• 융합신호처리학회논문지
• /
• 제21권1호
• /
• pp.38-48
• /
• 2020

기존에 건조되어 있거나, 신규로 건조되고 있는 선저를 검사하기 위해선 전문 잠수부가 직접 수중 하부에서 육안으로 검사를 한다. 하지만 직접 사람이 하는 일이기 때문에 인명사고, 충돌사고 등 많은 위험이 뒤따른다. 이를 해결하기 위해 Visual Inspection을 위한 선박 검사용 수중 드론 개발이 필요하다. 수중 드론에 적용되는 기술과 각 부품의 용도와 제작 과정, 펌웨어 개발과 같은 제작 방법에 관해 기술하였으며 수중에서 드론 자체의 주행 능력과 크롤러를 이용한 주행 능력을 측정하여 차이를 비교해보고 위치 추적 장치 테스트를 통해 실제 위치와의 오차를 확인하였다. 본 연구를 통해 제작된 수중 드론을 작업 현장에서 사용하였을 때 인명사고를 예방하고 경제적 효과와 안정성을 얻을 것으로 기대된다.

• 한국농업기계학회:학술대회논문집
• /
• 한국농업기계학회 2000년도 THE THIRD INTERNATIONAL CONFERENCE ON AGRICULTURAL MACHINERY ENGINEERING. V.III
• /
• pp.569-577
• /
• 2000

The position of tobacco leaves is affluence to the quality. To evaluate its quality, sample leaves was collected according to the position of attachment. In Korea, the position was divided into four classes such as high, middle, low and inside positioned leaves. Until now, the grade of standard sample was determined by human expert from korea ginseng and tobacco company. Many research were done by the chemical and spectrum analysis using NIR and computer vision. The grade of tobacco leaves mainly classified into 5 grades according to the attached position and its chemical composition. In high and low positioned leaves shows a low level grade under grade 3. Generally, inside and medium positioned leaf has a high level grade. This is the basic research to develop a real time tobacco leaves grading system combined with portable NIR spectrum analysis system. However, this research just deals with position recognition and grading using the color machine vision. The RGB color information was converted to HSI image format and the sample was all investigated using the bundle of tobacco leaves. Quality grade and position recognition was performed through well known general error back propagation neural network. Finally, the relationship about attached leaf position and its grade was analyzed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권6호
• /
• pp.3144-3164
• /
• 2019

Face recognition systems have several potential applications, such as security and biometric access control. Ongoing research is focused to develop a robust face recognition algorithm that can mimic the human vision system. Face pose, non-uniform illuminations, and low-resolution are main factors that influence the performance of face recognition algorithms. This paper proposes a novel method to handle the aforementioned aspects. Proposed face recognition algorithm initially uses 68 points to locate a face in the input image and later partially uses the PCA to extract mean image. Meanwhile, the AdaBoost and the LDA are used to extract face features. In final stage, classic nearest centre classifier is used for face classification. Proposed method outperforms recent state-of-the-art face recognition algorithms by producing high recognition rate and yields much lower error rate for a very challenging situation, such as when only frontal ($0^{\circ}$) face sample is available in gallery and seven poses ($0^{\circ}$, ${\pm}30^{\circ}$, ${\pm}35^{\circ}$, and ${\pm}45^{\circ}$) as a probe on the LFW and the CMU Multi-PIE databases.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.80-80
• /
• 2000

The GPS can provide accurate position information on the earth. But GPS receiver can't give position information inside buildings. DR(Dead-Reckoning) or INS(Inertial Navigation System) gives position information continuously indoors as well as outdoors, because they do not depend on the external navigation information. But in general, the inertial sensors severely suffer from their drift errors, the error of these navigation system increases with time. GPS and DR sensors can be integrated together with Kalman filter to overcome these problems. In this paper, we developed a personal navigation system which can be carried by person, using GPS and electronic pedometer. The person's footstep is detected by an accelerometer installed in vertical direction and the direction of movement is sensed by gyroscope and magnetic compass. In this case the step size is varying with person and changing with circumstance, so determining step size is the problem. In order to calculate the step size of detected footstep, the neural network method is used. The teaming pattern of the neural network is determined by human walking pattern data provided by 3-axis accelerometer and gyroscope. We can calculate person's location with displacement and heading from this information. And this neural network method that calculates step size gives more improved position information better than fixed step size.

• Nuclear Engineering and Technology
• /
• 제55권9호
• /
• pp.3277-3290
• /
• 2023

Nuclear power plant (NPP) operations with multiple objectives and devices are still performed manually by operators despite the potential for human error. These operations could be automated to reduce the burden on operators; however, classical approaches may not be suitable for these multi-objective tasks. An alternative approach is deep reinforcement learning (DRL), which has been successful in automating various complex tasks and has been applied in automation of certain operations in NPPs. But despite the recent progress, previous studies using DRL for NPP operations have limitations to handle complex multi-objective operations with multiple devices efficiently. This study proposes a novel DRL-based approach that addresses these limitations by employing a continuous action space and straightforward binary rewards supported by the adoption of a soft actor-critic and hindsight experience replay. The feasibility of the proposed approach was evaluated for controlling the pressure and volume of the reactor coolant while heating the coolant during NPP startup. The results show that the proposed approach can train the agent with a proper strategy for effectively achieving multiple objectives through the control of multiple devices. Moreover, hands-on testing results demonstrate that the trained agent is capable of handling untrained objectives, such as cooldown, with substantial success.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1458-1463
• /
• 2004

Tracking is one of the most important pre-required task for many application such as human-computer interaction through gesture and face recognition, motion analysis, visual servoing, augment reality, industrial assembly and robot obstacle avoidance. Recently, 3D information of object is required in realtime for many aforementioned applications. 3D tracking is difficult problem to solve because during the image formation process of the camera, explicit 3D information about objects in the scene is lost. Recently, many vision system use stereo camera especially for 3D tracking. The 3D feature based tracking(3DFBT) which is on of the 3D tracking system using stereo vision have many advantage compare to other tracking methods. If we assumed the correspondence problem which is one of the subproblem of 3DFBT is solved, the accuracy of tracking depends on the accuracy of camera calibration. However, The existing calibration method based on accurate camera model so that modelling error and weakness to lens distortion are embedded. Therefore, this thesis proposes 3D feature based tracking method using SVM which is used to solve reconstruction problem.

• International Journal of Contents
• /
• 제14권1호
• /
• pp.39-44
• /
• 2018

Today, cars have developed into intelligent automobiles that combine advanced control equipment and IT technology to provide driving assistance and convenience to users. These vehicles provide infotainment services to the driver, but this does not improve the safety of the driver. Accordingly, V2X communication, which forms a network between a vehicle and a vehicle, between a vehicle and an infrastructure, or between a vehicle and a human, is drawing attention. Therefore, various techniques for improving channel estimation performance without changing the IEEE 802.11p standard have been proposed, but they do not satisfy the packet error rate (PER) performance required by the C-ITS service. In this paper, we analyze existing channel estimation techniques and propose a new channel estimation scheme that achieves better performance than existing techniques. It does this by applying the updated matrix for the data pilot symbol to the construct data pilot (CDP) channel estimation scheme and by further performing the interpolation process in the frequency domain. Finally, through simulations based on the IEEE 802.11p standard, we confirmed the performance of the existing channel estimation schemes and the proposed channel estimation scheme by coded PER.

• 전기전자학회논문지
• /
• 제20권2호
• /
• pp.184-187
• /
• 2016

본 논문에서는 빌딩 자동제어용 센서 및 신호의 듀플리케이터(Duplicator)를 개발한다. 개발된 듀플리케이터는 빌딩 자동제어에 사용되는 여러 개의 센서를 원거리에 있는 빌딩자동화기기에 직접 연결하지 않고 데이터 수집장치와 데이터 송신장치간의 논리적 통신방식을 사용하여 센서 데이터를 그대로 복원한다. 이때 센서의 신호가 원거리로 전송될 때 발생될 수 있는 전기적인 오차를 줄일 수 있고 자동제어의 시공원가를 절감시킬 수 있다. 또한 논리적 통신방식으로 개방형 프로토콜을 사용하기 때문에 상용 HMI와 호환이 되어 확장성을 보장한다. 개발된 듀플리케이터의 성능을 평가하기 위하여 실제 환경에서 시운전을 수행하여 정상 동작함을 보였다. 또한 측정 오차율, 동작 온도, 동작 습도 등에 대하여 공인 시험기관의 장비를 사용하여 실험한 결과 우수한 성능을 나타내었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제5권10호
• /
• pp.1783-1798
• /
• 2011

Recently, a cryptographic construct, called fuzzy vault, has been proposed for crypto-biometric systems, and some implementations for fingerprint have been reported to protect the stored fingerprint template by hiding the fingerprint features. In this paper, we implement the fuzzy fingerprint vault, combining fingerprint verification and fuzzy vault scheme to protect fingerprint templates. To implement the fuzzy fingerprint vault as a complete system, we have to consider several practical issues such as automatic fingerprint alignment, verification accuracy, execution time, error correcting code, etc. In addition, to protect the fuzzy fingerprint vault from the correlation attack, we propose an approach to insert chaffs in a structured way such that distinguishing the fingerprint minutiae and the chaff points obtained from two applications is computationally hard. Based on the experimental results, we confirm that the proposed approach provides higher security than inserting chaffs randomly without a significant degradation of the verification accuracy, and our implementation can be used for real applications.

• Genomics & Informatics
• /
• 제14권4호
• /
• pp.187-195
• /
• 2016

In genetic association studies with high-dimensional genomic data, multiple group testing procedures are often required in order to identify disease/trait-related genes or genetic regions, where multiple genetic sites or variants are located within the same gene or genetic region. However, statistical testing procedures based on an individual test suffer from multiple testing issues such as the control of family-wise error rate and dependent tests. Moreover, detecting only a few of genes associated with a phenotype outcome among tens of thousands of genes is of main interest in genetic association studies. In this reason regularization procedures, where a phenotype outcome regresses on all genomic markers and then regression coefficients are estimated based on a penalized likelihood, have been considered as a good alternative approach to analysis of high-dimensional genomic data. But, selection performance of regularization procedures has been rarely compared with that of statistical group testing procedures. In this article, we performed extensive simulation studies where commonly used group testing procedures such as principal component analysis, Hotelling's $T^2$ test, and permutation test are compared with group lasso (least absolute selection and shrinkage operator) in terms of true positive selection. Also, we applied all methods considered in simulation studies to identify genes associated with ovarian cancer from over 20,000 genetic sites generated from Illumina Infinium HumanMethylation27K Beadchip. We found a big discrepancy of selected genes between multiple group testing procedures and group lasso.