A Robust Object Detection and Tracking Method using RGB-D Model

박서희,전준철,Park, Seohee,Chun, Junchul Korean Society for Internet Information 2017 인터넷정보학회논문지 Vol.18 No.4

Recently, CCTV has been combined with areas such as big data, artificial intelligence, and image analysis to detect various abnormal behaviors and to detect and analyze the overall situation of objects such as people. Image analysis research for this intelligent video surveillance function is progressing actively. However, CCTV images using 2D information generally have limitations such as object misrecognition due to lack of topological information. This problem can be solved by adding the depth information of the object created by using two cameras to the image. In this paper, we perform background modeling using Mixture of Gaussian technique and detect whether there are moving objects by segmenting the foreground from the modeled background. In order to perform the depth information-based segmentation using the RGB information-based segmentation results, stereo-based depth maps are generated using two cameras. Next, the RGB-based segmented region is set as a domain for extracting depth information, and depth-based segmentation is performed within the domain. In order to detect the center point of a robustly segmented object and to track the direction, the movement of the object is tracked by applying the CAMShift technique, which is the most basic object tracking method. From the experiments, we prove the efficiency of the proposed object detection and tracking method using the RGB-D model. 최근 지능형 CCTV는 빅 데이터, 인공지능 및 영상 분석과 같은 분야와 결합하여 다양한 이상 행위들을 탐지하고 보행자와 같은 객체의 전반적인 상황을 분석할 수 있으며, 이러한 지능형 영상 감시 기능에 대한 영상 분석 연구가 활발히 진행되고 있는 추세이다. 그러나 일반적으로 2차원 정보를 이용하는 CCTV 영상은 위상학적 정보 부족으로 인해 객체 오 인식과 같은 한계가 존재한다. 이러한 문제는 두 대의 카메라를 사용하여 생성된 객체의 깊이 정보를 영상에 추가함으로써 해결 할 수 있다. 본 논문에서는 가우시안 혼합기법을 사용하여 배경 모델링을 수행하고, 모델링 된 배경에서 전경을 분할하여 움직이는 객체의 존재 여부를 탐지한다. RGB 정보 기반 분할 결과를 이용하여 깊이 정보 기반 분할을 수행하기 위해 두 대의 카메라를 사용하여 스테레오 기반 깊이 지도를 생성한다. RGB 기반으로 분할된 영역을 깊이 정보를 추출하기 위한 도메인으로 설정하고, 도메인 내부에서 깊이 기반 분할을 수행한다. 강건하게 분할된 객체의 중심점을 탐지하고 방향을 추적하기 위해 가장 기본적인 객체 추적 방법인 CAMShift 기법을 적용하여 객체의 움직임을 추적한다. 실험을 통하여 제안된 RGB-D 모델을 이용한 객체 탐지 및 추적 방법의 우수성을 입증하였다.

CCTV 영상의 동적 객체 탐지 및 추적 기술 동향

박서희,전준철,Park, Seohee,Chun, Junchul 한국인터넷정보학회 2017 인터넷정보학회지 Vol.18 No.1

요통치료에서 보법의 영향

박서희(Seo Hee Park),장상철(Sang Chul Jang),나삼식(Sam Sik Na),안훈모(Hun Mo Ahn) 대한의료기공학회 2014 醫療氣功 Vol.14 No.1

This study is aimed to verify effects of Walking in treating hospitalized patients with low back pain in the Korean medicine hospital. The method of this study is choosing patients whose chief complaints were low back pain in the Korean medicine hospital in Gyeonggi Province. People who agreed to take Walking treatment were classified into experimental group and control group. Both experimental group and control group were treated with Korean medicine. In addition, experimental group performed Walking program more often. Changes of temperature difference taken by digital infrared thermal imaging(DITI) were used as measuring tools. The summary of this study's result is as in the following. There was significant difference in the experimental group which performed Walking treatments but not in the control group resulting from examining the change of temperature difference between left and right Yongchon. As known from the results above, Walking program is meaningful in curing the low back pain .

텍스트 스토리에서 이벤트 추출 및 표현 방법 연구

박서희(Seo-Hui Park),유혜연(Hye-Yeon Yu),정윤경(Yun-Gyung Cheong),배병철(Byung-Chull Bae) 한국정보과학회 2019 한국정보과학회 학술발표논문집 Vol.2019 No.6

상담사의 수퍼비전 경험에서 드러난 수퍼비전의 지속 요인

박서희(Seohee Park) 한국임상모래놀이치료학회 2019 상징과 모래놀이치료 Vol.10 No.1

In this qualitative research study, we explored the sustainability factors of counselors who received continuous counseling after qualification. We conducted individual, face-to face interviews with four counselors for two weeks between May and June 2018. The second and third data collection phases involved follow-up by phone or email where necessary. We applied topic analysis to our data and in the process arrived at four perspectives on the sustainability factors of supervision. The four themes were ‘Being helpful,’ ‘Being a good learning experience,’ ‘Recognizing the need for a third party,’ and ‘Discovering possibilities.’ ‘Being helpful’ is the factor of continuing the supervision and involves ‘Being psychologically stable,’ ‘Knowing one’s limits’ and ‘Deepening the understanding of clients.’ We classified ‘Being a good learning experience’ as ‘Reflecting on oneself,’ ‘Learning through participation and observation’ and ‘Understanding human beings.’ We defined ‘Recognizing the need for a third party’ as the ability of counselors to see their blind spots. The fourth persistent factor is ‘Discovering possibilities,’ and applies equally to both the counselor and the client as they discover their potential.

여고생의 사상체질 간 운동·식생활 습관

박서희(Seo Hee Park),최세영(Se Young Choi),고유선(You Sun Ko) 대한스포츠한의학회 2013 대한스포츠한의학회지 Vol.13 No.1

Objective This study is to examine exercise and eating habits according to female high school students sasang constitution. Method The Subjects are 339 students of girls high schools located in Seoul and Gunsan. For the classification of Sasang constitution, QSCCII was used. After the questionnaire survey was conducted targeting 261 students, apart from 78 students who were not classified among the subjects and crosstabulation analysis was performed to test the differences of exercise and eating habits by sasang constitution with SPSS 18.0 program. Result & Conclusion The results are as follows. First, the students of soeumin type is thought that their physical strength level was weak, compared with other two types and they were fat and usually felt tired. Second, the students of taeeumin type is thought that they were fat, compared with other two types and did not take nutritional supplements well. On the other hand, the students of soyangin type is thought that they had normal body shapes. Third, there were no significant differences in food intake patterns according to sasang constitution, but female high school students generally preferred eating vegetables. In eating meat, overall, they much preferred eating pork and chicken, but the preference for fish was relatively low. The above results show that there are differences in exercise and eating habits and can be provided as basic research materials for the development of proper living habits education program.

부스팅 기반 기계학습기법을 이용한 지상 미세먼지 농도 산출

박서희 ( Seohui Park ),김미애 ( Miae Kim ),임정호 ( Jungho Im ) 대한원격탐사학회 2021 大韓遠隔探査學會誌 Vol.37 No.2

미세먼지 (PM₁₀) 및 초미세먼지 (PM_2.5)는 인체에 흡수 가능하여 호흡기 질환 및 심장 질환과 같이 인체 건강에 악영향을 미치며, 심각할 경우 조기 사망에 영향을 줄 수 있다. 전 세계적으로 현장관측기반의 모니터링을 수행하고 있지만 미 관측지역에 대한 대기질 분포의 공간적인 한계점이 존재하여 보다 광범위한 지역에 대한 지속적이고 정확한 모니터링이 필요한 상황이다. 위성기반 에어로졸 정보를 사용함으로써 이러한 현장 관측자료의 한계점을 극복할 수 있다. 따라서 본 연구에서는 다양한 위성 및 모델자료를 활용하여 2019년도에 대해 한 시간 단위의 지상 PM₁₀ 및 PM_2.5 농도를 추정하였다. GOCI 위성의 관측영역을 포함하는 동아시아 지역에 대해 트리 기반 앙상블 방법을 사용하는 Boosting 기법인 GBRTs (Gradient Boosted Regression Trees)와 LightGBM (Light Gradient Boosting Machine)을 활용하여 모델을 구축하였다. 또한, 기상변수 및 토지피복변수의 사용유무에 따른 모델의 성능을 비교하기 위해 두 가지 festure set으로 나누어 테스트하였다. 두 기법 모두 주요 변수인 AOD (Aerosol Optical Depth), SSA (Single Scattering Albedo), DEM (Digital Eelevation Model), DOY(Day of Year), HOD (Hour of Day)와 기상변수 및 토지피복변수를 함께 사용한 Feature set 1을 사용하였을 때 높은 정확도를 보였다. Feature set 1에 대해 GBRT 모델이 LightGBM에 비해서약 10%의 정확도 향상을 보였다. 가장 정확도가 높았던 기상 및 지표면 변수를 포함한 Feature set1을 사용한 GBRT기반 모델을 최종모델로 선정하였으며 (PM₁₀: R² = 0.82 nRMSE = 34.9%, PM_2.5: R² = 0.75 nRMSE = 35.6%), 계절별 및 연평균 PM₁₀ 및 PM_2.5 농도에 대한 공간적인 분포를 확인해본 결과, 현장관측자료와 비슷한 공간 분포를 보였으며, 국가별 농도 분포와 계절에 따른 시계열 농도 패턴을 잘 모의하였다. Particulate matter (PM₁₀ and PM_2.5 with a diameter less than 10 and 2.5 μm, respectively) can be absorbed by the human body and adversely affect human health. Although most of the PM monitoring are based on ground-based observations, they are limited to point-based measurement sites, which leads to uncertainty in PM estimation for regions without observation sites. It is possible to overcome their spatial limitation by using satellite data. In this study, we developed machine learningbased retrieval algorithm for ground-level PM₁₀ and PM_2.5 concentrations using aerosol parameters from Geostationary Ocean Color Imager (GOCI) satellite and various meteorological parameters from a numerical weather prediction model during January to December of 2019. Gradient Boosted Regression Trees (GBRT) and Light Gradient Boosting Machine (LightGBM) were used to estimate PM concentrations. The model performances were examined for two types of feature sets-all input parameters (Feature set 1) and a subset of input parameters without meteorological and land-cover parameters (Feature set 2). Both models showed higher accuracy (about 10 % higher in R²) by using the Feature set 1 than the Feature set 2. The GBRT model using Feature set 1 was chosen as the final model for further analysis (PM₁₀: R² = 0.82, nRMSE = 34.9 %, PM_2.5: R² = 0.75, nRMSE = 35.6 %). The spatial distribution of the seasonal and annual-averaged PM concentrations was similar with in-situ observations, except for the northeastern part of China with bright surface reflectance. Their spatial distribution and seasonal changes were well matched with in-situ measurements.

사계절 꽃 이미지의 색채 특성 분석

박서희(Park Seo-hee),최경실(Choi Gyeong-sil) 한국색채학회 2010 한국색채학회 논문집 Vol.24 No.4

본 연구는 환경 색의 일부 요소인 계절 꽃의 색채 특성 분석을 목적으로 한다. 계절 꽃의 색채 특성을 분석함으로써 계절마다 꽃 이미지들이 어떻게 다르게 보일 수 있는지에 대해 유추해보고, 계절 꽃 색을 활용할 수 있는 기반을 마련하고자 한다. 계절 꽃의 색채 데이터들은 시즌마다 계절의 변화에 민감한 패션, 섬유, 제품, 인테리어 디자인 등의 전략적인 마케팅 자료로 사용될 수 있다. 연구 과정 중, 분석도구로 적용된 GOG (Gain-Offset-Gamma) 모델은 입력하는 색채 데이터 값과 웹상에 재현되는 색채 값의 오차를 산정해주는 프로그램으로 웹 이미지 컬러 분석에 적합한 프로그램이다. 총 3가지 실험 과정 - 색채 분석, 육안 관찰, 디지털 디스플레이 특성화 ? 을 거친 결과, 계절마다 꽃 잎의 색상, 채도, 명도, 톤이 모두 다르다는 것을 알 수 있었다. 이러한 계절 색의 차이점은 분야를 넘어서 계절의 영향을 받는 여러 디자인 분야에 도움이 될 것으로 예상된다. 그 이유는 이러한 계절 색의 변화가 결국 그 자연 환경에 사는 사람들의 선호 색과 연상 색 등에 영향력을 미칠 수 있기 때문이다. The purpose of this research is to analyze the color of seasonal flowers, which is one of the elements in the environmental colors. The bases to utilize the colors of seasonal flowers will be established by inferring ‘how the flower images would look different dependent on the season’, through the analysis of seasonal flowers color characteristics. The color data of seasonal flowers can be used as strategic marketing data in the industries such as fashion, textile, commodities and interior design, which industries are sensitive to the change of season. In addition, the process of reducing the data error through the color expression characteristic of digital color was employed in the suggestion of more accurate and efficient methodology for the color management. The GOG (Gain-Offset-Gamma) model applied in this research as the analyzing tool is the program which estimates the color value errors between the input color data and colors expressed on the Web. The program is compatible program in the analysis of Web image colors. Three experiments were done for the color analysis, bare-eye analysis and characterization of digital display. It was possible to learn from the results that the colors, chroma, brightness and tones are all different dependent on the season. It is expected that such difference in the seasonal colors will be helpful to various design areas affected by season beyond its area. The reason is that the change in such seasonal colors will eventually give impact to the preferred colors and associated colors of people living in that natural environment.

논문 : 공공 서비스 디자인을 위한 사용자 참여형 디자인 프로세스 모델 연구

박서희 ( Seo Hee Park ),노승완 ( Seung Wan Roh ) 디자인융복합학회 2012 디자인융복합연구 Vol.11 No.4

공공 서비스 디자인은 공익의 실현이라는 최종 목표를 통해, 시민들의 삶의 질을 높이고, 긍정적인 국가 이미지를 얻음으로써 또한 국가 경쟁력을 강화시키는 부가가치를 창출할 수 있는 잠재적 신시장성을 가진 디자인 분야로 급부상하고 있다. 이미 해외 디자인 선진국들은 이러한 서비스디자인에 대한 다양한 연구 자료들을 발표하고, 공공기관과 민간 기업의 체계적인 네트워크 형성을 통해 서비스 디자인 시장을 선도적으로 잘 주도해 나가고 있다. 사용자의 니즈(Needs)는 점차 복합적이고 전문화되고 있으며, 다양해지고 있다. 이러한 추세와 반대로 기존의 권위적이고 일방적이었던 서비스 제공의 문제점을 충분히 인식한 공공기관들은 디자인 관점에서 바라보는 공공 서비스의 차별화되고 고급화된 방법론에 대해 관심을 기울이고 있다. 그 이유는 공공 서비스 디자인의 최종 목표와 가치가사용자들의 경험 만족도를 합리적, 효과적으로 증대시키는 것이며, 이를 위한 사용자 중심 디자인 방법론이 다양하게 연구되고 있기 때문이다. 따라서 본 연구는 디자인 측면에서 공공 서비스 프로세스에 대한 기존의 해외 사례들을 통해 사용자들의 적극적인 참여도를 높일 수 있는 방법과 체계적이고 현실화 가능한 디자인 프로세스 모델을 제시하는데 그 의미가 있다. Public service design through the final goal which is the realization of public interests improve the quality of life for its citizens, a positive country image can be obtained. Furthermore it has emerged to have the potential availability of the new market design field. Overseas design developed countries announce a variety of these research design, presentation materials and make formation of a systematic network with public agencies and private companies in design market leading services continues to lead as well. Existing service providers now have been paying attention to the advanced methodologies of design perspective. The reason is the ultimate goal of public service design has value to users` reasonable satisfaction of the experience effectively increasing. There for, this study researched the design process for public services in terms of existing international practices through active participation of users and how that can increase the realization of systematic and design process as possible to present a model that makes sense.

한국판 보스톤 이름대기 검사 단축형(K-BNT-15)의 의미단서 고찰

박서희(Seo Hui Park),김향희(Hyang Hee Kim),서상규(Sang Kyu Seo),김한샘(Kim Han Saem) 한국언어치료학회 2016 한국언어치료학회 학술발표대회 논문집 Vol.2016 No.12

Purpose: Cueing is a common facilitation technique for impaired naming. When retrieving the words, semantic information is activated. Sematic cue are commonly used to retrieve information from person memory. The present study investigated the high frequency core meanings of the K-BNT-15 for verbal definitions task in normal elderly. In addition, the study confirmed the comparison of the high frequency core meanings for normal elderly and younger people. Methods: Fifty-six normal elderly and sixty-five younger people participated in the study. Normal elderly and younger people performed word definition tasks of K-BNT-15. Task performances were analyzed on the number of semantic features and divided into four semantic categories(functional, relational, physical, categorical) Results: The results were as follows. First, With regard to semantic categories of each K-BNT-15 words, ten of the 15 words showed higher ratio of relational definition. Second, It is found that there is a difference in high frequency core meanings of the K-BNT-15 between two age groups. Third, Core meanings of 10 words that showed difference between the two groups, normal elderly group showed significantly higher ratio in relational definition. Fourth, supplementary meanings of 10 words the showed difference between the two groups, normal elderly group showed significantly higher ratio in relational definition. Conclusions: The present study investigated the high frequency core meanings of the K-BNT-15 in normal elderly and compared with their core meanings with those of younger people. The results of the study can be utilized as a useful information in evaluation and treatment of the K-BNT-15.