• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 추계종합학술대회 논문집(3)
• /
• pp.69-72
• /
• 2000

In this paper, a novel algorithm for digital watermarking is proposed. We use two pattern keys from BCH (15, 7) code and one randomizing key. In the embedding process, optimal watermark coefficients are determined by statistical analysis of the DCT coefficients from the standpoint of HVS. In the detection, watermark coefficients are restored by correlation matching of the possible pattern keys and minimizing the estimation errors. Attacks tested in the experiments ate image enhancement and image compression (JPEG). Performance is evaluated by BER of the logo images and SNR/PSNR of the restored images. Our method has higher performance against JPEG attacks. Analysis for the performance is included.

• 한국지능시스템학회논문지
• /
• 제10권2호
• /
• pp.113-121
• /
• 2000

이 논문에서는 병렬구조 퍼지시스템(PSFS)에 기초한 카오스 시계열 데이터의 예측 알고리즘에 대해 연구하였다 병렬구조 퍼지시스템은 병렬로 연결된 여러개의 퍼지시스템에 의하여 구성되어있다. 병렬구조 퍼지시스템을 구성하고 있는 각 퍼지시스템은 다른 임베딩 차원과 시간지연을 가지고 과거의 데이터를 이용하여 동일한 데이터를 독립적으로 예측한다 퍼지시스템은 입출력 데이터를 클러스터링하여 모델링되는 MISO Sugeno 퍼지규칙에 의하여 특징지어진다. 각 퍼지시스템에 대한 최적 임베딩차원은 주어진 시간지연값에 대해서 최적의 성능을 갖도록 선정된다. 병렬구조 퍼지시스템은 각 구성요소 퍼지스템들의 예측값중에서 최대값과 최소값을 가지는 예측결과를 제외하고 나머지 값들을 평균하여 최종 예측 결과를 얻는다.

• 한국생산제조학회지
• /
• 제21권3호
• /
• pp.362-367
• /
• 2012

Though a micro pump is a crucial element in miniaturized bio-fluidic systems or drug delivery systems, most of the conventional micro pumps still have some limitations to miniaturize their controller system and to obtain the sufficient back pressure which can rise over the inner pressure of human body or experimental animals. In this paper, to overcome these limitation, a new PZT disc and its controller were designed and fabricated to get the sufficient flowrate and the back pressure with guaranteeing embeddability of the controller into pumping body. The amplitudes of the disc deflections were as large as 40 ${\mu}m$ at 200 V - 100 Hz condition. As results of experiments, the flow rate and the back pressure increase when the frequency increases. The obtainable maximum flow rate and back pressure are 5.2 ml/min at 95 Hz and 13.14 kPa at 90 Hz respectively.

• 인터넷정보학회논문지
• /
• 제23권2호
• /
• pp.45-52
• /
• 2022

With the development of artificial intelligence analysis methods, especially machine learning, various fields are widely expanding their application ranges. However, in the case of classical music, there still remain some difficulties in applying machine learning techniques. Genre classification or music recommendation systems generated by deep learning algorithms are actively used in general music, but not in classical music. In this paper, we attempted to classify opera among classical music. To this end, an experiment was conducted to determine which criteria are most suitable among, composer, period of composition, and emotional atmosphere, which are the basic features of music. To generate emotional labels, we adopted zero-shot classification with four basic emotions, 'happiness', 'sadness', 'anger', and 'fear.' After embedding the opera libretto with the doc2vec processing model, the optimal number of clusters is computed based on the result of the elbow method. Decided four centroids are then adopted in k-means clustering to classify unsupervised libretto datasets. We were able to get optimized clustering based on the result of adjusted rand index scores. With these results, we compared them with notated variables of music. As a result, it was confirmed that the four clusterings calculated by machine after training were most similar to the grouping result by period. Additionally, we were able to verify that the emotional similarity between composer and period did not appear significantly. At the end of the study, by knowing the period is the right criteria, we hope that it makes easier for music listeners to find music that suits their tastes.

• 대한임베디드공학회논문지
• /
• 제18권5호
• /
• pp.217-223
• /
• 2023

In this paper, we propose a new model for the conditional generation of music, considering key and rhythm, fundamental elements of music. MIDI sheet music is converted into a WAV format, which is then transformed into a Mel Spectrogram using the Short-Time Fourier Transform (STFT). Using this information, key and rhythm details are classified by passing through two Convolutional Neural Networks (CNNs), and this information is again fed into the Music Transformer. The key and rhythm details are combined by differentially multiplying the weights and the embedding vectors of the MIDI events. Several experiments are conducted, including a process for determining the optimal weights. This research represents a new effort to integrate essential elements into music generation and explains the detailed structure and operating principles of the model, verifying its effects and potentials through experiments. In this study, the accuracy for rhythm classification reached 94.7%, the accuracy for key classification reached 92.1%, and the Negative Likelihood based on the weights of the embedding vector resulted in 3.01.

• Structural Engineering and Mechanics
• /
• 제25권2호
• /
• pp.239-268
• /
• 2007

This paper deals with optimal stacking sequence design of laminate composite structures. The stacking sequence optimisation of laminate composites is formulated as a combinatorial problem and is solved using Simulated Annealing (SA), an algorithm devised based on inspiration of physical process of annealing of solids. The combinatorial constraints are handled using a correction strategy. The SA algorithm is strengthened by embedding Tabu search in order to prevent recycling of recently visited solutions and the resulting algorithm is referred to as tabu embedded simulated Annealing (TSA) algorithm. Computational performance of the proposed TSA algorithm is enhanced through cache-fetch implementation. Numerical experiments have been conducted by considering rectangular composite panels and composite cylindrical shell with different ply numbers and orientations. Numerical studies indicate that the TSA algorithm is quite effective in providing practical designs for lay-up sequence optimisation of laminate composites. The effect of various neighbourhood search algorithms on the convergence characteristics of TSA algorithm is investigated. The sensitiveness of the proposed optimisation algorithm for various parameter settings in simulated annealing is explored through parametric studies. Later, the TSA algorithm is employed for multi-criteria optimisation of hybrid composite cylinders for simultaneously optimising cost as well as weight with constraint on buckling load. The two objectives are initially considered individually and later collectively to solve as a multi-criteria optimisation problem. Finally, the computational efficiency of the TSA based stacking sequence optimisation algorithm has been compared with the genetic algorithm and found to be superior in performance.

• 디지털융복합연구
• /
• 제17권2호
• /
• pp.163-170
• /
• 2019

과거 10년은 웹의 발달로 인한 데이터가 폭발적으로 생성되었다. 데이터마이닝에서는 대용량의 데이터에서 무의미한 데이터를 구분하고 가치 있는 데이터를 추출하는 단계가 중요한 부분을 차지한다. 본 연구는 감성분석을 위한 재표현 방법과 속성선택 방법을 적용한 오피니언 마이닝 모델을 제안한다. 본 연구에서 사용한 재표현 방법은 백 오즈 워즈(Bag-of-words)와 Word embedding to vector(Word2vec)이다. 속성선택(Feature selection) 방법은 상관관계 기반 속성선택(Correlation based feature selection), 정보획득 속성선택(Information gain)을 사용했다. 본 연구에서 사용한 분류기는 로지스틱 회귀분석(Logistic regression), 인공신경망(Neural network), 나이브 베이지안 네트워크(naive Bayesian network), 랜덤포레스트(Random forest), 랜덤서브스페이스(Random subspace), 스태킹(Stacking)이다. 실증분석 결과, electronics, kitchen 데이터 셋에서는 백 오즈 워즈의 정보획득 속성선택의 로지스틱 회귀분석과 스태킹이 높은 성능을 나타냄을 확인했다. laptop, restaurant 데이터 셋은 Word2vec의 정보획득 속성선택을 적용한 랜덤포레스트가 가장 높은 성능을 나타내는 조합이라는 것을 확인했다. 다음과 같은 결과는 오피니언 마이닝 모델 구축에 있어서 모델의 성능을 향상시킬 수 있음을 나타낸다.

• 한국정보과학회논문지:데이타베이스
• /
• 제33권7호
• /
• pp.715-726
• /
• 2006

웹 서비스는 인터넷 상에 분산되어 있는 이질적인 응용들 사이의 연동 및 통합을 위한 표준화된 수단을 제공한다. 본 논문에서는 계층적인 연동 관계가 존재하는 복합 웹 서비스들에 대해 서비스 결과로 전달되는 내포 데이타를 활용하여 웹 서비스들의 호출 및 복귀 작업을 서버 및 통신 비용에 따라 효과적으로 분산 수행함으로써 웹 서비스 시스템의 전체적인 성능을 향상시킬 수 있는 방안을 제시한다. 본 논문에서는 내포 결과를 이용한 적법한 웹 서비스 호출 실행 계획 및 이에 대한 비용 기반 최적화 문제를 정의하고, 최적 호출 실행 계획을 찾기 위한 휴리스틱 탐색 방법과 효율적으로 수행될 수 있는 그리디 알고리즘을 제안한다. 실험 결과, 제안한 그리디 알고리즘은 빠른 시간 내에 최적 해에 가까운 효율적인 호출 실행 계획을 생성하며, 복잡한 웹 서비스 연동 관계에 대해서 우수한 확장성을 보였다.

• 전산구조공학
• /
• 제9권3호
• /
• pp.179-186
• /
• 1996

본 연구에서는 구조체의 위상학적 최적화를 위한 비선형 formulation(NLP)가 개발, 검토되었다. 이 NLP는 multiple-loading하에서 임의의 오브젝티브 함수, 응력, 변위 제약조건들을 쉽게 다룰 수가 있다. 또한 이 NLP는 해석과 최적화 디자인을 동시에 실시함으로써 요소 사이즈가 영으로 접근함에 따른 강성 매트릭스의 singularity를 피할 수 있다. 즉, 평형 방정식을 등제약조건으로 치환함으로써 강성 매트릭스 그 자체나 그의 역매트릭스를 구할 필요도 없어진다. 이 NLP는 multiple-loading conditon하에서 테스트되었으며, 이를 통해 이 NLP가 다양한 제약조건하에서 강력하게 작용함이 입증되었다.

• 한국산업융합학회 논문집
• /
• 제26권1호
• /
• pp.193-199
• /
• 2023

In this study, an analysis was conducted to analyze the structural stability of the piezoelectric power generation module and to determine the optimal burying hole interval for concrete, the installation site of the power generation module. A piezoelectric element refers to a functional ceramic having a piezoelectric direct effect that converts mechanical energy into electrical energy and a piezoelectric reverse effect. In the analysis of the piezoelectric power generation module, the load condition was applied with about 16 tons and a total of 10 wheels in consideration of the container trailer. The purpose was to evaluate the stability of major components of the piezoelectric power generation module through finite element analysis. In order to determine the optimal burying location of the concrete ground for burying the piezoelectric power generation module, the stability of the ground structure according to the distance of the holes was determined. As a result of the analysis, the maximum stress of the piezoelectric power generation module was generated in the support spring, showing a stress of about 276.7 MPa. It was found that the spacing of holes for embedding the piezoelectric power generation module should be set to a minimum of 100 mm or more.