• 제어로봇시스템학회논문지
• /
• 제17권2호
• /
• pp.93-101
• /
• 2011

For mobile robot, recognizing its current location is very important to navigate autonomously. Especially, loop closing detection that robot recognize location where it has visited before is a kernel problem to solve localization. A considerable amount of research has been conducted on loop closing detection and localization based on appearance because vision sensor has an advantage in terms of costs and various approaching methods to solve this problem. In case of scenes that consist of repeated structures like in corridors, perceptual aliasing in which, the two different locations are recognized as the same, occurs frequently. In this paper, we propose an improved method to recognize location in the scenes which have similar structures. We extracted salient regions from images using visual attention model and calculated weights using distinctive features in the salient region. It makes possible to emphasize unique features in the scene to classify similar-looking locations. In the results of corridor recognition experiments, proposed method showed improved recognition performance. It shows 78.2% in the accuracy of single floor corridor recognition and 71.5% for multi floor corridors recognition.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 추계학술대회
• /
• pp.225-228
• /
• 2022

한국건축정책학회의 건축 설계 대가 산정 방법은 설계 기간 중 투입된 시간을 고려한다. 그래서 설계 시간을 단축하면 설계 비용을 절감할 수 있다. 실제 건축학과 전공 학생들을 대상으로 설문 조사한 결과, 반복되는 도면 작성으로 인한 피로감과 도면 작성 시 고려할 사항이 많다는 점에서 어려움을 느끼고 있었다. 본 논문에서는 건축 설계 과정 중 도면화 단계에서 스케치를 규격화된 평면도로 자동 변환할 수 있는 프로그램과 사용자의 편리를 위한 웹앱을 구현한다. 제안하는 방법에서는 사용자가 스케치 이미지를 웹앱을 통해 등록하게 되면 프로그램은 도형 검출을 통해 도면 내 공간을 분리한다. 스케치에 표기된 치수와 공간 정보를 기반으로 외벽과 내벽을 표시하고 공간에 적절한 인테리어를 배치한다. 제안하는 방법을 통해 건축 설계 단계 중 도면화 과정의 시간을 줄여 설계 비용 절감을 기대할 수 있으며, 실제 건축 전공자 학생들 중 95.2%는 본 프로그램을 사용하고 싶다는 의향을 밝혔다.

• 대한기계학회논문집B
• /
• 제29권11호
• /
• pp.1220-1228
• /
• 2005

Fault detection and diagnostic (FDD) systems have the potential to reduce equipment downtime, service costs, and utility costs. In this study, model based algorithm and fuzzy algorithm were used to detect and diagnose various fault at System multi-air conditioner. various fault include the Refrigerant Low charging, Fouling of Indoor Heat Exchanger, Fouling of Outdoor Heat Exchanger A experimental verification was conducted in the 6HP System multi-air conditioner on an 8-floor building. Test results showed diagnosis result about 78 $\~$ 90$\%$ for given faults. This Study lays the foundation fur future work on develope the real-time fault detection and diagnosis system for the System multi-air conditioner.

• 대한임베디드공학회논문지
• /
• 제13권2호
• /
• pp.73-81
• /
• 2018

This paper describes a device platform which the blind can wear to keep path and to get surrounding information during their independent walking. Compared to the existing technologies, the proposed device could be used indoors and outdoors, and maps need not be provided in advance. It is composed of a glasses type device equipped with image sensors, and a portable device that analyzes sensor data for sighted guiding. RGB images and depth images are extracted to generate a walking map based on feature points. It also can cope with the risk of collision with bollard, color cone by applying vertical obstacle detection technology based on floor detection.

• 한국컴퓨터정보학회논문지
• /
• 제15권2호
• /
• pp.41-46
• /
• 2010

본 논문에서는 다양한 환경변수가 존재하는 실내에서 효율적인 객체를 검출하기 위한 방법을 제안한다. 일반적으로 실내 환경은 조명의 변화와 객체에 의해 발생된 그림자, 바닥면에 반사된 조명성분 등으로 인하여 정확한 객체 검출이 이루어지기 어려운 환경이다. 먼저 객체검출을 위한 배경영상을 생성한다. 영상 내에 객체가 존재하는 경우 이전에 생성된 배경영상과 현재 입력영상간의 유사도 비교를 통해 보정된 배경영상을 생성한다. 배경영상과 입력영상으로 생성한 평균영상과 보정된 배경영상을 이용하여 혼합영상을 생성한다. 마지막으로 혼합영상을 이용하여 입력된 영상으로부터 객체를 검출한다. 검출된 객체를 보완하기 위해 레이블링 과정을 통해 잡음 성분을 제거한 후 모폴로지 기법을 적용하여 객체영역 보완한다. 따라서 조명의 변화나 그림자와 같은 환경변수로부터 강인한 객체를 검출한다. 본 논문에서 제안한 시스템은 변형된 조명성분과 그림자 성분이 포함되어 있는 혼합영상을 사용하기 때문에 기존시스템보다 객체영역 검출이 더욱 효과적이다.

• Journal of information and communication convergence engineering
• /
• 제6권2호
• /
• pp.150-153
• /
• 2008

In the blind Maximum-likelihood (ML) detection for Orthogonal Space-Time Block Codes (OSTBC), the problem of ambiguity in determining the symbols has been a great concern. A possible solution to this problem is to apply semi-blind ML detection, i.e. the blind ML decoding with pilot symbols or training sequence. In order to increase the performance, the number of pilot symbols or length of training sequence should be increased. Unfortunately, this leads to a significantly decrease in system spectral efficiency. This work presents an approach to resolve the aforementioned issue by introducing a new method for constructing transmitted information symbols, in which transmitting information symbols drawn form different modulation constellations. Therefore, the ambiguity can be easily eliminated. In addition, computer simulation is implemented to verify the performance of the proposed approach.

• 치과방사선
• /
• 제24권1호
• /
• pp.115-124
• /
• 1994

The purpose of this study was to examine the differences in representation of a globular radiopaque mass on the pantomograms and Waters' views and to compare the efficacy of periapical radiograms, pantomograms and Waters' views in detection of defects on the internal walls of the maxillary sinus. This study was performed with dried human skull. For the study of difference of radiopaque mass shadow in the two views, rubber ball with a diameter of 10㎜ was used as the experimental lesion. It was placed successively on the internal wall of the anterior, posterior, medial, lateral walls and floor of the maxillary sinus. To examine the detectability of defects for radiographic techniques, defects were formed in the anterior, posterior, medial, lateral walls, and floor of the maxillary sinus. They were formed with 0.5㎜, 0.75㎜, 1.0㎜, 2.0㎜ and 3.0㎜ sized steel round burs with a slow speed dental handpiece. By subsequently plugging the holes with zinc oxide eugenol paste, radiopaque defects were produced. After that the periapical radiograms, the pantomograms and the Waters' views were taken each and every defect. The obtained results were as follows: 1. Rubber balls placed on each internal wall of the maxillary sinus were correctly depicted on the posterior wall and the floor in case of the pantomogram, and on the anterior wall and the medial wall in case of the Waters' view. 2. On the detectability of defects for each radiographic technique, radiolucent defects were detected in different places for each technique. Periapical radiogram could detect 1.0㎜ defect on the floor of the maxillary sinus, pantomogram could detect 2.0㎜ defect on every internal wall of the maxillary sinus, and Waters' view could detect 3.0㎜ defect on the anterior wall of the maxillary sinus. 3. On the detectability of defects for each radiographic technique, radiopaque defects were detected in different places for each technique. Periapical radiogram could clearly detect 0.5㎜ defect on the floor of the maxillary sinus, pantomogram could detect 0.5㎜ defect on every internal wall of the maxillary sinus, and finally Waters' view could detect 0.5㎜defect on the anterior wall of the maxillary sinus but 0.75㎜ defect on the anterior wall, lateral wall and floor of the maxillary sinus. As the result, the periapical radiogram is the most simple and satisfactory method for investigating in the maxillary sinus. The pantomogram is suitable method for screening of changes in the maxillary sinus. And the Waters' view is available for detect of lesion in the anterior wall of the maxillary sinus. For the purpose of accurate diagnosis and evaluation of lesion in the maxillary sinus, these techniques supplement each other.

• 한국정보과학회논문지:시스템및이론
• /
• 제35권1호
• /
• pp.44-52
• /
• 2008

본 논문에서는 모션켑쳐데이타를 이용한 두 캐릭터간의 빠른 충돌감지에 대한 연구를 논의한다. 본 연구의 목적이 군중 시뮬레이션이기 때문에, 제안한 알고리즘은 캐릭터를 실린더 형태로 모델링 한 후에 Rough한 충돌감지를 목표로 한다. 이를 위해 계층적인 바운딩 박스 데이타 구조인 MOBB를 제안한다. MOBB는 모션클립에 대한 시공간 바운딩 박스이며, 제안된 알고리즘에 대한 테스트 결과 2배 이상의 속도 향상이 있음을 밝힌다.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 1999년도 추계학술대회논문집
• /
• pp.321-332
• /
• 1999

It is important to detect objects movement for obstacle recognition and path searching of UCT(unmanned container transporters) with vision sensor. This paper shows the method to draw out objects and to trace the trajectory of the moving object using a CCD camera and it describes the method to recognize the shape of objects by neural network. We can transform pixel points to objects position of the real space using the proposed viewport. This proposed technique is used by the single vision system based on floor map.

• Journal of Radiation Protection and Research
• /
• 제16권1호
• /
• pp.15-24
• /
• 1991

CR-39 플라스틱 핵비적검출기를 라돈검출기로 내장한 멤브레인 필터컵(일명 : CR-39 라돈컵)을 이용하여 일부 국산건축자재의 라돈방출율을 측정하였다. 표준라듐선원을 이용한 라돈컵의 교정 실험을 수행하여 얻은 CR-39 검출기의 라돈검출인자는 $0.164{\pm}0.005(tracks \;cm^{-2}/Bq\;d\;m^{-3})$였으며 타 연구자들의 발표결과와 잘 일치하였다. 일부 건축자재(모래벽돌, 적벽돌, 화장석판, 콘크리트 덩어리, 건물바닥과 내벽)에 CR-39 라돈컵을 2개월 동안 기밀 부착하여 라돈방출율을 측정한 결과, 라돈방출율은 모래벽돌에서 평균$(75.0{\pm}5.5){\times}10^{-6}(Bq/m^2-s)$, 화강석판에서 $(6.8{\pm}2.9){\times}10^{-6}Bq/m^2-s)$로 10배정도의 차이를 보였으며, 모래벽돌>콘크리트바닥>콘크리트덩어리>건물벽>적벽돌>화강석 판의 순서를 보였다. 본 연구 결과, CR-39 라돈컵에 의해서도 건축자재의 라돈방출율을 효과적으로 측정할 수 있음이 입증하였다.