• Journal of KIISE
• /
• v.43 no.4
• /
• pp.404-410
• /
• 2016

Automated road map generation poses significant research challenges since GPS-based navigation systems prevail in most general vehicles. This paper proposes an automated detecting method for intersection points using GPS vehicle trajectory data without any background digital map information. The proposed method exploits the fact that the trajectories are generally split into several branches at an intersection point. One problem in previous work on this intersection detecting is that those approaches require stopping points and direction changes for every testing vehicle. However our approach does not require such complex auxiliary information for intersection detecting. Our method is based on partial trajectory matching among trajectories since a set of incoming trajectories split other trajectory cluster branches at the intersection point. We tested our method on a real GPS data set with 1266 vehicles in Gangnam District, Seoul. Our experiment showed that the proposed method works well at some bigger intersection points in Gangnam. Our system scored 75% sensitivity and 78% specificity according to the test data. We believe that more GPS trajectory data would make our system more reliable and applicable in a practice.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2013.06a
• /
• pp.263-266
• /
• 2013

표는 중요한 정보를 함축적으로 담고 있는 문서 요소로서 문서 영상에서 표의 내용과 구조를 분석하고 이해하려는 연구가 많이 진행되어 왔다. 이러한 표의 검출과 인식에 관한 기존의 연구들은 평판 스캐너로 취득한 문서 영상을 대상으로 이루어졌는데 최근에는 디지털 카메라와 스마트폰이 보급됨에 따라 평판 스캐너 대신 카메라를 이용한 표 인식의 필요성이 대두되고 있다. 따라서 본 논문에서는 카메라로 획득한 문서 영상에서 표 인식에 대한 알고리듬을 제안한다. 먼저 표가 선들의 집합으로 이루어져 있다는 가정 아래 문서 이미지에 존재하는 선을 이진화와 강인한 곡선 맞춤 알고리듬을 사용하여 검출한다. 검출된 선들의 교차점은 표의 요소일 수도 있으며 오검출의 결과일 수도 있는데 교차점 주변의 관찰 결과와 교차점 사이의 연관 관계를 에너지 식으로 표현하고 이 식을 최소화함으로써 각각의 교차점에 최적의 레이블을 할당한다. 얻어진 레이블은 표로 유일하게 변환되며 표의 구조를 셀 단위까지 추정할 수 있다. 다양한 표 영상에 대한 실험 결과를 통하여 제안한 방법이 문서영상의 기하학적인 왜곡에도 불구하고 영상에 존재하는 표를 성공적으로 인식함을 보여준다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.5
• /
• pp.2430-2436
• /
• 2013

This study suggests a method of detecting the still object, which becomes a cause of danger within the crossroad. The Inverse Perspective Transform was performed in order to make the object size consistent by being inputted the real-time image from CCTV that is installed within the crossroad. It established the detection area in the image with the perspective transform and generated the adaptative background image with the use of the moving information on object. The detection of the stop object was detected the candidate region of the stop object by using the background-image differential method. To grasp the appearance of truth on the detected candidate region, a method is proposed that uses the gradient information on image and EHD(Edge Histogram Descriptor). To examine performance of the suggested algorithm, it experimented by storing the images in the commuting time and the daytime through DVR, which is installed on the cross street. As a result of experiment, it could efficiently detect the stop vehicle within the detection region inside the crossroad. The processing speed is shown in 13~18 frame per second according to the area of the detection region, thereby being judged to likely have no problem about the real-time processing.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.363-366
• /
• 2016

In this paper, we propose a method for detecting pedestrian, problem-solving situations that occur in a cross. When a pedestrian crossing and other, there occurs a problem of detecting the other pedestrians for detecting a specific pedestrian in the image. The proposed method for solving the problem is as follows. First, select a specific pedestrian detected by bounding box, and extracts the area as a template. Detecting a pedestrian from the image using the HOG, and designated as a candidate region. The final choice of the pedestrian detected by comparison with a candidate pedestrian with the specific pedestrian extracted for template. In comparison, using the Template matching, Histogram comparison and LBP.

• Journal of Advanced Navigation Technology
• /
• v.17 no.1
• /
• pp.90-97
• /
• 2013

Recently, intelligent transportation system based on image processing has been developed. In this paper, we propose a collision detection algorithm based on the analysis of motion vectors and the staying periods of vehicles in intersections. Objects in the region of interest are extracted from the subtraction image between background images based on Gaussian mixture model and input images. Collisions and traffic jams are detected by analysing measured motion vectors of vehicles and their staying periods in intersections. Experiments are performed on video sequences actually recoded at intersections. Correct detection rate and false alarm rate are 85.7% and 7.7%, respectively.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.11a
• /
• pp.269-272
• /
• 2009

본 논문에서는 변이지도의 정확도 향상을 위하여, 비용 함수를 이용한 교차 일치성 검사 기법을 제안하고, 다양한 조건의 실험을 통하여 제안한 알고리듬이 효율적임을 보였다. 좌우 변이정보를 이용하는 교차 일치성 검사로 오정합을 검출하는 방법을 시도해왔다. 하지만 이러한 방법은 물체의 경계에서 발생하는 오정합을 찾기가 어렵다. 본 논문에서는 최종 변이의 신뢰도 향상을 위해 교차 일치성 검사의 정확도를 높이는 방법을 제안하였다. 일반적으로 영역 기반 스테레오정합 방법은 물체의 경계에서 정확도 높지 못하다. 이러한 문제점을 해결하기 위해 정합창의 크기를 늘리거나 특징 점을 이용한 적응적 가변 정합창을 적용하는 방법을 시도하였다. 하지만, 여전히 기존 교차 일치성 검사를 통한 오정합 검출은 부정확하다. 이러한 영역의 비용 함수 값들을 비교한 결과 첫 번째와 두 번째 값의 차이가 적거나 크게 나타난다. 제안한 방법은 기존 방법에 비해 오정합 검출 능력을 향상 시킨다. 제안한 방법의 결과를 확인하기 위해 스테레오 비전에서 많이 사용되는 영상을 적용하고 분석하였다. 또한, 기존 교차 일치성 검사 방법과 제안한 방법의 객관적으로 비교하기 위해 전체 영역에 대한 오차율 (error ratio)과 교차 일치성 검사로 유효하다고 판단된 변이 값 중 실제 변이 값과 일치하지 않은 변이값의 오차율을 비교하였다. 실험 결과 기존 방법에 비해 제안한 방법이 1~5%정도 낮은 오차율을 보였다

• The Journal of the Korea Contents Association
• /
• v.12 no.2
• /
• pp.78-85
• /
• 2012

This paper proposes an efficient algorithm to develop a 3D terrain tools which is essential in the development of 3D computer games. In particular, this paper proposes a cross detection technique to improve picking speed. In other words, this paper proposes a more efficient cross detection technique consisting of a ray and a parallelogram than a cross detection technique consisting of a ray and a triangle. So, we can confirm the faster picking speed. This paper uses a picking example among DirectX SDK samples to test it. In addition, this paper compares the number of function calls for cross detection using the existing techniques and the proposed technique. As a result, in this paper the proposed technique has fallen off to about a 50 percent than the existing techniques. And if it is calculated by times, in this paper the proposed technique was reduced to 1 to 2 seconds than the existing techniques. Additionally, in this paper 3D terrain tools are provide more improved algorithms for features such as texture splatting, height map control, object arrangement and realistic water effect. So, 3D terrain tools is available efficient in the development of real 3d computer games.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.8
• /
• pp.1-8
• /
• 2020

Air ducts installed for ventilation inside buildings accumulate contaminants during their service life. Robots are installed to clean the air duct at low cost, but they are still not fully automated and depend on manpower. In this study, an intersection detection algorithm for autonomous driving was applied to an air duct cleaning robot. Autonomous driving of the robot was achieved by calculating the distance and angle between the extracted point and the center point through the intersection detection algorithm from the camera image mounted on the robot. The training data consisted of CAD images of the duct interior as well as the cross-point coordinates and angles between the two boundary lines. The deep learning-based CNN model was applied as a detection algorithm. For training, the cross-point coordinates were obtained from CAD images. The accuracy was determined based on the differences in the actual and predicted areas and distances. A cleaning robot prototype was designed, consisting of a frame, a Raspberry Pi computer, a control unit and a drive unit. The algorithm was validated by video imagery of the robot in operation. The algorithm can be applied to vehicles operating in similar environments.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.2C
• /
• pp.194-199
• /
• 2009

This paper proposes a face detection algorithm that uses geometrical information on face and hair region. This information that face adjoins hair regions can be the important one for face detection. It is also kept in images with frontal, rotated and lateral face. The face candidates are founded by the analysis of skin regions after detecting the skin and hair color regions in an image. Next, the intersected lesions between face candidates and hair's are created. Finally, the face candidates that include the subsets of these regions turn out to be face. Experimental results showed the high detection rates for frontal and lateral faces as well as faces geometrically distorted.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.103-107
• /
• 2020

의존구조 말뭉치는 자연언어처리 분야에서 문장의 의존관계를 파악하는데 널리 사용된다. 이러한 말뭉치는 일반적으로 오류가 없다고 가정하지만, 현실적으로는 다양한 오류를 포함하고 있다. 이러한 오류들은 성능 저하의 요인이 된다. 이러한 문제를 완화하려고 본 논문에서는 XGBoost와 교차검증을 이용하여 이미 구축된 구문분석 말뭉치로부터 오류를 탐지하는 방법을 제안한다. 그러나 오류가 부착된 학습말뭉치가 존재하지 않으므로, 일반적인 분류기로서 오류를 검출할 수 없다. 본 논문에서는 분류기의 결과를 분석하여 오류를 검출하는 방법을 제안한다. 성능을 분석하려고 표본집단과 모집단의 오류 분포의 차이를 분석하였고 표본집단과 모집단의 오류 분포의 차이가 거의 없는 것으로 보아 제안된 방법이 타당함을 알 수 있었다. 앞으로 의미역 부착 말뭉치에 적용할 계획이다.