MR 머리 영상의 뇌 경계선 추출 및 디렉트 볼륨 렌더링

송주환(Ju Whan Song),권오봉(Ou Bong Gwun),이건(Kun Lee) 한국정보과학회 2002 정보과학회 컴퓨팅의 실제 논문지 Vol.8 No.6

본 논문은 MR 머리 영상 데이타를 디렉트 볼륨 렌더링하는 방법을 제안한다. MR 영상을 가시화하기 위해서는 서피스 렌더링을 많이 사용하나 이 방법은 면을 추출하는 과정에서 면 내부의 정보를 잃어버린다. 디렉트 볼륨 렌더링은 면 내부의 정보를 추출 할 수 있으나 데이타의 특성상 MR 머리 영상 데이타에 이 방법을 적용하기가 쉽지 않다. 이 논문에서는 MR 머리 영상 데이타를 뇌와 뇌 이외의 구성 요소로 분할한 다음에 뇌 복셀값을 증가시키고 원래의 영상과 다시 결합시켜 디렉트 볼륨 렌더링을 시도하였다. 뇌 경계선은 히스토그램 경계값, 모포로지 연산, 스네이크 알고리즘(snakes algorithm)을 이용하여 추출하였다. 추출된 뇌 경계선는 육안으로 추출한 것의 91~95%의 유사도를 보인다. 제안된 디렉트 볼륨 렌더링은 뇌와 뇌 이외의 구성 요소를 동시에 3차원 가시화하였다. This paper proposes a method which visualizes MRI head data in 3 dimensions with direct volume rendering. Though surface rendering is usually used for MRI data visualization, it has some limits of displaying little speckles because it loses the information of the speckles in the surfaces while acquiring the information. Direct volume rendering has ability of displaying little speckles, but it doesn't treat MRI data because of the data features of MRI. In this paper, we try to visualize MRI head data in 3 dimensions as follows. First, we separate the brain region from the head region of MRI head data, next increase the pixel level of the brain region, then combine the brain region with the increased pixel level and the head region without brain region, last visualizes the combined MRI head data with direct volume rendering. We segment the brain region from head region based on histogram threshold, morphology operations and snakes algorithm. The proposed segmentation method shows 91~95% similarity with a hand segmentation. The method rather clearly visualizes the organs of the head in 3 dimensions.

라즈베리 파이를 이용한 차량 추적 시스템에 관한 연구

송주환(Ju-Whan Song) 한국디지털콘텐츠학회 2020 한국디지털콘텐츠학회논문지 Vol.21 No.5

This study implemented a system that automatically finds the wanted vehicle and sends the vehicles location and photos to the server. We used Raspberry Pi to create a system to find the wanted vehicle, and we can find the wanted vehicle very quickly. In the future, we will add it to the vehicles black box so that we can find the wanted vehicle. The proposed system is performed by downloading the list of wanted vehicle numbers from the vehicle wanted server, finding the vehicle in real time, and if it is found, sending the vehicle number, photo, location information, etc. to the server. Finding the vehicle license plate from the image captured by the camera was processed using an image processing technique using OpenCV. To find the vehicle number on the license plate, the tesseract library was used after preprocessing the image. The license plate recognition and number recognition resulted in good results with recognition rates of 96% and 64%, respectively.

판별분석을 이용한 효율적인 3차원 모델 검색

송주환(Ju-Whan Song),최성희(Seong-Hee Choi),권오봉(Ou-Bong Gwun) 대한전자공학회 2008 電子工學會論文誌 IE (Industry electronics) Vol.45 No.2

본 논문은 통계 기법인 판별 분석 함수를 이용하여 효율적으로 3차원 모델을 검색하는 시스템을 구현하였다. 제안한 방법은 판별분석 함수를 이용하여 색인으로 검색 하는 기법으로, 색인의 생성은 Osada의 D2방법에 의해 추출된 128개의 특징벡터에 통계치(범위, 최소값, 평균, 표준편차, 왜도, 척도)를 변수로 판별분석 함수의 값을 색인 값으로 생성하였다. 쿼리 모델 검색 시 1차 검색으로 쿼리와 저장된 클래스(동종의 모델 그룹)의 색인을 비교하여 상위 2%이내(98% 이상)의 클래스를 추출하여 추출된 클래스에 속하는 모델만을 검색하였다. 이 방법은 검색시간을 단축시키는 효율적인 검색 기법임을 구현을 통해 알 수 있었다. 제안한 방법은 기존의 방법(Osada)보다 3차원 모델 검색 시간을 57%로 단축시켰으며, 쿼리 모델 검색 시 유사모델이 최초로 발견되는 정확도(precision)가 0.362로 기존의 방법보다 44.8%의 효율이 있었음을 알 수 있었다. This study established the efficient system that retrieves the 3D model by using a statistical technique called the function of discriminant analysis. This method was suggested to search index, which was formed by the statistics of 128 feature vectors including those scope, minimum value, average, standard deviation, skewness and scale. All of these were sampled with Osada's D2 method and the statistics as a factor effecting a change turned the value of discriminant analytic function into that of index. Through the primary retrieval on the model of query, the class above the top 2% was drawn out by comparing the query with the index of previously saved class from the group of same models. This method was proved an efficient retrieval technique that saved its procedural time. It shortened the retrieval time for 3D model by 57% faster than the existing Osada's method, and the precision that similar models were found in the first place was recorded 0.362, which revealed it more efficient by 44.8%.

구면좌표계 기반에서 3차원 모델 검색

송주환(Ju-Whan Song),최성희(Seong-Hee Choi) 대한전자공학회 2009 電子工學會論文誌 IE (Industry electronics) Vol.46 No.1

본 논문에서는 구면 좌표계 기반에서 3차원 모델을 검색하는 새로운 알고리즘을 제안한다. 3차원 모델 위의 임의의 점들의 좌표(x, y, z)를 구하고, 이 좌표들을 구면좌표계의 좌표로 변환한다. 이 샘플들의 위도(zenith)의 분포를 3차원 모델의 특징으로 정의한다. 임의의 샘플 좌표를 구하기 위해 우리는 Osada가 제안한 방법을 사용하였고, 좌표축을 정규화하기 위하여 PCA 알고리즘을 사용하였다. 데이터는 프린스턴 대학의 벤치마크 데이터를 사용하였으며 Vranic이 제안한 depth buffer-based feature vector 알고리즘과 비교하였고, 본 논문에서 제안한 방법이 정확도에서 12.6% 더 정확하게 모델을 검색하였다. In this paper, we propose a new algorithm for 3D model retrieval based on spherical coordinate system. We obtains sample points in a polygons on 3D model. We convert a point in cartesian coordinates(x, y, z) to it in spherical coordinate. 3D shape features are achieved by adopting distribution of zenith of sample point in spherical coordinate. We used Osada's method for obtaining sample points on 3D model and the PCA method for the pose standardization 3D model. Princeton university's benchmark data was used for this research. Experimental results numerically show the precision improvement of proposed algorithm 12.6% in comparison with Vranic's depth buffer-based feature vector algorithm.

신문사 RSS를 활용한 실시간뉴스 어플리케이션 설계 및 구현

송주환(Ju-Whan Song) 한국디지털콘텐츠학회 2018 한국디지털콘텐츠학회논문지 Vol.19 No.4

In order to read newspaper articles, the use of paper newspapers is decreasing and smartphone are increasingly used. As a result, the number of news apps continues to increase. Many of the news apps in the Android Play Store fall into two categories. The first is an app that is developed by a specific newspaper company and distributes only the articles of the newspaper company. The rest is an app that shows a list of newspapers and shows the homepage when a newspaper is selected. In this paper, we have designed and implemented a Real-Time News app for collecting articles from many newspapers and providing them in real time. Newspapers provide up-to-date articles with RSS feeds. The server program stores them in the DB, and transmits the articles requested in the Real-Time News app in real time. In order to see the latest news, it is possible to collect the articles of each newspaper without visiting the websites of the various newspapers, and it is possible to reduce the mobile data usage used to access each website.

반사 대칭을 이용한 3차원 오브젝트 검색에 관한 연구

송주환(Ju-Whan Song),최성희(Seong-Hee Choi),권오봉(Ou-Bong Gwun) 대한전자공학회 2007 電子工學會論文誌 IE (Industry electronics) Vol.44 No.4IE

멀티미디어 활용이 다양해지면서 3차원 오브젝트 데이터 처리(저장, 편집, 검색 등)에 대한 관심이 높아지고 있다. 본 논문은 대칭면을 활용한 3차원 오브젝트의 검색 시스템을 제안?구현하였다. 검색기법으로는 투영선을 이용한 반사대칭축을 이용하였다. 반사대칭에 의해 대칭면을 구하고, 대칭면에 대한 깊이버퍼를 구하였으며, 깊이버퍼에 푸리에 변환을 적용하여 오브젝트에 대한 특징벡터를 생성하여 검색하였다. 질의에 의한 샘플은 콘스탄츠 대학(Konstantz University)의 벤치마크 데이터를 이용하여 유사 대칭성의 모델을 추출하였다. 대부분의 물체들이 대칭의 특성을 갖는 오브젝트임을 고려해 볼 때 본 논문이 제안한 반사대칭면을 이용한 3차원 오브젝트 검색에 의한 방법은 우수한 검색 시스템임을 알 수 있었다. Due to the diverse utilization of multimedia, interests are increasing towards processing (saving, editing, searching, etc) three dimensional object data. This paper proposes and realizes a retrieval system for three dimensional objects using reflective symmetry. For the retrieval method, a reflective symmetric axis with a projector is used. The symmetric plane is calculated by the reflection symmetry, and the depth buffer is calculated for the symmetric plane. Then, by applying the Fourier Transform to the depth buffer, the feature vector for the object is generated and retrieved. For the sample of inquiry, the model of similar symmetry was extracted using the benchmark data from Konstantz University. Considering that most of the objects have symmetrical characteristics, the proposed method of retrieving three dimensional objects using a reflective symmetric surface is an outstanding retrieval system.

펠티어 모듈과 PID 제어를 이용한 온도 정밀제어 시스템 개발

나성주(Sung-Ju Na),송주환(Ju-Whan Song) 한국디지털콘텐츠학회 2022 한국디지털콘텐츠학회논문지 Vol.23 No.3

This study designed and developed a temperature precision control system for a small chamber that can be used for plant research using Raspberry Pi and Peltier module. The existing system that goes into a small chamber was a simple on/off control. This greatly affected the temperature, which is a factor closely related to plant growth. The temperature was not maintained, and the vibrating phenomenon around the target temperature was found. In order to solve this point, this study calculated each temperature and the control amount every 1 minute through PID control to control the Peltier module. As a result, when the external temperature vibrates in a decimal point unit at 22 degrees Celsius, the chamber implemented in this study was confirmed that the internal temperature of 20.2 degrees was maintained for about 20 minutes. Through this study, it will be of great help to researchers by realizing factor precision control even in a small chamber.

라즈베리파이 기반 원격 식물 모니터링 시스템 개발

나성주(Sung-Ju Na),송주환(Ju-Whan Song) 한국디지털콘텐츠학회 2021 한국디지털콘텐츠학회논문지 Vol.22 No.4

This study proposes a monitoring system that can remotely check and transmit environmental conditions of plants in real time by measuring temperature, humidity, carbon dioxide, etc. which are factors affected by plants using a Raspberry Pi. Introducing an implementation example that enables you to check the environmental status of plants on a smartphone in real time through MQTT. It utilizes the Raspberry Pi, which is more open than the existing closed Arduino system and makes remote access even easier. The environment in which plants are placed is measured through various sensors connected to the Raspberry Pi. It is a system that collects environmental data, stores it in a web database, and checks it with a smartphone through MQTT in real time. This study is expected to reduce the cost of constructing an expensive Smart Farm and relieve the burden of farmers.

라즈베리 파이 기반의 IoT 식물재배기 구현

이수아(Su-A Lee),송주환(Ju-Whan Song) 한국디지털콘텐츠학회 2021 한국디지털콘텐츠학회논문지 Vol.22 No.5

As modern peoples interest in health has increased, the number of people who grow vegetables themselves has increased. However, there are a number of difficulties that arise when raising them directly. In this paper, we propose the plant growth chamber that anyone can manage more easily. Various plant growth chamber are available on the market, but most of them are limited because they are remotely controlled using Bluetooth communication. The system proposed in this paper measures the environment inside the plant growth chamber with a sensor connected to the Raspberry Pi, monitors the cultivation environment through the Android app, and enables real-time streaming of the inside of the plant growth chamber with a camera, so that the condition of the crop can be grasped. In addition, Raspberry Pi and Android app communicate with each other through a socket to remotely control the actuator so that a plant growth chamber can be managed from outside.

다중해상도 복셀화 기법

김아미(A-Mi Kim),송주환(Ju-Whan Song) 한국정보기술학회 2012 한국정보기술학회논문지 Vol.10 No.6

In this paper we introduce a new method about voxelization. Unlike existing methods, polygons are projected onto the principal axis and then we set the bounding volume to satisfy the minimal coverage criteria of discrete surface representation. Solid voxelization is performed by filling the model’s surface with a voxel in each case the vertex intersects the planes. In this process, lists of associated vertices and polygons are stored in a hash table, which is used to reduce the time to consume and the space to store in high resolution voxelization. In a voxelization experiment with PSB data, proposed methods have shown a better performance than J.H.’s plane depth based method in 67% of models. As the resolution is high, process time is slowly increased.