http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
2단계 신경망과 계층적 프레임 탐색 방법을 이용한 MPEG 비디오 분할
김주민,최영우,정규식,Kim, Joo-Min,Choi, Yeong-Woo,Chung, Ku-Sik 한국정보과학회 2002 정보과학회논문지 : 소프트웨어 및 응용 Vol.29 No.1
본 논문에서는 MPEG 비디오 데이터의 컷(cut)과 디졸브(dissolve)를 검출하여 샷(shot) 단위로 분할하고 각 샷의 카메라 동작 또는 객체 움직임의 형태를 분류하는 방법을 제안하고자 한다. 정확한 샷의 위치와 카메라, 객체의 세분화된 동작을 구별하기 위한 전단계의 연구에서[1] 우선 MPEG 데이터의 I(Intra) 프레임의 DC(Direct Current) 계수를 분석하여 픽처 그룹을 Shot(장면이 바뀐 경우), Move(카메라 동작 또는 객체가 움직인 경우), Static(영상의 변화가 거의 없는 경우)으로 세분화하여 분류하였다. 이 과정에서 2단계 구조의 신경망을 구성하고 여러 종류의 특징을 서로 다른 해상도에서 추출하여 결합시키는 방법을 제안하였다. 다음 단계로 Shot 또는 Move로 분류된 픽처 그룹의 P(Predicted), B(Bi-directional) 프레임을 선별적, 계층적으로 탐색하여 컷의 정확한 발생 위치와 카메라 동작 또는 객체 움직임의 종류를 결정하는 방법을 제안한다. P, B 프레임의 매크로 블록의 종류별 분포를 통계적으로 이용하여 컷의 발생 위치를 검출하여, P, B 프레임의 매크로 블록 종류와 움직임 벡터를 동시에 사용하는 신경망을 구성하여 디졸브, 카메라 동작, 객체 움직임의 종류를 검출한다. 본 논문에서 제안하는 방법은 MPEG 데이터의 압축을 풀지 않은 상태에서 I 프레임의 DC 계수만을 사용하여 픽처 그룹을 분류하며, 분류된 픽처 그룹 내에서 일부의 P, B 프레임만을 계층적으로 선택하여 탐색함으로서 처리 시간을 감소시키고자 하였다. 세 종류의 서로 다른 비디오 데이터를 사용한 실험에서 93.9-100.0%로 픽처 그룹을, 96.1-100.0%로 컷을 검출하였다. 또한 두 종류의 비디오 데이터를 사용한 실험에서 90.13% 및 89.28%의 정확성으로 카메라 동작 또는 객체 움직임을 분류하였다. In this paper, we are proposing a hierarchical segmentation method that first segments the video data into units of shots by detecting cut and dissolve, and then decides types of camera operations or object movements in each shot. In our previous work[1], each picture group is divided into one of the three detailed categories, Shot(in case of scene change), Move(in case of camera operation or object movement) and Static(in case of almost no change between images), by analysing DC(Direct Current) component of I(Intra) frame. In this process, we have designed two-stage hierarchical neural network with inputs of various multiple features combined. Then, the system detects the accurate shot position, types of camera operations or object movements by searching P(Predicted), B(Bi-directional) frames of the current picture group selectively and hierarchically. Also, the statistical distributions of macro block types in P or B frames are used for the accurate detection of cut position, and another neural network with inputs of macro block types and motion vectors method can reduce the processing time by using only DC coefficients of I frames without decoding and by searching P, B frames selectively and hierarchically. The proposed method classified the picture groups in the accuracy of 93.9-100.0% and the cuts in the accuracy of 96.1-100.0% with three different together is used to detect dissolve, types of camera operations and object movements. The proposed types of video data. Also, it classified the types of camera movements or object movements in the accuracy of 90.13% and 89.28% with two different types of video data.
치어기 넙치(Paralichthys olivaceus) 사료 내 돈모분(Pig Bristle Meal)의 어분대체 가능성 평가
김유정 ( Youjeong Kim ),신재형 ( Jaehyeong Shin ),권황원 ( Hwangwon Kwon ),이소연 ( Soyoon Lee ),김주민 ( Joo-min Kim ),김민기 ( Min-gi Kim ),김정대 ( Jeong-dae Kim ),이경준 ( Kyeong-jun Lee ) 한국수산과학회(구 한국수산학회) 2018 한국수산과학회지 Vol.51 No.2
This study was conducted to evaluate dietary hydrolyzed pig bristle meal (PBM) for juvenile olive flounder Paralichthys olivaceus. In Experiment 1 (EXP-1), six experimental diets were prepared to contain 0, 3, 6, 9, 12 and 15% PBM (designated Con, PBM3, PBM6, PBM9, PBM12 and PBM15, respectively). Triplicate groups of olive flounder (initial body weight, 8.69 g) were fed the diets to apparent satiation for 8 weeks during the optimal water temperature season (20.5±2.12℃). All PBM supplemented groups except for PBM3 showed significantly lower growth performance and feed utilization compared to the control group. The protein digestibility of PBM3, PBM6, and PBM9 diets did not significantly differ from that of the control diet. In Experiment 2, 1% mono-calcium phosphate was added into the experimental diets used in Exp-1. Triplicate groups of olive flounder (10.6 g) were fed the diets to apparent satiation for 8 weeks during the low water temperature season (12.5±1.12℃). The growth performances and feed utilization of fish fed all diets except for PBM15 diet did not significantly differ from those of the control diet. This study indicates that hydrolyzed PBM can replace fish meal by up to 12% with limiting amino acids and mono-calcium phosphate in diets for juvenile olive flounder.
김진우(Jin-woo Kim),김동선(Dong-sun Kim),김주민(Joo-min Kim),김대원(Daewon Kim) 대한전기학회 2010 정보 및 제어 심포지엄 논문집 Vol.2010 No.4
In this paper, we propose an evaluation model for a virtual robot contest to design a reliable and balanced virtual robot contest. Since it is important to ensure public confidence, we define evaluation indices based on the standard of software quality assessment, that is ISO 9126. We validate the evaluation model through two processes. Firstly, we design a virtual robot contest concerning the escape from obstacles by simulation including the ODE. Through the experience of designing the contest, we modify some evaluation indices. Secondly, we modify some evaluation indices in the model through the evaluation of existing virtual robot contests such as "Robo Champ by the MSRDS". Finally we make the reliable evaluation model of virtual robot contests.
김진우(Jin-woo Kim),김주민(Joo-min Kim),김대원(Dae-won Kim) 대한전기학회 2009 대한전기학회 학술대회 논문집 Vol.2009 No.7
This paper is proposed to make a virtual reality 3D operation model including a robot model and an environment model for a robot contest. First of all, we make the robot model using Open Dynamics Engine(ODE), and the environment model similar to the robot contest. These ODE models are programmed by the C-language. Lastly, we perform simulations to verify the robot model and the environment model. As a result, it shows the possibility of models for the robot contest.
안드로이드 플랫폼 상에서 동기화가 고려된 통합 커서의 설계 및 구현
김경환,하주호,원종필,이의성,김주민,손진호,Kim, Kyung-Hwan,Ha, Jo-Ho,Won, Jong-Pil,Lee, Uee-Song,Kim, Joo-Min,Son, Jin-Ho 대한임베디드공학회 2011 대한임베디드공학회논문지 Vol.6 No.3
Android platform provides a content provider and a cursor mechanism to access the internal SQLite engine. Content providers not only store and retrieve data but also make it accessible to applications. Applications can only share data through content provider, since there's no common storage area that Android packages can access. Cursor is an interface that provides random read-write access to the result set returned by a database query. However, this cursor possesses two major limitations. First, a cursor does not support a join clause among cursors, since the cursor can only access a single table in the content provider. Second, the cursor is not capable of creating user-customized field in the predefined content providers. In this paper, we propose the unified cursor architecture that merges several cursors into a single virtual cursor. Cursor translation look-aside buffer (TLB), column windowing mechanism and virtual data management are the three major techniques we have adopted to implement our structure. And we also propose a delayed synchronization method between an application and a proposed unified cursor. An application can create a user-customized field and sort multiple tables using a unified cursor on Android platform.
개미 집단 최적화 기법을 이용한 이동 로봇 최적 경로 생성 알고리즘 개발
고종훈(Jong-Hoon Ko),김주민(Joo-Min Kim),김대원(Dae-Won Kim) 대한전기학회 2009 대한전기학회 학술대회 논문집 Vol.2009 No.7
In this paper proposes a new algorithm for path planning using the ant colony optimization algorithm. The proposed algorithm is a new hybrid algorithm that composes of the features of the ant colony algorithm method and the Maklink graph method. At first, paths are produced for a mobile robot in a static environment, and then, the midpoints of each obstacles nodes are found using the Maklink graph method. Finally, the shortest path is selected by the ant colony optimization algorithm.
김진홍(Jin-Hong Kim),우창균(Chang-Gyun Woo),김주민(Joo-Min Kim),박기홍(Ki-Hong Park) 한국정보기술학회 2021 Proceedings of KIIT Conference Vol.2021 No.11
탄소중립 시대가 도래함에 따라 다양한 분야에서 전동화를 이룬 자동차가 개발 · 출시되고 있다. 이에 자율주행 차량 시대도 현실화 되면서 컴퓨터 비전기술 기반의 객체 탐지기술과 그림자 제거의 기술은 중요시되고 있다. 본 논문에서는 객체탐지의 효율성을 개선하기 위한 기법을 분석하고, 향후 그림자 제거 기술까지 적용하여 효율적인 차선인식율 향상을 위한 방법을 제안한다. As the era of carbon neutrality approaches, automobiles that have been electrified in various fields were developed. Accordingly, as the era of self-driving vehicles has become a reality, object detection technology based on computer vision technology and shadow removal technology are becoming important. In this paper, a technique for improving the efficiency of object detection is analyzed, and a method for efficient lane recognition rate is proposed by applying shadow removal technology in the future.