A Traffic Surveillance System using Dynamic Saliency Map and SVM Boosting

이민호,이정우,이우노 제어·로봇·시스템학회 2010 International Journal of Control, Automation, and Vol.8 No.5

This paper proposes a traffic surveillance system that can efficiently detect an interesting object and identify vehicles and pedestrians in real traffic situations. The proposed system consists of a moving object detection model and an object identification model. A dynamic saliency map is used for analyzing dynamics of the successive static saliency maps, and can localize an attention area in dy-namic scenes to focus on a specific moving object for traffic surveillance purposes. The candidate local areas of a moving object are followed by a blob detection processing including binarization, morphological closing and labeling methods. For identifying a moving object class, the proposed system uses a hybrid of global and local information in each local area. Although the global feature analysis is a compact way to identify an object and provide a good accuracy for non-occluded objects, it is sensitive to image translation and occlusion. Therefore, a local feature analysis is also considered and combined with the global feature analysis. In order to construct an efficient classifier using the global and local features, this study proposes a novel classifier based on boosting of support vector machines. The proposed object identification model can identify a class of moving object and discard unexpected candidate area which does not include an interesting object. As a result, the proposed road surveillance system is able to detect a moving object and identify the class of the moving object. Experimental results show that the proposed traffic surveillance system can successfully detect specific moving objects.

[컴퓨터지능 및 지능시스템] DTW 알고리즘을 이용한 다중 이동 객체 추적 방법

왕교선(Kyo Sun Wang),주영훈(Yung Hoon Joo) 대한전기학회 2019 전기학회논문지 Vol.68 No.5

In this paper, we propose a method to track multiple moving objects using DTW algorithm for images received from CCTV Cameras. The proposed method first extracts a moving object using GMM background modeling to extract a moving object from the input image, and then removes and extends the shadow and noise of the extracted moving object using morphology. The moving object thus obtained is recognized by using the labeling method, and the labeling is merged using the morphology to identify the problem that the recognized moving object is divided into several labels. Then, when the recognized moving object is detected in the designated ROI, the HSV and RGB values of the moving object are extracted and the extracted data is sequentially stored in the DTW data base. Next, we propose a method of classifying the moving object as similar to the DTW by judging the similarity between the stored DTW DB and the moving object. The moving objects classified by the proposed method are continuously compared and tracked. If it is judged that the moving objects to be tracked are overlapped, it is possible to compare the remaining objects. It also suggests how tracking can be done if the moving object is unwrapped. Finally, the applicability of the proposed method is verified through various experiments.

A dynamic multi-objective evolutionary algorithm based on prediction

Wu Fei,Chen Jiacheng,Wang Wanliang 한국CDE학회 2023 Journal of computational design and engineering Vol.10 No.1

The dynamic multi-objective optimization problem (DMOP) is a common problem in optimization problems; the main reasons are the objective’s conflict and environment changes. In this paper, we provide a prediction approach based on diversity screening and special point prediction (DSSP) to tackle the dynamic optimization issue. First, we introduce a decision variable clustering and screening strategy that clusters the decision space of the non-dominated solution set to find the cluster centroids and then employs a decision variable screening strategy to filter out solutions that have an impact on the distribution of individuals. This approach can broaden the range of dynamic multi-objective optimization algorithms. Second, an approach for predicting special points is suggested. The algorithm’s convergence is improved following environmental changes by forecasting the special point tracking Pareto front in the object space. Finally, the forward-looking center points are used to predict the non-dominated solution set and eliminate the useless individuals in the population. The prediction strategy can help the solution set converge while maintaining its diversity, which is compared with the four other state-of-the-art strategies. Our experimental results demonstrate that the proposed algorithm, DSSP, can effectively tackle DMOPs.

수학 문제 해결 과정의 의사소통에 대한 기호학적 분석

김수민,김선희 대한수학교육학회 2018 학교수학 Vol.20 No.1

본 연구는 의사소통에서 하나의 기호 운반체로 여러 가지 생각들이 교환되는 것을 미시적으로 살펴보기 위해 대상체를 세분하여 문제 해결 과정을 기호학적으로 분석하였다. 두 명의 대학생들에게 경사도 문제를 제시하고 그들의 의사소통을 문제 해결 단계별로 구분하여 직접적 대상체, 역동적 대상체, 기호 운반체의 역할을 살펴보았다. 학생들은 문제의 답이 되는 최종적인 해석체를 형성하기 위해 여러 대상체들을 생성하였고 그 과정에서 동일한 기호 운반체에 대해 직접적 대상체와 역동적 대상체를 생성하였다. 그리고 의사소통에서 의견이 일치하지 않은 경우, 발신자는 문제 해결을 위해 수신자가 역동적 대상체를 수정할 수 있도록 노력함을 확인할 수 있었다. In order to microscopically examine the exchange of many thoughts as a single sign-vehicle in communication, this study semiotically analyzed the problem-solving process by subdividing sign-objects. After presenting a gradient problem to two university students and dividing their communication into each problem-solving stage, the roles of immediate object, dynamic object, and sign-vehicle were examined. The students formed diverse objects for forming interpretant, and in the process, they formed the immediate object and dynamic object regarding the same sign-vehicle. And in case when opinions were not agreed in the problem-solving process through communication, for the correct problem solving, the sender made efforts, for changing sign-vehicles so that the receiver could modify the dynamic object.

이동 객체 데이타베이스에서 KDB-tree의 동적 분할 정책

임덕성(Duksung Lim),이창헌(Changheun Lee),홍봉희(Bonghee Hong) 한국정보과학회 2006 정보과학회논문지 : 데이타베이스 Vol.33 No.4

Moving object databases manage a large amount of past location data which are accumulated as the time goes. To retrieve fast the past location of moving objects, we need index structures which consider features of moving objects. The KDB-tree has a good performance in processing range queries. Although we use the KDB-tree as an index structure for moving object databases, there has an over-split problem in the spatial domain since the feature of moving object databases is to increase the time domain. Because the over-split problem reduces spatial regions in the MBR of nodes inverse proportion to the number of splits, there has a problem that the cost for processing spatial-temporal range queries is increased. In this paper, we propose the dynamic split strategy of the KDB-tree to process efficiently the spatial-temporal range queries. The dynamic split strategy uses the space priority splitting method for choosing the split domain, the recent time splitting policy for splitting a point page to maximize the space utilization, and the last division policy for splitting a region page. We compare the performance of proposed dynamic split strategy with the 3DR-tree, the MV3R-tree, and the KDB-tree. In our performance study for range queries, the number of node access in the MKDB-tree is average 30% less than compared index structures. 시간의 흐름에 따라 누적되는 대용량의 과거 위치를 관리하는 이동 객체 데이타베이스에서 이동 객체의 과거 위치를 효율적으로 검색하기 위해서는 이동 객체의 특성을 고려한 색인 구조가 필요하다. 그러나, 영역 질의 성능이 우수한 다차원 색인인 KDB-tree를 이동 객체 데이타베이스에 적용할 경우 시간 도메인이 증가하는 이동 객체 데이타베이스의 특성으로 인해 공간 도메인 가중 분할이 발생한다. 공간도메인 가중 분할은 하나의 노드가 차지하는 MBR의 공간 영역이 분할 횟수에 반비례하게 감소되어 시공간 영역 질의 처리시 색인의 검색 비용을 증가시키는 문제가 있다. 이 논문에서는 이동 객체 데이타베이스에서 시공간 영역 질의를 효율적으로 처리하기 위한 KDB-tree의 동적 분할 정책을 제안한다. 동적 분할 정책은 공간 우선 분할 방법을 적용하는 분할 도메인 선정 방법과 포인터 페이지에서 공간 활용도를 최대화 시킬 수 있는 최근 시간 분할 정책, 영역 페이지에서 적용되는 최후 시간 분할 정책으로 구성된다. 제안한 동적 분할 정책의 성능을 평가하기 위해 3DR-tree, MV3R-tree, KDB-tree와의 성능을 비교한다. 영역 질의를 위한 성능 평가에서 동적 분할 정책을 적용한 MKDB-tree는 기존 색인에 비해 평균 30%이상의 노드 접근 회수를 감소시킨다.

2D Image Recognition of Partially Occluded Objects Based on Adaptive Window Positioning and Dynamic Time Warping

Mohammed Almuashi,Siti Z. Mohd Hashim,Dzulkifli Mohamad,Mohammed Hazim Alkawaz 보안공학연구지원센터 2016 International Journal of Multimedia and Ubiquitous Vol.11 No.7

The boundaries of computer vision are especially challenged by partial occluded ob-ject recognition. In this paper, we proposed a new algorithm that used to recognize the partially occluded objects in 2-D images. Two different cases of occlusion are considered. Firstly, when an object missing some parts. Secondly, objects are overlapping each other. This approach uses contour of an object. Adaptive Window Positioning (AWP) uses to extract features from a contour. The proposed method uses the orientation field as fea-tures of the fragment. Dynamic Time Warping (DTW) works for matching and find the similarity between sub-images. As experiment result, the proposed recognition algorithm works without prior segmentation and robust to identify missing and overlapping objects and it can woks with strength occlusion.

RGB-D Dense Visual Odometry through Pixel Level Segmentation in Dynamic Environments

Changsuk Oh,Youngseok Jang,H.Jin Kim 제어로봇시스템학회 2019 제어로봇시스템학회 국제학술대회 논문집 Vol.2019 No.10

Estimating a camera pose in dynamic environments is one of the challenging problems in Visual Odometry. We propose an RGB-D Dense Visual Odometry (Dense-VO) system which uses preprocessed images that passed the Convolutional Neural Network (CNN). The algorithm adopts the CNN that tracks the designated dynamic object. The tracked dynamic object is excluded when the Dense-VO estimates the camera motion by minimizing photometric error between consecutive images. The system was tested in two datasets which includes a dynamic object. The proposed approach containing the preprocessing procedure estimates the camera trajectory with less drift in a dynamic environment.

자율협력주행을 위한 역할 기반 동적정보 서비스 평가 방법

노창균,김형수,임이정 한국ITS학회 2022 한국ITS학회논문지 Vol.21 No.1

안전한 자율주행을 위하여 차량 센서에만 의존하는 Stand-alone 방식의 한계를 극복하기 위 한 방법으로, 노변의 인프라와 자율주행차간 정보를 교환하는 ‘자율협력주행' 방식의 기술 개 발이 이루어지고 있다. 이 과정에서 협력의 대상이 되는 동적정보는 통신 데이터 손실 측면의 평가방법이 일반적이지만, 정보로서 역할 중심의 평가방법이 필요하다. 본 연구에서는 자율협 력주행에서 동적정보 서비스 역할의 적정성을 평가하기 위하여 역할 기반 평가방법을 제안하 였다. 평가 척도로 검출률, 검출 소요시간, LDM 처리시간을 제안하였고, 평가방법론을 검증하 기 위하여 실제 도로에서 보행자 정보를 대상으로 실증 실험을 시행하였다. 실험 결과로는 검 출률 99%, 소요시간 200ms/건, 처리시간 19ms/건을 얻었다. 향후 제안된 동적정보 서비스 평가 방법이 관련 정보제공 서비스의 평가에 활용되기를 기대한다. The technology implementation method was diversified into an ‘autonomous cooperative driving’ method to overcome the limitations of a stand-alone autonomous vehicle with vehicle sensor-based autonomous driving. The autonomous cooperative driving method involves exchanging information between roadside infrastructure and autonomous vehicles. In this process, the concept of dynamic information (LDM), a target of cooperation, was established. But, evaluation methods and standards for dynamic information have not been established. Therefore, this study, a dynamic information evaluation method based on information on pedestrians within the moving objects. In addition, autonomous cooperative driving was demonstrated, and dynamic information was also verified through the evaluation method. The significance of this study is that it established the dynamic information evaluation methodology for autonomous cooperative driving for the first time. Based on this, this study is expected to contribute to the application of safe autonomous cooperative driving technology to the field.

교육목표의 우선순위를 고려한 동적 강의계획서 구성 시스템

김호숙 ( Ho Sook Kim ),김형석 ( Hyoung Seok B. Kim ) 한국컴퓨터교육학회 2009 컴퓨터교육학회 논문지 Vol.12 No.2

본 논문은 선행 학습과 후행 학습 관계인 교과의 선 후수 관계가 뚜렷하고, 학습자의 이해 정도를 교수자가 실시간으로 파악하기 쉬운 컴퓨터 교육에서 발생할 수 있는 정적인 교육 실행 계획의 문제점을 해결하기 위하여, 교육목표에 대한 우선순위를 기반으로 물리적인 교육 시간의 변화 및 학습자의 수준 변화에 따른 동적 강의계획서 구성 시스템을 제안한다. 이를 활용하여 출발점 행동이 서로 다른 두 집단을 대상으로 수업을 진행한 결과, 제안된 시스템 활용이 교육의 전이도가 높고 해당 교과목에서 반드시 다루어야 하는 대표 주제를 중심으로 수업을 진행하는데 도움을 주어 우선순위가 높은 교육목표의 달성에 효과적임을 확인하였다. In this paper, we propose a new dynamic syllabus composition system to solve the problems of a static syllabus which can appear in the field of computer education, where the relationship between pre-post study subjects is clear and teachers may grasp easily the degree of understanding of learners in real time. Our dynamic syllabus composition system is designed to be adjusted according to the physical change of the amount of education and the level of learners, which is based on the priority of educational objects. The result of instance performed on two groups of different entering behavior shows that the proposed method enhances the degree of transfer of education and helps us teach a class around the representative subject which has to be dealt with in the class, so that it is very effective for the achievement of educational objects prior to others.

공간 네트워크에서 이동객체의 위치정보관리를 위한 동적 분산 그리드 기법

김영창(YoungChang Kim),홍승태(SeungTae Hong),조경진(KyungJin Jo),장재우(JaeWoo Chang) 한국정보과학회 2009 정보과학회 컴퓨팅의 실제 논문지 Vol.15 No.12

최근 공간 네트워크에서 대용량 이동객체의 위치정보를 관리하기 위한 DS-GRID(distributed S-GRID)가 제안되었다[1]. 그러나 DS-GRID는 균일 크기의 그리드 셀을 이용하기 때문에, 실제 응용에서 빈번히 발생하는 이동객체의 쏠림 현상을 효율적으로 관리하지 못하는 단점을 지닌다. 이를 해결하기 위해, 본 논문에서는 이동객체의 밀도에 따라 그리드 셀을 동적으로 분할하는 동적 분산 그리드 기법을 제안한다. 아울러 이를 위한 k-최근접 질의처리 알고리즘을 제안한다. 마지막으로 성능 평가를 통해 이동객체의 쏠림 현상이 발생하였을 경우, 제안하는 동적 분산 그리드 기법이 검색 및 업데이트 성능 측면에서 DS-GRID보다 우수함을 입증한다. Recently, a new distributed grid scheme, called DS-GRID(distributed S-GRID), has been proposed to manage the location information of moving objects in a spatial network[1]. However, because DS-GRID uses uniform grid cells, it cannot handle skewed data which frequently occur in the real application. To solve this problem, we propose a dynamic distributed grid scheme which splits a grid cell dynamically based on the density of moving objects. In addition, we propose a k-nearest neighbor processing algorithm for the proposed scheme. Finally, it is shown from the performance analysis that our scheme achieves better retrieval and update performance than the DS-GRID when the moving objects are skewed.