• 한국정보통신학회논문지
• /
• 제19권2호
• /
• pp.435-442
• /
• 2015

본 논문에서는 지능형 다중 화상감시시스템에 응용할 수 있는 움직이는 물체 추적 및 보행자/차량 인식 방법을 제안한다. 지능형 다중 화상감시시스템은 다수의 고정형 카메라와 한 대의 PTZ 카메라로 구성되며, 고정형 카메라에서 검출된 움직이는 물체들을 PTZ 카메라로 팬/틸트/줌 제어하고, 보행자인지 또는 차량인지를 자동으로 인식한다. 넓은 영역을 감시하는 고정된 카메라에서 검출된 물체는 너무 작고, 변별력이 떨어지는 문제가 있다. 이러한 문제를 극복하기 위해 PTZ 카메라를 통한 특정 움직이는 물체를 팬/틸트/줌인 제어함으로써 움직이는 물체의 변별력과 감시성능을 높일 수 있다. 제안된 시스템은 움직이는 물체를 추적하는 기능 외에 SVM 학습알고리즘을 이용하여 검출된 물체가 보행자 또는 차량인지를 판단할 수도 있다. 그리고 추적에러를 줄이기 위해 기존의 고정된 카메라와 PTZ 카메라간의 캘리브레이션 방법을 개선한다. 다양한 실험결과를 통하여 제안한 시스템의 효용성을 입증하였다.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제23권3호
• /
• pp.148-155
• /
• 2017

본 논문에서는 스마트폰 내 가속도, 자기장, 자이로스코프 센서들을 이용해 사용자의 걸음과 걸음 수를 인식하는 시스템을 개발하였다. 센서 데이터 분석을 통해 사용자의 걸음을 스마트폰을 손에 든 상황과 주머니에 넣은 상황에서의 걸음 패턴으로 분류하고 이를 추출할 수 있는 알고리즘을 사용하여 걸음 수 인식의 정확성을 개선하였다. 알고리즘을 적용한 결과 손에든 상황에서 96%, 주머니에 넣은 상황에서 95.5% 수준의 걸음 수 인식 정확도를 보였으며, 나머지 터치 스크린, 위아래 반복 흔들기, 앉아서 일어서기, 오른쪽 왼쪽 흔들기와 같은 행위로 인해 발생하는 6%의 오차를 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권8호
• /
• pp.3657-3682
• /
• 2018

This paper presents a Bluetooth and pedestrian dead reckoning (PDR)-based indoor fusion localization approach (BtPDR) using smartphones. A Bluetooth and PDR-based indoor fusion localization approach can localize the initial position of a smartphone with the received signal strength (RSS) of Bluetooth. While a smartphone is moving, BtPDR can track its position by fusing the localization results of PDR and Bluetooth RSS. In addition, BtPDR can adaptively modify the parameters of PDR. The contributions of BtPDR include: a Bluetooth RSS-based Probabilistic Voting (BRPV) localization mechanism, a probabilistic voting-based Bluetooth RSS and PDR fusion method, and a heuristic search approach for reducing the complexity of BRPV. The experiment results in a real scene show that the average positioning error is < 2m, which is considered adequate for indoor location-based service applications. Moreover, compared to the traditional PDR method, BtPDR improves the location accuracy by 42.6%, on average. Compared to state-of-the-art Wireless Local Area Network (WLAN) fingerprint + PDR-based fusion indoor localization approaches, BtPDR has better positioning accuracy and does not need the same offline workload as a fingerprint algorithm.

• 대한안전경영과학회지
• /
• 제21권4호
• /
• pp.51-58
• /
• 2019

The LBS(Location Based Service) technology plays an important role in reducing wastes of time, losses of human lives and economic losses by detecting the user's location in order by suggesting the optimal evacuation route of the users in case of safety accidents. We developed an algorithm to estimate indoor location, movement path and indoor location changes of smart phone users based on the built-in sensors of smartphones and the dead-reckoning algorithm for pedestrians without a connection with smart devices such as Wi-Fi and Bluetooth. Furthermore, seven different indoor movement scenarios were selected to measure the performance of this algorithm and the accuracy of the indoor location estimation was measured by comparing the actual movement route and the algorithm results of the experimenter(pedestrian) who performed the indoor movement. The experimental result showed that this algorithm had an average accuracy of 95.0%.

• 자동차안전학회지
• /
• 제13권4호
• /
• pp.7-13
• /
• 2021

This paper presents an vulnerable road user (VRU) classification and tracking algorithm using vision and LiDAR sensor fusion method for urban autonomous driving. The classification and tracking for vulnerable road users such as pedestrian, bicycle, and motorcycle are essential for autonomous driving in complex urban environments. In this paper, a real-time object image detection algorithm called Yolo and object tracking algorithm from LiDAR point cloud are fused in the high level. The proposed algorithm consists of four parts. First, the object bounding boxes on the pixel coordinate, which is obtained from YOLO, are transformed into the local coordinate of subject vehicle using the homography matrix. Second, a LiDAR point cloud is clustered based on Euclidean distance and the clusters are associated using GNN. In addition, the states of clusters including position, heading angle, velocity and acceleration information are estimated using geometric model free approach (GMFA) in real-time. Finally, the each LiDAR track is matched with a vision track using angle information of transformed vision track and assigned a classification id. The proposed fusion algorithm is evaluated via real vehicle test in the urban environment.

• 정보처리학회논문지B
• /
• 제11B권3호
• /
• pp.317-326
• /
• 2004

본 논문에서는 이동 카메라에서 취득한 영상에서 컬러 정보를 이용하여 다수의 보행자를 검출하고 특정 보행자를 추적하는 방법을 제안한다. 먼저 연속한 동영상 입력에 대해 BMA(Block Matching Algorithm)을 이용하여 움직임 벡터를 추출하고 움직임 보상을 한 후 차 영상을 생성한다. 다음은 이진 영상으로 변환한 후 불필요한 잡음 능을 제거하친, 프로젝션을 수행하여 보행자를 검출한다. 만약 검출된 보행자가 서로 인접하거나 겹쳐졌을 경우 RGB 컬러 정보를 이용하여 분리시킨다. 검출된 다수의 보행자로부터 특정 보행자를 추적하기 위해 보행자 가운데 영역의 RGB 컬러 정보를 이용하여 추적한다. 제안된 방법에 대하여 비디오 카메라로 녹화한 영상을 컴퓨터에서 입력받아 검출과 추적 실험을 수행한 결과, 검출 성공률이 97％, 검출 실패율이 3％로 나타났고 추적 또한 우수함을 입증하였다.

• Journal of Positioning, Navigation, and Timing
• /
• 제12권2호
• /
• pp.91-100
• /
• 2023

The existing indoor localization method using Wi-Fi fingerprinting has a high collection cost and relatively low accuracy, thus requiring integrated correction of convergence with other technologies. This paper proposes a new method that significantly reduces collection costs compared to existing methods using Wi-Fi fingerprinting. Furthermore, it does not require labeling of data at collection and can estimate pedestrian travel paths even in large indoor spaces. The proposed pedestrian movement path estimation process is as follows. Data collection is accomplished by setting up a feature area near an indoor space intersection, moving through the set feature areas, and then collecting data without labels. The collected data are processed using Kernel Linear Discriminant Analysis (KLDA) and the valley point of the Euclidean distance value between two data is obtained within the feature space of the data. We build learning data by labeling data corresponding to valley points and some nearby data by feature area numbers, and labeling data between valley points and other valley points as path data between each corresponding feature area. Finally, for testing, data are collected randomly through indoor space, KLDA is applied as previous data to build test data, the K-Nearest Neighbor (K-NN) algorithm is applied, and the path of movement of test data is estimated by applying a correction algorithm to estimate only routes that can be reached from the most recently estimated location. The estimation results verified the accuracy by comparing the true paths in indoor space with those estimated by the proposed method and achieved approximately 90.8% and 81.4% accuracy in two experimental spaces, respectively.

• 한국철도학회논문집
• /
• 제9권3호
• /
• pp.271-276
• /
• 2006

Insight into behaviour of pedestrians as welt as tools to assess passenger flow condition is important in such instances as planning and geometric design of railway station under regular and safety-critical circumstances. Algorithm for passenger flow analysis based on DEM (Discrete Element Method) is newly developed. There are lots of similarity between particle-laden two phase flow and passenger flow. The velocity component of 1st phase corresponds to the unit vector of calculation cell, each particle to passenger, volume fraction to population density and the particle velocity to the walking velocity, etc. And, the walking velocity of passenger is also represented by the function of population density. Key algorithms are developed to determine the position of passenger, population density and numbering to each passenger. To verify the effectiveness of new algorithm, passenger flow analysis for the basic models of railway station is conducted.

• Structural Monitoring and Maintenance
• /
• 제1권1호
• /
• pp.89-110
• /
• 2014

Structural Health Monitoring (SHM) is an effective alternative to conventional inspections which are time-consuming and subjective. SHM can detect damage early and reduce maintenance cost and thereby help reduce the likelihood of catastrophic structural events to infrastructure such as bridges. After reviewing the Damage Index Method (DIM), an Iterative Damage Index Method (IDIM) is proposed to improve the accuracy of damage detection. These two damage detection techniques are compared based on damage on two structures, a simply supported beam and a pedestrian bridge. Compared to the traditional damage detection algorithm, the proposed IDIM is shown to be less arbitrary and more accurate.

• 한국공간구조학회논문집
• /
• 제12권4호
• /
• pp.99-108
• /
• 2012

In recently, sky-bridge are often applied to high-rised adjacent buildings for pedestrian bridge. the seisnic response control of adjacent buildings have been studied and magneto-rheological(MR) fluid dampers have been applied to seismic response control. In this study, vibration control effect of the MR damper connected adjacent buildings has been investigated. Adjacent building structures with different natural frequencies were used as example structures. Two typed of control methods, displacement based or velocity based, are applied to determinate control force of MR damper. In this numerical analysis, it has been shown that displacement-based control algorithm is more effective than velocity-based control algorithm for seismic response control of adjacent buildings. And, when displacement-based control method is applied to control of adjacent buildings, the control of building occurred large displacement is more efficient in reducing the seismic response.