• 한국멀티미디어학회논문지
• /
• 제19권6호
• /
• pp.1003-1013
• /
• 2016

This paper proposes a framework of superpixel-based vehicle detection method using plane normal vector in disparity space. We utilize two common factors for detecting vehicles: Hypothesis Generation (HG) and Hypothesis Verification (HV). At the stage of HG, we set the regions of interest (ROI) by estimating the lane, and track them to reduce computational cost of the overall processes. The image is then divided into compact superpixels, each of which is viewed as a plane composed of the normal vector in disparity space. After that, the representative normal vector is computed at a superpixel-level, which alleviates the well-known problems of conventional color-based and depth-based approaches. Based on the assumption that the central-bottom of the input image is always on the navigable region, the road and obstacle candidates are simultaneously extracted by the plane normal vectors obtained from K-means algorithm. At the stage of HV, the separated obstacle candidates are verified by employing HOG and SVM as for a feature and classifying function, respectively. To achieve this, we trained SVM classifier by HOG features of KITTI training dataset. The experimental results demonstrate that the proposed vehicle detection system outperforms the conventional HOG-based methods qualitatively and quantitatively.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권6호
• /
• pp.2735-2751
• /
• 2020

The emergence of Software Defined Networking (SDN) overcomes the limitations of traditional networking architectures. There are some advantages in SDN which are centralized global network view, programmability, and separation of the data plane and control plane. Due to the limitation of data plane storage capacity in SDN, it is necessary to process the redundancy rules of switch. In this paper, we propose a method for active detection and processing of redundant rules. We use the result generated by the customized probe package to detect redundant rules. And by checking the forwarding behavior of probe packets in the data plane, the redundancy rules are further processed. Furthermore, in order to quickly check the dynamic networks, we propose an incremental algorithms for rapidly evolve the network strategies. We conduct simulation experiments on Matlab to verify the feasibility of the algorithm. The influence of some parameters on the result are discussed.

• 제어로봇시스템학회논문지
• /
• 제14권3호
• /
• pp.236-241
• /
• 2008

This paper presents a method to recognize the plane regions for movement of walking robots. When the autonomous agencies using stereo camera or laser scanning sensor is under unknown 3D environment, the mobile agency has to detect the plane regions to decide the moving direction and perform the given tasks. In this paper, we propose a very fast method for plane detection using normal vector of a triangle by 3 vertices defined on a small circular region. To reduce the effect of noises and outliers, the triangle rotates with respect to the center position of the circular region and generates a series of triangles with different normal vectors based on different three points on the boundary of the circular region. The vectors for several triangles are normalized and then median direction of the normal vectors is used to test the planarity of the circular region. The method is very fast and we prove the performance of algorithm for real range data obtained from a stereo camera system.

• Structural Engineering and Mechanics
• /
• 제81권2호
• /
• pp.195-203
• /
• 2022

This paper presents the quantification and location damage detection of plane and space truss structures in a two-phase method to reduce the computations efforts significantly. In the first phase, a proposed damage indicator based on the residual force vector concept is used to get the suspected damaged members. In the second phase, using damage quantification as a variable, a teaching-learning based optimization algorithm (TLBO) is used to obtain the damage quantification value of the suspected members obtained in the first phase. TLBO is a relatively modern algorithm that has proved distinguished in solving optimization problems. For more verification of TLBO effeciency, the classical particle swarm optimization (PSO) is used in the second phase to make a comparison between TLBO and PSO algorithms. As it is clear, the first phase reduces the search space in the second phase, leading to considerable reduction in computations efforts. The method is applied on three examples, including plane and space trusses. Results have proved the capability of the proposed method to precisely detect the quantification and location of damage easily with low computational efforts, and the efficiency of TLBO in comparison to the classical PSO.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제4권10호
• /
• pp.465-472
• /
• 2015

본 논문에서는 시차공간상의 평면검출 방법을 제안하고 그 성능을 평가한다. 다양한 표면을 평면으로 근사하고 검출함으로써 시차공간에 나타난 장면을 간소화하고 수식화하여 다루기 쉽도록 한다. 또한 시차공간에서 근사적으로 구한 평면은 3차원 공간상에서 실측 크기로 표현 가능하고 장애물 검출 및 카메라 위치 추정에 활용할 수 있다. 먼저 스테레오 매칭 기술을 이용해 두 개의 영상으로부터 2차원 공간상에 좌표쌍마다 시차값을 가지는 시차공간을 생성한다. x 또는 y축의 전체적인 추이를 반영하도록 돕는 선 단순화 기법을 이용하여 시차값의 접선 기울기를 추정한다. 기울기 쌍의 조합에 따라 10개의 라벨을 시차공간의 좌표쌍에 부여한다. 상하좌우 방향으로 인접하고 동일한 라벨을 가지는 좌표쌍을 연결하여 군집을 생성하고 최소자승법을 이용해 각 군집에 대한 평면식을 추정한다. 시차공간 내에서 평면식을 만족하는 점들이 가장 많은 평면을 검출하고 이를 시차공간을 가장 잘 간소화한 N개의 평면으로 선택한다. 평면검출의 성능을 정량적으로 평가하였고 그 결과는 3차원 원뿔과 원통에서 각각 97.9%, 86.6% 품질을 보였다. 스테레오 비전 알고리즘의 성능을 평가하기 위해 대표적으로 이용되는 Middlebury와 KITTI 실험데이터로부터 제안된 평면검출 방법은 훌륭하게 평면을 검출하였다.

• 제어로봇시스템학회논문지
• /
• 제21권2호
• /
• pp.130-136
• /
• 2015

In this paper, we propose a robust curved lane marking detection method. Several lane detection methods have been proposed, however most of them have considered only straight lanes. Compared to the number of straight lane detection researches, less number of curved-lane detection researches has been investigated. This paper proposes a new curved lane detection and tracking method which is robust to various illumination conditions. First, the proposed methods detect straight lanes using a robust road feature image. Using the geometric relation between a vehicle camera and the road plane, several circle models are generated, which are later projected as curved lane models on the camera images. On the top of the detected straight lanes, the curved lane models are superimposed to match with the road feature image. Then, each curve model is voted based on the distribution of road features. Finally, the curve model with highest votes is selected as the true curve model. The performance and efficiency of the proposed algorithm are shown in experimental results.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 제13회 신호처리 합동 학술대회 논문집
• /
• pp.675-678
• /
• 2000

This paper presents a parametric line equation on the log-polar mapped plane to detect the straight lines in an original image. The log-polar edge image used in Hough transform is constructed by combining the edge images of both fovea and periphery. The foveal edge image detected by a Sobel mask on the Cartesian plane is transformed to the log-polar plane by forward mapping but the edge detection of the peripheral region is obtained by directly applying the newly developed mask to the log-polar plane. This paper also proposes a analytic method then determining a border between the fovea and the periphery regions.

• 전기학회논문지
• /
• 제67권12호
• /
• pp.1648-1656
• /
• 2018

The configuration of redundant inertial sensors are very important when considering navigation performance and fault detection and isolation (FDI) performance. By constructing a redundant sensor system using multiple inertial sensors, it is possible to improve the navigation performance and fault detection and isolation performance, which are highly related to the sensor configuration and allocation. In order to deploy multiple MEMS inertial measurement units effectively, a configuration and allocation methods considering navigation performance, fault detection and isolation performance, and lever arm effect in one plane are presented, and the performance is analyzed through simulation in this research. From the results, it is confirmed that the proposed configuration and allocation method can improve navigation, FDI, and lever arm effect rejection performances more effectively by more than 70%.

• 전자공학회논문지SC
• /
• 제46권5호
• /
• pp.29-37
• /
• 2009

본 논문은 움직이는 차 안에서 교통 표지판을 실시간으로 검출하는 영상 기반 시스템을 기술한다. 제안된 시스템은 표지판 검출을 실시간으로 수행하기 위하여 아다부스트 알고리즘을 기본 구조로 가지고 있고, 대부분의 교통 표지판이 지표면으로부터 수직으로 세워져 있는 점에 착안하여 수직면 값을 후보군 검출 과정에서 사용하였다. 기존의 아다부스트 알고리즘은 실시간 검출을 위하여 유용하지만, 특징으로써 누적 영상들만을 사용하므로 복잡한 도로 환경에서는 검출율이 저하된다. 본 논문에서는 이러한 단점을 해결하기 위하여 후보군 검출을 위한 특징으로써 수직면 값을 이용함으로써 검출 후보군의 신뢰도를 높이고, 또한 특징 원형의 종류가 추가된 누적 영상들을 사용함으로써 표지판 검출 성능을 향상시킨다. 실험 결과에서는 본 논문에서 제안된 방법이 실제 도로 환경에서 기존의 아다부스트 방법보다 검출율이 향상되었음을 보인다.

• 디지털융복합연구
• /
• 제14권11호
• /
• pp.319-324
• /
• 2016

실내에서 카메라를 이용한 로봇 응용이나 가상현실(Virtual Reality) 응용의 경우 평면을 찾고 추정하는 기술은 매우 중요한 기술이다. RGB-D 카메라의 경우 실내의 평면에서 질감 정보가 없는 평면에서도 3차원 관측 데이터를 얻을 수 있지만, 이미지 영역에서 점군 데이터(Point-cloud Data)를 처리하기 위해서는 많은 연산량이 필요하다. 더군다나 현재 관측되고 있는 평면의 개수가 몇 개인지 미리 알 수 없으며, 평면으로 검출(Plane Detection) 하더라도 강인하게 3차원에서 평면을 추정(Plane Estimation)하려면 추가적인 연산이 필요하다. 본 논문에서는 연속 데이터를 이용해 실시간으로 평면의 개수를 선택하며 평면을 추정하는 방법을 제시하고자 한다. 실험 결과를 통해 제안하는 방법이 전체 데이터를 처리하는 것에 비해 약 22배의 속도 개선을 가져 올 수 있음을 보였다.