• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.329-331
• /
• 2018

지속적인 딥러닝 기반의 영상처리 기술의 발전으로 객체분류나 객체검출 문제에 대해서 뛰어난 성능 보이고 있다. 하지만 객체추적 문제에서는 성능이 좋은 추적기는 실시간 동작이 불가능하고 딥러닝 기반의 객체추적도 단일 객체에만 고려한 기법이 많기 때문에 개선할 필요가 있다. 전처리로 검출된 객체영역과 kalman filter를 통해 예측된 추적영역 간의 embedding feature 비교를 통해 동일인물인지 판단하여 고유 ID를 부여하고 추적한다. 객체끼리 교차하거나 가려지는 상황에서 추적을 실패하게 되는데 이 후에 지속적인 추적을 위해 IoU 비교를 통해 후보 추적기로 남겨두는 과정을 거친다. 실험 결과 실시간 동작여부와 객체끼리 교차하거나 프레임 밖으로 나갔다가 다시 나타나는 경우에도 추적이 가능함을 확인하였다.

• Journal of Advanced Navigation Technology
• /
• v.27 no.6
• /
• pp.884-887
• /
• 2023

Recently, research has been actively conducted to overcome the limitations of high-priced single sensors and reduce costs through the convergence of low-cost multi-variable sensors. This paper estimates state variables through asynchronous Kalman filters constructed using CVXPY and uses Cvxpylayers to compare and learn state variables estimated from CVXPY with true value data to estimate filter parameters of low-cost sensors fusion.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.4 s.36
• /
• pp.47-56
• /
• 2005

With the development of wearable(ubiquitous) computers, those traditional interfaces between human and computers gradually become uncomfortable to use, which directly leads to a requirement for new one. In this paper, we study on a new interface in which computers try to recognize the gesture of human through a digital camera. Because the method of recognizing hand gesture through camera is affected by the surrounding environment such as lighting and so on, the detector should be a little sensitive. Recently, Viola's detector shows a favorable result in face detection. where Adaboost learning algorithm is used with the Haar features from the integral image. We apply this method to hand area detection and carry out comparative experiments with the classic method using skin color. Experimental results show Viola's detector is more robust than the detection method using skin color in the environment that degradation may occur by surroundings like effect of lighting.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.8 no.4
• /
• pp.43-52
• /
• 1998

We extract a hand region by using color information, which is an important feature for human vision to distinguish objects. Because pixel values in images are changed according to the luminance and lighting source, it is difficult to extract a hand region exactly without previous knowledge. We generate a hand skin model at learning stage, and extract a hand region from images by using the model. We also use a Kalman filter to consider changes of pixel values in a hand skin model. A Kalman filter restricts a search area for extracting a hand region at next frame also. The validity of the proposed method is proved by implementing the hand-region acquisition module.

• Journal of Digital Contents Society
• /
• v.10 no.2
• /
• pp.291-298
• /
• 2009

In this paper we proposed a vision based realtime hand gesture recognition method. To extract skin color, we translate RGB color space into YCbCr color space and use CbCr color for the final extraction. To find the center of extracted hand region we apply practical center point extraction algorithm. We use Kalman filter to tracking hand region and use HMM(Hidden Markov Model) algorithm (learning 6 type of hand gesture image) to recognize it. We demonstrated the effectiveness of our algorithm by some experiments.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.5
• /
• pp.289-295
• /
• 2021

Recently, owing to the development of autonomous vehicles, research on precisely determining the position of a moving object has been actively conducted. Previous research mainly used the fusion of GNSS/IMU (Global Positioning System / Inertial Navigation System) and sensors attached to the vehicle through a Kalman filter. However, in recent years, new technologies have been used to determine the location of a moving object owing to the improvement in computing power and the advent of deep learning. Various techniques using RNN (Recurrent Neural Network), LSTM (Long Short-Term Memory), and NARX (Nonlinear Auto-Regressive eXogenous model) exist for such learning-based positioning methods. The purpose of this study is to compare the precision of existing filter-based sensor fusion technology and the NARX-based method in case of GNSS signal blockages using simulation data. When the filter-based sensor integration technology was used, an average horizontal position error of 112.8 m occurred during 60 seconds of GNSS signal outages. The same experiment was performed 100 times using the NARX. Among them, an improvement in precision was confirmed in approximately 20% of the experimental results. The horizontal position accuracy was 22.65 m, which was confirmed to be better than that of the filter-based fusion technique.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.3
• /
• pp.42-52
• /
• 2022

Bus Information System (BIS) services are expanding nationwide to small and medium-sized cities, including large cities, and user satisfaction is continuously improving. In addition, technology development related to improving reliability of bus arrival time and improvement research to minimize errors continue, and above all, the importance of information accuracy is emerging. In this study, accuracy performance was evaluated using LSTM, a machine learning method, and compared with existing methodologies such as Kalman filter and neural network. As a result of analyzing the standard error for the actual travel time and predicted values, it was analyzed that the LSTM machine learning method has about 1% higher accuracy and the standard error is about 10 seconds lower than the existing algorithm. On the other hand, 109 out of 162 sections (67.3%) were analyzed to be excellent, indicating that the LSTM method was not entirely excellent. It is judged that further improved accuracy prediction will be possible when algorithms are fused through section characteristic analysis.

• Smart Media Journal
• /
• v.11 no.8
• /
• pp.84-92
• /
• 2022

Precision livestock monitoring promises greater management efficiency for farmers and higher welfare standards for animals. Recent studies on video-based animal activity recognition and tracking have shown promising solutions for understanding animal behavior. To achieve that, surveillance cameras are installed diagonally above the barn in a typical cattle farm setup to monitor animals constantly. Under these circumstances, tracking individuals requires addressing challenges such as occlusion and visual appearance, which are the main reasons for track breakage and increased misidentification of animals. This paper presents a framework for multi-cattle tracking in closed barns with appearance and motion models. To overcome the above challenges, we modify the DeepSORT algorithm to achieve higher tracking accuracy by three contributions. First, we reduce the weight of appearance information. Second, we use an Ensemble Kalman Filter to predict the random motion information of cattle. Third, we propose a supplementary matching algorithm that compares the absolute cattle position in the barn to reassign lost tracks. The main idea of the matching algorithm assumes that the number of cattle is fixed in the barn, so the edge of the barn is where new trajectories are most likely to emerge. Experimental results are performed on our dataset collected on two cattle farms. Our algorithm achieves 70.37%, 77.39%, and 81.74% performance on HOTA, AssA, and IDF1, representing an improvement of 1.53%, 4.17%, and 0.96%, respectively, compared to the original method.

• Structural Engineering and Mechanics
• /
• v.83 no.4
• /
• pp.537-549
• /
• 2022

In this paper, a novel multi-feature model predictive control (MPC) framework with real-time and adaptive performances is proposed for intelligent structural control in which some drawbacks of the algorithm including, complex control rule and non-optimality, are alleviated. Hence, Linear Programming (LP) is utilized to simplify the resulted control rule. Afterward, the Whale Optimization Algorithm (WOA) is applied to the optimal and adaptive tuning of the LP weights independently at each time step. The stochastic control rule is also achieved using Kalman Filter (KF) to handle noisy measurements. The Extreme Learning Machine (ELM) is then adopted to develop a data-driven and real-time control algorithm. The efficiency of the developed algorithm is then demonstrated by numerical simulation of a twenty-story high-rise benchmark building subjected to earthquake excitations. The competency of the proposed method is proven from the aspects of optimality, stochasticity, and adaptivity compared to the KF-based MPC (KMPC) and constrained MPC (CMPC) algorithms in vibration suppression of building structures. The average value for performance indices in the near-field and far-field (El earthquakes demonstrates a reduction up to 38.3% and 32.5% compared with KMPC and CMPC, respectively.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.65-68
• /
• 2004

복잡한 문제 학습을 위해 여러 가지 형태의 모듈라 네트워크의 구조가 제시되어 왔다. 그 중 엑스퍼트 네트워크와 게이팅 네트워크로 구성된 Mixtures of Experts network은 복잡한 문제를 단순한 문제들로 분해하고, 각각의 엑스퍼트 네트워크가 분해된 단순한 문제를 학습하여 결과를 도출함으로써, 국소적 지역해의 위험을 방지하고 보다 정확한 학습을 가능하게 한다. 그러나 엑스퍼트 네트워크의 수렴은 게이팅 네트워크의 수렴에 많은 영향을 받게 되고, 모든 복잡한 데이터에 대한 엑스퍼트 네트워크의 기여도를 학습하는 게이팅 네트워크는 역전파 알고리즘에 의한 학습 방법으로는 수렴 속도가 떨어진다. 본 논문에서는 게이팅 네트워크를 칼만필터로 학습하여 복잡한 문제에 대한 강건성은 유지하고 보다 빠른 수렴이 가능한 방법을 제시하고자한다.