• 전자공학회논문지CI
• /
• 제47권2호
• /
• pp.11-20
• /
• 2010

본 논문에서는 피부색 정보를 이용한 얼굴 후보 검출 방법과 얼굴의 구조적 특징을 이용한 얼굴 확인 방법으로 구성된 얼굴 검출 시스템을 제안한다. 먼저 제안하는 얼굴 후보 검출 방법은 피부색 영역과 피부색의 주변 영역에 대한 이미지 분할과 병합 알고리듬을 이용한다. 이미지 분할과 병합 알고리듬의 적용은 복잡한 이미지에 존재하는 다양한 얼굴들을 후보로 검출할 수 있다. 그리고 제안하는 얼굴 확인 방법은 얼굴을 지역적인 특징에 따라 분류 가능한 부분 얼굴 분류기를 사용하여 얼굴의 구조적 특징을 판단하고, 얼굴과 비-얼굴을 구별한다. 부분 얼굴 분류기는 학습 과정에서 얼굴 이미지만을 사용하고, 비-얼굴 이미지는 고려하지 않기 때문에 적은 수의 훈련 이미지를 사용한다. 실험 결과 제안한 얼굴 후보 검출 방법은 기존의 방법보다 평균 9.55% 많은 얼굴을 후보로 검출하였다. 그리고 얼굴/비-얼굴 분류 실험에서 비-얼굴에 대한 분류율이 99%일 때 기존의 분류기보다 평균 4.97% 높은 얼굴 분류율을 달성 하였다.

• 전기학회논문지
• /
• 제67권2호
• /
• pp.293-300
• /
• 2018

We propose a moving object extraction and tracking method for improvement of animal identification and tracking technology. First, we propose a method of merging separated moving objects into a moving object by using FCM (Fuzzy C-Means) clustering algorithm to solve the problem of moving object loss caused by moving object extraction process. In addition, we propose a method of extracting data from a moving object and a method of counting moving objects to determine the number of clusters in order to satisfy the conditions for performing FCM clustering algorithm. Then, we propose a method to continuously track merged moving objects. In the proposed method, color histograms are extracted from feature information of each moving object, and the histograms are continuously accumulated so as not to react sensitively to noise or changes, and the average is obtained and stored. Thereafter, when a plurality of moving objects are overlapped and separated, the stored color histogram is compared with each other to correctly recognize each moving object. Finally, we demonstrate the feasibility and applicability of the proposed algorithms through some experiments.

• 한국멀티미디어학회논문지
• /
• 제22권12호
• /
• pp.1404-1414
• /
• 2019

Educational books commonly use some copyrighted images with various kinds of deformation for helping students understanding. When using several copyrighted images made by merging or editing distortion in legal, we need to pay a charge to original copyright holders for each image. In this paper, we propose an efficient matching algorithm by separating each copyrighted image with the merged and edited type including rotation, illumination change, and change of size. We use the Oriented FAST and Rotated BRIEF (ORB) method as a basic feature matching scheme. To improve the matching accuracy, we design a new MIN-MAX similarity in matching stage. With the distorted dataset, the proposed method shows up-to 97% of precision in experiments. Also, we demonstrate that the proposed similarity measure also outperforms compared to other measure which is commonly used.

• 방송공학회논문지
• /
• 제14권4호
• /
• pp.397-408
• /
• 2009

In this paper, we present a non-static color histogram method to retrieve clothing images that are similar to a query clothing. Given clothing area, our method automatically extracts major colors by using the octree-based quantization approach[16]. Then, a color palette that is composed of the major colors is generated. The feature of each clothing, which can be either a query or a database clothing image, is represented as a color histogram based on its color palette. We define the match color bins between two possibly different color palettes, and unify the color palettes by merging or deleting some color bins if necessary. The similarity between two histograms is measured by using the weighted Euclidean distance between the match color bins, where the weight is derived from the frequency of each bin. We compare our method with previous histogram matching methods through experiments. Compared to HSV cumulative histogram-based approach, our method improves the retrieval precision by 13.7 % with less number of color bins.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제15권10호
• /
• pp.761-765
• /
• 2009

본 연구는 HAP(High Altitude Platform) 기반 네트워크 구성에서 최적의 이동 기지국의 위치와 적용범위를 찾는 것을 목적으로 한다. 이를 위하여 지상 노드들을 BIRCH(Balanced Iterative Reducing and Clustering Using Hierarchies) 알고리즘을 응용하여 클러스터링(Clustering) 하였다. BIRCH 알고리즘의 특징인 계층적 구조를 통해 CF(Clustering-Feature) 트리를 만들어 모바일 노드들을 관리하였고, 최대 반경과 노드 수 제약조건으로 분할과 합병 과정을 수행하도록 하였다. 제주도를 기반으로 한 모빌리티 모델을 만들어 시뮬레이션 작업을 수행하였으며, 제약 조건에 만족하는 이동 기지국의 최적위치와 적용범위를 확인했다.

• 한국인터넷방송통신학회논문지
• /
• 제18권1호
• /
• pp.223-229
• /
• 2018

본 논문에서는 이미지에서 검출된 각 연결성분들의 위상적 지속구간 정보를 그래프 기반 이미지 세그멘테이션에 결합하여 보다 안정적인 이미지 세그멘테이션 기법을 제안한다. 이미지의 밝기 또는 색상정보 등을 이용하여 모스 함수를 정의하고 이의 레벨세트로부터 각 연결성분의 위상적 지속구간을 구한다. 각 연결성분이 생성되고 긴 지속구간을 갖는 연결성분에 적절히 병합되는 과정을 영 차원 호몰로지 군의 관점에서 설명한다. 다양한 특성을 갖는 이미지들에 대하여 짧은 지속구간을 갖는 연결성분들을 지속구간이 긴 인근 성분에 적절히 병합시키는 과정을 통하여 보다 안정적인 이미지 세그멘테이션 결과들 얻을 수 있음을 실험으로 확인한다.

• 천문학회보
• /
• 제37권1호
• /
• pp.40.1-40.1
• /
• 2012

In the hierarchical universe, galaxy merger is predicted to be frequent, and thus it is an important element for understanding galaxy evolution. In particular, star formation is greatly enhanced during the merger. The aim of this study is to understand the position and rate change of star formation caused by equal-mass edge-on mergers. We use the GADGET2- N-body/SPH code, and fully consider gas cooling, star formation, and supernova feedback. We show the star formation rate (SFR), and the magnitude and color evolution of the merger remnants for 18 different configurations varying orbit elements and inclinations of host galaxies against orbit planes. Then we construct the mock images of the remnants and investigate on how equal-mass galaxy merger affects the SFR and color/magnitude evolution while considering dust reddening. We conclude that over 90% mass of SF in equal-mass merger is in the central region. SF in tidal feature involves a small fraction of new stars and thus is difficult to detect unless deep imaging is performed. Around 55 ${\pm}$ 5 percent of gas turns into stars until the final coalescence which typically corresponds to 0.8, 1.2, and 2.5 Gyr for direct, parabolic, and elliptical orbit, respectively. This result is roughly consistent with Cox et al. 2000. We plan to implement this result into semi-analytic model of galaxy formation. Caveats and future work on merging conditions are discussed.

• 정보처리학회논문지B
• /
• 제14B권2호
• /
• pp.141-144
• /
• 2007

문장이 길어질수록 구문분석의 정확률이 급격히 떨어지므로, 문장을 분할하여 각각의 분할단위로 구문분석을 수행한 후 각 구문분석결과를 합쳐 완성된 구문트리를 만드는 것이 일반적이다. 이 때 주로 절 단위로 문장이 분할되고, 각 절의 구문분석결과를 통합하게 되는데, 통합 과정에서 절-절 간의 의존관계 설정에 많은 오류가 생긴다. 이러한 절 간의 의존관계의 애매성을 해결하기 위하여, 본 논문은 기계학습을 이용하여 절-절 간의 의존관계를 분석해 본다. Support Vector Machines(SVM)을 사용하여 성능을 평가하고, 본 논문에서 실험한 방법과 기존의 방법들의 성능을 비교해 본 결과, 절-절 간의 의존관계 설정에 있어서 $8.88{\sim}15.35%$의 성능향상을 보였다.

• Applied Microscopy
• /
• 제46권2호
• /
• pp.93-99
• /
• 2016

The thickness of specimen is an important factor in microscopic researches. Thicker specimen contains more information, but it is difficult to obtain well focused image with precise details due to optical limit of conventional microscope. Recently, a microscope unit that combines improved illumination system, which allows real time three-dimensional (3D) image and automatic z-stack merging software. In this research, we evaluated the usefulness of this unit in observing thick samples; Golgi stained nervous tissue and ground prepared bone, tooth, and non-transparent small sample; zebra fish teeth. Well focused image in thick samples was obtained by processing z-stack images with Panfocal software. A clear feature of neuronal dendrite branching pattern could be taken. 3D features were clearly observed by oblique illumination. Furthermore, 3D array and shape of zebra fish teeth was clearly distinguished. A novel combination of two channel oblique illumination and z-stack imaging process increased depth of field and optimized contrast, which has a potential to be further applied in the field of neuroscience, hard tissue biology, and analysis of small organic structures such as ear ossicles and zebra fish teeth.

• 전자공학회논문지B
• /
• 제31B권8호
• /
• pp.137-147
• /
• 1994

In this paper,we propose a knowledge-based system for locating Destination Address Block(DAB) by analyzing the structure of Korean envelope images. In the proposed system the preprocessing steps such as adaptive binarization connected component extraction and deskewing are carried out first for the effective structure analysis of the envelope image. Then DAB containing address name and zipcode parts of the input envelope image is extracted by an iterative procedure based on the knowledge acquired from the statistical feature analysis of the various envelope images. Most of the system for slocating address blocks on envelopes have extracted DAB by segmenting an envelope image into several candidate blocks followed by selecting one among the candidate blocks. Because it is very difficult to segment a Korean envelope image into several blocks due to the specific writing habits that the addresses on the envelope are written in close proximity to each other the proposed iterative procedure determines DAB by splitting or merging the connected components and verifies the determined DAB without segmentation and selection. Experiments with a great number of the live envelopes provided from Seoul Mail Center in Koorea were carried out. The results reveal that the proposed system is very effective for address block location on Korean envelopes.