• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1997년도 춘계학술대회 학술발표 논문집
• /
• pp.124-127
• /
• 1997

In this paper, we propose the new face detection and tracking angorithm in sequential images which have complex background. In order to apply face deteciton algorithm efficently, we convert the conventional RGB coordiantes into CIE coordonates and make the input images insensitive to luminace. And human face shapes and colors are learned using ueural network's backpropagation. For variable face size, we make mosaic size of input images vary and get the face location with various size through neural network. Besides, in sequential images, we suggest face motion tracking algorithm through image substraction processing and thresholding. At this time, for accurate face tracking, we use the face location of previous. image. Finally, we verify the real-time applicability of the proposed algorithm by the simple simulation.

• 한국정보통신학회논문지
• /
• 제18권3호
• /
• pp.510-518
• /
• 2014

본 연구에서는 유체 속에서의 로봇의 방향전환 메커니즘의 성능을 개선하고 최적화하기 위하여 물 속 자연환경에 최적화되어 있는 물고기의 CST(CST:C-shape sharp turn) 패턴을 모방하여 물고기 로봇의 꼬리 관절 궤적을 신경회로망(neural network)을 사용하여 제안하였다. 물고기의 CST 패턴을 모방하기 위해 CST 패턴을 순차적으로 기록한 정보를 수치적으로 변환하여 좌표 데이터를 생성하고 함수화하였다. 함수화된 모션 함수를 물고기 로봇의 상대 관절각으로 변환하였으나, 구해진 상대 관절 궤적은 잉어의 순차적 기록에 의해 구해진 각도이므로 분해능이 떨어져 실제 물고기 로봇의 제어에 적용하기 어렵다. 그러므로 상대 관절 궤적을 일반화 기능이 뛰어난 신경회로망을 사용하여 보간하고 물고기 로봇에 적용하였다. 모의실험을 통하여 신경회로망을 이용한 상대 관절 궤적 함수가 고차의 다항식 궤적 함수에 비하여 물고기 로봇의 CST 모션에 더 좋은 성능을 나타냄을 확인하였다.

• 디지털융복합연구
• /
• 제13권10호
• /
• pp.287-293
• /
• 2015

본 논문에서는 신호 처리 기술과 가중 퍼지소속함수 기반 신경망 (Neural Network with Weighted Fuzzy Membership Functions; NEWFM)을 이용하여 간질을 검출하는 방안을 제안하였다. 신호 처리 기술로는 웨이블릿 변환(Wavelet Transform), 점증적 증가 방법, 위상공간 재구성(Phase Space Reconstruction)을 이용하였다. 신호 처리 기술의 첫 번째 단계에서는 웨이블릿 변환을 이용하여 뇌파로부터 웨이블릿 계수를 추출하였다. 두 번째 단계에서는 점증적 증가 방법을 이용하여 웨이블릿 계수로부터 첨점(Peak)을 추출하였다. 세 번째 단계에서는 위상공간 재구성을 이용하여 추출된 첨점으로부터 3차원 다이어그램을 생성하였다. NEWFM의 입력으로 사용할 16개의 특징을 추출하기 위하여 유클리드 거리와 통계적 방법을 이용하였다. 이들 16개의 특징을 NEWFM의 입력으로 사용하여 97.5%, 100%, 95%의 정확도, 특이도, 민감도를 각각 구하였다.

• 한국정보통신학회논문지
• /
• 제7권6호
• /
• pp.1304-1311
• /
• 2003

실세계환경에서 물체를 추적하는 기술은 영상의 지속적인 변화 및 영상데이터 방대함과 처리속도의 문제로 인하여 해결하기 어려운 문제이다. 특히 해상과 같은 환경에서는 더욱 어려운 현실이다. 본 논문에서는 복잡한 환경에서 물체를 추적하고 탐지하기 위한 방법으로 자기조직화 신경망을 사용하여 구성하였다. 본 논문에서의 접근 방법은 코호넨의 자기 조직화 신경망 분석 기법과 영역확장 기법 및 에너지 최소화함수를 이용하여 물체 추적시스템을 구성하였다. 자기조직화 신경망은 하나의 프레임 내에서 이동하는 물체의 중심점을 탐지할 수 있다. 그리고 연속적인 영상에서 이전에 탐지되어진 뉴런의 위치를 이용하여 물체를 추적할 수 있다. 자기조직화 신경망을 이용한 물체 추적의 실험결과 다양한 환경의 변화에서도 물체의 추적이 가능함을 알 수 있었다.

• Spatial Information Research
• /
• 제2권2호
• /
• pp.175-188
• /
• 1994

본 연구에서는 최근 많은 분야데서 패턴인식을 위한 효과적인 기법으로 이용되고 있는 신경망 기법을 원격탐사자료의 군집화 기법으로서 적용하고자 하였다. 이를 위해 선택된 신경망 모델은 경쟁학습 신경망이며 이를 구성하는 각종 변수들을 재구성하여 원격탐사자료의 군집화를 위한 신경망모델을 설정하였다. 본 신경망을 이용한 군집화 기법은 항공기를 이용하여 획득된 원격탐사자료를 이용하여 순차적(sequential)군집화 기법 K 평균 군집화 기법과 비교되었다. 계산시간은 순차적 기법이나 K 평균기법에 비하여 더 많이 소요되나 정확도면에 있어서는 비교적 우수한 결과를 나타냈다.

• International Journal of Computer Science & Network Security
• /
• 제21권11호
• /
• pp.23-30
• /
• 2021

Sentiment Analysis has become very important field of research because posting of reviews is becoming a trend. Supervised, unsupervised and semi supervised machine learning methods done lot of work to mine this data. Feature engineering is complex and technical part of machine learning. Deep learning is a new trend, where this laborious work can be done automatically. Many researchers have done many works on Deep learning Convolutional Neural Network (CNN) and Long Shor Term Memory (LSTM) Neural Network. These requires high processing speed and memory. Here author suggested two models simple & bidirectional deep leaning, which can work on text data with normal processing speed. At end both models are compared and found bidirectional model is best, because simple model achieve 50% accuracy and bidirectional deep learning model achieve 99% accuracy on trained data while 78% accuracy on test data. But this is based on 10-epochs and 40-batch size. This accuracy can also be increased by making different attempts on epochs and batch size.

• 전자공학회논문지B
• /
• 제31B권5호
• /
• pp.141-148
• /
• 1994

Both the neural network classifier utilizing multi-layer perceptron and the linear tree classifier composed of hierarchically structured linear discriminating functions can form arbitrarily complex decision boundaries in the feature space and have very similar decision making processes. In this paper, a new method for automatically choosing the number of neurons in the hidden layers and for initalzing the connection weights between the layres and its supporting theory are presented by mapping the sequential structure of the linear tree classifier to the parallel structure of the neural networks having one or two hidden layers. Experimental results on the real data obtained from the military ship images show that this method is effective, and that three exists no siginificant difference in the classification acuracy of both classifiers.

• 한국멀티미디어학회논문지
• /
• 제22권12호
• /
• pp.1457-1465
• /
• 2019

3D convolution is to stack multiple consecutive frames to form a cube, and then apply the 3D convolution kernel in the cube. In this structure, each feature map of the convolutional layer is connected to multiple adjacent sequential frames in the previous layer, thus capturing the motion information. However, due to the changes of pedestrian posture, motion and position, the convolution at the same place is inappropriate, and when the 3D convolution kernel is convoluted in the time domain, only time domain features of three consecutive frames can be extracted, which is not a good enough to get action information. This paper proposes an action recognition method based on feature fusion of 3D convolutional neural network. Based on the VGG16 network model, sending a pre-acquired optical flow image for learning, then get the time domain features, and then the feature of the time domain is extracted from the features extracted by the 3D convolutional neural network. Finally, the behavior classification is done by the SVM classifier.

• Communications for Statistical Applications and Methods
• /
• 제27권3호
• /
• pp.313-326
• /
• 2020

Deep Learning is the most important key to the development of Artificial Intelligence (AI). There are several distinguishable architectures of neural networks such as MLP, CNN, and RNN. Among them, we try to understand one of the main architectures called Recurrent Neural Network (RNN) that differs from other networks in handling sequential data, including time series and texts. As one of the main tasks recently in Natural Language Processing (NLP), we consider Neural Machine Translation (NMT) using RNNs. We also summarize fundamental structures of the recurrent networks, and some topics of representing natural words to reasonable numeric vectors. We organize topics to understand estimation procedures from representing input source sequences to predict target translated sequences. In addition, we apply multiple translation models with Gated Recurrent Unites (GRUs) in Keras on English-Korean sentences that contain about 26,000 pairwise sequences in total from two different corpora, colloquialism and news. We verified some crucial factors that influence the quality of training. We found that loss decreases with more recurrent dimensions and using bidirectional RNN in the encoder when dealing with short sequences. We also computed BLEU scores which are the main measures of the translation performance, and compared them with the score from Google Translate using the same test sentences. We sum up some difficulties when training a proper translation model as well as dealing with Korean language. The use of Keras in Python for overall tasks from processing raw texts to evaluating the translation model also allows us to include some useful functions and vocabulary libraries as well.

• 산업경영시스템학회지
• /
• 제41권1호
• /
• pp.84-93
• /
• 2018

In this paper, we propose an Elman recurrent neural network to predict and analyze a time series of power energy consumption. To this end, we consider the volatility of the time series and apply the sample variance and the detrended fluctuation analyses to the volatilities. We demonstrate that there exists a correlation in the time series of the volatilities, which suggests that the power consumption time series contain a non-negligible amount of the non-linear correlation. Based on this finding, we adopt the Elman recurrent neural network as the model for the prediction of the power consumption. As the simplest form of the recurrent network, the Elman network is designed to learn sequential or time-varying pattern and could predict learned series of values. The Elman network has a layer of "context units" in addition to a standard feedforward network. By adjusting two parameters in the model and performing the cross validation, we demonstrated that the proposed model predicts the power consumption with the relative errors and the average errors in the range of 2%~5% and 3kWh~8kWh, respectively. To further confirm the experimental results, we performed two types of the cross validations designed for the time series data. We also support the validity of the model by analyzing the multi-step forecasting. We found that the prediction errors tend to be saturated although they increase as the prediction time step increases. The results of this study can be used to the energy management system in terms of the effective control of the cross usage of the electric and the gas energies.