• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.119-121
• /
• 2019

부가적 백색 잡음 모델(additive white Gaussian noise, AWGN에서 학습된 깊은 신경만 (deep neural networks)을 이용한 잡음 제거기는 제거하려는 잡음이 AWGN인 경우에는 뛰어난 성능을 보이지만 실제 카메라 잡음에 대해서 잡음 제거를 시도하였을 때는 성능이 크게 저하된다. 본 논문은 U-Net 구조의 깊은 인공신경망 모델에 residual block을 결합함으로서 실제 카메라 영상에서 기존 알고리즘보다 뛰어난 성능을 지니는 신경망을 제안하다. 제안한 방법을 통해 Darmstadt Noise Dataset에서 PSNR과 SSIM 모두 CBDNet 대비 향상됨을 확인하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.916-919
• /
• 2018

심층신경망 모델은 우수한 성능을 갖고 있음에도 불구하고 모델이 어떤 판단 과정을 통해 결론을 내렸는지 파악하기 어렵다. 그에 따라 판단에 대한 근거가 중요한 분야에서는 심층신경망 모델을 적용한 실제 사례를 찾기 어렵다. 인공신경망 모델을 해석하기 어렵다는 문제를 해결하기 위해 내부 구조를 이용하여 규칙을 추출하는 decompositional 접근법이 제안되었으나 기존의 연구는 대부분 은닉층이 1개인 다층 퍼셉트론 모델에서 규칙을 생성하는 것을 가정하고 있다. 오늘날 사용하는 심층신경망 모델은 일반적으로 여러 은닉층을 가지고 있기 때문에 기존의 접근법을 그대로 적용할 경우 규칙 불확실성에 따라 잘못된 규칙을 추출하는 문제가 발생한다. 본 논문은 decompositional 접근법에 존재하는 규칙 불확실성 문제를 완화하고 깊이가 깊은 심층신경망 모델에 규칙을 추출하는 방법을 제안한다. 제안한 접근법은 실제 활성화 값을 통해 지식을 추출하며, 이를 통해 규칙 불확실성 문제를 완화할 수 있었다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.22 no.5
• /
• pp.17-24
• /
• 2017

Recently Deep-learning, Technologies using Large or Deep Artificial Neural Networks, have Shown Remarkable Performance, and the Increasing Size of the Network Contributes to its Performance Improvement. However, the Increase in the Size of the Neural Network Leads to an Increase in the Calculation Amount, which Causes Problems Such as Circuit Complexity, Price, Heat Generation, and Real-time Restriction. In This Paper, We Propose and Test a Method to Reduce the Number of Network Connections by Effectively Pruning the Redundancy in the Connection and Showing the Difference between the Performance and the Desired Range of the Original Neural Network. In Particular, we Proposed a Simple Method to Improve the Performance by Re-learning and to Guarantee the Desired Performance by Allocating the Error Rate per Layer in Order to Consider the Difference of each Layer. Experiments have been Performed on a Typical Neural Network Structure such as FCN (full connection network) and CNN (convolution neural network) Structure and Confirmed that the Performance Similar to that of the Original Neural Network can be Obtained by Only about 1/10 Connection.

• Journal of the Korean Geosynthetics Society
• /
• v.17 no.4
• /
• pp.199-212
• /
• 2018

This research concerns the prediction method for ground movement and wall member force due to determination structural stability check and failure check during deep excavation construction. First, research related with excavation influence parameters is conducted. Then, numerical analysis for various excavation conditions were conducted using Finite Element Method and Beam-column elasto-plasticity method. Excavation analysis database was then constructed. Using this database, development of ANN (artificial neural network) was performed for each ground movements and using structural member forces. By comparing the numerical analysis results with ANN's prediction, it is validated that development of ANN can be used efficient for prediction of ground movement and structural member forces in deep excavation site.

• Journal of the Korean Geotechnical Society
• /
• v.20 no.3
• /
• pp.53-65
• /
• 2004

This paper presents the prediction of deep excavation-induced ground surface movements using artificial neural network(ANN) technique, which is of prime importance in the damage assessment of adjacent buildings. A finite element model, which can realistically replicate deep excavation-induced ground movements, was employed to perform a parametric study on deep excavations with emphasis on ground movements. The result of the finite element analysis formed a basis for the Artificial Neural Network(ANN) system development. It was shown that the developed ANN system can be effective for a first-order prediction of ground movements associated with deep-excavation.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.36-39
• /
• 2020

최근 딥 러닝을 이용한 방법들이 이미지 분류에서 뛰어난 성능을 보임에 따라, 복잡한 특징을 담고 있는 얼굴 이미지에 대해 이를 적용하려는 시도가 늘어나고 있다. 특히, 이미지로부터 주요한 특징들을 추출하여 간결하게 이미지를 대표할 수 있는 이미지 기술자 (Image descriptor)를 딥 러닝을 통해 생성하는 연구가 인기를 끌고 있다. 이는 딥 러닝 끝 단에 있는 Fully-connected layer 의 출력으로 얻을 수 있으며 이미지의 의미론적 상관관계를 이용하여 학습된다. 구체적으로, 이미지 기술자는 실수형 벡터 데이터로서, 한 장의 이미지를 수치화 하여 비슷한 이미지 사이에는 벡터 거리가 가깝게, 서로 다른 이미지 사이에는 벡터 거리가 멀게 구성된다. 본 연구에서는 미리 학습된 인공 신경망을 통과시켜 얻은 얼굴 이미지 기술자를 활용하여 멤버 분류를 위한 두 개의 인공 신경망을 학습하는 것을 목표로 한다. 제안된 방법을 검증하기 위해 얼굴 인식에 널리 사용되는 벤치 마크 데이터셋을 활용하였고, 그 결과 제안된 방법이 높은 정확도로 멤버를 분류할 수 있다는 것을 확인하였다.

• KIISE Transactions on Computing Practices
• /
• v.23 no.5
• /
• pp.322-327
• /
• 2017

The classification problem in the field of Natural Language Processing has been studied for a long time. Continuing forward with our previous research, which classifies large-scale text using Convolutional Neural Networks (CNN), we implemented Recurrent Neural Networks (RNN), Long-Short Term Memory (LSTM) and Gated Recurrent Units (GRU). The experiment's result revealed that the performance of classification algorithms was Multinomial Naïve Bayesian Classifier < Support Vector Machine (SVM) < LSTM < CNN < GRU, in order. The result can be interpreted as follows: First, the result of CNN was better than LSTM. Therefore, the text classification problem might be related more to feature extraction problem than to natural language understanding problems. Second, judging from the results the GRU showed better performance in feature extraction than LSTM. Finally, the result that the GRU was better than CNN implies that text classification algorithms should consider feature extraction and sequential information. We presented the results of fine-tuning in deep neural networks to provide some intuition regard natural language processing to future researchers.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2021.06a
• /
• pp.355-356
• /
• 2021

딥러닝 기술과 하드웨어의 발전으로 다양한 분야에서 인공신경망과 관련한 연구가 활발히 진행되고 있다. 비디오 코덱 부분에서도 딥러닝 기술을 적용하는 부호화 기술이 많이 연구되고 있다. 본 논문은 최근 완료된 VVC 에 채택된 신경망 기반의 기술인 MIP(Matrix Weighted Intra Prediction)를 확장하여 보다 깊은 계층의 모델로 학습된 새로운 화면내 예측 모델을 제안한다. 기존 VVC 의 MIP 의 성능과 비교하기 위하여 기존 MIP 모델과 제안하는 다중완전연결계층(Fully Connected Layer) 화면내 예측 모델을 HEVC(High Efficiency Video Coding)에 적용하여 그 성능을 비교하였다. 실험결과 제안기법은 VVC MIP 대비 0.08 BD-rate 성능 향상을 보였다.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.11
• /
• pp.2780-2785
• /
• 1997

Although KBANN(knowledge-based artificial neural network) has been shown to be more effective than other machine learning algorithms, KBANN doesn't have the theory refinement capability because the topology of the network can't be altered dynamically. Although TopGen algorithm was proposed to extend the ability of KABNN in this respect, it also had some defects due to the connection of hidden nodes from all input nodes and the use of beam search. An algorithm, which could solve this TopGen's defects by adding the hidden nodes connected from only related input nodes and using hill-climbing search with backtracking, is proposed.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.10
• /
• pp.407-420
• /
• 2021

Deep neural networks are widely used to solve various problems. However, the deep neural network with a deep hidden layer frequently has a vanishing gradient or exploding gradient problem, which is a major obstacle to learning the deep neural network. In this paper, we propose a parametric activation function to alleviate the vanishing gradient problem that can be caused by nonlinear activation function. The proposed parametric activation function can be obtained by applying a parameter that can convert the scale and location of the activation function according to the characteristics of the input data, and the loss function can be minimized without limiting the derivative of the activation function through the backpropagation process. Through the XOR problem with 10 hidden layers and the MNIST classification problem with 8 hidden layers, the performance of the original nonlinear and parametric activation functions was compared, and it was confirmed that the proposed parametric activation function has superior performance in alleviating the vanishing gradient.