• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.12
• /
• pp.1919-1925
• /
• 2022

In this paper, we propose an image-to-image translation approach based on the conditional generative adversarial network for semantic segmentation. Semantic segmentation is the task of clustering parts of an image together which belong to the same object class. Unlike the traditional pixel-wise classification approach, the proposed method parses an input RGB image to its corresponding semantic segmentation mask using a pixel regression approach. The proposed method is based on the Pix2Pix image synthesis method. We employ residual connections-based convolutional neural network architectures for both the generator and discriminator architectures, as the residual connections speed up the training process and generate more accurate results. The proposed method has been trained and tested on the NYU-depthV2 dataset and could achieve a good mIOU value (49.5%). We also compare the proposed approach to the current methods in semantic segmentation showing that the proposed method outperforms most of those methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.11a
• /
• pp.66-69
• /
• 2017

본 논문에서는 깊은 합성 곱 신경망 (Deep Convolutional Neural Network) 를 이용해서 SAR (Synthetic Aperture Radar) 영상의 반전 잡음 (speckle noise) 성분을 제거하는 기법을 제안하고자 한다. Deep Convolutional Neural Network는 이미지의 데이터 특성에 적합한 딥 러닝 방법이고, 이는 SAR 위성영상의 반전 잡음 제거에 사용해도 효과적이다. 반전 잡음 필터 모델 추정을 위한 학습은 임의로 반전 잡음을 합성한 트레이닝 이미지들과 원본 트레이닝 이미지들을 이용한 회귀모델을 통해 진행된다. 학습을 통해 얻은 반전 잡음 필터는 기존 알고리즘에 비해 우수한 외곽선 보존 성능을 나타냄을 확인하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.659-660
• /
• 2021

본 논문에서는 데카르트 좌표계 기반으로 노드를 압축함으로써 SR(Super-resolution) 기반 연기 합성을 효율적으로 처리할 수 있는 방법을 제안한다. 제안하는 방법은 다운 스케일링과 이진화를 통하여 연기 시뮬레이션의 계산 공간을 효율적으로 줄이고, 데카르트 좌표계 축을 기준으로 쿼드트리의 말단 노드를 압축함으로써 네트워크의 입력으로 전달하는 데이터 개수를 줄인다. 학습에 사용된 데이터는 COCO 2017 데이터셋이며, 인공신경망은 VGG19 기반 네트워크를 사용한다. 컨볼루션 계층을 거칠 때 데이터의 손실을 막기 위해 잔차(Residual)방식과 유사하게 이전 계층의 출력 값을 더해주며 학습한다. 결과적으로 제안하는 방법은 이전 결과에 비해 네트워크로 전달해야 하는 데이터가 압축되어 개수가 줄어드는 결과를 얻었으며, 그로 인해 네트워크 단계에서 필요한 I/O 과정을 효율적으로 처리할 수 있게 되었다.

• Proceedings of The KACE
• /
• 2017.08a
• /
• pp.47-49
• /
• 2017

인공 신경망 기반 자연어 처리 시스템들에서 단어를 벡터로 변환할 때, 크게 색인 및 순람표를 이용하는 방법과 합성곱 신경망이나 회귀 신경망을 이용하는 방법이 있다. 이 때, 전자의 방법을 사용하려면 시스템이 수용 가능한 어휘집이 정의되어 있어야 하며 새로운 단어를 어휘집에 추가하기 어렵다. 반면 후자의 방법을 사용하면 단어를 구성하는 문자들을 바탕으로 벡터 표현을 생성하기 때문에 어휘집이 필요하지 않지만, 추가적인 인공 신경망 구조가 필요하기 때문에 모델의 복잡도와 파라미터의 수가 증가한다는 단점이 있다. 본 연구에서는 위 두 방법의 한계를 극복하고자 Bag of Characters를 응용하여 단어를 구성하는 문자들의 집합을 바탕으로 벡터 표현을 생성하는 방법을 제안한다. 제안된 방법은 문자를 기반으로 동작하기 때문에 어휘집을 정의할 필요가 없으며, 인공 신경망 구조가 사용되지 않기 때문에 시스템의 복잡도도 증가시키지 않는다. 또한, 단어의 벡터 표현에 단어를 구성하는 문자들의 정보가 반영되기 때문에 Out-Of-Vocabulary 단어에 대한 성능도 어휘집을 사용하는 방법보다 우수할 것으로 기대된다.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.1
• /
• pp.119-127
• /
• 1998

In this paper, the design of a VLSI-based multilayer neural network is presented, which can be used as a dedicated hardware for character-type segmentation and character-element recogniti on consuming large processing time in conventional software-based Hangul printed-character recognition systems. Also the architecture and its design of a neural coprocessor interfacing the neural network with a host computcr and controlling thc neural network are presented. The architecture, behavior, and performance of the proposed neural coprocessor are justified using VHDL modeling and simulation. Experimental results show the successful rates of character-type segmentation and character-element recognition is competitive to those of software-based Hangul printed-character recognition systems with retaining high-speed.

• Korea Information Processing Society Review
• /
• v.29 no.1
• /
• pp.43-50
• /
• 2022

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.43-45
• /
• 2021

The short-term quantitative precipitation prediction (QPF) system is important socially and economically to prevent damage from severe weather. Recently, many studies for short-term QPF model applying the Deep Neural Network (DNN) has been conducted. These studies require the sophisticated pre-processing because the mistreatment of various and vast meteorological data sets leads to lower performance of QPF. Especially, for more accurate prediction of the non-linear trends in precipitation, the dataset needs to be carefully handled based on the physical and dynamical understands the data. Thereby, this paper proposes the following approaches: i) refining and combining major factors (weather radar, terrain, air temperature, and so on) related to precipitation development in order to construct training data for pattern analysis of precipitation; ii) producing predicted precipitation fields based on Convolutional with ConvLSTM. The proposed algorithm was evaluated by rainfall events in 2020. It is outperformed in the magnitude and strength of precipitation, and clearly predicted non-linear pattern of precipitation. The algorithm can be useful as a forecasting tool for preventing severe weather.

• Journal of Korean Society of Transportation
• /
• v.19 no.6
• /
• pp.171-182
• /
• 2001

Travel Time is an important characteristic of traffic conditions in a road network. Currently, there are so many road users to get a unsatisfactory traffic information that is provided by existing collection systems such as, Detector, Probe car, CCTV and Anecdotal Report. This paper presents the results achieved with Data Fusion Model, Hybrid Neuro Fuzzy System for on - line estimation of travel times using RTMS(Remote Traffic Microwave Sensor) and Probe Data in the signalized arterial road. Data Fusion is the most important process to compose the various of data which can present real value for traffic situation and is also the one of the major process part in the TIC(Traffic Information Center) for analyzing and processing data. On-line travel time estimation methods(FALEM) on the basis of detector data has been evaluated by real value under KangNam Test Area.

• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.1115-1122
• /
• 2018

In this paper, we analyze the performance degradation of previous 1-Bit convolutional neural network method and introduce ways to mitigate it. Previous work applies 32-Bit operation to first and last layers. But our method applies 32-Bit operation to second layer too. We also show that nonlinear activation function can be removed after binarizing inputs and weights. In order to verify the method proposed in this paper, we experiment the object detection neural network for korean license plate detection. Our method results in 96.1% accuracy, but the existing method results in 74% accuracy.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.130-138
• /
• 2021

This paper presents an effective structure by applying various normalization to Convolutional Neural Networks (CNN) for seismic event classification. Normalization techniques can not only improve the learning speed of neural networks, but also show robustness to noise. In this paper, we analyze the effect of input data normalization and hidden layer normalization on the deep learning model for seismic event classification. In addition an effective model is derived through various experiments according to the structure of the applied hidden layer. As a result of various experiments, the model that applied input data normalization and weight normalization to the first hidden layer showed the most stable performance improvement.