적합성시험에서 EFSM모델에 대한 최소길이의 관찰중심적 시험케이스 자동생성에 관한 연구

김태형 연세대학교 대학원 1995 국내석사

인공지능을 이용한 감성조명 LED 제어기 설계에 관한 연구

이재홍 한국해양대학교 2011 국내석사

Today the light of mankind since the dawn of mankind and that they were living "lives" on the huge influence across space is a dynamic energy. Because of this, when all human beings with the light and you can feel more comfortable with a stable mind to the beauty and the pursuit of happiness is to have instincts. Moreover, the spectral characteristics of the sun and the light rhythm of the day, seasonal changes in light of all the plants to produce various physiological responses is the psychic effect of color temperature as the lighting plan to intentionally use the space more comfortable to can be instrumental in creating. For this reason, most modern people spend a lot of emotion in an indoor space, light is needed. Recently has been the culprit of environmental destruction incandescent, fluorescent, LED light regulation of the use of the demand is growing. Indoor activities of modern society is growing more and more LED lighting as the interior lighting spotlight and into a digital signal can be controlled simply because the LED lighting control technology is also a growing trend. LED lighting of the traditional incandescent bulbs, fluorescent lamps can achieve higher efficiency compared to the addition since it has the advantage of long life light sources reduce maintenance costs due to replacement can be. LED 50,000 hours lamp life by 50 times when compared to incandescent bulbs, fluorescent lights, 6 times, 4 times in lamps which use it 8 hours a day, assuming the case, life is known to be a whopping 17 years. In addition, the LED current in one direction by the light-emitting device that is dependent on the DC current response rate is very fast, very high recently improved the efficiency of the light fixture to replace the existing policy was to elicit. In this paper, a fuzzy control system using a combination of external environmental factors, respectively, the conversion of quantitative uncertainty information into the LED lighting is designed to express algorithms, available in indoor circadian control circuit was designed and fabricated LED. Factor in the external environment temperature, humidity, and light intensity values to the controller through the sensor to accept these values and the optimal values for lighting the fuzzy control algorithm is converted into human emotion to feel comfortable to express through the dimming control LED lighting.

인공지능을 이용한 자동 싱크대 높낮이 제어 시스템 구현에 관한 연구

박상용 한국해양대학교 대학원 2013 국내석사

According to the press release of 1990, most 40% of back pain patients were housewife. But nowadays, the rate of patients has been increasing in number of students and office workers. This result is deduced from whole family doing housekeeping work by increasing of husband and wife both working, Not only wife's things. Also, children do some part of housekeeping work like washing dish and somethings instead of their mother. modern sinks are set same pattern of the height, So children have to set extra chairs or boxes so as to do washing dish and washing fruits on the sink. but this situation makes chilren dangerous and cause possibility of falling down. Also children need to clean up their working place. Of course, the original sink foot has the ability to adjust the height while supporting the bottom. For this reason, if you rotate the device, the height of the sink will increase and if you rotate in the opposite direction, the height is low. Adults can learn how to use easily. But the children still difficult to use due to lack understanding. Therefore, a new sink for height adjustment is required to solve above problems. The purpose of this paper is You should be able to automatically adjust the height of the sink depending on the individual's height. If the height of the sink is too high or too low, people will feel the pain in the waist while using the sink. That's why this control system can prevent back pain and this paper is a study of the automatic height control system.

인공 지능을 이용한 식물 공장 LED 조명 제어 시스템 구현에 관한 연구

류창완 한국해양대학교 대학원 2013 국내석사

Currently, LED plant factory have been using colors by fixed red and blue colors spectrums. This has the disadvantage that can not be applied to organic, depending on the situation around. In other words, if the type of plants according to the growth stage or the appropriate light environment can not be controlled, but rather hinder or promote the growth of plants, can cause unnecessary energy consumption. Growth progresses, the environmental conditions suitable for the growth period of the plant will vary, but always maintain the same environmental conditions, plant growth promotion and growth of plants can be inhibited in order to save energy when appropriate according to the step-by-step growth of plants formulation of the light source is needed. Because it has been using for secondary source energy is not the main source of energy in photosynthesis, also used in LED plant factory directly, depending on the external environment affect the growth of plants, do not give. In addition, LED unit price formation and also expensive because typically tend to be reluctant to use. In this paper, we present a system that is controlled by the organic as well as the use of LED lighting as the main energy source rather than a secondary energy source, according to the surrounding environment (temperature, illumination, humidity, soil conditions, etc.) Plant Factory Lighting System has been implemented. Existing fixed wavelength of LED, without using Variable LED wavelengths, and changes the system in the environment to be controlled organically by introducing the fuzzy inference system has constructed.

퍼지 추론을 이용한 자동 롤러 쉐이드 제어 시스템에 관한 연구

임진강 한국해양대학교 대학원 2012 국내석사

Currently the world is facing the energy problems. So energy saving have become a big issue. But the building's energy usage is increasing by the higher, larger, specialization buildings. 22% of total energy usage is lighting energy. So reducing the lighting energy is large part of reducing building's energy usage. Current, using light to save energy are being research actively. Among the ways to use daylight, there are automated roller shading system. That adjust roller shade height through the sun's change. Automated roller shading system can block direct sunlight and enter useful sunlight. In this paper, for knowing the usefulness of automated roller shading system using fuzzy inference, build automated roller shading system of two cases(using fuzzy inference and using sun's changing) and compare system using the fuzzy inference with system using the sun's changing. As a result, fuzzy control system has higher intensity of illumination and uniformity factor than using sun's changing automated roller shade system.

도어락에 적용한 화자 종속형 음성 인식 시스템 구현에 관한 연구

정성훈 한국해양대학교 대학원 2005 국내석사

본 논문은 휴먼 인터페이스의 개념을 적용한 도어락 출입 통제시스템에 음성인식 모듈 적용 사례를 연구한 것이다. 현재 개발된 음성 인식 알고리즘은 DTW(Dynamic Time Warping), HMM(Hidden Markov Model), NN( Neural Network)이 있으며, 이 중 본 논문에서는 화자 종속형 알고리즘을 연구하였다. 우선, 다량의 데이터를 처리할 수 있는 하드웨어 선정에 TI(Texas Instruments)사의 DSP (TMS320C32)를 사용하였으며, 여러 가지 잡음처리 및 특징 추출을 모두 적용하여 테스트를 시도하였다. 특히, 본 논문에서 제안한 시스템 은 잡음에도 높은 인식률과 정확성, 빠른 수렴속도를 가지는 RLS(Recursive Least Square) 적응 필터를 사용하였다. 이 알고리즘은 필터계수가 최적의 값에 수렴하도록 파라미터 값이 바뀔 때 마다 계산을 하여 에러률을 최소화하였고, FIR 필터 구조 중 트랜스버셜(직접형)을 사용하여 계산량의 감소와 수렴속도의 개선을 가져왔다. 그리고 인식 알고리즘 테스트는 화자 종속형 DTW 이용하여 구현하였다. 테스트 적용 방법은 (주)에디텍의 SITK-32C 및 본 연구실에서 자체 설계된 보드로 테스트를 하였다. 인식 명령어 실험은 도어락 명령어(6개)로 사용하였으며, 인식 결과 화자 종속형 시스템에는 잡음 환경을 고려하여 평균 97%이상의 높은 인식률이 나왔으며, 기존의 화자 종속형 시스템 보다 FRR(본인거부률)이 2.74%가 향상되었고, 음성 인식시스템에서 가장 중요한 요소인 FAR(타인수락률)은 1.2%의 개선으로 안정된 보안 시스템임을 입증하였다. 본 논문 결과 효율적인 시스템을 구현하기 위해서 다음과 같은 내용을 유추할 수 있다. 첫 번째로 잡음 환경이 인식률에 상당한 영향을 끼치며, 이를 해결하기 위해 RLS 적응 필터 사용이 필수적이며, 두 번째로 음성 인식 알고리즘으로는 DTW가 가장 효율이 좋은 것으로 나타났으며, 끝으로 특징 추출 단계는 MFCC가 가장 이상적으로 생각된다. 마지막으로 향후 과제로 가변 어휘 기술을 이용하여 음성 DB를 구성하는 것이 필수적이라 생각한다.

퍼지이론을 이용한 거리기반 조명제어시스템에 관한 연구

남영철 한국해양대학교 대학원 2018 국내석사

최근 들어 지구 온난화 및 고유가로 인한 에너지 위기로 전 세계적으로 이산화탄소 배출규제가 본격화되고 에너지 소비에 따른 지구환경을 보존하기 위한 대표적인 국제적인 GEF(Green Energy Family) 활동은 이산화탄소 배출 금지를 위한 교토의정서(Kyoto protocol), RoHS(Restriction of Hazardous Substances directive)에서는 무 수은 조명 사용억제, WEEE(Waste Electrical and Electronice Equipment)에서는 조명 통신융합으로 폐기물 최소화를 목적으로 폐기물 회수를 요구하는 등 다각적 노력을 경지하고 있다. 현재 기존에 상용화 되어 있는 LED(Light Emitting Diode) 조명기기의 경우, LED 동작 시퀀스가 고정이 되어 있는 상태로 판매되고 있다. 이와 같은 상태로는 외부 환경 요인이 고려되지 않고 오직 장소에만 적용되는 조명 환경 용도로서의 기능만을 수행한다. 4계절의 특성을 가진 우리나라에서 외부 환경 요인의 변화는 실질적으로 최적의 조명 환경 조성이 어렵게 되므로 외부 환경 요인의 변화에 좀 더 유기적이고 능동적으로 적응할 수 있도록 그 외부 환경 값을 실시간으로 입력받아 실시간으로 최적 조명 값이 반영될 수 있도록 해야 될 필요성이 있다는 결론을 얻게 되었다. 본 논문에서는 기존의 외부 환경 요인(조도, 피사체와의 거리 등)에 의하여 실시간으로 변동되는 환경 데이터를 마이크로프로세서를 활용하여 외부 환경 요인을 확인하고 또한 퍼지 추론 시스템을 접목하여 RGB LED 모듈 조명 제어가 가능한 제어기를 구성하였다. 이를 위하여 퍼지 제어 알고리즘을 설계하고, 퍼지 제어 시스템을 구성하였다. 외부 환경 요소인 피사체와의 거리, 조도 값을 센서로 통해 입력 받고 이 값들을 퍼지 제어 알고리즘을 통하여 최적 조명 값으로 변환하여 RGB LED 모듈 디밍 제어를 통하여 표현하고자 한다. 현재 기존의 외부 환경요인인 온도, 습도 등의 외부 환경 요인은 일반적으로 가장 많이 사용되기 때문에 본 논문에서는 조도 센서와 초음파 거리센서를 이용하고 실시간으로 거리와 조도를 확인할 수 있는 CLCD(Character Liquid Crystal Display)를 추가하여 RGB LED 모듈 조명 제어기를 구현하였다. 조도의 경우 민감한 외부 환경 요인이기 때문에 일반적인 조도계로는 정확한 측정이 어렵다. 그러므로 일반적인 조명 제어 시스템과 퍼지 제어시스템을 이용한 결과 값의 명확한 비교가 어렵게 되므로 본 논문에서는 Matlab 시뮬레이션을 통하여 결과 값을 비교 분석하였다. 본 논문에 사용된 마이크로프로세서는 Atmel사의 ATmega128A이며, RGB LED 제어기 프로그래밍 완료 후에는 독자적으로 동작할 수 있도록 SMPS(Switching Mode Power Supply) 및 전원 변환 부를 구성하였다. 조도 및 피사체와의 거리 값으로 색상 변화를 제어하고, 이들 입력 값을 실시간으로 확인할 수 있도록 CLCD를 추가하였다. Recently Kyoto for worldwide emissions regulations in earnest and representative international GEF(Green Energy Family) activity emissions prohibited in order to preserve the global environment due to the energy consumed by the energy crisis caused by global warming and high oil prices Protocol(Kyoto protocol), RoHS(Restriction of Hazardous Substance directive), and WEEE(Waste Electrical and Electronic Equipment), which require the recovery of waste in order to minimize waste by integrating lighting and communication. In the case of LED(Liquid Emitting Diode) lighting devices which have been commercialized at present, LED operating sequences are being sold in a fixed state. In such a state, only external environment factors (illumination, distance to the subject, etc.) are taken into consideration and only the function as an illumination environment application is applied to only a place. Changes in external environmental factors in Korea, which have four distinct seasonal characteristics, make it difficult to create an optimal lighting environment. Therefore, it was concluded that there is a need to input the external environment value in real time so that the optimal illumination value can be reflected in real time so that it can adapt more organically and actively to the change of external environment factors. In this paper, environmental data which is changed in real time by external environment factors (illumination, distance to the object, etc.) are checked by using microprocessor and external environment factors are confirmed. In addition, RGB LED module illumination control is applied by combining fuzzy inference system A possible controller was constructed. For this, a fuzzy control algorithm is designed and a fuzzy control system is constructed. The distance to the subject and the illuminance value from the external environment element are inputted through the sensor, and these values are converted into the optimum illumination value through the fuzzy control algorithm to express through RGB LED module dimming control. The external environmental factors such as temperature and humidity, which are external factors of the existing environment, are generally used most commonly. In this paper, CLCD(Character Liquid Crystal Display), which can check distance and illuminance in real time using an illuminance sensor and an ultrasonic distance sensor, We implemented a module lighting controller. In the case of Illuminance, it is difficult to measure accurately with a Illuminometer because it is a sensitive external environmental factor. Therefore, it is difficult to make a clear comparison of results using general lighting control system and fuzzy control system. Therefore, in this paper, we compared and analyzed the results through Matlab simulation. The microprocessor used in this paper is Atmel's ATmega128A. After the RGB LED controller programming is completed, the SMPS(Switching Mode Power Supply)and power converter are configured to operate independently. CLCD is added to control the color change by the distance between the object and the illuminance, and to check these input values in real time.

인공 지능을 이용한 LED 조명 제어시스템에 관한 연구

정성욱 韓國海洋大學校 大學院 2015 국내석사

SVM을 이용한 다중 생체 인식 시스템의 설계 및 구현에 관한 연구

김정훈 한국해양대학교 대학원 2007 국내박사

This study intends to design and implement the multi-modal biometric system by applying the RFID recognition system presently has been greatly issued in the ubiquitous society to the multi-modal biometric system using the fingerprint and the speech widely used in the biometrics, in order to improve FAR and FRR of the existing single biometric system. The recognition system is designed with TI’s DSP TMS- 320C32, TMS320VC5509 and AVR so that it can be implemented as embedded subsystems and is connected to the RFID system in serial while the fingerprint recognition system and the speech recognition system are connected in parallel. The hardware is built with the algorithms excellent in real-time for efficient real-time implementation. Among them, SVM algorithm, which is the most essential in this study, distinguishes the registered from the impostors by using the distance value obtained two recognition systems. This integrate system improves FAR and FRR. In this study, we performed the SVM performance test the number of vectors within the support vector is the least, which is 27(10.3%), with the Bspline kernel function among 7 kernel functions; therefore, it is verified that the kernel function of this study gives the best efficiency. We compared with other algorithms(weight sum, fuzzy integral, decision tree) in order to evaluate the performance of the multi- modal biometric system using the linear Bslpine function, and the result shows that it is much more efficient with maximum 12% difference.

인공지능 지문인식보드를 이용한 개인정보 조회시스템에 관한 연구

김재완 한국해양대학교 대학원 2007 국내석사

This paper extracted characteristic points(end points, divergent points) and the central point of fingerprints by applying the neural network algorithm. With the back-propagation algorithm of neural network algorithms, ambiguity of the central point was minimized and thus the central point was extracted. Centered on the central point, lengths between the central point and end points(and divergent points) were measured respectively. During the pre-processing procedure, Butterworth Low-Pass Filter was used to remove noises of fingerprint image. It was learned that the frequency domain was more effective in processing fingerprint images than the space domain. So Butterworth Low-Pass Filter, which processes fingerprint images in the frequency domain, was adopted in this research. Smoothing, binarization, sessionization, and histograms equalization were extracted. Then, the data of orientation were used as input data for the neural network in order to extract the central point of fingerprint. TI's DSP(TMS320VC5509) was used as the main board of the fingerprint recognition system, and ATMEL's AVR(ATmega16L) was used as the control board. MFC, which was chosen as personal information inquiry system, manages and displays detailed personal data on PC screen by connecting the fingerprint recognition system through serial communication. In the personal information inquiry system, data sources are made and registered at ODBC Data Sources manager before making programs. Fifty fingerprints of 10 people(five fingerprints of each person) were used to check the recognition rate. Verification results were retrieved by comparing one fingerprint of a person with forty-five fingerprints of the other persons(FAR(0.1)) and one fingerprint of a person with the rest four images of that person(FRR(4.5)).