• Journal of Korea Multimedia Society
• /
• v.9 no.12
• /
• pp.1529-1541
• /
• 2006

Moving objects have been widely employed in traffic and logistic applications. Spatio-temporal aggregations mainly describe the moving object's behavior in the spatial data warehouse. The previous works usually express the object moving in some certain region, but ignore the object often moving along as the trajectory. Other researches focus on aggregation and comparison of trajectories. They divide the spatial region into units which records how many times the trajectories passed in the unit time. It not only makes the storage space quite ineffective, but also can not maintain spatial data property. In this paper, a spatio-temporal aggregation index structure for moving object trajectory in constrained network is proposed. An extended B-tree node contains the information of timestamp and the aggregation values of trajectories with two directions. The network is divided into segments and then the spatial index structure is constructed. There are the leaf node and the non leaf node. The leaf node contains the aggregation values of moving object's trajectory and the pointer to the extended B-tree. And the non leaf node contains the MBR(Minimum Bounding Rectangle), MSAV(Max Segment Aggregation Value) and its segment ID. The proposed technique overcomes previous problems efficiently and makes it practicable finding moving object trajectory in the time interval. It improves the shortcoming of R-tree, and makes some improvement to the spatio-temporal data in query processing and storage.

• Journal of Korea Multimedia Society
• /
• v.13 no.5
• /
• pp.651-662
• /
• 2010

This paper describes a method to identify objects for autonomous navigation of an outdoor mobile robot. To identify objects, the robot recognizes the object from an image taken by moving robot on outdoor environment. As a beginning, this paper presents the candidates for a segment of region to building of artificial object, sky and trees of natural objects. Then we define their characteristics individually. In the process, we segment the regions of the objects included by preprocessing using multiple features. Multiple features are HSI, line segments, context information, hue co-occurrence matrix, principal components and vanishing point. An analysis of building identifies the geometrical properties of building facet such as wall region, windows and entrance. The building as intersection in vertical and horizontal line segment of vanishing point extracts the mesh. The wall region of building detect by merging the mesh of the neighbor parallelograms that have similar colors. The property estimates the number of story and rooms in the same floors by merging skewed parallelograms of the same color. We accomplish the result of image segmentation using multiple features and the geometrical properties analysis of object through experiments.

• Journal of Broadcast Engineering
• /
• v.26 no.2
• /
• pp.184-196
• /
• 2021

This paper proposes statistics adaptive linear regression-based object size prediction method for object detection. YOLOv2 and YOLOv3, which are typical deep learning-based object detection algorithms, designed the last layer of a network using statistics adaptive exponential regression model to predict the size of objects. However, an exponential regression model can propagate a high derivative of a loss function into all parameters in a network because of the property of an exponential function. We propose statistics adaptive linear regression layer to ease the gradient exploding problem of the exponential regression model. The proposed statistics adaptive linear regression model is used in the last layer of the network to predict the size of objects with statistics estimated from training dataset. We newly designed the network based on the YOLOv3tiny and it shows the higher performance compared to YOLOv3 tiny on the UFPR-ALPR dataset.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.626-626
• /
• 2015

본 연구에서는 하천공간정보의 웹서비스를 위해 SOAP(Simple Object Access Protocol) API 및 REST(Representation State Transfer) API로 제공하는 HydroG-OneFlow 웹서비스를 개발하였다. HydroG-OneFlow는 GML 기반의 서비스를 제공하며 GetBasin, GetGeoVariable 및 GetData 등의 기본서비스로 구성된다. GML은 GIS S/W의 벡터 GML 포맷과 공간정보 오픈플랫폼 서비스인 브이월드 데이터 API에서 제공하는 GML 포맷을 참고하여 하천공간 벡터정보를 제공할 수 있도록 GML을 구성하였다. GDM 공간 데이터에 대한 벡터정보 ML 수용 수준을 향상시킬 수 있도록 벡터구조의 점, 선, 면 정보에 대하여 GML의 PointPropertyType, CurvePropertyType, SurfacePropertyType을 도입하였다. 또한 일반적인 공간자료에서는 Multi 객체에 대한 지원도 필요하다. 현 GDM 데이터베이스에서도 OGC 표준의 MultiPoint, MultiLineString, MultiPolygon을 지원하고 있다. 이를 위하여 GML의 상응 요소인MultiPointPropertyType, MultiCurvePropertyType, MultiSurfacePropertyType을 하천공간정보 벡터 스키마에 도입하여 활용하였다. 클라이언트 서버 통신은 메시지 교환프로토콜인 SOAP을 사용하여 서버의 객체를 직접 호출하여 이루어진다. 서버는 서버의 제공 서비스를 WSDL(Web Service Description Language)를 통하여 게시하고 클라이언트는 이 기준(Criteria)을 참고하여 접근한다. GetData의 경우 Type(GRID or VECTOR), GDM(Geospatial Data Model) 여부(true or false), LayerName, BasinID, GenTime을 인자로 받아 GeoData에서 검색된 정보를 반환한다. SOAP버전은 1.1과 1.2를 지원하여 접근하는 클라이언트에서 선택할 수 있도록 개발하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.2713-2718
• /
• 2003

Some of manufacturing tasks such as sawing task often requires continuous impulsive motion. In case of sawing task, such impulsive motions can be observed between the teeth of the saw and the object. The amount of the external impulse exerted on the object has been treated as an important control parameter. The purpose of this work is to introduce a new concept of an effective mass in sawing task and to suggest an external impulse model in sawing task. A normalized impulse ellipsoid reflecting the velocity direction is employed to visualize the impact geometry. Experiments are performed for soft and hard workpieces to justify the external impulse model in the sawing task. It is demonstrated through simulation and experiment that the proposed external impulse model is effective to characterize the impact property.

• IE interfaces
• /
• v.17 no.2
• /
• pp.226-232
• /
• 2004

In shipbuilding area, data sharing is one of the crucial issues. Recently, for collaborative design, ship structural CAD systems adopt the database as its primary storage. Database is useful to deal with the large amount of design information among the heterogeneous design department and design stage. To make the database-based CAD system object-oriented database(OODB) and object-relational database(ORDB) can be used. It is important to select proper database because the CAD system performance mainly depends on access performance of database. In this research, using prototype CAD system from other research, access performance of OODB and ORDB form CAD system was evaluated. STEP application protocol was used as the database schema and experiment was made in query by property and query by region. The results give some idea of how to choose the database for CAD systems.

• Proceedings of the KOSOMBE Conference
• /
• v.1990 no.11
• /
• pp.5-10
• /
• 1990

The object detection capabilities of ultrasonic imaging systems are limited by the ability of the detection process to distinguish the resolved object signals from backscattered speckle noise. It has been shown that the phase component of the Fourier transform of the speckle noise is random. Based on this property. we propose a new algorithm for distinguishing between speckle and specular targets. The proposed algorithm is implemented by taking the Fourier transform of the received signal, low-pass filtering the phase, and taking the inverse Fourier transform of the filtered phase to enhance specular reflectors and reduce speckle in the image. Simulations and experiments using phantoms confirm the algorithm yielding significant reduction of speckle noise.

• Korean Journal of Child Studies
• /
• v.28 no.4
• /
• pp.245-264
• /
• 2007

This study analyzed physical knowledge displayed in the utterances of 10 Korean children 23 - 76 months of age. Utterances were blocked into four-month time periods, except the first two months and classified into 5 categories including properties of physical objects and materials, friction, buoyancy, gravity, and the motion of physical objects. All five types of physical expressions were found in the data of the earliest period the percent age of the properties and motion was higher than the percent of friction, buoyancy, and gravity. Data suggested two transitional points of change in physical knowledge : 2-year-old's descriptions were based only on perceptually salient physical features 3- to 4-year-olds described only end-state results while children older than 4 included systematic causality.

• Proceedings of the KOSOMBE Conference
• /
• v.1996 no.11
• /
• pp.99-102
• /
• 1996

The ultimate object of FES is on the recovering function of body and shape demaged from desease or injury to original state. On this study, object is recovering of gait function of the disabled who, especially, have gait disturbance. Paralyzed muscle from the central nerve disable, if peripheral nerves which be in the lower part of the harmed are activated, muscle contraction is possible. The traumatic trouble, peripheral nerves aren't connected to a central nerve but origin of peripheral nerve cells which are in the lower part of the harmed are alive, react on stimulation. We design 4-channel stimulator, being based on standard stimuli pattern. stimulator is manufactured with compact size and light weight to portable.

• Journal of Korea Multimedia Society
• /
• v.25 no.10
• /
• pp.1433-1447
• /
• 2022

Recently, as the demand for physical security technology to prevent leakage of technical and business information of companies and public institutions increases, the high tech companies are operating X-ray security checkpoints at building entrances to protect their intellectual property and technology. X-ray security checkpoints are operated to detect cameras and storage media that may store or leak important technologies in the bags of people entering and leaving the building. In this study, we propose an X-ray security checkpoint system that automatically detects a storage medium in an X-ray image using a deep learning based object detection method. The proposed system consists of an edge computing unit and a cloud-computing unit. We employ the RetinaNet for automatic storage media detection in the X-ray security checkpoint images. The proposed approach achieved mAP of 95.92% on private dataset.