• 한국음향학회지
• /
• 제18권7호
• /
• pp.101-111
• /
• 1999

본 논문에서는 정확한 콘크리트 설계강도 분류를 위해 인공지능 기법에 바탕을 둔 증거축적방법에 의한 초음파신호의 패턴인식방법을 제안하였다. 이를 위해 우선 초음파신호의 특징파라메터로 분산, 영점교차횟수, 평균주파수, 자기회귀모델계수 및 선형 켑스트럼계수를 추출하였다. 추출된 특징파라메터들의 각각의 특성을 알아보고, 하나의 특징파라메터로 설계강도의 정확한 분류가 어렵다는 것을 보였다. 이러한 문제점을 해결하기 위하여 추출된 다수의 특징파라메터들을 이용하여 설계강도 분류를 증거축적방법을 통해 수행하였다. 또한, 이 증거축적방법을 콘크리트 패턴인식에 적용하기 위해 퍼지매핑 함수를 도입하였다. 본 논문에서 제안한 알고리즘이 다수의 특징파라메터들을 효율적으로 이용하여 92%의 패턴인식률을 보였으며, 이는 기존의 패턴 분류 알고리즘보다 콘크리트 설계강도를 보다 정확하게 분류함을 확인하였다.

• Journal of Electrical Engineering and Technology
• /
• 제12권1호
• /
• pp.363-372
• /
• 2017

This paper presents a novel face recognition algorithm based on the deep convolution neural network and key point detection jointed local binary pattern methodology to enhance the accuracy of face recognition. We firstly propose the modified face key feature point location detection method to enhance the traditional localization algorithm to better pre-process the original face images. We put forward the grey information and the color information with combination of a composite model of local information. Then, we optimize the multi-layer network structure deep learning algorithm using the Fisher criterion as reference to adjust the network structure more accurately. Furthermore, we modify the local binary pattern texture description operator and combine it with the neural network to overcome drawbacks that deep neural network could not learn to face image and the local characteristics. Simulation results demonstrate that the proposed algorithm obtains stronger robustness and feasibility compared with the other state-of-the-art algorithms. The proposed algorithm also provides the novel paradigm for the application of deep learning in the field of face recognition which sets the milestone for further research.

• 한국IT서비스학회지
• /
• 제17권1호
• /
• pp.135-153
• /
• 2018

E-Card Service for Child Nutrition Program is one of the main public policy services nowadays. In case of inconvenience during the use of the e-cards, it is recommended to cooperate with related organizations in order to promptly handle and provide guidance, and thoroughly manage child feeding service such as hygiene, nutrition and kindness etc. To do so, it is very important to provide food service that meets local actual conditions and children's needs in a cost effective manner for the underage who are worried about the poorly-fed by understanding the pattern of child feeding e-card service. Hence. this paper aims to investigate how child feeding e-card service efficiently provides meals according to the local situation and children's needs through big data analysis and to propose a method of identifying welfare conditions according to the purpose of service with actual application examples. The results suggest that, first of all, this study is able to judge appropriateness of public institution's policy in a timely and repetitive manner through non-standard data analysis such as Naver News and transaction data. Secondly, this paper proposes a multi-layered analysis framework, which performs online open data analysis to detect policy issues, visualizes retrieval and preprocessing of real data, and performs abnormal pattern recognition. These will be worthy of reference to other similar projects.

• Journal of Power Electronics
• /
• 제11권4호
• /
• pp.449-455
• /
• 2011

Electric drive vehicles (EDV) have received much attention recently because of their environmental and energy benefits. In an EDV, the motor drive system directly influences the performance of the propulsion system. However, the available DC voltage is limited, which limits the maximum speed of the motors. At high speeds, the inverter voltage increases if the square wave (SW) voltage (six-step operation) is used. Although conventional direct torque control (DTC) has several advantages, it cannot work in the six-step mode required in high-speed applications. In this paper, a single-mode seamless DTC for AC motors is proposed. In this scheme, the trajectory of the reference flux changes continuously between circular and hexagonal paths. Therefore, the armature voltage changes smoothly from a high-frequency switching pattern to a square wave pattern without torque discontinuity. In addition, because multi-mode controllers are not used, implementation is more straightforward. Simulation results show the voltage pattern changes smoothly when the motor speed changes, and consequently, torque control without torque discontinuity is possible in the field weakening area even with a six-step voltage pattern.

• 로봇학회논문지
• /
• 제3권2호
• /
• pp.91-98
• /
• 2008

This paper describes the recognition method of moving objects in mobile robot with an omnidirectional camera. The moving object is detected using the specific pattern of an optical flow in omnidirectional image. This paper consists of two parts. In the first part, the pattern of an optical flow is investigated in omnidirectional image. The optical flow in omnidirectional image is influenced on the geometry characteristic of an omnidirectional camera. The pattern of an optical flow is theoretically and experimentally investigated. In the second part, the detection of moving objects is presented from the estimated optical flow. The moving object is extracted through the relative evaluation of optical flows which is derived from the pattern of optical flow. In particular, Focus-Of-Expansion (FOE) and Focus-Of-Contraction (FOC) vectors are defined from the estimated optical flow. They are used as reference vectors for the relative evaluation of optical flows. The proposed algorithm is performed in four motions of a mobile robot such as straight forward, left turn, right turn and rotation. Experimental results using real movie show the effectiveness of the proposed method.

• 전기의세계
• /
• 제14권3호
• /
• pp.10-17
• /
• 1965

This paper presents a method of the voltage-reactive power control in the long and short range operations and introduces a conception, "optimum control pattern." The optimum control pattern, aiming at the over-all system control, is defined as the optimum voltage distribution which minimizes the system operating cost under the conditions that the specified power be supplied and the system voltage be kept within the specified bounds. The following procedure was adopted to obtain this optimum pattern. In the first place, a power system was divided into three blocks, namely, load, substation and generator. Lagrange's, multiplier method is applied to each block in turn, paying attention only at the operating voltage distribution. Phase angles at each bus are then modified so that the continuity of active power is maintained. This procedure may be called "block relaxation method with Lagrange's multipliers." In a long range operation, this control pattern determines the optimum installation capacity of reactive power sources. In a short range operation, it also gives the reference state of real time control and the optimum switching capacity of reactive power souces. The real time control problem is also studied from the standpoint of cooperation of control devices such as generators, shunt capacitors, shunt reactors and ratio load controllers. A proposal for the real time control will contribute to the automation of power system operation in the near future. in the near future.

• Bulletin of the Korean Chemical Society
• /
• 제31권11호
• /
• pp.3371-3381
• /
• 2010

In this study, quantitative and pattern recognition analysis for the quality evaluation of Magnoliae Flos using HPLC/UV was developed. For quantitative analysis, eleven major bioactive lignan compounds were determined. The separation conditions employed for HPLC/UV were optimized using ODS $C_{18}$ column ($250{\times}4.6\;mm$, $5\;{\mu}m$) with isocratic elution of acetonitrile and water with 1% acetic acid as the mobile phase at a flow rate of 1.0 mL/min and a detection wavelength of 278 nm. These methods were fully validated with respect to the linearity, accuracy, precision, recovery, and robustness. The HPLC/UV method was applied successfully to the quantification of eleven major compounds in the extract of Magnoliae Flos. The HPLC analytical method for pattern recognition analysis was validated by repeated analysis of twenty one reference samples corresponding to seven different species of Magnoliae Flos and nine samples purchased from market. The results indicate that the established HPLC/UV method is suitable for the quantitative analysis and quality control of multi-components in Magnoliae Flos.

• Bulletin of the Korean Chemical Society
• /
• 제32권1호
• /
• pp.239-246
• /
• 2011

In this study, quantitative and pattern recognition analysis for the quality evaluation of Cimicifugae Rhizoma using HPLC/UV was developed. For quantitative analysis, three major bioactive phenolic compounds were determined. The separation conditions employed for HPLC/UV were optimized using ODS $C_{18}$ column ($250{\times}4.6mm$, $5{\mu}M$) with isocratic elution of acetonitrile and water with 0.1% phosphoric acid as the mobile phase at a flow rate of 1.0 mL/min and a detection wavelength of 323 nm. These methods were fully validated with respect to the linearity, accuracy, precision, recovery, and robustness. The HPLC/UV method was applied successfully to the quantification of three major compounds in the extract of Cimicifugae Rhizoma. The HPLC analytical method for pattern recognition analysis was validated by repeated analysis of twelve reference samples corresponding to five different species of Cimicifugae Rhizoma and seventeen samples purchased from markets. The results indicate that the established HPLC/UV method is suitable for the quantitative analysis and quality control of multi-components in Cimicifugae Rhizoma.

• Geomechanics and Engineering
• /
• 제1권1호
• /
• pp.53-74
• /
• 2009

In recent years, several computer-aided pattern recognition and data mining techniques have been developed for modeling of soil behavior. The main idea behind a pattern recognition system is that it learns adaptively from experience and is able to provide predictions for new cases. Artificial neural networks are the most widely used pattern recognition methods that have been utilized to model soil behavior. Recently, the authors have pioneered the application of genetic programming (GP) and evolutionary polynomial regression (EPR) techniques for modeling of soils and a number of other geotechnical applications. The paper reviews applications of pattern recognition and data mining systems in geotechnical engineering with particular reference to constitutive modeling of soils. It covers applications of artificial neural network, genetic programming and evolutionary programming approaches for soil modeling. It is suggested that these systems could be developed as efficient tools for modeling of soils and analysis of geotechnical engineering problems, especially for cases where the behavior is too complex and conventional models are unable to effectively describe various aspects of the behavior. It is also recognized that these techniques are complementary to conventional soil models rather than a substitute to them.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.771-789
• /
• 2019

Text detection has been a popular research topic in the field of computer vision. It is difficult for prevalent text detection algorithms to avoid the dependence on datasets. To overcome this problem, we proposed a novel unsupervised text detection algorithm inspired by bootstrap learning. Firstly, the text candidate in a novel form of superpixel is proposed to improve the text recall rate by image segmentation. Secondly, we propose a unique text sample selection model (TSSM) to extract text samples from the current image and eliminate database dependency. Specifically, to improve the precision of samples, we combine maximally stable extremal regions (MSERs) and the saliency map to generate sample reference maps with a double threshold scheme. Finally, a multiple kernel boosting method is developed to generate a strong text classifier by combining multiple single kernel SVMs based on the samples selected from TSSM. Experimental results on standard datasets demonstrate that our text detection method is robust to complex backgrounds and multilingual text and shows stable performance on different standard datasets.