• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권3호
• /
• pp.835-854
• /
• 2012

In this paper, we propose a novel positioning system for indoor navigation which helps a user navigate easily to desired destinations in an unfamiliar indoor environment using his mobile phone. The system requires only the user's mobile phone with its basic equipped sensors such as a camera and a compass. The system tracks user's positions and orientations using a vision-based approach that utilizes $360^{\circ}$ panoramic images captured in the environment. To improve the robustness of the vision-based method, we exploit a digital compass that is widely installed on modern mobile phones. This hybrid solution outperforms existing mobile phone positioning methods by reducing the error of position estimation to around 0.7 meters. In addition, to enable the proposed system working independently on mobile phone without the requirement of additional hardware or external infrastructure, we employ a modified version of a fast and robust feature matching scheme using Histogrammed Intensity Patch. The experiments show that the proposed positioning system achieves good performance while running on a mobile phone with a responding time of around 1 second.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제28권10호
• /
• pp.764-772
• /
• 2001

본 논문은 지문의 특징점들이 융선으로 서로 연결되어진 구조 정보를 활용한 지문 매칭 알고리즘에 대해 연구하였다. 매칭에 이용한 특징은 분기점과 단점의 위치 및 분기점과 단점에서 융선의 진행 방향 등이다. 동일 융선 상에 존재하는 특징점들에 대한 연결정보를 추출하고, 이를 이용하여 원문지문과 입력지문간의 좌표 변환의 기준이 되는 특징점 쌍을 검출한다. 서로 일치하는 한 쌍의 특징을 이용해 입력지문을 이동 회전하여 원문지문과 일치시킨 후 각 특징들의 위치, 융선 방향이 일치하는 개수에 따라 지문의 동일여부를 판단하였다. 제안된 알고리즘은 회전과 이동에 무관한 지문인식이 가능하다. 89명으로부터 얻은 445매의 지문영상에 대한 실험결과, 지문매칭의 처리속도가 31%향상되었다.

• 한국자동차공학회논문집
• /
• 제16권6호
• /
• pp.134-140
• /
• 2008

Preceding vehicle detection is a crucial issue for driver assistance system as well as for autonomous vehicle guidance function and it has to be performed with high reliability to avoid any potential collision. The vision-based preceded vehicle detection systems are regarded promising for this purpose because they require little infrastructure on a highway. However, the feasibility of these systems in passenger car requires accurate and robust sensing performance. In this paper, an preceded vehicle detection system is developed using stereo vision sensors. This system utilizes feature matching, epipoplar constraint and feature aggregation in order to robustly detect the initial corresponding pairs. After the initial detection, the system executes the tracking algorithm for the preceded vehicles including a leading vehicle. Then, the position parameters of the preceded vehicles or leading vehicles can be obtained. The proposed preceded vehicle detection system is implemented on a passenger car and its performances is verified experimentally.

• Journal of Electrical Engineering and Technology
• /
• 제13권2호
• /
• pp.800-811
• /
• 2018

This paper presents a regenerative braking control scheme for Switched Reluctance Machine (SRM) drive in Electric Vehicles (EVs). The main purpose is to maximize the recovered energy during battery charging by taking into account the nonlinear physical characteristics of the Switched Reluctance Machine. The proposed regenerative braking method employs the back-EMF in the generation process as a complicated position-dependent voltage source. The proposed maximum power recovery (MPR) operation of the regenerative braking is first based on the maximization of the extracted power from the machine and then the maximization of the power transferred to the battery. The maximum power extraction (MPE) from SRM is based on maximizing the energy conversion ratio by the calculation of the optimum PWM switching duty cycle, turn-on, and turn-off angles. By using the impedance matching theorem that allows the maximum power transfer (MPT) of the MPE, the proposed MPR is achieved. The parametric averaged value modeling of the machine phase currents in the chopping control mode is used for MPR realization. By following this model, a nonlinear equivalent input resistance is derived for the battery internal resistance matching. The effectiveness of the proposed regenerative braking method is demonstrated through simulation results and experimental implementation.

• 제어로봇시스템학회논문지
• /
• 제12권9호
• /
• pp.876-882
• /
• 2006

This study presents an ultrasonic location awareness system for the ubiquitous computing with absolute position. The flight time of ultrasonic waves is determined by a period detecting technique which is able to extend the sensing range compared with traditional methods. For location awareness, ultrasonic waves are sent successively from each ultrasonic transmitter and synchronized by radio frequency (RF) signal, where the transmitting part is fixed and the receiving part is movable. To expand the recognizing range, cell matching technique and coded ultrasonic technique are introduced. The experimentation for various distances is accomplished to verify the used period detecting technique of U-SAT system. The positioning accuracy by using cell matching is also verified by finding the locations of settled points and the usability of coded ultrasonic technique is verified. As a result, the possibility of ultrasonic location awareness system for the ubiquitous computing can be discussed as a pseudo-satellite system with low cost, a high update rate, and relatively high precision, in the places where CPS is not available.

• 대한전자공학회논문지
• /
• 제26권4호
• /
• pp.120-126
• /
• 1989

최근 집적도의 증가와 더불어 집적회로 제작과 점검을 자동화하기 위하여 영상처리방법을 이용한 집적회로의 인식 알고리듬이 개발되고 있다. 집적회로를 인식하기 위해서는 프레임 영상으로 부터 특징을 얻고 이것으로써 칩내의 모든 프레임 영상을 정합하여야한다. 본 논문은 layout 정보를 나타내는 꼭지점들의 위치와 형태를 직선화 알고리듬을 이용하여 추출한다. 부분적으로 겹치는 이웃 프레임들을 얻어진 꼭지점의 위치와 형태를 특징으로 하여 정합함으로써 꼭지점의 전체적 좌표와 형태를 추출한다.

• 스마트미디어저널
• /
• 제11권8호
• /
• pp.84-92
• /
• 2022

Precision livestock monitoring promises greater management efficiency for farmers and higher welfare standards for animals. Recent studies on video-based animal activity recognition and tracking have shown promising solutions for understanding animal behavior. To achieve that, surveillance cameras are installed diagonally above the barn in a typical cattle farm setup to monitor animals constantly. Under these circumstances, tracking individuals requires addressing challenges such as occlusion and visual appearance, which are the main reasons for track breakage and increased misidentification of animals. This paper presents a framework for multi-cattle tracking in closed barns with appearance and motion models. To overcome the above challenges, we modify the DeepSORT algorithm to achieve higher tracking accuracy by three contributions. First, we reduce the weight of appearance information. Second, we use an Ensemble Kalman Filter to predict the random motion information of cattle. Third, we propose a supplementary matching algorithm that compares the absolute cattle position in the barn to reassign lost tracks. The main idea of the matching algorithm assumes that the number of cattle is fixed in the barn, so the edge of the barn is where new trajectories are most likely to emerge. Experimental results are performed on our dataset collected on two cattle farms. Our algorithm achieves 70.37%, 77.39%, and 81.74% performance on HOTA, AssA, and IDF1, representing an improvement of 1.53%, 4.17%, and 0.96%, respectively, compared to the original method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권4호
• /
• pp.1066-1079
• /
• 2023

Multi-target tracking based on the detector is a very hot and important research topic in target tracking. It mainly includes two closely related processes, namely target detection and target tracking. Where target detection is responsible for detecting the exact position of the target, while target tracking monitors the temporal and spatial changes of the target. With the improvement of the detector, the tracking performance has reached a new level. The problem that always exists in the research of target tracking is the problem that occurs again after the target is occluded during tracking. Based on this question, this paper proposes a DeepSORT model based on SIFT features to improve ship tracking. Unlike previous feature extraction networks, SIFT algorithm does not require the characteristics of pre-training learning objectives and can be used in ship tracking quickly. At the same time, we improve and test the matching method of our model to find a balance between tracking accuracy and tracking speed. Experiments show that the model can get more ideal results.

• 한국통신학회논문지
• /
• 제25권1B호
• /
• pp.128-140
• /
• 2000

본 논문에서는 이웃 탐색점에서의 평균 절대치 오차 (mean absoulte error, MAE) 및 탐색여역 줄임을 이용한 고속 블록 정합 알고리듬 제안하였다. 이 알고리듬은 두 단계로 구성되어있다. 첫 번째 단계에서는 탐색영역을 3$\times$3 크기의 영역으로 겹치지 않게 나눈 뒤, 각 영역의 중심 탐색점에 대하여 블록 정합을 행하여 MAE를 구하고, 이들 중 가장 작은 MAE를 기준 MAE로 정한다. 그리고, 두 번째 단계에서는 각 영역의 중심 탐색점에서의 MAE를 이용하여 각 3$\times$3 영역의 나머지 탐색점에서의 MAE의 최소 범위를 구한 뒤, 최소 범위가 기준 MAE로 결정된 탐색점 근처에 존재할 가능성이 매우 큼을 이용하여 기준 MAE로 결정된 탐색점을 중심으로 탐색영역의 크기를 줄인 뒤, 블록 정합이 필요한 탐색점에 대하여서만 블록 정합을 행함으로써 고속으로 움직임을 추정하였다. 모의 실험을 통하여 본 제안한 방법이 우수한 움직임 추정 성능을 유지하면서도 많은 계산량의 감소를 얻을 수 있음을 확인하였다.

• 한국측량학회지
• /
• 제24권4호
• /
• pp.367-376
• /
• 2006

본 연구는 IKONOS 위성영상의 메타자료와 지상기준점으로부터 라인변화에 따른 외부표정요소의 최적 다항식 모델을 결정하였다. 외부표정요소의 위치요소는 일차다항식, 회전요소는 상수항을 최적 다항식으로 결정하였다. 그리고 결정된 외부표정요소의 위치결정 정확도는 기존의 검증된 방법 즉, 최소제곱법에 의한 RPC 편의보정 방법과 비교하였다, 그 결과, 두 방법에 의한 정확도는 거의 차이가 없었다. 결정된 외부표정요소로부터 정합선을 수립한 후, 수치표고모델을 작성하여, RPC 편의보정 방법이 적용된 ERDAS IMAGINE에 의한 DEM과 비교 검토하였다. 그 결과로써, 제안방법과 ERDAS IMAGINE으로 작성한 수치표고모델의 높이차이는 허용 표준편차 범위 내에 존재하였다. 따라서 본 연구에서 제작한 수치표고모델은 기존의 검증된 방법과 유사한 결과 정확도를 나타냈다.