• 제목/요약/키워드: Frame Classification

검색결과 260건 처리시간 0.027초

학령훈기 남아의 상반신 체형유형분석 - 만 l1~12세 남아를 대상으로 - (Somatotype Classification in the Upper Half of Body of Elementary School Boys at the Ages 11 to 12)

  • 여혜린
    • 복식
    • /
    • 제53권3호
    • /
    • pp.63-72
    • /
    • 2003
  • The purposes of this study were to classify the upper half of body somatotype and analyze the characteristics of each somatotype. The subjects of survey were 272 elementary school boys of 11 to 12 years old living in Pusan and Kyungsangnam-do. Datas were collected through 36 anthropometric measurements and 7 photographic measurements. They were analyzed by factor analysis, cluster analysis and analysis of variance. The results of the study were as follows : 1. According to the factor analysis. seven factors were extracted from measurements of the upper half of body and those factors comprised 79.62% of total variance. Specially factor 1 was characterized sectional size and factor 2 was characterized longitudinal size comprised 58.83% of total variance. 2. According to the cluster analysis, the upper half of body somatotype was classified four types : Boys in type 1 had quite high stature and big frame, broadest and most sloping shoulders, flattest chest and belly, quite protruded shoulder blades boys in type 2 had quite short stature and small frame, quite broad and most rising shoulder, most protruded belly, quite protruded shoulder blades boys in type 3 had shortest stature, smallest frame, narrowest and quite rising shoulders, most protruded chest, flattest shoulder blade and quite flat belly : boys in type 4 had highest stature, biggest frame, most protruded shoulder blades and quite protruded chest and belly.

프레임 분류와 합성필터의 변형을 이용한 적은 지연을 갖는 음성 부호화기의 성능 (Improving LD-CELP using frame classification and modified synthesis filter)

  • 임은희;이주호;김형명
    • 한국통신학회논문지
    • /
    • 제21권6호
    • /
    • pp.1430-1437
    • /
    • 1996
  • 중간 주파수 대역(8kbps) 이하에서 적은 지연을 갖는 벡터여기 선형예측 음성 부호화기(LD-CELP)에 대하여 고려한다. 합성필터를 입력 프레임의 종류에 따라 변화시켜 음성 부호화기의 성능을 향상시키고자 한다. 먼저 프레임을 유성음과 무성음 그리고 개시 프레임으로 분류한다. 유성음과 무성음 프레임에서는 합성필터의 스펙트럼 포락을 음운의 특성에 적합하도록 변화시킨다. 개시 프레임에서는 합성필터의 성격을 바꾸어주기 위하여 바이어스 필터를 이용한다. 제안된 부호화기는 다른 적은 지연을 갖는 벡터여기 선형예측 음성 부호화기들에 비하여 비슷한 지연시간을 갖으면서 더 나은 음질을 제공하였다.

  • PDF

철골공사의 품질관리 업무 프로세스 모델 개발 (A Work Process Model for a Quality Management in Steel Frame Work)

  • 김종성;김형중;변은정;구교진;현창택
    • 한국건설관리학회논문집
    • /
    • 제8권3호
    • /
    • pp.150-158
    • /
    • 2007
  • 건축물이 고층화, 대형화되면서 철골구조를 기본골조로 채택하는 경우가 많아지고 있으며, 철골공사의 품질관리는 사전계획단계부터 이루어지며, 단계별 참여자간의 업무 구분 및 연계가 매우 중요하다 이에 본 연구는 철골공사의 품질관리를 보다 원활하게 수행하기 위하여, 기존 철골공사 품질관리업무를 기반으로 업무간 상호관계 및 입출력 정보를 고려한 품질관리 업무 프로세스 모델을 개발하고자 한다.

Anomaly detection of isolating switch based on single shot multibox detector and improved frame differencing

  • Duan, Yuanfeng;Zhu, Qi;Zhang, Hongmei;Wei, Wei;Yun, Chung Bang
    • Smart Structures and Systems
    • /
    • 제28권6호
    • /
    • pp.811-825
    • /
    • 2021
  • High-voltage isolating switches play a paramount role in ensuring the safety of power supply systems. However, their exposure to outdoor environmental conditions may cause serious physical defects, which may result in great risk to power supply systems and society. Image processing-based methods have been used for anomaly detection. However, their accuracy is affected by numerous uncertainties due to manually extracted features, which makes the anomaly detection of isolating switches still challenging. In this paper, a vision-based anomaly detection method for isolating switches, which uses the rotational angle of the switch system for more accurate and direct anomaly detection with the help of deep learning (DL) and image processing methods (Single Shot Multibox Detector (SSD), improved frame differencing method, and Hough transform), is proposed. The SSD is a deep learning method for object classification and localization. In addition, an improved frame differencing method is introduced for better feature extraction and a hough transform method is adopted for rotational angle calculation. A number of experiments are conducted for anomaly detection of single and multiple switches using video frames. The results of the experiments demonstrate that the SSD outperforms the You-Only-Look-Once network. The effectiveness and robustness of the proposed method have been proven under various conditions, such as different illumination and camera locations using 96 videos from the experiments.

동화사 수마제전의 건축적 특징 (Architectural Characteristic of SooMaJaiJeon in DongHwaSa)

  • 이경수
    • 한국산업융합학회 논문집
    • /
    • 제26권1호
    • /
    • pp.69-78
    • /
    • 2023
  • For this purpose, the research is to study the architectural characteristic of SooMaJaiJeon which is one of the DongHwaSa in the traditional wooden architecture by dividing it into three section-bracket, roof structure and frame structure. This study is largely divided into four stage-section do subject, research and actual measurement and conclusion. The whole process was consistently executed through detailed steps. The com position of this study is as follows. The 1st chapter-the purpose, background, method, object and range of the research. The 2nd chapter-the history of SooMajaiJeon. the 3rd chapter-the structure of Dapo-style bracket has generally considered, the frame structure of Dapo-style, vertical and horizontal member and podium, the characteristic of bracket with member and the structure, design of bracket, roof structure. In the 4th chapter, the conclusion of this study has been summarized, Dapo-style is the building that has deep symbolism and structural characteristic of traditional wooden architecture. The frame structure has a dominant regional characteristic and a typical part of typological classification in SooMaJaiJeon.

웨이브렛 변환 영역에서의 질감 유사성을 이용한 차량검지 및 차종분류 (Vehicle Detection Classification Using Textural Similarity in Wavelet Transformed Domain)

  • 임채환;박종선이창섭김남철
    • 대한전자공학회:학술대회논문집
    • /
    • 대한전자공학회 1998년도 추계종합학술대회 논문집
    • /
    • pp.959-962
    • /
    • 1998
  • In this paper, we propose an efficient vehicle detection and classification algorithm for an electronic toll collection, which is based on shadow robust vehicle presence test. In order to improve the performance of vehicle presence test, we use correlation coefficients between wavelet transformed input and reference images, which takes advanage of textural similarity. We compare the performance of the vehicle presence test with those of some conventional approaches that use variance of frame difference. Experimental results from field test show that the proposed vehicl detection and classification algorithm performs well even under abrupt intensity change due to the characteristics of sensor and occurrence of shadow.

  • PDF

IEEE 802.11ac 변조 방식의 딥러닝 기반 분류 (Deep learning-based classification for IEEE 802.11ac modulation scheme detection)

  • 강석원;김민재;최승원
    • 디지털산업정보학회논문지
    • /
    • 제16권2호
    • /
    • pp.45-52
    • /
    • 2020
  • This paper is focused on the modulation scheme detection of the IEEE 802.11 standard. In the IEEE 802.11ac standard, the information of the modulation scheme is indicated by the modulation coding scheme (MCS) included in the VHT-SIG-A of the preamble field. Transmitting end determines the MCS index suitable for the low signal to noise ratio (SNR) situation and transmits the data accordingly. Since data field decoding can take place only when the receiving end acquires the MCS index information of the frame. Therefore, accurate MCS detection must be guaranteed before data field decoding. However, since the MCS index information is the information obtained through preamble field decoding, the detection rate can be affected significantly in a low SNR situation. In this paper, we propose a relatively robust modulation classification method based on deep learning to solve the low detection rate problem with a conventional method caused by a low SNR.

MFCC 특징벡터와 신경회로망을 이용한 프레임 기반의 수중 천이신호 식별 (Frame Based Classification of Underwater Transient Signal Using MFCC Feature Vector and Neural Network)

  • 임태균;김일환;김태환;배건성
    • 대한전자공학회:학술대회논문집
    • /
    • 대한전자공학회 2008년도 하계종합학술대회
    • /
    • pp.883-884
    • /
    • 2008
  • This paper presents a method for classification of underwater transient signals using, which employs a binary image pattern of the mel-frequency cepstral coefficients(MFCC) as a feature vector and a neural network as a classifier. A feature vector is obtained by taking DCT and 1-bit quantization for the square matrix of the MFCC sequences. The classifier is a feed-forward neural network having one hidden layer and one output layer, and a back propagation algorithm is used to update the weighting vector of each layer. Experimental results with some underwater transient signals demonstrate that the proposed method is very promising for classification of underwater transient signals.

  • PDF

지능형 교통 시스템을 위한 형태학적 차량 분류 알고리즘 (Morphological Vehicle Classification Algorithm for Intelligent Transportation System)

  • 김기석
    • 한국멀티미디어학회논문지
    • /
    • 제5권1호
    • /
    • pp.10-17
    • /
    • 2002
  • 제한된 도로 여건 하에서 대중 교통 활성화를 위해 전용차로 운영을 위한 지능형 무인 관리 시스템의 필요성이 대두되고 있다. 본 논문에서는 수리 형태학적 영상 처리 및 인식 기법을 적용하여 차량 검지 자동화 시스템을 연구하였다. 배경과 분리된 차량 객체 영상을 추출하였으며, 형태학적 골격을 분석하여 골격 히스토그램으로부터 차종 분류를 위한 새로운 유일 가중 골격 특징을 추출하는 알고리즘을 제안하였다. 실험을 통해 제안한 차종 분류 알고리즘이 승용차, 트럭 등의 차종 인식에 효과적임을 볼 수 있었다.

  • PDF

딥러닝을 이용한 사용자 구분 및 위치추적 알고리즘 (User classification and location tracking algorithm using deep learning)

  • 박정탁;이솔;박병서;서영호
    • 한국정보통신학회:학술대회논문집
    • /
    • 한국정보통신학회 2022년도 춘계학술대회
    • /
    • pp.78-79
    • /
    • 2022
  • 본 논문에서는 RGB-D 카메라를 이용하여 획득한 다수 사용자의 정규화된 스켈레톤의 신체 비율 분석을 통해 각 사용자의 구분 및 위치를 추적하는 기법을 제안한다. 이를 위해 3D 포인트 클라우드로부터 각 사용자의 3D 스켈레톤을 추출한 뒤 신체 비율 정보를 저장한다. 이후 저장된 신체 비율 정보를 전체 프레임에서 출력된 신체 비율 데이터와 유사도를 비교하여 전체 영상에서의 사용자 구분 및 위치추적 알고리즘을 제안한다.

  • PDF