• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.947-948
• /
• 2008

Recently, many efforts have been made for research and application of object tracking system. However, introduced object tracking algorithms have limitations to adopt a realtime object tracking system with multi camera. In this paper, we present a novel background generation and target object recognition algorithm for realtime object tracking system with multi camera and implemented it.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2007년도 추계종합학술대회
• /
• pp.119-122
• /
• 2007

본 논문에서는 모바일 폰 카메라의 프레임 단위 영상 신호 처리 과정에서 필요한 나눗셈 연산을 위한 나눗셈기 설계 방법을 제안한다. 나눗셈기의 내부 데이터 처리 방법에는 직렬 방식과 병렬 방식이 있다. 직렬방식은 실시간 연산이 가능한 반면에 많은 비교기와 Buffer Memory의 사용으로 인해 하드웨어 사이즈가 크다. 병렬방식은 실시간 연산을 할 수 없지만 하나의 비교기를 공유해서 연산함으로 직렬방식에 비해 하드웨어 크기를 줄일 수 있다. 이미지 처리를 위한 프레임 단위 연산은 실시간 연산을 필요로 하지 않으므로 하드웨어 자원으 효율성을 위해 직렬방식 나눗셈기를 적용한다. 입출력 조건을 동일하게 해서 병렬방식과 직렬방식의 나눗셈을 구현하여 하드웨어 크기를 비교 했을 때 동일한 동작 주파수에서 직렬방식의 나눗셈기가 병렬방식의 나눗셈기의 대락 1/8 정도의 하드웨어 크기를 가지는 것을 확인하였다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1996년도 춘계학술대회 논문집
• /
• pp.254-258
• /
• 1996

In this paper, we proposed realtime visual tracking system of moving object for 2D target using extended Kalman Filter Algorithm. A targeting marker are recongnized in each image frame and positions of targer object in each frame from a CCD camera while te targeting marker is attached to the tip of the SCARA robot hand. After the detection of a target coming into any position of the field-of-view, the target is tracked and always made to be located at the center of target window. Then, we can track the moving object which moved in inter-frames. The experimental results show the effectiveness of the Kalman filter algorithm for realtime tracking and estimated state value of filter, predicting the position of moving object to minimize an image processing area, and by reducing the effect by quantization noise of image

• 제어로봇시스템학회논문지
• /
• 제9권6호
• /
• pp.448-455
• /
• 2003

A system design technique of automatic thickness measurement and defect inspection system, which measures the thickness of the ALC(Autoclaved Lightweight Concrete) block and inspects the defect on a realtime basis is proposed. The image processing system was established with a CCD camera, an image grabber, and a personal computer without using assembled measurement equipment. The image obtained by this system was analyzed by a devised algorithm, specially designed for the enhanced measurement accuracy. For the realization of the proposed algorithm, the preprocessing method that can be applied to overcome uneven lighting environment, an enhanced edge decision method using 8 edge-pairs with irregular and rough surface, the unit length decision method in uneven condition with rocking objects, and the curvature calibration method of camera using a constructed grid are developed. The experimental results, show that the required measurement accuracy specification is sufficiently satisfied using our proposed method.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 1999년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.80-83
• /
• 1999

An efficient modeling and management of a large amount of surface data for a wide rage of geographic information play an important role in determining the functionality of 3D geographic information system. It has been put many efforts to design and manage an effective way to enhence the manipulation of the data by considering geometry type and data structures. Recently, DEM(Data Elevation Model) and TIN(Triangulated Irregular Network) are used for representing surface data. In this paper, we propose a 3D data processing method. The method utilizes the major properties of DEM and TIN, respectively. Furthermore, by approximating DEM with a TIN of an appropriate resolution, we can support a fast and realistic surface modeling. We implement the structure with the following 4 level stages. The first is an optimal resolution of DEM which represent all of wide range of geographic data. The second is the full resolution DEM which is a subarea of original data generated by user's selection in our implemeatation. The third is the TIN approximation of this data with a proper resolution determined by the relative position with the camera. And the last step is multi-resolution TIN data whose resolution is dynamically decided by considering which direction user take notice currently. Specially, the TIN of the last step is designed for realtime camera navigation. By using the structure we implemented realtime surface clipping, efficient approximation of height field and the locally detailed surface LOD(Level of Detail). We used the initial 10-meter sampling DEM data of Seoul, KOREA and implement the structure to the 3D Virtual GIS based on the Internet.

• 한국GIS학회:학술대회논문집
• /
• 한국GIS학회 1999년도 추계학술대회 발표요약문
• /
• pp.66-71
• /
• 1999

For given 3D geographic data which is usually of DEM(Data Elevation Model) format, we have to represent and manipulate the data in various ways. For example, we have to draw a part of them in drawing canvas. To do this we give users a way of selecting area they want to visualize. And we have to give a base tool for users to select the local area which can be chosen for some geographic operation. In this paper, we propose a 3D data processing method for representation and manipulation. The method utilizes the major properties of DEM and TIN(Triangular Irregular Network), respectively. Furthermore, by approximating DEM with a TIN of an appropriate resolution, we can support a fast and realistic surface modeling. We implement the structure with the following 4 level stages. The first is an optimal resolution of DEM which represent all of wide range of geographic data. The second is the full resolution DEM which is a subarea of original data generated by user's selection in our implemeatation. The third is the TIN approximation of this data with a proper resolution determined by the relative position with the camera. And the last step is multi-resolution TIN data whose resolution is dynamically decided by considering which direction user take notice currently. Specialty, the TIN of the last step is designed for realtime camera navigation. By using the structure we implemented realtime surface clipping, efficient approximation of height field and the locally detailed surface LOD(Level of Detail). We used the initial 10-meter sampling DEM data of Seoul, KOREA and implement the structure to the 3D Virtual GIS based on the Internet.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.95.2-95
• /
• 2002

1. Introduction of Camera Detection system Camera Detection system is an equipment that can detect realtime traffic information by image processing techniques. This information can be used to analyze and control road traffic flow. It is also used as a method to detect and control traffic flow for ITS(Intelligent Transportation System). Traffic information includes speed, head way, traffic flow, occupation time and length of queue. There are many detection systems for traffic data. But video detection system can detect multiple lanes with only one camera and collect various traffic information. So it is thought to be the most efficient method of all detection system. Though the...

• 한국멀티미디어학회논문지
• /
• 제20권4호
• /
• pp.606-613
• /
• 2017

When the planar area is captured by the depth camera, the shape of the plane in the captured image has perspective projection distortion according to the position of the camera. We can correct the distorted image by the depth information in the plane in the captured area. Previous depth information based perspective distortion correction methods fail to satisfy the real-time property due to a large amount of computation. In this paper, we propose the method of applying the conversion table selectively by measuring the motion of the plane and performing the correction process by parallel processing for correcting perspective projection distortion. By appling the proposed method, the system for correcting perspective projection distortion correct the distorted image, whose resolution is 640x480, as 22.52ms per frame, so the proposed system satisfies the real-time property.

• 방송공학회논문지
• /
• 제21권5호
• /
• pp.782-791
• /
• 2016

영상에서 배경으로부터 객체를 추출하는 영상 segmentation 알고리즘은 물체 인식 및 추적 등 다양한 응용분야에서 활용될 수 있다. 본 논문에서는 고정된 카메라에서 다수의 초기 프레임을 참조하여 실시간 객체 segmentation 방법을 제안한다. 먼저 객체와 배경을 분류하는 확률모델을 제안하였으며 초기 프레임 동안에 카메라의 color consistency와 focus 특성을 분석하여 안정적인 segmentation 성능을 증가시켰다. 또한 분류된 객체에서 human의 skeleton 특성을 이용하여 추출 결과를 보정하는 방법을 제안한다. 마지막으로 제안된 알고리즘은 객체 segmentation 실시간 처리를 위하여 복잡도를 최소화하므로 다양한 mobile 단말에 확대 적용 가능하다.

• 한국인터넷방송통신학회논문지
• /
• 제8권3호
• /
• pp.35-45
• /
• 2008

카메라를 이용한 실시간 시정측정은 인간의 시정 감각과 유사하고 현실성 있는 시정자료를 획득할 수 있으며 기존 고가의 광학기기를 이용한 측정방법을 대체할 수 있는 새로운 측정방식이다. 영상으로부터 깊이정보나 3차원 구조의 추출 등을 통해 시정을 측정하려는 시도가 있으나, 단순하고 빠른 처리가 요구되는 실시간 시정측정과 이동물체가 많이 나타나는 경우에는 많은 문제점을 갖고 있다. 본 논문에서는 시정의 감소는 영상에서 지수적인 대비의 감소로 나타난다는 점에 착안하여 영상으로부터 이동영역 필터를 이용하여 대비를 추출하고 영상대비와 시정간의 상관관계를 수학적으로 모델링함으로써 쉽고 빠르게 시정을 측정할 수 있는 방법을 제안한다. 이동영역 필터는 영상으로부터 시정측정에 영향을 주는 하늘과 물체의 이동영역을 효과적으로 제거하기 위해 사용된다. 제안된 방법은 카메라를 통해 입력된 영상으로부터 실시간 시정측정이 가능할 뿐만 아니라 도로와 같이 차량의 이동이 많은 경우에도 안정적인 시정측정이 가능하다.