• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.13 no.2
• /
• pp.157-163
• /
• 2013

In this paper, we propose a human detection technique using thermal imaging camera. The proposed method is useful at night or rainy weather where a visible light imaging cameras is not able to detect human activities. Under the observation that a human is usually brighter than the background in the thermal images, we estimate the preliminary human regions using the statistical confidence measures in the gray-level, brightness histogram. Afterwards, we applied Gaussian filtering and blob labeling techniques to remove the unwanted noise, and gather the scattered of the pixel distributions and the center of gravities of the blobs. In the final step, we exploit the aspect ratio and the area on the unified object region as well as a number of the principal components extracted from the object region images to determine if the detected object is a human. The experimental results show that the proposed method is effective in environments where visible light cameras are not applicable.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.4
• /
• pp.682-691
• /
• 2017

In this paper, we propose the fusion design methodology of both pedestrian detection and object tracking system realized with the aid of HOG-PCA based RBFNN pattern classifier. The proposed system includes detection and tracking parts. In the detection part, HOG features are extracted from input images for pedestrian detection. Dimension reduction is also dealt with in order to improve detection performance as well as processing speed by using PCA which is known as a typical dimension reduction method. The reduced features can be used as the input of the FCM-based RBFNNs pattern classifier to carry out the pedestrian detection. FCM-based RBFNNs pattern classifier consists of condition, conclusion, and inference parts. FCM clustering algorithm is used as the activation function of hidden layer. In the conclusion part of network, polynomial functions such as constant, linear, quadratic and modified quadratic are regarded as connection weights and their coefficients of polynomial function are estimated by LSE-based learning. In the tracking part, object tracking algorithms such as mean shift(MS) and cam shift(CS) leads to trace one of the pedestrian candidates nominated in the detection part. Finally, INRIA person database is used in order to evaluate the performance of the pedestrian detection of the proposed system while MIT pedestrian video as well as indoor and outdoor videos obtained from IC&CI laboratory in Suwon University are exploited to evaluate the performance of tracking.

• Journal of Internet Computing and Services
• /
• v.9 no.4
• /
• pp.115-131
• /
• 2008

Small mobile devices have become commonplace in our everyday life, from cellular phones to PDAs. Discovering dense regions for the mobile devices is one of the problems of grate practical importance. It can be used in monitoring movement of vehicles, concentration of troops, etc. In this paper, we propose a novel algorithm on continuously clustering a large set of mobile objects. We assume that a mobile object reports its position only if it is too far away from the expected position and thus the location data received may be imprecise. To compute the location of each individual object could be costly especially when the number of objects is large. To reduce the complexity of the computation, we want to first cluster objects that are in proximity into a group and treat the members in a group indistinguishable. Each individual object will be examined only when the inaccuracy causes ambiguity in the final results. We conduct extensive experiments on various data sets and analyze the sensitivity and scalability of our algorithms.

• KIISE Transactions on Computing Practices
• /
• v.20 no.11
• /
• pp.598-603
• /
• 2014

In wireless sensor networks, two methods have been generally used to track continuously moving object: a user query-based method and a periodic report-based method. Although the former method generates more overhead as a result of the user queries, the former one is also an energy-efficient method that does not transfer unnecessary information. For the user query-based method, a virtual tree that consist of sensor nodes is used to perform the user query and the sensor reporting. The tree stores the information of the mobile objects, and the stored information triggers a report b the user query. However, in case of a fast-moving object, the tracking accuracy decreases as a result of the time delay of the end-to-end repeated query. In order to solve this problem, we propose a query-relay method that reduces the time delay for mobile object tracking. In the proposed method, the nodes in the tree relay the query to adjacent nodes according to the movement of mobile object that is tracked. When the query messages are relayed. The end-to-end querying time delay is reduced. and a simulation shows that our method is superior to existing ones in terms of tracking accuracy.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2007.04a
• /
• pp.169-173
• /
• 2007

Nowadays the aerial photos is using to get the information around our spatial environment and it increases by geometric progression in many fields. The aerial photos need in a simple object such as cartography and ground covey classification and also in a social objects such as the city plan, environment, disaster, transportation etc. However, the shadow, which includes when taking the aerial photos, makes a trouble to interpret the ground information, and also users, who need the photos in their field tasks, have restriction. This study, for removing the shadow, uses the single image and the image without the source of image and taking situation. Also, this study present clustering algorism based on HIS color model that use Hue, Saturation and Intensity, especially this study used I(intensity) to extract shadow area from image. And finally by filtering in Fourier frequency domain creates the intrinsic image which recovers the 3-D color information and removes the shadow.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.2
• /
• pp.115-120
• /
• 2015

Recently, in automotive technology area, intelligent safety systems have been actively accomplished for drivers, passengers, and pedestrians. Also, many researches are focused on development of autonomous vehicles. This paper propose the application of LiDAR sensors, which takes major role in perceiving environment, terrain classification, obstacle data clustering method, and local map building for autonomous driving. Finally, based on these results, planning for lane change path that vehicle tracking possible were created and the reliability of path generation were experimented.

• Proceedings of the IEEK Conference
• /
• 2000.06d
• /
• pp.222-225
• /
• 2000

Human frequently communicate non-linguistic information with gesture. So, we must develop efficient and fast gesture recognition algorithms for more natural human-computer interaction. However, it is difficult to recognize gesture automatically because human's body is three dimensional object with very complex structure. In this paper, we suggest a method which is able to detect key frames and frame changes, and to classify image sequence into some gesture groups. Gesture is classifiable according to moving part of body. First, we detect some frames that motion areas are changed abruptly and save those frames as key frames, and then use the frames to classify sequences. We symbolize each image of classified sequence using Principal Component Analysis(PCA) and clustering algorithm since it is better to use fewer components for representation of gestures. Symbols are used as the input symbols for the Hidden Markov Model(HMM) and recognized as a gesture with probability calculation.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2016.07a
• /
• pp.195-197
• /
• 2016

시멘틱웹을 구축하는 표준언어인 RDF (Resource Description Framework)는 언어의 그래프 기반 특성으로 인해 일반적인 방식들로는 효과적인 저장과 추출이 어렵다. 더욱이 대용량 RDF 데이터의 저장과 추출에는 성능문제가 더욱 커지므로 많은 연구들이 이루어지고 있다. 본 논문에서는 SPARQL을 지원하면서 RDF 파일들을 효과적으로 저장하고 검색할 수 있는 저장방식에 대해 연구한 결과를 제시한다. RDF 데이터를 전처리를 통해 RDF의 트리플(주어:subject, 술어:property, 목적어:Object)에서 중복되는 주어(S)나 목적어(O)를 묶고, 사용자가 SPARQL 형식으로 검색했을 때 이용자가 주어부분을 변수로 두었는지 아니면 서술어 부분을 변수로 두어 찾는지에 따라 검색어와 유사한 단어 클러스터를 찾아준다. 동일 단어에 대해 여러 번 검색되던 부분을 한 번 검색으로 처리할 수 있기 때문에 효율이 높아진다.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.15-18
• /
• 2001

동영상에서 움직이는 객체 검출은 동영상의 내용을 표현하고 유사한 동영상을 검색하는 데 있어 중요한 특징간을 추출하는 방법으로 사용된다. 그러나 복잡하게 카메라가 움직이는 동영상에서 움직이는 객체 검출은 아직까지 어려운 과제이다. 본 논문에서는 복잡한 카메라의 움직임이 있는 환경에서 움직이는 객체를 강인하게 검출하는 방법을 제안한다. 움직이는 객체 검출 방법은 입력 영상을 색상간의 클러스터링을 이용하여 각 영역으로 구분하는 Mean Shift 알고리즘과 인접한 프레임에서 구분된 영역을 대응시켜 영역의 모션 벡터를 구하는 영역 매칭, 유사한 궤적을 가지는 영역들의 클러스터링을 이용하여 객체를 검출하는 궤적 클러스터링 알고리즘을 사용한다. 제안한 영역 기반 알고리즘은 기존의 픽셀이나 블록 기반의 방법보다 움직이는 객체를 정확하게 검출하였다. 실험 결과 복잡하게 움직이는 카메라의 환경 속에서 움직이는 객체를 강인하게 검출하였다.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.05a
• /
• pp.487-490
• /
• 2002

AGV is very useful equipment to transfer containers in automated container terminal. AGV must have Obstacle Detection System (ODS) fur port automation. ODS needs the function to classify some specified object from background in acquired data. And it must be able to track classified moving objects. Finally, ODS could determine its next action for safe deriving whether it should do emergency stop or speed down, or it should change its deriving lane. For these functions, ODS can have many different kinds of algorithm. In this paper, we present one of them under developing.