• 제목/요약/키워드: Feature point extraction

검색결과 265건 처리시간 0.03초

특징 추출을 이용한 다중 영상 정합 및 융합 연구 (Multimodality Image Registration and Fusion using Feature Extraction)

  • 우상근;김지현
    • 한국컴퓨터정보학회논문지
    • /
    • 제12권2호
    • /
    • pp.123-130
    • /
    • 2007
  • 본 논문에서는 소동물 생체내 실험시 서로 다른 장비에서 획득된 영상의 융합 및 정합을 위한 방법을 제안한다. 마우스의 꼬리 정맥에 $[[^{18}F]FDG$를 주사하여 60분 섭취후 서로 다른 장비에서 동일한 위치의 영상을 획득하기 위하여 아크릴 재질의 소동물 가이드에 기준마크를 설정하고 microPET과 CT 영상을 획득하였다. MicroPET으로 획득된 리스트모드(list-mode) 데이터는 Fourier Rebinning(FRB) 방법을 사용하여 사이노그램(Sinogram)으로 변환 후 4 번의 반복횟수를 가지는 Ordered Subset Expectation Maximization(OSEM) 알고리즘으로 재구성하였다. MicroPET 영상획득후 PET/CT의 CT를 이용하여 CT영상을 획득하였다. MicroPET 영상에서 폐영역을 정확히 찾아내는 어려움이 있어. 해부학적 정보를 제공하는 CT 영상을 이용하여 폐 영역을 구분하였다. 영상 융합을 위한 불일치 부분을 해결하기 위하여 기준마크의 정보와 폐 영역의 정보를 이용하여 회전과 이동정보를 가지는 어파인 (affine) 변환 행렬 구하여 영상 정합에 사용하였다. 이 방법은 정량적 정확성과 영상 해석의 정확성을 개선할 것으로 기대된다.

  • PDF

주성분 분석법을 이용한 회귀다항식 기반 모델 및 패턴 분류기 설계 (Design of Regression Model and Pattern Classifier by Using Principal Component Analysis)

  • 노석범;이동윤
    • 한국정보전자통신기술학회논문지
    • /
    • 제10권6호
    • /
    • pp.594-600
    • /
    • 2017
  • 본 논문에서는 매우 높은 차원을 가진 데이터에서 의미 있는 특징 벡터 추출하여 입력 공간의 차원을 줄이기 위하여 주성분 분석법을 사용하였다. 주성분 분석법을 이용하여 축소된 차원을 가진 입력 데이터를 이용하여 회귀 다항식의 입력벡터로 사용하는 모델과 패턴 분류기의 설계 방법을 제안하였다. 제안된 모델 및 패턴 분류기는 매우 단순한 구조를 가진 회귀다항식을 기반으로 설계하여 모델 및 패턴 분류기의 과적합 문제를 해결 하고자 하였다. 제안된 설계방법을 적용하여 설계된 모델과 패턴 분류기의 성능을 비교 및 평가하기 위하여, 다양한 기계 학습 데이터 집합을 사용하였다.

구조물의 안전성 모니터링을 위한 통계/확률기반 적응형 임계치 설정 알고리즘 개발 (Development of Statistical/Probabilistic-Based Adaptive Thresholding Algorithm for Monitoring the Safety of the Structure)

  • 김태헌;박기태
    • 한국구조물진단유지관리공학회 논문집
    • /
    • 제20권4호
    • /
    • pp.1-8
    • /
    • 2016
  • 최근의 건축물은 복합적인 기능과 형태를 보이고 있으며, 크기가 거대해짐에 따라 구조물 건전성 감시(Structural Health Monitoring)기술의 수요 또한 증가하고 있다. 구조물마다 고유한 동특성을 가지고 있으며, 다양한 외력의 영향을 받기 때문에 구조물의 건전성을 평가하는 다양한 방법들이 연구되고 있다. 전문가에 의지하여 접근 가능한 지점에 대한 육안 검사 및 비파괴 검사를 벗어나 사각지대가 없는 온라인 계측 시스템의 구비와 함께 자동으로 위험요소를 검출하는 시스템이 요구되고 있다. 본 연구에서는 비선형적인 구조물의 응답을 고려하기 위해 관리도 기법, 평균제곱근편차, 일반 극치 분포 등과 같은 통계적 기법을 이용하여 이상거동을 판별에 활용할 수 있는 신호 특징 추출과 적응형 임계치 설정 알고리즘을 제안하였으며, 강제진동 실험과 실제 운용중에 있는 구조물의 지진 계측 시스템의 가속도 응답을 이용하여 성능을 검증하였다.

웨이블릿 변환과 인공신경망을 이용한 결함분류 프로그램 개발과 용접부 결함 AE 신호에의 적용 연구 (Development of Defect Classification Program by Wavelet Transform and Neural Network and Its Application to AE Signal Deu to Welding Defect)

  • 김성훈;이강용
    • 비파괴검사학회지
    • /
    • 제21권1호
    • /
    • pp.54-61
    • /
    • 2001
  • 웨이블릿 변환과 인공신경망을 이용하여 AE 신호를 분류하는 소프트웨어 패키지를 개발하였다. 웨이블릿 변환으로는 연속 웨이블릿 변환과 이산 웨이블릿 변환을 모두 고려하였으며, 인공신경망의 모델로는 오류 역전파 인공신경망을 사용하였다. 분류에 사용된 AE 신호는 용접부에 인공결함을 가진 시편의 3점 굽힘시험에서 발생한 신호이다. 개발된 소프트웨어 패키지를 이용하여 이 신호를 웨이블릿 변환시켜 생성된 시간-주파수 평면상에서 특징값을 추출하고 이를 인공신경망에 학습하여 인공신경망 분류기를 설계하고 검증하였다. 본 연구에서 개발된 소프트웨어 패키지를 이용한 AE 신호 분류법이 유용함을 보이고, 또한 연속 웨이블릿 변환과 이산 웨이블릿 변환에 의한 분류 결과를 비교하였다.

  • PDF

Accuracy Estimation of Electro-optical Camera (EOC) on KOMPSAT-1

  • Park, Woon-Yong;Hong, Sun-Houn;Song, Youn-Kyung
    • Korean Journal of Geomatics
    • /
    • 제2권1호
    • /
    • pp.47-55
    • /
    • 2002
  • Remote sensing is the science and art of obtaining information about an object, area or phenomenon through the analysis of data acquired by a device that is not in contact with the object, area, or phenomenon under investigation./sup 1)/ EOC (Electro -Optical Camera) sensor loaded on the KOMPSAT-1 (Korea Multi- Purpose Satellite-1) performs the earth remote sensing operation. EOC can get high-resolution images of ground distance 6.6m during photographing; it is possible to get a tilt image by tilting satellite body up to 45 degrees at maximum. Accordingly, the device developed in this study enables to obtain images by photographing one pair of tilt image for the same point from two different planes. KOMPSAT-1 aims to obtain a Korean map with a scale of 1:25,000 with high resolution. The KOMPSAT-1 developed automated feature extraction system based on stereo satellite image. It overcomes the limitations of sensor and difficulties associated with preprocessing quite effectively. In case of using 6, 7 and 9 ground control points, which are evenly spread in image, with 95% of reliability for horizontal and vertical position, 3-dimensional positioning was available with accuracy of 6.0752m and 9.8274m. Therefore, less than l0m of design accuracy in KOMPSAT-1 was achieved. Also the ground position error of ortho-image, with reliability of 95%, is 17.568m. And elevation error showing 36.82m was enhanced. The reason why elevation accuracy was not good compared with the positioning accuracy used stereo image was analyzed as a problem of image matching system. Ortho-image system is advantageous if accurate altitude and production of digital elevation model are desired. The Korean map drawn on a scale of 1: 25,000 by using the new technique of KOMPSAT-1 EOC image adopted in the present study produces accurate result compared to existing mapping techniques involving high costs with less efficiency.

  • PDF

딥러닝을 이용한 핸드크림의 마찰 시계열 데이터 분류 (Deep Learning-based Approach for Classification of Tribological Time Series Data for Hand Creams)

  • 김지원;이유민;한상헌;김경택
    • 산업경영시스템학회지
    • /
    • 제44권3호
    • /
    • pp.98-105
    • /
    • 2021
  • The sensory stimulation of a cosmetic product has been deemed to be an ancillary aspect until a decade ago. That point of view has drastically changed on different levels in just a decade. Nowadays cosmetic formulators should unavoidably meet the needs of consumers who want sensory satisfaction, although they do not have much time for new product development. The selection of new products from candidate products largely depend on the panel of human sensory experts. As new product development cycle time decreases, the formulators wanted to find systematic tools that are required to filter candidate products into a short list. Traditional statistical analysis on most physical property tests for the products including tribology tests and rheology tests, do not give any sound foundation for filtering candidate products. In this paper, we suggest a deep learning-based analysis method to identify hand cream products by raw electric signals from tribological sliding test. We compare the result of the deep learning-based method using raw data as input with the results of several machine learning-based analysis methods using manually extracted features as input. Among them, ResNet that is a deep learning model proved to be the best method to identify hand cream used in the test. According to our search in the scientific reported papers, this is the first attempt for predicting test cosmetic product with only raw time-series friction data without any manual feature extraction. Automatic product identification capability without manually extracted features can be used to narrow down the list of the newly developed candidate products.

학습을 이용한 손 자세의 강인한 추정 (Robust Estimation of Hand Poses Based on Learning)

  • 김설호;장석우;김계영
    • 한국정보통신학회논문지
    • /
    • 제23권12호
    • /
    • pp.1528-1534
    • /
    • 2019
  • 최근 들어, 3차원의 깊이 카메라의 대중화로 인해서 RGB 영상에서 수행되던 연구에 새로운 관심과 기회가 생겼지만 사람의 손 자세의 추정은 여전히 어려운 주제 중의 하나로 분류되고 있다. 본 논문에서는 다양하게 입력되는 3차원의 깊이 영상으로부터 사람의 손의 자세를 학습 알고리즘을 이용하여 강인하게 추정하는 방법을 제안한다. 제안된 접근 방법에서는 먼저 뼈대 기반의 손 모델을 생성한 다음, 생성된 손 모델을 3차원의 포인트 클라우드 데이터에 정렬한다. 그런 다음, 랜덤 포레스트 기반의 학습 알고리즘을 이용하여 정렬된 손 모델로부터 손의 자세를 강인하게 추정한다. 본 논문의 실험 결과에서는 제안된 접근 방법이 다양한 실내외의 환경에서 촬영된 입력 영상으로부터 사람의 손의 자세를 강인하고 빠르게 추정한다는 것을 보여준다.

차량검출 GMM 2.0을 적용한 도로 위의 차량 검출 시스템 구축 (On-Road Car Detection System Using VD-GMM 2.0)

  • 이옥민;원인수;이상민;권장우
    • 한국통신학회논문지
    • /
    • 제40권11호
    • /
    • pp.2291-2297
    • /
    • 2015
  • 본 연구에서는 레이더 검지 시스템과 통합하여 적용하기 위해 도로 위를 이동하는 자동차의 영상을 입력 받아 자동차를 검출하는 방법을 제안한다. 입력 영상의 제약조건이 있다. 도로 위에서 아래 방향을 비스듬히 내려 보는 고정된 시야를 가져야한다는 점이다. 주어진 영상 중 도로 영역만을 이용하기 위해 도로 영역을 관심영역으로 검출해 적용한다. 서론에서는 도로 영역 내에서 차량 검출을 위해 사용한 모션 히스토리 이미지 추출 방법, SIFT(Scale-Invariant Feature Transform) 알고리즘, 히스토그램 분석 등을 적용한 실험결과와 이에 대한 한계점을 제시했다. 이를 해결하기 위해서 가우시안 혼합 모델(GMM, Gaussian Mixture Model)의 응용을 제안한다. 가우시안 혼합 모델 알고리즘을 응용한 차량 검출 GMM(VDGMM, Vehicle Detection GMM)과 이를 차량 검출에 더 최적화한 차량 검출 GMM 2.0을 설명하고, 차량 검출 GMM 2.0을 적용한 실험결과 및 결론을 제시한다. 도로 영역 검출 없이 GMM을 적용한 결과는 정확율, 재현율, F1이 각각 9%, 53%, 15%이었고, 도로 영역 검출 후 차량 검출 GMM 2.0을 적용한 결과는 각각 85%, 77%, 80%로 많은 차이를 보였다.

기계학습에 기반한 생의학분야 전문용어의 자동인식 (Machine-Learning Based Biomedical Term Recognition)

  • 오종훈;최기선
    • 한국정보과학회논문지:소프트웨어및응용
    • /
    • 제33권8호
    • /
    • pp.718-729
    • /
    • 2006
  • 일정 분야의 문서들에서 그 분야 특정을 반영하는 전문용어를 자동으로 인식하는 연구에 대한 관심이 증가하고 있다. '전문용어 인식'은 문서에서 전문용어가 될 수 있는 언어적 단위를 파악하는 '용어 추출' 과정과 '용어추출' 과정에서 얻어진 용어목록 중 해당분야의 전문용어를 고르는 '전문용어 선택' 과정으로 구성된다. '전문용어 선택' 과정은 용어목록을 전문용어의 특정에 따라 순위화한 후 타당한 전문용어를 파악하는 작업으로 정의된다. 따라서 전문용어 선택 문제는 용어목록의 순위화 작업과 순위화된 목록에서 전문용어와 비전문용어 간의 경계를 인식하는 작업으로 정의된다. 기존의 전문용어 선택 기법은 주로 용어의 빈도수 등과 같은 통계적 특정만을 이용하였다. 하지만 통계적 특정만으로는 효과적으로 전문용어를 선택하기 어렵다. 본 논문의 논제는 전문용어 선택에서 다양한 전문용어의 특정을 고려하고 이들 중 전문용어 선택에서 효과적인 특정을 찾으려는 것이다. 순위화 문제는 다양한 전문용어 특정을 도출하고 이들을 기계학습방법으로 통합하여 해결한다. 경계인식 문제는 전문용어와 비전문용어의 이진 분류 문제로 정의하고 기계학습방법으로 해결한다. 본 논문의 기법은 경계인식측면에서 78-86%의 정확률과 87% -90%의 재현율을 나타내었으며, 순위화 측면에서 89%-92%의 11포인트 평균정확률을 나타내었다. 또한 기존 연구보다 최고 26% 의 성능향상을 보였다.

시각정보획득과정에 나타난 주사판정과 성별 주시특성 - 지하철 홀 공간을 대상으로 - (Scanning Determination & Observation Features by Sex shown in the Process of Acquiring Visual Information - With the Object of Subway Station Hall Space -)

  • 김종하;최계영
    • 한국실내디자인학회논문집
    • /
    • 제23권6호
    • /
    • pp.115-124
    • /
    • 2014
  • This study has carried out scanning tests in order to figure out the features of scanning search by sex of space users, with the result of which the validity of data has been estimated. In this research, the scanning patterns were set up for verifying the typology of scanning paths and then the reason for determining scanning paths and the validity of estimation method were reviewed. Since the observation features depends on sex, the analysis of visual activities for acquiring any information in a space will reveal the intention and purpose of space users. The findings by analyzing the features of scanning pattern by sex which were found at the determination of scanning patterns can be defined as the followings. First, for estimating the process of space-information search, the movement distance at each point of continuative-observation data from the angle of eye-movement has been extracted, on the ground of which the fixation and movement of eye have been defined for the establishment of scanning-cut characteristics. Second, the scanning times were estimated for the extraction of effective observation data that would be used for comparative analysis, which showed that men had more data (3,398.2/64.4%) than women (2,998.2/55.6%). This enables the acknowledgment that the scanning cut of men was relatively less, which indicates that men will acquire more information on space than women in the process of observing any space. Third, men's scanning times (58.0 times/2.02 seconds) were less than those of women (71.9 times/1.39 seconds) while the scanning time of the former was longer than that of the latter, which shows the feature that it takes longer for men than women in scanning while the scanning times of the former is less than those of the latter. Fourth, the observation features can be determined that the combination of this result with the predominance character by sex for a general viewpoint to be employed indicates that while men employ mixed-scanning for observation activities to acquire space-information spending for longer time, women, by concentrated-scanning, focus on a single point for shorter time or stay at one location for a considerably long time for space-information acquirement.