• International Journal of Contents
• /
• 제10권1호
• /
• pp.1-11
• /
• 2014

In the current generation of smart mobile devices, object tracking is one of the most important research topics for computer vision. Because human face tracking can be widely used for many applications, collecting a dataset of face videos is necessary for evaluating the performance of a tracker and for comparing different approaches. Unfortunately, the well-known benchmark datasets of face videos are not sufficiently diverse. As a result, it is difficult to compare the accuracy between different tracking algorithms in various conditions, namely illumination, background complexity, and subject movement. In this paper, we propose a new dataset that includes 91 face video clips that were recorded in different conditions. We also provide a semi-automatic ground-truth generation tool that can easily be used to evaluate the performance of face tracking systems. This tool helps to maintain the consistency of the definitions for the ground-truth in each frame. The resulting video data set is used to evaluate well-known approaches and test their efficiency.

• 한국멀티미디어학회논문지
• /
• 제13권2호
• /
• pp.225-236
• /
• 2010

본 논문은 카메라 영상 정보와 기울기 센서 정보를 통합한 바이모달 응급상황 인식방법을 제안한다. 제안된 방법은 어느 한 센서가 오작동 하거나 사용자가 착용형 기울기 센서를 착용하지 않거나, 영상 획득의 어려움이 있는 욕실과 같은 곳에 있는 경우에도 응급 상황을 감지하여 센서 간에 상호 협력과 보완을 함으로써 응급 상황을 인식할 수 있다. 본 논문에서는 HMM 학습 및 인식을 통해 걷는 동작, 바닥에 앉는 동작, 소파에 앉는 동작, 눕는 동작, 기절 동작을 판단할 수 있도록 하였다. 영상의 특징 벡터와 기울기 센서의 특징 벡터를 결합하여 학습하고 인식했을 때, 인식률의 향상을 가져올 수 있었다. 또한 다양한 조명의 변화에도 적응적 배경 모델을 통해 움직이는 객체를 강건하게 검출할 수 있어서 높은 인식률을 유지할 수 있었다.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2015년도 춘계학술대회
• /
• pp.586-589
• /
• 2015

손동작 식별 룰을 통한 컴퓨터의 프레젠테이션 제어 시스템을 제안한다. 발표자의 손 동작 식별을 위해 (일반적인 웹캠을 사용하여) 이미지를 입력받아 하르 분류기를 이용하여 사용자의 얼굴영역을 추출한다. YCbCr 컬러모델을 이용하여 손 영역을 추출한 후에 사용자의 얼굴과 손의 무게중심을 이용하여 손의 현재 움직임 상태와 위치를 판별 하였다. 사용자의 손이 모션 감지 룰에 적용되어 프레젠테이션 제어 명령이 실행된다. 제안하는 시스템은 모션 식별 룰을 이용하여 부가적인 기기를 사용하지 않고 배경의 복잡도에 독립적인 프레젠테이션을 제어가 가능한 시스템이다. 실험은 어두운 실내 분위기인 조도범위(lx) 15-20-30에서 프레젠테이션 실험을 통해 안정적인 제어동작을 확인하였다.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제55권7호
• /
• pp.313-318
• /
• 2006

An automatic sample centering system is underway at the protein crystallography beam line of the Pohang Light Source to improve the efficiency of the crystal screening process. A sample pin which contains a protein crystal is mounted on a goniometer head. Then the crystal should be moved to the center of X-ray beam by controlling the motorized goniometer to obtain diffraction data. Since the X-ray beam is located at the center of the image obtained from the CCD camera when the image of the sample pin is in focus, an auto-focusing algorithm is a very important part in the auto-sample-centering system. However the results of applying several well-known auto focusing algorithms directly to the images are not satisfactory owing to the following factors: misalignment of CCD camera, non-uniform cryo-stream in the background of the image and the supporter of the loop. The performance of an auto-focusing algorithm can be increased if the algorithm is applied to only the loop region identified. Non-uniform cryo-stream and a various illumination condition and a stain, which is shown in the image, are main obstacles to loop region identification. In this paper, a simple loop region identification algorithm, which can solve these problems, is proposed and the effective ness of the proposed scheme is shown by applying the auto-focusing algorithm to the loop region identified.

• World Technopolis Review
• /
• 제6권1호
• /
• pp.1.1-1.34
• /
• 2017

This paper aims to illuminate the role of serial context among industry, academia and government, taking much care of the role of society and community in the sustainable regional planning and practice. This paper is composed of five chapters, each of them dealing with different aspects. In In chapter 1, we give the little long introduction of the time of mutation and significance of locus that explains the conceptual background and framework of this paper. In chapter 2, we elaborate on the mutation and metamorphosis of structural, social, and natural changes in the world and especially Japan. In chapter 3 and chapter 4, the main chapters of this paper, we describe the evolution of academic town in megalopolis, the revitalization of technopolis, and the creative local culture of the World Heritage for the regional renaissance in Japan. In chaper 5, we conclude this paper. As for this illumination, we should pay more due regards to the locus, orbit and iconography of region to develop better hosting environment and habitat for global innovation of industry-academia-government collaboration's serial contexts through sustainable tourism and tourism sustainability. Especially, at the time of natural and social mutation, we could not look over the heavy and sudden natural unexpected changes, the deep structural social and community changes in Japan, and war and terrorism in Asia on the global scene for sustainable rejuvenation.

• 한국멀티미디어학회논문지
• /
• 제20권7호
• /
• pp.986-993
• /
• 2017

Object Tracking is a technique for tracking moving objects over time in a video image. Using object tracking technique, many research are conducted such a detecting dangerous situation and recognizing the movement of nearby objects in a smart car. However, it still remains a challenging task such as occlusion, deformation, background clutter, illumination variation, etc. In this paper, we propose a novel deep visual object tracking method that can be operated in robust to many challenging task. For the robust visual object tracking, we proposed a Convolutional Neural Network(CNN) which shares weight of the convolutional layers. Input of the CNN is a three; first frame object image, object image in a previous frame, and current search frame containing the object movement. Also we propose a method to consider the motion of the object when determining the current search area to search for the location of the object. Extensive experimental results on a authorized resource database showed that the proposed method outperformed than the conventional methods.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.136-141
• /
• 2009

This paper proposed a framework for handover method in continuously tracking a person of interest across cooperative pan-tilt-zoom (PTZ) cameras. The algorithm here is based on a robust non-parametric technique for climbing density gradients to find the peak of probability distributions called the mean shift algorithm. Most tracking algorithms use only one cue (such as color). The color features are not always discriminative enough for target localization because illumination or viewpoints tend to change. Moreover the background may be of a color similar to that of the target. In our proposed system, the continuous person tracking across cooperative PTZ cameras by mean shift tracking that using color and shape histogram to be feature distributions. Color and shape distributions of interested person are used to register the target person across cameras. For the first camera, we select interested person for tracking using skin color, cloth color and boundary of body. To handover tracking process between two cameras, the second camera receives color and shape cues of a target person from the first camera and using linear color calibration to help with handover process. Our experimental results demonstrate color and shape feature in mean shift algorithm is capable for continuously and accurately track the target person across cameras.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권11호
• /
• pp.5436-5458
• /
• 2017

Automatic Video Surveillance System (AVSS) has become important to computer vision researchers as crime has increased in the twenty-first century. As a new branch of AVSS, baggage detection has a wide area of security applications. Some of them are, detecting baggage in baggage restricted super shop, detecting unclaimed baggage in public space etc. However, in this paper, a detection & classification framework of baggage is proposed. Initially, background subtraction is performed instead of sliding window approach to speed up the system and HSI model is used to deal with different illumination conditions. Then, a model is introduced to overcome shadow effect. Then, occlusion of objects is detected using proposed mirroring algorithm to track individual objects. Extraction of rotational signal descriptor (SP-RSD-HOG) with support plane from Region of Interest (ROI) add rotation invariance nature in HOG. Finally, dynamic human body parameter setting approach enables the system to detect & classify single or multiple pieces of carried baggage even if some portions of human are absent. In baggage detection, a strong classifier is generated by boosting similarity measure based multi layer Support Vector Machine (SVM)s into HOG based SVM. This boosting technique has been used to deal with various texture patterns of baggage. Experimental results have discovered the system satisfactorily accurate and faster comparative to other alternatives.

• 제어로봇시스템학회논문지
• /
• 제22권1호
• /
• pp.1-7
• /
• 2016

ADAS (Advanced Driver Assistance Systems) requires not only real-time robust lane detection, both straight and curved, but also predicting upcoming steering direction by detecting the curvature of lanes. In this paper, a curvature lane detection algorithm is proposed to enhance the accuracy and detection rate based on using inverse perspective images and Gaussian Mixture Model (GMM) to segment the lanes from the background under various illumination condition. To increase the speed and accuracy of the lane detection, this paper used template matching, RANSAC and proposed post processing method. Through experiments, it is validated that the proposed algorithm can detect both straight and curved lanes as well as predicting the upcoming direction with 92.95% of detection accuracy and 50fps speed.

• 한국통신학회논문지
• /
• 제30권12C호
• /
• pp.1175-1180
• /
• 2005

본 논문에서는 비전 기반의 손동작 검출 및 추적 시스템을 제안하고자 한다. 기존의 손동작 인식 시스템은 정적인 관측 환경에서 배경을 제거함으로 손을 검출하는 단순한 방법을 사용함으로써, 카메라의 움직임, 조명의 변화 등에 의해 견실하지 못하였다. 그러므로 본 논문에서는 기하학적 구조에 의하여 손의 외형을 인식하여 검출할 수 있는 통계적 방법을 제안하였다. 또한 카메라의 각도에 의한 손이 겹쳐 보이는 문제를 줄이기 위하여 다중 카메라를 사용하였으며 비동기식 다중 관측으로 시스템의 범용성을 향상시키었다. 실험 결과 제안된 방법이 기존의 외관을 이용한 방법보다 $3.91\%$ 개선된 $99.28\%$의 인식률을 나타내어 제안한 방법의 효율성을 입증하였다.