• 한국컴퓨터정보학회논문지
• /
• 제22권12호
• /
• pp.55-61
• /
• 2017

In this paper, we propose a novel tracking method using target separation and detection that are based on discriminative correlation filter (DCF), which is studied a lot recently. 'Retainability' is one of the most important factor of tracking. There are some factors making retainability of tracking worse. Especially, fast movement and occlusion of a target frequently occur in image data, and when it happens, it would make target lost. As a result, the tracking cannot be retained. For maintaining a robust tracking, in this paper, separation of a target is used so that normal tracking is maintained even though some part of a target is occluded. The detection algorithm is executed and find new location of the target when the target gets out of tracking range due to occlusion of whole part of a target or fast movement speed of a target. A variety of experiments with various image data sets are conducted. The algorithm proposed in this paper showed better performance than other conventional algorithms when fast movement and occlusion of a target occur.

• 한국통신학회논문지
• /
• 제38B권9호
• /
• pp.736-743
• /
• 2013

적외선 탐색 및 추적 시스템에서 원거리에 표적이 존재할 경우 표적의 크기가 매우 작고, 해무와 같은 클러터와 다양한 센서 잡음으로 인해 표적의 검출이 매우 어렵다. 특히 표적의 화소 값과 유사한 잡음이나 클러터가 존재하는 경우 일반적인 임계화 기법을 적용하는 경우 표적의 오검출 위험이 매우 높다. 이러한 이유로 본 논문에서는 영상의 밝기 정보와 표적에 대한 사전 정보를 이용하여 최적의 표적 검출 결과를 도출하기 위한 적응적 임계화 기법을 제안한다. 소형 표적을 강조하기 위하여 인간 시각 시스템을 반영한 CSF(Contrast Sensitivity Function)를 적용하고, 표적이 강조된 영상에서 영상의 밝기 정보와 거리 정보를 이용하여 표적을 검출한다. 다양한 환경 조건에서 획득된 적외선 영상에 대한 실험 결과들은 제안 알고리즘의 견실한 성능을 보여준다.

• 산업경영시스템학회지
• /
• 제43권4호
• /
• pp.229-235
• /
• 2020

Investors aim to maximize the return rate for their own investment, utilizing various information as possible as they can access. However those investors, especially individual investors, have limitations of interpretation of the domain-specific information or even the acquisition of the information itself. Thus, individual investors tend to make decision affectively and frequently, which may cause a loss in returns. This study aims to analyze analysts' target price and to suggest the strategy that could maximize individual's return rate. Most previous literature revealed that the optimistic bias exists in the analysts' target price and it is also confirmed in this study. In this context, this study suggests the upper limit of target rate of returns and the optimal value named 'alpha(α)' which performs the adjustment of proposed target rate to maximize excess earning returns eventually. To achieve this goal, this study developed an optimization problem using linear programming. Specifically, when the analysts' proposed target rate exceeds 30%, it could be adjusted to the extent of 59% of its own target rate. As apply this strategy, the investors could achieve 1.2% of excess earning rate on average. The result of this study has significance in that the individual investors could utilize analysts' target price practically.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권10호
• /
• pp.5112-5128
• /
• 2016

Tracking human body is an important problem in computer vision field. Tracking failures caused by occlusion can lead to wrong rectification of the target position. In this paper, a robust human tracking algorithm is proposed to address the problem of occlusion, rotation and improve the tracking accuracy. It is based on Tracking-Learning-Detection framework. The key auxiliary information is used in the framework which motivated by the fact that a tracking target is usually embedded in the context that provides useful information. First, face localization method is utilized to find key face location information. Second, the relative position relationship is established between the auxiliary information and the target location. With the relevant model, the key face information will get the current target position when a target has disappeared. Thus, the target can be stably tracked even when it is partially or fully occluded. Experiments are conducted in various challenging videos. In conjunction with online update, the results demonstrate that the proposed method outperforms the traditional TLD algorithm, and it has a relatively better tracking performance than other state-of-the-art methods.

• 지능정보연구
• /
• 제9권1호
• /
• pp.195-209
• /
• 2003

데이터마이닝의 목적은 알려지지 않은 유용한 정보를 얻어 경영에 활용하려는 것이다. 데이터마이닝의 한 기법인 연관규칙도 마찬가지의 목적을 가지고 있다. 하지만 연관규칙 기법으로 생성된 결과는 모두 같기 때문에 타겟 마케팅에 적용 시 그 마케팅 전략은 같을 수밖에 없다. 본 논문에서는 연관규칙을 이용하여 타겟 마케팅에 적용 시에 데이터마이닝의 목적에 부합하는 새로운 규칙을 제안한다. 새로운 규칙이란 연관규칙과 같이 고객이 구매한 항목에 대해 관심을 가져 구매 항목간의 규칙을 생성하는 것뿐만 아니라 구매하지 않은 항목에 대해서도 관심을 가짐으로써 구매하지 않은 항목간의 규칙을 생성하여 타겟 마케팅에 필요한 정보를 생성하는 것을 말한다. 이러한 정보를 이용하면 타겟 항목을 바로 마케팅 하는 전략뿐만 아니라 타겟 항목을 판매하기 위하여 다른 항목을 마케팅 하는 전략도 가능하게 된다. 그 이유는 새로운 규칙 에 의해 생성된 다른 항목의 마케팅이 타겟 항목의 판매를 촉진시키기 때문이다. 본 논문에서는 타겟 마케팅에 적용하기 위한 새로운 규칙과 전략을 설명하고 예제를 통하여 실제 데이터베이스에서 기존 전략 외에 새로운 전략을 생성시키는 과정을 설명한다.

• International Journal of Internet, Broadcasting and Communication
• /
• 제12권3호
• /
• pp.1-8
• /
• 2020

This paper studied a method of estimating target information using a radar in wireless communication. Position information on the target can be estimated angle, distance and velocity. The velocity information can be estimated since the Doppler frequency is changed in the moving target. The signal incident on the receiving array antenna is multiplied by the delay time and the reference signal to represent the output signal. This output signal is estimated by applying FFT (Fast Fourier Transform) after calculating signal correlation through correlation integrator. Since the output signal must be calculated within the correlator, it should be processed with the Dwell time. The correlation signal of the correlation integrator outside this Dwell time is indicated by the velocity measurement error. The FFT is applied to the signal that has passed through the correlated integrator in order to estimate the distance of the signal. The Doppler resolution must be improved because the FFT estimates target information using the Doppler information. The Doppler resolution decreases with increasing the integration time. The velocity information estimation should have no spread of the velocity. As a result of the simulation, there was no spread of the target velocity in this study.

• 한국컴퓨터정보학회논문지
• /
• 제21권3호
• /
• pp.65-71
• /
• 2016

To reduce the expenses for development a novel drug, systems biology has been studied actively. Target prediction, a part of systems biology, contributes to finding a new purpose for FDA(Food and Drug Administration) approved drugs and development novel drugs. In this paper, we propose a classification model for predicting novel target genes based on relation between target genes and disease related genes. After collecting known target genes from TTD(Therapeutic Target Database) and disease related genes from OMIM(Online Mendelian Inheritance in Man), we analyzed the effect of target genes on disease related genes based on PPI(Protein-Protein Interactions) network. We focused on the distinguishing characteristics between known target genes and random target genes, and used the characteristics as features for building a classifier. Because our model is constructed using information about only a disease and its known targets, the model can be applied to unusual diseases without similar drugs and diseases, while existing models for finding new drug-disease associations are based on drug-drug similarity and disease-disease similarity. We validated accuracy of the model using LOOCV of ten times and the AUCs were 0.74 on Alzheimer's disease and 0.71 on Breast cancer.

• 한국산학기술학회논문지
• /
• 제11권3호
• /
• pp.978-983
• /
• 2010

무선 센서 네트워크에서 음향 표적의 식별은 환경 감시, 침입 감시, 다중 표적 분리 등에서 많이 연구된다. 무선 센서 네트워크의 센서 노드에서 사용하는 기존의 신호 처리기법은 표적으로부터 수신된 신호의 에너지를 계산하여 표적의 존재 유무만을 기지국으로 전송하는 방법과 수신 신호를 압축하여 전송하는 방법이 많이 사용되었다. 전자의 경우 표적의 감시를 위한 무선 센서 네트워크에서는 표적의 정보가 한정적이므로 적합하지 않고 후자의 경우는 센서 노드에서의 신호처리 및 전송에 소모되는 에너지가 높아 센서의 생존시간이 줄어들게 된다. 따라서 본 논문에서는 표적의 감시를 위한 무선 센서 네트워크에서 필요한 시간정보와 표적의 주파수 정보를 포함하는 센서 노드에서의 특징 추출 기법을 제안한다. 본 논문에서는 웨이블릿 변환을 이용하여 추출된 웨이블릿 상수에서 표적의 시간 정보와 잡음이 제거된 표적의 식별 정보를 추출함으로서 센서 노드에서 에너지 효율적인 신호처리를 구현하고 추출된 특징을 전송하여 통신에 소모되는 에너지를 원신호 대비 28%로 줄이는 알고리듬을 제안한다.

• 제어로봇시스템학회논문지
• /
• 제21권9호
• /
• pp.843-848
• /
• 2015

In this paper, The ARR-TSE (Automatic Range Restore - Triangulation based target Size Estimator) algorithm is presented for IIR (Imaging Infrared) seeker. The target size is important information for the IIR target tracking. The TSE (Triangulation based target Size Estimator) algorithm has suitable performance to estimate target size for static IIR target. but, the performance of the algorithm can be decreased by noise. In order to decrease influence of noise, we propose the ARR-TSE algorithm. The performance of proposed method is tested at target intercept scenario. The simulation results show that the proposed algorithm has the accurate target size estimating performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권12호
• /
• pp.4292-4307
• /
• 2021

For the research field of night infrared surveillance video, the target imaging in the video is easily affected by the light due to the characteristics of the active infrared camera and the classical ViBe algorithm has some problems for moving target extraction because of background misjudgment, noise interference, ghost shadow and so on. Therefore, an improved ViBe algorithm (I-ViBe) for moving target extraction in night infrared surveillance video is proposed in this paper. Firstly, the video frames are sampled and judged by the degree of light influence, and the video frame is divided into three situations: no light change, small light change, and severe light change. Secondly, the ViBe algorithm is extracted the moving target when there is no light change. The segmentation factor of the ViBe algorithm is adaptively changed to reduce the impact of the light on the ViBe algorithm when the light change is small. The moving target is extracted using the region growing algorithm improved by the image entropy in the differential image of the current frame and the background model when the illumination changes drastically. Based on the results of the simulation, the I-ViBe algorithm proposed has better robustness to the influence of illumination. When extracting moving targets at night the I-ViBe algorithm can make target extraction more accurate and provide more effective data for further night behavior recognition and target tracking.