• 제목/요약/키워드: HAR

검색결과 160건 처리시간 0.025초

Digital Image Quality Assessment Based on Standard Normal Deviation

  • Park, Hyung-Ju;Har, Dong-Hwan
    • International Journal of Contents
    • /
    • 제11권2호
    • /
    • pp.20-30
    • /
    • 2015
  • We propose a new method that specifies objective image quality factors by evaluating an image quality measurement model using random images. In other words, No-Reference variables are used to evaluate the quality of an original image without using any reference for comparison. 1000 portrait images were collected from a web gallery with votes constituting over 30 recommendation values. The bottom-up data collecting process was used to calculate the following image quality factors: total range, average, standard deviation, normalized distribution, z-score, preference percentage. A final grade is awarded out of 100 points, and this method ranks and grades the final estimated image quality preference in terms of total image quality factors. The results of the proposed image quality evaluation model consist of the specific dynamic range, skin tone R, G, B, L, A, B, and RSC contrast. We can present the total for the expected preference points as the average of the objective image qualities. Our proposed image quality evaluation model can measure the preferences for an actual image using a statistical analysis. The results indicate that this is a practical image quality measurement model that can extract a subject's preferred image quality.

서울 지하철 내 공기 중 먼지의 특성에 관한 연구 (A study on Characteristics of Airborne Dusts in Seoul Subway Stations)

  • 김진경;백남원
    • 한국환경보건학회지
    • /
    • 제30권2호
    • /
    • pp.154-160
    • /
    • 2004
  • The purpose of this study was to evaluate airborne concentrations and characteristics of TSP, IPM, TPM and RPM in Seoul subway stations. Sampling was performed at 14 stations from April 11 to 29, 2002. Size-selective dust concentrations and metal concentrations were measured by gravimetric method and ICP-AES, respectively. The geometric mean of TSP, IPM, TPM and RPM concentrations in Seoul subway stations were 176$\mu\textrm{g}$/㎥, 348$\mu\textrm{g}$/㎥, 158$\mu\textrm{g}$/㎥ and 104$\mu\textrm{g}$/㎥, respectively. Dust concentrations in pathway were the highest and those in lobby were the lowest. The size distribution of dusts was significantly different by location of collection. When the deposition rate into pulmonary gas exchange region was estimated by size distribution, the deposition rate of dust collected from platform was higher than those of dust collected from lobby and pathway. The lower the basement levels were, the higher the deposition rates of dusts into tracheobronchial region and gas exchange region were. Copper and iron concentrations measured in platform higher were than those in other areas.

무선 매체 접근 제어 프로토콜 상에서의 음성/데이타 통합 시스템을 위한 뉴로 퍼지 제어기 설계 (Design of a NeuroFuzzy Controller for the Integrated System of Voice and Data Over Wireless Medium Access Control Protocol)

  • 최원석;김응주;김범수;임묘택
    • 대한전기학회:학술대회논문집
    • /
    • 대한전기학회 2001년도 하계학술대회 논문집 D
    • /
    • pp.1990-1992
    • /
    • 2001
  • In this paper, a NeuroFuzzy controller (NFC) with enhanced packet reservation multiple access (PRMA) protocol for QoS-guaranteed multimedia communication systems is proposed. The enhanced PRMA protocol adopts mini-slot technique for reducing contention cost, and these minislot are futher partitioned into multiple MAC regions for access requests coming from users with their respective QoS (quality-of-service) requirements. And NFC is designed to properly determine the MAC regions and access probability for enhancing the PRMA efficiency under QoS constraint. It mainly contains voice traffic estimator including the slot information estimator with recurrent neural networks (RNNs) using real-time recurrent learning (RTRL), and fuzzy logic controller with Mandani- and Sugeno-type of fuzzy rules. Simulation results show that the enhanced PRMA protocol with NFC can guarantee QoS requirements for all traffic loads and further achieves higher system utilization and less non real-time packet delay, compared to previously studied PRMA, IPRMA, SIR, HAR, and F2RAC.

  • PDF

여자 중$\cdot$고등학교학생의 의복 선호도에 관한 연구 (A Study on Middle and High School Girls' Tendencies in Selecting Clothes)

  • 이선재
    • 한국의류학회지
    • /
    • 제6권1호
    • /
    • pp.39-49
    • /
    • 1982
  • This treatise deals with teen-agers' tendencies in selecting clothes, upon the recent liberalization measures of dress restriction. in regard to girl students' uniforms inmiddle and high schools, effective 1983. Results of my poll of 430 inquirees on the subject is outlined as follows; The surveyees have a preference for wearing box-style upper garments with a round-neck line and ribbon/tie collar. They also prefer to wear shirt cuffs mainly in set-in sleeve style. A straight skirt silhouetting the body line is anticipated to be longer in length than the knee line by 2 or 3 cm's. A dressy or sporty look would be most welcome, too. The ensemble of blue jeans and T-shirt blouse would be usual wear during school attendance. The reason why they like to choose trousers is considered to stem from the unisex look in fashion rather economical and practical purposes. Clothes material tends to be natural fabrics, single-colored and non-patterned. Color needs to be in a subdued mood, tuned with the same color and of neutral tint. In styling, emphasis is to be placed on femininity. In particular, the middle school girls wish to intensify 'Har-mony' in style while the high school misses are concerned with 'Individuality.'

  • PDF

영상 구성 파라미터 추출을 위한 융합 분석 알고리듬 연구 (Convergence Analysis Algorithm Study for Extracting Image Configuration Parameters)

  • 맹채정;하동환
    • 한국과학예술포럼
    • /
    • 제37권3호
    • /
    • pp.125-134
    • /
    • 2019
  • 본 연구는 영상콘텐츠 제작과정에서 배경음악 선정의 자동화를 위하여 영상의 특성을 분류, 분석할 수 있는 프로그램을 구성하였다. 연구 결과 및 내용은 다음과 같다. 영상의 특성은 '주제 범주', '감정', '픽셀 움직임 속도', '색상', '등장인물' 로 선정하며, '주제 범주'와 '감정'은 Microsoft사의 Azure Video Indexer를, '픽셀 움직임 속도'는 Optical flow, '색상'은 Image Histogram, '등장인물'은 CNN (Convolutional Neural Network)을 활용하여 데이터를 추출하였다. 이러한 본 연구의 결과는 최근 주목을 받고있는 '인터넷 1인 방송 크리에이터'들의 콘텐츠 제작과정에서 배경음악 매칭을 위한 영상 특성 분석이 이루어졌다는 점에서 의의가 있다.

A Study on the Research of Job Characteristics on Organizational Commitment, Resilience and Organizational Citizenship Behavior for Korean Government-funded Research Institutes in the Field of Science and Technology

  • KOH, Sung-Joo;YU, Jae Har;LEE, Chun-Su
    • 동아시아경상학회지
    • /
    • 제10권2호
    • /
    • pp.43-54
    • /
    • 2022
  • Purpose - Government-funded research institutes are important national institutions socially and nationally but academic studies on Korean government-funded research institutes are scarce, especially in areas of human resources and organization. This study investigated the effects of job characteristics on organizational commitment, resilience, and organizational citizenship behavior for organizational members of government-funded research institutes in the field of science and technology. Research design, data, and methodology - Literature review on the effects of job characteristics on organizational commitment, resilience and organizational citizenship behavior for organizational members of government-funded research institutes in the field of science and technology. Based on the review, exploratory propositions were proposed to conduct future empirical study. Result - In this study, based on the results of previous studies, it was presumed that job characteristics would affect organizational commitment, and organizational commitment would affect resilience and organizational citizenship behavior. In addition, proposition on the mediating role of resilience on both organizational commitment and organizational citizenship behavior was formulated. Conclusion - It was propositioned that job characteristics would affect organizational commitment, and organizational commitment would affect resilience and organizational citizenship behavior. Resilience, along with the direct effect of organizational commitment on organizational citizenship behavior, would play a role in mediating organizational commitment and organizational citizenship behavior.

교통정보제공을 위한 노변방송시스템 구축에 관한 연구 (A study of the highway advisory radio system implementation for traffic information service)

  • 정성학
    • 한국컴퓨터정보학회논문지
    • /
    • 제14권6호
    • /
    • pp.153-164
    • /
    • 2009
  • 본 연구의 목적은 일반국도를 대상으로 첨단교통정보서비스(ATIS: Advanced Traffic Information Services)를 제공하는 일환으로 노변방송시스템 구축방안을 제시하는데 있다. 노변방송은 소통 정보 제공을 통한 통행의 분산유도 뿐 만 아니라 전방의 홍수, 폭설, 낙석, 도로유실, 붕괴 등과 같은 자연재해 및 긴급상황 발생시 관련 정보를 시의 적절하게 제공함으로써 운전자의 안전한 도로운행을 지원하게 된다. 따라서 노변방송은 대부분의 국민들에게 편리하게 사용할 수 있는 대국민 교통정보서비스 구축으로 안전지향형 도로관리체계에 일조할 것이다.

Anomaly detection of smart metering system for power management with battery storage system/electric vehicle

  • Sangkeum Lee;Sarvar Hussain Nengroo;Hojun Jin;Yoonmee Doh;Chungho Lee;Taewook Heo;Dongsoo Har
    • ETRI Journal
    • /
    • 제45권4호
    • /
    • pp.650-665
    • /
    • 2023
  • A novel smart metering technique capable of anomaly detection was proposed for real-time home power management system. Smart meter data generated in real-time were obtained from 900 households of single apartments. To detect outliers and missing values in smart meter data, a deep learning model, the autoencoder, consisting of a graph convolutional network and bidirectional long short-term memory network, was applied to the smart metering technique. Power management based on the smart metering technique was executed by multi-objective optimization in the presence of a battery storage system and an electric vehicle. The results of the power management employing the proposed smart metering technique indicate a reduction in electricity cost and amount of power supplied by the grid compared to the results of power management without anomaly detection.

mmWave 레이더 기반 사람 행동 인식 딥러닝 모델의 경량화와 자원 효율성을 위한 하이퍼파라미터 최적화 기법 (Hyperparameter optimization for Lightweight and Resource-Efficient Deep Learning Model in Human Activity Recognition using Short-range mmWave Radar)

  • 강지헌
    • 대한임베디드공학회논문지
    • /
    • 제18권6호
    • /
    • pp.319-325
    • /
    • 2023
  • In this study, we proposed a method for hyperparameter optimization in the building and training of a deep learning model designed to process point cloud data collected by a millimeter-wave radar system. The primary aim of this study is to facilitate the deployment of a baseline model in resource-constrained IoT devices. We evaluated a RadHAR baseline deep learning model trained on a public dataset composed of point clouds representing five distinct human activities. Additionally, we introduced a coarse-to-fine hyperparameter optimization procedure, showing substantial potential to enhance model efficiency without compromising predictive performance. Experimental results show the feasibility of significantly reducing model size without adversely impacting performance. Specifically, the optimized model demonstrated a 3.3% improvement in classification accuracy despite a 16.8% reduction in number of parameters compared th the baseline model. In conclusion, this research offers valuable insights for the development of deep learning models for resource-constrained IoT devices, underscoring the potential of hyperparameter optimization and model size reduction strategies. This work contributes to enhancing the practicality and usability of deep learning models in real-world environments, where high levels of accuracy and efficiency in data processing and classification tasks are required.

디지털 사진영상의 크기조절과정에서 유실되는 정보를 이용한 비트심도의 확장 (Research for Bit-depth Conversion Development by Detection Lost Information to Resizing Process for Digital Photography)

  • 조두희;비벡마이크;백준기;하동환
    • 한국콘텐츠학회논문지
    • /
    • 제9권4호
    • /
    • pp.189-197
    • /
    • 2009
  • 디지털 사진영상은 8비트의 음영으로 이루어져 있다. 이것은 0부터 255까지, 총 256 단계의 레벨값으로 나눠지며 자연수에 해당하는 음영만을 표현하고 소수점 이하 실수 자리의 음영은 표현할 수 없다. 하지만 영상의 크기조절과정을 수행할 때 보간 수식에 의하여 소수점이하자리가 발생하게 되지만 최종적인 출력에서는 '버림연산'에 의하여 사라지게 된다. 본 연구에서는 이렇게 버려지는 소수점 이하 자리를 이용하여 사진영상의 비트 심도를 확장할 수 있는 방법을 제시한다. 이것은 크기조절 이후 콘트라스트 조절에 사용하여 사진영상의 해상도 저하를 최소화하는데 사용한다. SFR 측정을 통하여 포토샵 결과와 비교할 경우 본 연구에서 제시하는 방법이 원본의 특성을 더 잘 유지함을 알 수 있다.