• Journal of Korea Spatial Information System Society
• /
• v.11 no.2
• /
• pp.45-53
• /
• 2009

The fast progress on multimedia data acquisition technologies has enabled collecting vast amount of videos in real time. Although the amount of information gathered from these videos could be high in terms of quantity and quality, the use of the collected data is very limited typically by human-centric monitoring systems. In this paper, we propose a framework for analyzing long traffic video using series of content-based analyses tools. Our framework suggests a method to integrate theses analyses tools to extract highly informative features specific to a traffic video analysis. Our analytical framework provides (1) re-sampling tools for efficient and precise analysis, (2) foreground extraction methods for unbiased traffic flow analysis, (3) frame property analyses tools using variety of frame characteristics including brightness, entropy, Harris corners, and variance of traffic flow, and (4) a visualization tool that summarizes the entire video sequence and automatically highlight a collection of frames based on some metrics defined by semi-automated or fully automated techniques. Based on the proposed framework, we developed an automated traffic flow analysis system, and in our experiments, we show results from two example traffic videos taken from different monitoring angles.

• Journal of the Korean Society for Precision Engineering
• /
• v.28 no.11
• /
• pp.1297-1308
• /
• 2011

The light-weight design of UTM (Urban Transit Maglev)-02 car-body frames are performed, based on initial configuration. The thicknesses of fourteen sub-structures are defined as design variables and the loading condition is considered according to weight of sub-structures, electronic and pneumatic modules and passengers. For efficient and robust process of design optimization, objective function and constraints are approximated by response surface approximation. Structural analysis is performed at some sampling points to construct the approximated objective function and constraints composed of design variables. Design space is changed to find many optimal candidates and best optimal design can be found eventually. The Matlab Optimization Toolbox is used to find optimal value and sensitivity analysis about each design variable is also performed.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.9A
• /
• pp.920-927
• /
• 2006

There are three critical considerations in developing a PCM telemetry encoder to be installed in an artillery projectile. The first is the performance consideration, such as sampling rate and data transmission rate. The second is the size consideration due to the severely limited installation space in an artillery projectile and the last is the power consumption consideration due to limitations of the munition's power supply. To meet these three considerations, the best alternative is a one-chip solution. Using a commercially available TMS320F2812 DSP, we have implemented a 30-channel PCM telemetry encoder to process randomized data frames, composed of 16-channel analog data, 14-channel digital data and 2-frame synchronization channels per data frame at 10Mbps transmitting baud rate.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.11
• /
• pp.5624-5638
• /
• 2016

A novel method based on Surf points is proposed to detect and lock-track single ground target in aerial videos. Videos captured by moving cameras contain complex motions, which bring difficulty in moving object detection. Our approach contains three parts: moving target template detection, search area estimation and target tracking. Global motion estimation and compensation are first made by grids-sampling Surf points selecting and matching. And then, the single ground target is detected by joint spatial-temporal information processing. The temporal process is made by calculating difference between compensated reference and current image and the spatial process is implementing morphological operations and adaptive binarization. The second part improves KALMAN filter with surf points scale information to predict target position and search area adaptively. Lastly, the local Surf points of target template are matched in this search region to realize target tracking. The long-term tracking is updated following target scaling, occlusion and large deformation. Experimental results show that the algorithm can correctly detect small moving target in dynamic scenes with complex motions. It is robust to vehicle dithering and target scale changing, rotation, especially partial occlusion or temporal complete occlusion. Comparing with traditional algorithms, our method enables real time operation, processing $520{\times}390$ frames at around 15fps.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.9
• /
• pp.4438-4460
• /
• 2017

The Kernelized Correlation Filters (KCF) tracker has caused the extensive concern in recent years because of the high efficiency. Numerous improvements have been made successively. However, due to the abrupt motion between the consecutive image frames, these methods cannot track object well. To cope with the problem, we propose an extended KCF tracker based on swarm intelligence method. Unlike existing KCF-based trackers, we firstly introduce a swarm-based sampling method to KCF tracker and design a unified framework to track smooth or abrupt motion simultaneously. Secondly, we propose a global motion estimation method, where the exploration factor is constructed to search the whole state space so as to adapt abrupt motion. Finally, we give an adaptive threshold in light of confidence map, which ensures the accuracy of the motion estimation strategy. Extensive experimental results in both quantitative and qualitative measures demonstrate the effectiveness of our proposed method in tracking abrupt motion.

• Journal of Fashion Business
• /
• v.21 no.6
• /
• pp.106-121
• /
• 2017

This study was to identify underlying dimensions of service encounter quality, to test difference in service encounter quality between interpersonal service(IPS) and self-service technology(SST) and to predict service effectiveness and behavioral intentions in the fashion retail context. A field experimental study was designed for collecting data. Repeat sampling frames were involved in two types of service encounters: interpersonal service and self-service technology within the store environment. Thirty participants served as subjects. Sample represents more females than males, and age was ranged from 20 to 33 years old (Mean=24.2). The result suggested that service encounter quality consisted of competence, dedication, and listening. There was a significant mean difference on listening factor of service encounter quality between IPS and SST. For the interpersonal service, dedication and listening had significant effects on service effectiveness. For the self-service technology, competence and listening had significant effects on service effectiveness. In the IPS condition, the service effectiveness significantly affected the revisit intention, whereas it was not significantly related to the revisit intention in the SST condition. This study discussed managerial implications for fashion retailers seeking to effectively manage service quality by specifying interpersonal service versus self-service technology in the retail environments.

• Journal of Korea Multimedia Society
• /
• v.24 no.7
• /
• pp.880-889
• /
• 2021

In everyday life, recognizing people's emotions from their frames is essential and is a popular research domain in the area of computer vision. Visual emotion has a severe class imbalance in which most of the data are distributed in specific categories. The existing methods do not consider class imbalance and used accuracy as the performance metric, which is not suitable for evaluating the performance of the imbalanced dataset. Therefore, we proposed a method for recognizing visual emotion using balanced data augmentation to address the class imbalance. The proposed method generates a balanced dataset by adopting the random over-sampling and image transformation methods. Also, the proposed method uses the Focal loss as a loss function, which can mitigate the class imbalance by down weighting the well-classified samples. EfficientNet, which is the state-of-the-art method for image classification is used to recognize visual emotion. We compare the performance of the proposed method with that of conventional methods by using a public dataset. The experimental results show that the proposed method increases the F1 score by 40% compared with the method without data augmentation, mitigating class imbalance without loss of classification accuracy.

• Journal of the Korean Society of Industry Convergence
• /
• v.21 no.5
• /
• pp.219-225
• /
• 2018

In this paper, for the testing of tensile strength of dead-end clamp used in transmission line, resulting values were estimated by designing and producing the horizontal version of widely-used vertical tensile tester. Tensile strength test of dead-end clamp for transmission line is essential for quality test of products. Moreover, tensile tester is an equipment that requires high level of reliability which needs to be examined through sampling tests commensurate with total inspection. Frames of tensile tester were made up of H-beams so that it can endure more than 20 [tons] of load capability and the test was implemented for 60[seconds] applying five types of tension. In consequence, the tester could withstand up to 21,600[kg] of weight as well as all types of tension. This newly developed horizontal tensile tester can be utilized in figuring out properties of various materials by estimating tensile strength of materials such as metal, rubber and fiber.

• ETRI Journal
• /
• v.44 no.2
• /
• pp.286-299
• /
• 2022

Human activity recognition in real time is a challenging task. Recently, a plethora of studies has been proposed using deep learning architectures. The implementation of these architectures requires the high computing power of the machine and a massive database. However, handcrafted features-based machine learning models need less computing power and very accurate where features are effectively extracted. In this study, we propose a handcrafted model based on three-dimensional sequential skeleton data. The human body skeleton movement over a frame is computed through joint positions in a frame. The joints of these skeletal frames are projected into two-dimensional space, forming a "movement polygon." These polygons are further transformed into a one-dimensional space by computing amplitudes at different angles from the centroid of polygons. The feature vector is formed by the sampling of these amplitudes at different angles. The performance of the algorithm is evaluated using a support vector machine on four public datasets: MSR Action3D, Berkeley MHAD, TST Fall Detection, and NTU-RGB+D, and the highest accuracies achieved on these datasets are 94.13%, 93.34%, 95.7%, and 86.8%, respectively. These accuracies are compared with similar state-of-the-art and show superior performance.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2021.06a
• /
• pp.318-321
• /
• 2021

최근 정보통신의 발달과 함께 인터넷에 접속하는 사용자 수와 그에 따른 비디오 데이터의 전송량이 늘어나는 추세이다. 이렇게 늘어나는 많은 비디오 데이터를 관리하고 분석하기 위해서 최근에는 딥 러닝 기법을 많이 활용하게 된다. 일반적으로 비디오 데이터에 딥 러닝 모델을 학습할 때 컴퓨터 자원의 한계로 인해 전체 비디오 프레임에서 균등한 간격 또는 무작위로 프레임을 선택하는 방법을 많이 사용한다. 하지만 학습에 사용되는 비디오 데이터는 항상 시간 축에 따라 같은 문맥을 담고 있는 Trimmed 비디오라고 가정할 수가 없다. 만약 같지 않은 문맥을 지닌 Untrimmed 비디오에서 균등한 간격 또는 무작위로 프레임을 선택해서 사용하게 된다면 비디오의 범주와 관련이 없는 프레임이 샘플링 될 가능성이 있기 때문에 모델의 학습 및 최적화에 전혀 도움이 되지 않는다. 이를 해결하기 위해 우리는 각 비디오 프레임에서 심층 특징을 추출하여 평균값을 계산하고 이와 각 추출된 심층특징들과 코사인 유사도를 계산해서 얻은 유사도 점수를 바탕으로 Untrimmed 비디오에서 의미 있는 비디오 프레임을 추출하는 기법을 제안한다. 그리고 Untrimmed 비디오로 구성된 데이터셋으로 유명한 ActivityNet 데이터셋에 대해서 대표적인 2가지 프레임 샘플링 방식(균등한 간격, 무작위)과 비교하여 우리가 제안하는 기법이 Untrimmed 비디오에서 효과적으로 비디오의 범주에 해당하는 의미 있는 프레임 추출이 가능함을 보일 것이다. 우리가 실험에 사용한 코드는 https://github.com/titania7777/VideoFrameSampler에서 확인할 수 있다.