• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.08a
• /
• pp.293-296
• /
• 2000

본 논문에서는 얼굴의 구조적 특성과 누적 히스토그램을 이용하여 다양한 정보를 포함하고 있는 얼굴의 6가지 표정을 인식하는 알고리즘을 기술하였다. 표정 인식을 위해 특징점 추출 전처리 과정으로 입력 영상으로부터 에지 추출, 이진화, 잡음 제거, 모폴로지 기법을 이용한 팽창, 레이블링 순으로 적용한다. 본 논문은 레이블 영역의 크기를 이용해 1차 특징점 영역을 추출하고 가로방향의 누적 히스토그램 값과 대칭성의 구조적인 관계를 이용하여 2차 특징점 추출 과정을 거쳐 정확하게 눈과 입을 찾아낸다. 또한 표정 변화를 정량적으로 측정하기 위해 추출된 특징점들의 눈과 입의 크기, 미간 사이의 거리 그리고 눈에서 입까지의 거리 정보를 이용하여 표정을 인식한다. 1, 2차 특징점 추출 과정을 거치므로 추출률이 매우 높고 특징점들의 표정에 따른 변화 거리를 이용하므로 표정 인식률이 높다. 본 논문은 안경 착용 영상과 같이 복잡한 얼굴 영상에서도 표정 인식이 가능하다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.345-348
• /
• 2007

자동차 번호판 인식은 카메라의 발달과 무인자동차 주차 시스템, 불법 주정차 단속 등 응용 서비스의 증가로 부각되고 있는 텔레매틱스 분야의 핵심 기술이다. 특히 우리나라의 번호판은 영업용과 비영업용의 도색이 틀리고, 현재 4종류의 번호판 체계를 갖고 있다. 따라서 번호판 인식은 이들 번호판을 종류별로 분류하고 인식해야 되는 어려움이 있다. 본 논문은 레이블링 기법으로 번호판 종류를 분류하고, 지역 글자 인식에서 뭉개짐 현상이 발생하는 경우, 기존의 신경망에서 인식치 못하는 것을 차영상과 신경망을 이용하여 인식률 향상을 이루었다.

• Journal of KIISE:Databases
• /
• v.34 no.6
• /
• pp.518-527
• /
• 2007

It is an important issue to find out the difference between RDF documents, because RDF documents are changed frequently. When RDF documents contain blank nodes, we need a matching technique for blank nodes in the change detection. Blank nodes have a nested form and they are used in most RDF documents. A RDF document can be modeled as a graph and it will contain many subtrees. We can consider a change detection problem as a minimum cost tree matching problem. In this paper, we propose a change detection technique for RDF documents using the labeling scheme for blank nodes. We also propose a method for improving the efficiency of general triple matching, which used predicate grouping and partitioning. In experiments, we showed that our approach was more accurate and faster than the previous approaches.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.10a
• /
• pp.607-610
• /
• 2007

In this paper, proposes the monitoring system it will be able to trace the executed of each threads in OpenMP based a parallel program. The monitoring system of existing in uses each threads label information and the analysis technique which uses the access-history was most. This has the problem which raises the time and space complexity which is caused by with massive information creation. In this paper, only the thread which includes interest information it creates tracking information with the target. And it provides information which is intuitive to the user it provides the visualization system for to a same time. The visualization model is composed the images-information of a base. This does to be it will be able to understandable a program execute situation using an image processing technique. Therefore, this paper provides the parallel program an effective debugging environment.

• KIISE Transactions on Computing Practices
• /
• v.21 no.2
• /
• pp.160-165
• /
• 2015

Traditional speech emotion recognition techniques recognize emotions using a general training model based on the voices of various people. These techniques can not consider personalized speech character exactly. Therefore, the recognized results are very different to each person. This paper proposes an adaptive speech emotion recognition framework made from user's' immediate feedback data using a prompted labeling technique for building a personal adaptive recognition model and applying it to each user in a mobile device environment. The proposed framework can recognize emotions from the building of a personalized recognition model. The proposed framework was evaluated to be better than the traditional research techniques from three comparative experiment. The proposed framework can be applied to healthcare, emotion monitoring and personalized service.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.826-829
• /
• 2024

3D 사람 자세 추정 기술은 다양한 응용 분야에서의 높은 활용성으로 인해 대량의 학습 데이터가 수집되어 딥러닝 모델 연구가 진행되어 온 반면, 동물 자세 추정의 경우 3D 동물 데이터의 부족으로 인해 관련 연구는 극히 미진하다. 본 연구는 동물 자세 추정을 위한 예비연구로서, 3D 학습 데이터가 없는 상황에서 단일 이미지로부터 3D 사람 자세를 추정하는 딥러닝 기법을 제안한다. 이를 위하여 사전 훈련된 다중 시점 학습모델을 사용하여 2D 자세 데이터로부터 가상의 다중 시점 데이터를 생성하여 훈련하는 연역적 학습 기반 교사-학생 모델을 구성하였다. 또한, 키포인트 깊이 정보 대신 2D 이미지로부터 레이블링 된 순서 깊이 정보에 기반한 손실함수를 적용하였다. 제안된 모델이 동물데이터에서 적용 가능한지 평가하기 위해 실험은 사람 데이터를 사용하여 이루어졌다. 실험 결과는 제안된 방법이 기존 단안 이미지 기반 모델보다 3D 자세 추정의 성능을 개선함을 보여준다.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.5
• /
• pp.93-102
• /
• 2007

Metadata and ontology can be used to retrieve related information through the inference mure accurately and simply on the Semantic Web. RDF and RDF Schema are general languages for representing metadata and ontology. An enormous number of keywords on the Semantic Web are very important to make practical applications of the Semantic Web because most users prefer to search with keywords. In this paper, we consider a resource as a unit of query results. And we classily queries with keyword conditions into three patterns and propose indexing techniques for keyword-search considering both metadata and ontology. Our index maintains resources that contain keywords indirectly using conceptual relationships between resources as well as resources that contain keywords directly. So, if user wants to search resources that contain a certain keyword, all resources are retrieved using our keyword index. We propose a structure of table for storing RDF Schema information that is labeled using some simple methods.

• Journal of KIISE:Databases
• /
• v.32 no.5
• /
• pp.526-535
• /
• 2005

Ontology is a methodology on describing specific concepts and their relationships, and it is being considered important more and more as semantic web and variety of knowledge management systems are being highlighted. Ontology uses the relationships among concerts to represent some concrete semantics of specific concept. When we want to get some useful information from ontology, we severely have to process the transitive relationships because most of relationships among concepts represent transitivity. Technically, it causes recursive calls to process such transitive closure queries with heavy costs. This paper describes the efficient technique for processing transitive closure queries in ontology. To the purpose of it, we examine some approaches of current systems for transitive closure queries, and propose a technique by graph labeling scheme. Basically, we assume large size of ontology, and then we show that our approach gives relative efficiency in processing of transitive closure, queries.

• Journal of KIISE:Software and Applications
• /
• v.32 no.7
• /
• pp.625-635
• /
• 2005

In this paper, we propose a hybrid approach for segmenting the lungs efficiently and automatically in chest CT images. The proposed method consists of the following three steps. first, lungs and airways are extracted by two- and three-dimensional automatic seeded region growing and connected component labeling in low-resolution. Second, trachea and large airways are delineated from the lungs by two-dimensional morphological operations, and the left and right lungs are identified by connected component labeling in low-resolution. Third, smooth and accurate lung region borders are obtained by refinement based on image subtraction. In experiments, we evaluate our method in aspects of accuracy and efficiency using 10 chest CT images obtained from 5 patients. To evaluate the accuracy, we Present results comparing our automatic method to manually traced borders from radiologists. Experimental results show that proposed method which use connected component labeling in low-resolution reduce processing time by 31.4 seconds and maximum memory usage by 196.75 MB on average. Our method extracts lung surfaces efficiently and automatically without additional processing like hole-filling.

• Journal of the Korean Society for Marine Environment & Energy
• /
• v.9 no.3
• /
• pp.168-175
• /
• 2006

The cultural grounds of lave r have been surveyed using SPOT-5 satellite images. The facilities of laver cultivation area in the coastal waters of Korea were calculated. 10 m resolution multispectral images of SPOT-5 are adopted for the southern are a of Jebu Island, Hwaseong city to develop an automatic detection approach of laver nets that consists of the following: band difference technique, canny edge detector and morphological analysis: The number of satellite-based facilities was relatively high as compared with the licensed number in 2005, 676,749 chaek and 572,745 chaek(柵, unit of measure for laver farm), respectively. The ratio of a law abiding facility was very low at 52.9%. These data could be applied to control its national production keeping a stable market price for the government body.