• Journal of the Optical Society of Korea
• /
• 제11권3호
• /
• pp.108-112
• /
• 2007

Spectroscopic ellipsometry (SE) has become an important tool in scatterometry based nano-structure 3D profiling. In this paper, we propose a novel 3D nano object recognition method by use of phase sensitive scatterometry. We claims that only phase sensitive scatterometry can provide a reasonable 3D nano-object recognition capability since phase data gives much higher sensitive 3D information than amplitude data. To show the validity of this approach, first we generate various $0^{th}$ order SE spectrum data ($\psi$ and ${\Delta}$) which can be calculated through rigorous coupled-wave analysis (RCWA) algorithm and then we calculate correlation values between a reference spectrum and an object spectrum which is varied for several different object 3D shape.

• 한국멀티미디어학회논문지
• /
• 제23권10호
• /
• pp.1236-1249
• /
• 2020

In a most recent object tracking research work, applying Convolutional Neural Network and Recurrent Neural Network-based strategies become relevant for resolving the noticeable challenges in it, like, occlusion, motion, object, and camera viewpoint variations, changing several targets, lighting variations. In this paper, the LSTM Network-based Tracking association method has proposed where the technique capable of real-time multi-object tracking by creating one of the useful LSTM networks that associated with tracking, which supports the long term tracking along with solving challenges. The LSTM network is a different neural network defined in Keras as a sequence of layers, where the Sequential classes would be a container for these layers. This purposing network structure builds with the integration of tracking association on Keras neural-network library. The tracking process has been associated with the LSTM Network feature learning output and obtained outstanding real-time detection and tracking performance. In this work, the main focus was learning trackable objects locations, appearance, and motion details, then predicting the feature location of objects on boxes according to their initial position. The performance of the joint object tracking system has shown that the LSTM network is more powerful and capable of working on a real-time multi-object tracking process.

• 한국융합학회논문지
• /
• 제10권4호
• /
• pp.33-38
• /
• 2019

최근 다양한 LBS(location-based service) 서비스를 지원하기 위해 실제 공간 네트워크를 고려한 연구가 활발하게 진행 중이다. 이를 위해, 도로 네트워크에서 데이터 처리를 위한 실험 데이터가 다수 존재한다. 그러나 이러한 이동 객체의 궤적을 처리하기 위한 데이터는 이용하기에 적합하지 않다. 따라서 본 논문에서는 도로 네트워크 환경에서 궤적 데이터를 처리할 수 있는 색인 구조와 궤적 데이터 생성 알고리즘을 제안한다. 또한, 제안하는 구조와 알고리즘의 우수성을 입증하기 위해, 샌프란시스코 맵으로부터 만들어진 데이터를 이용하여 제안하는 알고리즘을 통해 에지 기반의 궤적 데이터를 생성됨을 보인다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2013년도 춘계학술발표대회
• /
• pp.45-48
• /
• 2013

In this paper, we proposed an Indexing Structure System (ISS) based on Apache Hadoop in Wireless Sensor Network (WSN). Nowadays sensors data continuously keep growing that need to control. Data constantly update in order to provide the newest information to users. While data keep growing, data retrieving and storing are face some challenges. So by using the ISS, we can maximize processing quality and minimize data retrieving time. In order to design ISS, Indexing Types have to be defined depend on each sensor type. After identifying, each sensor goes through the Indexing Structure Processing (ISP) in order to be indexed. After ISP, indexed data are streaming and storing in Hadoop Distributed File System (HDFS) across a number of separate machines. Indexed data are split and run by MapReduce tasks. Data are sorted and grouped depend on sensor data object categories. Thus, while users send the requests, all the queries will be filter from sensor data object and managing the task by MapReduce processing framework.

• 한국콘텐츠학회논문지
• /
• 제16권9호
• /
• pp.451-459
• /
• 2016

객체 검출은 다양한 컴퓨터 비전 응용 분야의 핵심 기술이지만 조명 변화와 기하학적 왜곡에 강인성을 갖기 위해서는 막대한 계산이 필요한 기술이다. 최근에 이 문제를 분류기의 토대로 체계화함으로써 효과적으로 해결하기 위한 접근법들이 소개되고 있다. 그 중 무작위 fern 알고리즘은 단순한 구조와 높은 인식 성능으로 많은 관심을 받고 있다. 그러나 기존의 무작위 fern 알고리즘은 화소간의 밝기 차이만으로 특징을 추출하고 있어 대조, 조명 변화와 같은 밝기 변화나 잡음에 대해 취약점을 갖는다. 본 논문에서는 기존의 무작위 fern의 단점을 개선하기 위해 패치의 기하학적 구조를 반영할 수 있는 깊이 정보를 결합시킨 결합형 무작위 fern을 새로이 제안하고 이를 이용한 객체 검출기의 성능 개선 방안을 제시한다. 모의실험을 통해 결합형 무작위 fern이 기존 방식보다 조명의 영향이나 잡음에 강인함을 보인다.

• 전기전자학회논문지
• /
• 제17권3호
• /
• pp.371-377
• /
• 2013

본 논문에서는 FAST와 BRIEF 알고리즘을 기반으로 하는 증강현실을 구현하기 위해서 효율적인 병렬 구조를 제안한다. 객체 인식 알고리즘으로 잘 알려진 SURF 알고리즘은 객체인식에 강인하지만 연산 량이 많아 실시간으로 구현하기에 어려운 단점을 가지고 있다. FAST와 BRIEF 알고리즘을 활용하여 객체를 인식하였고, 임베디드 환경에서 성능을 향상하기 위해 기존의 OpenMP 라이브러리를 사용한 병렬구조를 개선하여 속도를 약 70%에서 100%로 향상 시켰다.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제50권9호
• /
• pp.439-446
• /
• 2001

This paper considers the design technique of the real-time control algorithm to implement the electronic interlocking system which is the most important station control system in railway signal field. The proposed technique consists of the structure design and the detail design which are based on the ROOM(Real-Time Object-Oriented Modeling). The structure design is designed with a modeling using the heuristic search technique which, at first, catch and make out the specific requested condition, and then, is designed on the requested condition. The detail design can be implemented if it may get the satisfying values through the repetitive modeling after comparing and examining the data obtained from the structure design in order for the more reliable and accurate system to be implemented. The technique proposed in this paper is implemented with C++ language which is easy to be transferred and compatible with the existing interfaces, and also the operating system is designed and simulated on the VRTX which is a real-time operating system. This proposed technique is applied to the typical station model in order to prove the validity as verifying the performance of the modeled station.

• 한국정보처리학회논문지
• /
• 제6권5호
• /
• pp.1153-1165
• /
• 1999

In various fields of information technology, it is growing up the needs about dynamic content management systems to store and manage SGML(Standard Generalized Markup language) documents in a database system. In this paper, we consider the issue of storing SGML documents that having complex hierarchical structure into a database system, and then propose a data model based on ODMG(Object Database Management Group) object model in order to store SGML documents without loss of information. Because the proposed data model reflects physical element structure and logical entity structure of SGML documents, it is able to store the SGML document in a database system at the system at the element- level granularity without any information loss. And also the proposed data model can be adapted among ODMG-compliant object database management systems. Finally, we will discuss on the implementation details of SGML repository system supports the functionality of automatic database schema creation for any DTD(Document Type Definition0, the functionality of storing the SGML document, the functionality of dynamic document assembly from stored database objects to SGML document, and the functionality of indexing and searching for database objects.

• 한국안전학회지
• /
• 제7권2호
• /
• pp.14-23
• /
• 1992

A methodology for developing an object description system based on functional-structure of chemical plant is proposed. A knowledge base for chemical plant fault diagnosis is also organized in a generic fashion using the heuristic knowledge of human operators. A plant can be seen as a hierarchical set of subsystems. Each subsystem is called a SCOPE. The state of the plant and the behavior of each subsystem is managed by the SCOPES. A computer-based system based on thls methodology and knowledge base has been developed and applied to the subprocess of ethylene plant to evaluate the effectiveness of the methodology.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2008년도 추계학술발표대회
• /
• pp.878-881
• /
• 2008

To index the object's trajectory is an important aspect in moving object database management. This paper implements an optimizing index structure named Rend 3D R-tree based on 3D R-Tree. This paper demonstrates the time period update method to reconstruct the MBR for the moving objects in order to decrease the dead space that is produced in the closed time dimension of the 3D R-tree, then a rend method is introduced for indexing both current data and history data. The result of experiments illustrates that given methods outperforms 3D R-Tree and LUR tree in query processes.