상내 혈관벽의 윤곽선을 따라 움직이는 것으로 시각화될 수 있다. 제안 방법에서 백혈구는 혈관벽의 윤곽선을 따라 움직인다는 구속조건을 사용하며 시공간 영상 해석방법의 사용에 의해 백혈구 운동을 검출한다. 생성된 시공간 영상은 특수한 목적의 방향 선택 필터에 의해 처리되고 후속의 분류처리가 행해진다. 이 후속의 분류처리는 단순한 임계값 및 윤곽선 처리에 의해 획득된 모든 성분중에서 백혈구 궤적 성분을 선택하고 분류한다. 실험 결과, 제안 방법은 복수개의 백혈구 흔적이 서로 교차할 때에도 백혈구 운동을 안정하게 검출할 수 있음을 보여준다.