• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.6
• /
• pp.1263-1268
• /
• 2018

In this paper, a video analysis is proposed to implement video surveillance system with deep learning object detection and probabilistic data association filter for tracking multiple objects, and suggests its implementation using GPU. The proposed video analysis technique involves object detection and object tracking sequentially. The deep learning network architecture uses ResNet for object detection and applies probabilistic data association filter for multiple objects tracking. The proposed video analysis technique can be used to detect intruders illegally trespassing any restricted area or to count the number of people entering a specified area. As a results of simulations and experiments, 48 channels of videos can be analyzed at a speed of about 27 fps and real-time video analysis is possible through RTSP protocol.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.5C
• /
• pp.355-364
• /
• 2012

Target processing mechanisms are necessary to collect target information, real-time data fusion, and tactical environment recognition for cooperative engagement ability. Among these mechanisms, the target tracking starts from predicting state of speed, acceleration, and location by using sensors' measurements. However, it can be a problem to give the reliability because the measurements have a certain uncertainty. Thus, a technique which uses multiple sensors is needed to detect the target and increase the reliability. Also, data fusion technique is necessary to process the data which is provided from heterogeneous sensors for target tracking. In this paper, a target tracking algorithm is proposed based on probabilistic data association(PDA) by fusing radar and ESM sensor measurements. The radar sensor's azimuth and range measurements and the ESM sensor's bearing-only measurement are associated by the measurement fusion method. After gating associated measurements, state estimation of the target is performed by PDA filter. The simulation results show that the proposed algorithm provides improved estimation under linear and circular target motions.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.18-22
• /
• 2019

FAQ 분류는 자주 묻는 질문을 범주화하고 사용자 질의에 대해 가장 유사한 클래스를 추론하는 방식으로 진행된다. FAQ 데이터셋은 클래스가 다수 존재하기 때문에 클래스 간 포함 및 연관 관계가 존재하고 특정 데이터가 서로 다른 클래스에 동시에 속할 수 있다는 특징이 있다. 그러나 최근 FAQ 분류는 다중 클래스 분류 방법론을 적용하는 데 그쳤고 FAQ 데이터셋의 특징을 모델에 반영하는 연구는 미미했다. 현 분류 방법론은 이러한 FAQ 데이터셋의 특징을 고려하지 못하기 때문에 정답으로 해석될 수 있는 예측도 오답으로 여기는 경우가 발생한다. 본 논문에서는 신뢰성이 부족한 FAQ 데이터셋에서도 분류를 잘 하기 위해 손실 함수를 조정하는 정규화 기법을 소개한다. 이 정규화 기법은 클래스 간 포함 및 연관 관계를 반영할 수 있도록 오답을 예측한 경우에도 예측 강도에 비례하여 손실을 줄인다. 이는 오답을 높은 확률로 예측할수록 데이터의 신뢰성이 낮을 가능성이 크다고 판단하여 학습을 강하게 하지 않게 하기 위함이다. 실험을 위해서는 다중 클래스 분류에서 가장 좋은 성능을 보이고 있는 모형인 BERT를 이용했으며, 비교 실험을 위한 정규화 방법으로는 통상적으로 사용되는 라벨 스무딩을 채택했다. 실험 결과, 본 연구에서 제안한 방법은 기존 방법보다 성능이 개선되고 보다 안정적으로 학습이 된다는 것을 확인했으며, 데이터의 신뢰성이 부족한 상황에서 효과적으로 분류를 수행함을 알 수 있었다.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.1
• /
• pp.1-8
• /
• 2016

In the recently, violent crime and accidental crime has been generated continuously. Consequently, people anxiety has been heightened. The Closed Circuit Television (CCTV) has been used to ensure the security and evidence for the crimes. However, the video captured from CCTV has being used in the post-processing to apply to the evidence. In this paper, we propose a flexible multi-level models for estimating whether dangerous behavior and the environment and context for pedestrians. The situation analysis builds the knowledge for the pedestrians tracking. Finally, the decision step decides and notifies the threat situation when the behavior observed object is determined to abnormal behavior. Thereby, tracking the behavior of objects in a multi-region, it can be seen that the risk of the object behavior. It can be predicted by the behavior prediction of crime.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.1-7
• /
• 2021

In this paper, we propose a feature selection technique for multi-label classification. Many existing feature selection techniques have selected features by calculating the relation between features and labels such as a mutual information scale. However, since the mutual information measure requires a joint probability, it is difficult to calculate the joint probability from an actual premise feature set. Therefore, it has the disadvantage that only a few features can be calculated and only local optimization is possible. Away from this regional optimization problem, we propose a feature selection technique that constructs a low-rank space in the entire given feature space and selects features with sparsity. To this end, we designed a regression-based objective function using Nuclear norm, and proposed an algorithm of gradient descent method to solve the optimization problem of this objective function. Based on the results of multi-label classification experiments on four data and three multi-label classification performance, the proposed methodology showed better performance than the existing feature selection technique. In addition, it was showed by experimental results that the performance change is insensitive even to the parameter value change of the proposed objective function.

• Journal of the Society of Naval Architects of Korea
• /
• v.60 no.4
• /
• pp.248-255
• /
• 2023

In order to implement the autonomous navigation function, it is essential to track an object within a certain radius of the ship's route. This paper proposes the Multiple Probabilistic Data Association Filter (MPDAF), which can track multiple ships by extending Probabilistic Data Association Filter (PDAF), an existing single object tracking algorithm, using radar data obtained from real marine environments. The proposed MPDAF algorithm was developed to address the problem of tracking multiple objects in a complex environment where there can be significant uncertainty in the number and identification of objects to be tracked. Using real-world radar data provided by the German aerospace center (DLR), it has been verified that the proposed algorithm can track a large number of objects with a small position error.

• Journal of Internet Computing and Services
• /
• v.21 no.2
• /
• pp.49-55
• /
• 2020

Information theoretic learning (ITL) methods based on random symbols (RS) use a set of random symbols generated according to a target distribution and are designed nonparametrically to minimize the cost function of the Euclidian distance between the target distribution and the input distribution. One drawback of the learning method is that it can not utilize the input power statistics by employing a constant stepsize for updating the algorithm. In this paper, it is revealed that firstly, information potential input (IPI) plays a role of input in the cost function-derivative related with information potential output (IPO) and secondly, input itself does in the derivative related with information potential error (IPE). Based on these observations, it is proposed to normalize the step-size with the statistically varying power of the two different inputs, IPI and input itself. The proposed algorithm in an communication environment of impulsive noise and multipath fading shows that the performance of mean squared error (MSE) is lower by 4dB, and convergence speed is 2 times faster than the conventional methods without step-size normalization.

• Korean Journal of Psychosomatic Medicine
• /
• v.30 no.1
• /
• pp.46-53
• /
• 2022

Objectives : This study aimed to assess the possible association of dental pain with suicidal ideation among adolescents by analysing data from the 2018 Korean Youth Risk Behavior Survey, a nationwide online survey. Methods : Of 62,823 adolescent middle and high school students in Korea, 60,040 participants were selected for analysis, after excluding cases with missing values. Participants were given a questionnaire about their self-evaluation of health including dental pain and suicidal ideation. Logistic regression analysis demonstrated the relationships between dental pain and suicidal ideation after controlling for potential confounding factors. Results : The proportion of Korean adolescents reporting suicidal ideation was 13.3%. The proportion of adolescents who experienced dental pain was 23.4%. Compared to adolescents who did not report dental pain, adolescents who reported experiencing dental pain were significantly more likely to experience suicidal ideation (OR=1.94, p<0.001). In two multivariate models, the relationships between dental pain and suicidal ideation (AOR=1.24, p<0.001) were statistically significant. Conclusions : Dental pain was associated with increased risk of suicidal ideation among Korean adolescents, even when controlling for sociodemographic factors and other health conditions. It is necessary to consider screening adolescent patients who present with dental pain for suicidal ideation.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.1
• /
• pp.62-72
• /
• 2010

PM OLED is used in many applications as one of the display for the next generation. The most essential problems are the power dissipation and the short life time in applying PM OLED into a commercial application. Many efforts are made in developing the panel and in improving the circuit for expanding the current market wider. The life time in PM OLED is expanded by lessening the power dissipation of the circuit for the magnitude of the driving current is lowered. It is possible to minimize the power dissipation from improving the driving technology. The classical technology, Row-to-Row driving, is that row is selected one by one while applying the column current input individually. The multi-line driving is a new technology which is to select multiple rows simultaneously while applying the column current as a whole. However, the solution of the multi-line driving is NP-complete problem. The efficiency is dependant on the sort of picture and the driving condition. This paper presents the new efficient multi-line driving which is that the multiple lines are selected by applying column current together after grouping the simultaneous driving group applying the gnew efficient muthe coi-line dr coefficient. Bengrouping the several rows which has the higher coi-line dr coefficient, the more efficient driving is realized to present the high quality image and to lessen the power dissipation and to stretch the life time in the PM OLED.