부산시 반송동 정책이주지 주택 변화 연구

하남구,우신구 (사) 한국도시재생학회 2022 도시재생 Vol.8 No.1

This study is to see the changing process of housing in Bansong-dong policy migration area started from 1960s and try to analyze the cause and effect. Through document survey, field survey and in-depth interview, this study traced the changing process and categorized types of housing. From 1969 to 2021, changing process of housing in Bansong-dong policy migration area can be divided into four stages : formation period, extension period, stagnation period, and declining period. After exploring the 4 stages of changing process of housing, the characteristics of transformation can be analyzed as following. First, the interior spaces of housing has be changed through constant addition and extension. Second, the front space of housing has been the focus of spatial change according to the lot-street relation and house-street relation. Third, the spaces of housing has been organized three-dimensionally to overcome small and narrow interior space. Fourth, movement of corridor and stairs has been carefully designed to obtain individuality of separate households in a house. The survey and analysis of this study would by used as basic research data for local-based research to estimate the value and meaning of policy migration area and for urban regeneration project to conserve and manage the reclined residential area formed during modernization and industrialization period of Busan.

부산시 정책이주지의 변화과정 연구 - 금정구 서동 정책이주지를 중심으로 -

하남구,우신구 대한건축학회지회연합회 2021 대한건축학회연합논문집 Vol.23 No.3

This study shows the process of changes in Seo-dong Policy Migration area of Geumjeong-gu in Busan. Considering the urban policies, population change and chronological incidents, this study divided 4 different stages of change as ‘Formation Period of policy migration area’, ‘Expansion Period according to needs and requirements’, ‘Stagnant Period following Act on Temporary Measures’ and ‘Decline Period with New Town Plan’. After exploring the policies, spatial structure and architectural changes of each period, following conclusions have been found. First, without detailed plan and infra structures, policy migration in Busan in the 1960s and 1970s produced another low quality residential area excluding urban poor in the downtown to the outskirts of the city with many problems in living, working and education. Second, simple planning of repetitive back-to-back small standard lot (50.98m2) without considering topography produced monotonous residential area without adequate street, park and public facility. Third, following urban expansion with industrial district nearby area and urban policies to consider the people in policy migration area had caused huge housing demand and legal/illegal extension, expansion and new construction of of housing. Fourth, the people in the policy migration area have been responding to the changing social/economic/policy environment with producing maximum building coverage ratio and fllor area ratio, creating various types of housing and functional variation of spaces utilizing small standard lot, merging lots or dividing lot.

그룹 이동성 모델에서 전방향 벡터를 이용한 정확한 파티션 시간 예측

하남구(Namkoo Ha),이병직(Byeongjik Lee),황성호(Sungho Hwang),한기준(Kijun Han) 한국정보과학회 2004 한국정보과학회 학술발표논문집 Vol.31 No.1A

애드혹 네트워크에서 그룹 이동성을 고려할 경우, 주요한 이슈 중 하나가 그룹의 파티션 문제이다. 파티션 시간의 예측을 위해서, 기존의 모델들은 파티션이 발생하는 두 그룹의 대칭적 이동을 가정한다. 그러나, 실제 상황에서 파티션은 다양한 방향으로 일어나므로 대칭적 방향만을 고려한다면, 정확한 파티션 시간을 예측할 수 없다. 본 논문은 대칭적 이동뿐만 아니라, 모든 방향을 고려한 네트워크 파티션 분석 모델을 제시하고 수학적인 분석을 통해 이를 증명한다. 따라서, 정확한 파티션 시간의 예측이 가능하다.

부산 감천2동 경사지 마을의 특성에 관한연구

하남구(Ha, Nam-Gu),우신구(Woo, Shin-Koo) 대한건축학회 2011 대한건축학회 학술발표대회 논문집 - 계획계/구조계 Vol.31 No.1(계획계)

무선 센서 네트워크에서 효율적인 타겟 트래킹 방법

하남구(Namkoo Ha) 대한전자공학회 2010 대한전자공학회 학술대회 Vol.2010 No.6

아미동 비석마을의 공간구조

정회영,우신구,하남구 대한건축학회 2018 대한건축학회논문집 Vol.34 No.2

Ami-dong, known as Tombstone Village, was generated by refugees after Korean war on the site of Japanese public cemetery which hadbeen originally constructed during Japanese occupation. Since then, the village has been expanded as low-income class neighborhood duringModernization and Industrialization Period from 1960s to 1980s. This study try to find the impact of the Japanese cemetery on the spatialstructure of the village by analyzing and comparing the street structure and spatial layout of the Japanese cemetery and current TombstoneVillage. The final objective of this study is to confirm the continuity of built environment, historical value and place identity of of thevillage. 비석마을이라는 이름으로 알려져 있는 부산 아미동의 주거지역은 일제강점기 일본인 공동묘지에 한국 전쟁 이후 부산으로 모여든 피난민들이 정착하면서 주거지역으로 형성되었다. 이후 1960년대에서 1980년대까지 산업화 시기를 거치면서 이 지역은 저소득층 주거지로 크게 확장되었다. 본 연구는 일제강점기 공동묘지가 현재의 주거지에 어떤 영향을 끼쳤는지 알아보기 위해 공동묘지와 현재 주거지의 가로구조와 공간배치를 비교하고 분석하였다. 본 연구의 궁극적 목적은 건조환경의 지속성, 마을의 역사적 가치와 장소적 정체성을 확인하는 것이다.

단일 영상 비균일 블러 제거를 위한 다중 학습 구조

정형주,장현성,하남구,연윤모,권구용,손광훈 한국멀티미디어학회 2019 멀티미디어학회논문지 Vol.22 No.10

We present a novel deep learning architecture for obtaining a latent image from a single blurry image, which contains dynamic motion blurs through object/camera movements. The proposed architecture consists of two sub-modules: blur image restoration and optical flow estimation. The tasks are highly related in that object/camera movements make cause blurry artifacts, whereas they are estimated through optical flow. The ablation study demonstrates that training multi-task architecture simultaneously improves both tasks compared to handling them separately. Objective and subjective evaluations show that our method outperforms the state-of-the-arts deep learning based techniques.

안개영상의 의미론적 분할 및 안개제거를 위한 심층 멀티태스크 네트워크

송태용,장현성,하남구,연윤모,권구용,손광훈 한국멀티미디어학회 2019 멀티미디어학회논문지 Vol.22 No.9

Image semantic segmentation and dehazing are key tasks in the computer vision. In recent years, researches in both tasks have achieved substantial improvements in performance with the development of Convolutional Neural Network (CNN). However, most of the previous works for semantic segmentation assume the images are captured in clear weather and show degraded performance under hazy images with low contrast and faded color. Meanwhile, dehazing aims to recover clear image given observed hazy image, which is an ill-posed problem and can be alleviated with additional information about the image. In this work, we propose a deep multi-task network for simultaneous semantic segmentation and dehazing. The proposed network takes single haze image as input and predicts dense semantic segmentation map and clear image. The visual information getting refined during the dehazing process can help the recognition task of semantic segmentation. On the other hand, semantic features obtained during the semantic segmentation process can provide cues for color priors for objects, which can help dehazing process. Experimental results demonstrate the effectiveness of the proposed multi-task approach, showing improved performance compared to the separate networks.

시간에 따라 변화하는 빗줄기 장면을 이용한 딥러닝 기반 비지도 학습 빗줄기 제거 기법

조재훈,장현성,하남구,이승하,박성순,손광훈 한국멀티미디어학회 2019 멀티미디어학회논문지 Vol.22 No.1

Single image rain removal is a typical inverse problem which decomposes the image into a background scene and a rain streak. Recent works have witnessed a substantial progress on the task due to the development of convolutional neural network (CNN). However, existing CNN-based approaches train the network with synthetically generated training examples. These data tend to make the network bias to the synthetic scenes. In this paper, we present an unsupervised framework for removing rain streaks from real-world rainy images. We focus on the natural phenomena that static rainy scenes capture a common background but different rain streak. From this observation, we train siamese network with the real rain image pairs, which outputs identical backgrounds from the pairs. To train our network, a real rainy dataset is constructed via web-crawling. We show that our unsupervised framework outperforms the recent CNN-based approaches, which are trained by supervised manner. Experimental results demonstrate that the effectiveness of our framework on both synthetic and real-world datasets, showing improved performance over previous approaches.

적외선 표적 모델링을 위한 3차원 복합 열해석 기법 연구

장현성(Hyunsung Jang),하남구(Namkoo Ha),이승하(Seungha Lee),최태규(Taekyu Choi),김민아(Minah Kim) 한국정보과학회 2017 정보과학회논문지 Vol.44 No.4

적외선 표적의 정밀한 모델링을 위해서는 정확한 표면온도 계산이 필요하다. 본 논문에서는 전도, 대류, 복사를 고려한 복합 열해석 모듈을 소프트웨어로 구현하고, 이를 통하여 표적 재질 및 자세, 환경 요건에 따른 표적의 표면온도 해석을 수행 하였다. 구현된 결과는 상용 소프트웨어인 OKTAL-SE 와의 비교를 통하여 결과의 신뢰성을 검증하였다. 그 결과 자체 검증이 완료된 상용 소프트웨어인 OKTAL-SE 와 약 1% 이내의 오차를 보였다. 계산된 온도 결과를 바탕으로 적외선 표적 모델링을 수행하였으며 OKTAL-SE와의 연동을 통해 적외선 신호 해석을 수행하였다. The spectral radiance received by an infrared (IR) sensor is mainly influenced by the surface temperature of the target itself. Therefore, the precise temperature prediction is important for generating an IR target image. In this paper, we implement the combined three-dimensional surface temperature prediction module against target attitudes, environments and properties of a material for generating a realistic IR signal. In order to verify the calculated surface temperature, we are using the well-known IR signature analysis software, OKTAL-SE and compare the result with that. In addition, IR signal modeling is performed using the result of the surface temperature through coupling with OKTAL-SE.