• 대한원격탐사학회지
• /
• 제5권2호
• /
• pp.133-145
• /
• 1989

This paper describes an implementation of linear feature extraction algorithms for satellite images and a method of automatic GCP(Ground Control Point) filing using the extracted linear feature. We propose a new linear feature extraction algorithm which uses magnitude and direction information of edges. The result of applying the proposed algorithm to satellite images are presented and compared with those of the other algorithms. By using the proposed algorithm, automatic GCP filing was successfully performed.

• 한국통신학회논문지
• /
• 제23권8호
• /
• pp.1998-2009
• /
• 1998

본 논문에서는 배경의 움직임이 유발되는 능동 CCD 카메라를 통하여 실시간으로 포착되는 영상 데이터를 대상으로 카메라의 사전 설치 정보나 좌표 보정(calibration) 없이 강체(rigid body) 혹은 비 강체(non-rigid body)의 움직이는 물체를 추출하고 이의 이동 방향을 판단하여, 추적하는 효율적인 알고리즘을 제안한다. 이동 물체의 영역분할을 위하여 동체의 형태를 규정하는 특징 점을 추출하고, 시간에 따른 특징 점의 이동 벡터로 구성된 특정 플로우 필드(feature flow field)를 구한 후 이들을 다차원 특정 공간상에서 군집화(clustering)함으로써 동체를 추출한다. 제안하는 IRMAS(lncremenatal Rotational Minimum Angle Search)에 의하여 군집화된 특정점들의 볼록 다각형(convex hull)올 구함으로써 이동 물체의 개괄적인 외곽 형태를 재 구성한다. 또한, 이동 궤적의 갑작스러운 변화를 가져올 수 있는 동작 특성을 가지는 이동 물체의 효과적인 추적을 목적으로 개선된 선형 예측기를 사용하였다. 이동 궤적 예측기는 기존의 선형 예측기의 차수를 이동의 변화도에 따라 적응적으로 조정함으로써 예측 오차를 감소시켜, 빠른 속도로 이동 궤적에 수렴한다.

• 건축역사연구
• /
• 제18권3호
• /
• pp.27-44
• /
• 2009

In Korea, the direction of houses are typically determined by considering the directional orientation and shape of the mountain range rather than ignoring the geographical feature of the mountain range. Traditional villages of Korea are known to have very particular ways of adopting the geomantic surroundings of natural environment. This is very true especially have a high mountain in the back and a lower mountain in front. At the same time, most of the houses tend to prefer south as a man direction so that they can receive more sun light. However, if the mountain range faces north, it will not be easy to determine the directional orientation of houses. This paper, therefore, tries to identify how the houses of villages facing north, direst the orientation. This, the northern village, solves the problem by facing all direction rather than one major direction. The houses of the villages facing north, tend to revise the direction by changing the back mountain(주산) or front mountain(인산) that helps them change the direction towards he range of eastern or western direction. As a result, the houses tend to the direction towards east and wes compared to north and south. The directional orientation of houses was clearly distributed or concentrated by depending of the shape and directional orientation of the mountain range. This kind of research let us know the relationship between the natural north direction, the direction of geomantic surrounding, and the direction of houses in traditional Korean villages.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2005년도 Proceedings of ISRS 2005
• /
• pp.575-578
• /
• 2005

Topographical images, in case of aerial or satellite images, are usually similar in colors and textures, and complex in shapes. Thus we have to use shape features of images for efficiently retrieving a query image from topographical image databases. In this paper, we propose a shape feature extraction method which is suitable for topographical images. This method, which improves the existing projection in the Cartesian coordinates, performs the projection operation in the polar coordinates. This method extracts three attributes, namely the number of region pixels, the boundary pixel length of the region from the centroid, the number of alternations between region and background, along each angular direction of the polar coordinates. It extracts the features of complex shape objects which may have holes and disconnected regions. An advantage of our method is that it is invariant to rotation/scale/translation of images. Finally we show the advantages of our method through experiments by comparing it with CSS which is one of the most successful methods in the area of shape feature extraction

• 한국정보처리학회논문지
• /
• 제7권5호
• /
• pp.1361-1369
• /
• 2000

For the implementation of 3D Sound Localization system, the binaural filtering by HRTFs is generally employed. But the HRTF filter is of high order and its coefficients for all directions have to be stored, which imposes a rather large memory requirement. To cope with this, research works have centered on obtaining low dimensional HRTF representations without significant loss of information and synthesizing the original HRTF efficiently, by means of feature extraction methods for multivariate dat including PCA. In these researches, conventional linear PCA was applied to the frequency domain HRTF data and using relatively small number of principal components the original HRTFs could be synthesized in approximation. In this paper we applied neural network based nonlinear PCA model (NLPCA) and the nonlinear PLS repression model (NLPLS) for this low dimensional HRTF modeling and analyze the results in comparison with the PCA. The NLPCA that performs projection of data onto the nonlinear surfaces showed the capability of more efficient HRTF feature extraction than linear PCA and the NLPLS regression model that incorporates the direction information in feature extraction yielded more stable results in synthesizing general HRTFs not included in the model training.

• 한국멀티미디어학회논문지
• /
• 제20권5호
• /
• pp.748-757
• /
• 2017

In order to recognize the facial expressions, good features that can express the facial expressions are essential. It is also essential to find the characteristic areas where facial expressions appear discriminatively. In this study, we propose a directional LBP feature for facial expression recognition and a method of finding directional LBP operation and feature region for facial expression classification. The proposed directional LBP features to characterize facial fine micro-patterns are defined by LBP operation factors (direction and size of operation mask) and feature regions through AdaBoost learning. The facial expression classifier is implemented as a SVM classifier based on learned discriminant region and directional LBP operation factors. In order to verify the validity of the proposed method, facial expression recognition performance was measured in terms of accuracy, sensitivity, and specificity. Experimental results show that the proposed directional LBP and its learning method are useful for facial expression recognition.

• 한국컴퓨터정보학회논문지
• /
• 제11권3호
• /
• pp.151-159
• /
• 2006

본 논문에서는 두 대의 카메라 영상으로부터 얼굴의 포즈를 추정하는 방법을 제안한다. 제안된 방법은 먼저 두 얼굴 영상으로부터 대응되는 눈썹, 눈, 입의 특징점을 추출한 다음, 스테레오 비전의 삼각법에 의해 특징점에 대한 3차원 위치를 계산한다. 그 다음에는 특징점으로 부터 삼각형을 생성하고 그 삼각형에 수직 방향을 계산함으로써 얼굴의 포즈를 계산한다. 계산된 얼굴의 포즈를 3D 얼굴 모델에 적용해 본 결과 본 논문에서 제안된 방법이 정확한 얼굴 포즈를 추정할 수 있음을 알 수 있었다.

• ETRI Journal
• /
• 제37권2호
• /
• pp.283-294
• /
• 2015

A model that precisely forecasts how much wind power is generated is critical for making decisions on power generation and infrastructure updates. Existing studies have estimated wind power from wind speed using forecasting models such as ANFIS, SMO, k-NN, and ANN. This study applies a projected clustering technique to identify wind power patterns of wind turbines; profiles the resulting characteristics; and defines hourly and daily power patterns using wind power data collected over a year-long period. A wind power pattern prediction stage uses a time interval feature that is essential for producing representative patterns through a projected clustering technique along with the existing temperature and wind direction from the classifier input. During this stage, this feature is applied to the wind speed, which is the most significant input of a forecasting model. As the test results show, nine hourly power patterns and seven daily power patterns are produced with respect to the Korean wind turbines used in this study. As a result of forecasting the hourly and daily power patterns using the temperature, wind direction, and time interval features for the wind speed, the ANFIS and SMO models show an excellent performance.

• 복식
• /
• 제56권7호
• /
• pp.148-162
• /
• 2006

The purpose of this study is to examine the concept of convergence which is one of the trendy issues as a new digital paradigm of Integrative thinking in 21st century, to analyze the plastic feature and internal meaning of convergence expressed in fashion design, and to grasp the cultural symbolism through this aesthetic analysis. Because there have been considerable discussions on convergence, centering on industrial product area associated with media, I will proceed my study on the basis of them. For this, the documentary study and practical case study have been executed. This study will be helpful to find a direction of future fashion design trend. Convergence in digital stage can be defined as a phenomenon which different functions of product move towards one direction for greater efficiency, and not only as a technical integration between functions of product, but also an extension of area. Convergence can be classified by their use as (1) convergence for convenient daily life (2) convergence with intelligent scientific technology (3) convergence for entertainment on the basis of sensual experience. The plasticity of convergence designs feature as a open dynamic structure which potentiate transformation and their internal meaning can be inquired such qualities as integrative multiplicity, efficiency, mobility, intelligence. Specially convergence fashion design has protection qualify resulting from wearability on body. Ultimately convergence fashion design as a future digital paradigm can be thought as both eco-friendly design and human-centered design from positive technology-based viewpoint, because it is easy to transform according to our environment, convenient to reserve, and efficient to enhance spatial usibility.

• 제어로봇시스템학회논문지
• /
• 제18권9호
• /
• pp.887-893
• /
• 2012

In this paper, we propose a new approach to extract directional features based on directional patterns of each region in fingerprint images. The proposed approach computes the center of gravity to extract features from fingerprint images of various shapes. According to it, we divide a fingerprint image into four regions and compute the directional values of each region. To extract directional features of each region from a fingerprint image, we spilt direction values of ridges in a region into 18 classes and compute frequency distribution of each region. Through the result of our experiment using FVC2002 DB database acquired by electronic devices, we show that directional features are effectively extracted from various fingerprint images of exceptional inputs which lost all or part of singularities. To verify the performance of the proposed approach, we explained the process to model Arch, Left, Right and Whorl class using the extracted directional features of four regions and analyzed the classification result.