• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.8
• /
• pp.4255-4269
• /
• 2019

This paper proposes an image generation method using a Multi Discriminator Generative Adversarial Net (MDGAN) as a next generation 2D game sprite creation technique. The proposed GAN is an Autoencoder-based model that receives three areas of information-color, shape, and animation, and combines them into new images. This model consists of two encoders that extract color and shape from each image, and a decoder that takes all the values of each encoder and generates an animated image. We also suggest an image processing technique during the learning process to remove the noise of the generated images. The resulting images show that 2D sprites in games can be generated by independently learning the three image attributes of shape, color, and animation. The proposed system can increase the productivity of massive 2D image modification work during the game development process. The experimental results demonstrate that our MDGAN can be used for 2D image sprite generation and modification work with little manual cost.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2191-2208
• /
• 2022

Motion blur in PET (Positron emission tomography) images induced by respiratory motion will reduce the quality of imaging. Although exiting methods have positive performance for respiratory motion correction in medical practice, there are still many aspects that can be improved. In this paper, an improved 3D unsupervised framework, Res-Voxel based on U-Net network was proposed for the motion correction. The Res-Voxel with multiple residual structure may improve the ability of predicting deformation field, and use a smaller convolution kernel to reduce the parameters of the model and decrease the amount of computation required. The proposed is tested on the simulated PET imaging data and the clinical data. Experimental results demonstrate that the proposed achieved Dice indices 93.81%, 81.75% and 75.10% on the simulated geometric phantom data, voxel phantom data and the clinical data respectively. It is demonstrated that the proposed method can improve the registration and correction performance of PET image.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.283-286
• /
• 2021

4차 산업혁명을 맞이해 최근 산업 및 기술 영역에서는 인공지능을 이용한 생산력 향상, 자동화 등 딥러닝의 보편화가 빠르게 진행되고 있다. 또한, 딥러닝의 성능을 도출하기 위해서는 수많은 양의 학습용 데이터가 필요하며 그 데이터의 양은 딥러닝 모델의 성능과 정비례한다. 이에 본 작품은 최신형 영상처리 Library인 Albumentations를 이용하여 영상처리 알고리즘을 이용하여 이미지를 증강하고, 이미지 데이터 크롤링 기능을 통해 Web에서 영상 데이터를 수집을 자동화하며, Label Pix를 연동하여 수집한 데이터를 라벨링 한다. 더 나아가 라벨링 된 데이터의 증강까지 포함하여 다양한 증강 자동화를 한 인터페이스에 집적시켜 딥러닝 모델을 생성할 때 데이터 수집과 전처리를 수월하게 한다. 또한, Neural Net 기반의 AdaIN Transfer를 이용하여 이미지를 개별적으로 학습하지 않고 Real time으로 이미지의 스타일을 옮겨올 수 있도록 하여 그림 데이터의 부족 현상을 해결한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.675-679
• /
• 2020

최근 이미지 분류의 성능 향상을 위해 깊은 레이어와 넓은 채널을 가지는 모델들이 제안되어져 왔다. 높은 분류 정확도를 보이는 모델을 제안하는 것은 과한 컴퓨팅 파워와 계산시간을 요구한다. 본 논문에서는 이미지 분류 기법에서 사용되는 딥 뉴럴 네트워크 모델에 있어, 프루닝 방법을 통해 상대적으로 불필요한 가중치를 제거함과 동시에 분류 정확도 하락을 최소로 하는 동적 필터 프루닝 방법을 제시한다. 원샷 프루닝 기법, 정적 필터 프루닝 기법과 다르게 제거된 가중치에 대해서 소생 기회를 제공함으로써 더 좋은 성능을 보인다. 또한, 재학습이 필요하지 않기 때문에 빠른 계산 속도와 적은 컴퓨팅 파워를 보장한다. ResNet20 에서 CIFAR10 데이터셋에 대하여 실험한 결과 약 50%의 압축률에도 88.74%의 분류 정확도를 보였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.323-323
• /
• 2021

본 연구는 국지적으로 발생하는 침수상황을 빠르게 감지하고 대처하기 위하여 다채널 실시간 CCTV 영상을 무인 모니터링하고 자동으로 감지하기 위한 영상분석 기술을 개발하는 것을 목적으로 한다. 이에 다양한 공간에서 촬영된 학습 및 검증을 위한 데이터를 구축하였고, 대표적인 CNN 계열 분류모델을 중심으로 딥러닝 모델을 개발하였다. 5가지 CNN 알고리즘으로 시험결과, ResNet-50 모델의 분류 정확도가 87.5%로 가장 우수한 성능을 보였다. 공간적으로는 실외, 도로공간에서 82% 이상의 분류성능을 보였고, 실내공간에서는 양질의 학습데이터 부족으로 분류성능이 떨어지는 것으로 나타났다. 본 연구성과는 지능형 CCTV 기술 발전과 방재 목적의 다목적 활용으로, 향후 홍수피해 저감을 위한 보조적인 수단으로 활용되길 기대한다.

• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.558-561
• /
• 2021

심리학에서 사람의 심리 상태를 알아보기 위해 사용되는 검사 방법 중, 집-나무-사람 검사(HTP Test)는 피실험자가 그린 집, 나무, 사람을 포함하는 그림을 사용하여 피실험자의 심리를 분석하는 투영 검사법이다. 본 논문에서는 딥러닝 모델을 이용해 HTP Test 에 사용되는 그림을 분석하는 시스템을 제안하며, 성능 평가를 통해 심리학에서의 딥러닝 모델 적용 가능성을 확인한다. 또한 그림 데이터 분석에 적합한 사전 훈련 모델을 개발하기 위해, ImageNet 과 스케치 데이터셋으로 사전 훈련하여 성능을 비교한다. 본 논문에서 제안하는 시스템은 크게 감정 분석을 위한 이미지 객체 추출부, 추출된 객체로 피실험자의 감정을 분류하는 감정 분류부로 구성되어 있다. 객체 추출과 이미지 분류 모두 CNN(Convolution Neural Network) 기반의 딥러닝 모델을 사용하며, 이미지 분류 모델은 서로 다른 데이터셋으로 모델을 사전 훈련한 후, 훈련 데이터셋으로 전이 학습하여 모델의 성능을 비교한다. 그림 심리 분석을 위한 HTP test 스케치 데이터셋은, HTP Test 와 동일하게 피실험자가 3 개 클래스의 집, 나무, 사람의 그림을 그려 자체 수집하였다.

• Nuclear Engineering and Technology
• /
• v.55 no.10
• /
• pp.3716-3724
• /
• 2023

Loose parts monitoring and detecting alarm type in real Nuclear Power Plant have challenges such as background noise, insufficient alarm data, and difficulty of distinction between alarm data that occur during start and stop. Although many signal processing methods and alarm determination algorithms have been developed, it is not easy to determine valid alarm and extract the meaning data from alarm signal including background noise. To address these issues, this paper proposes a denoising autoencoder-based majority vote classification. Training and test data are prepared by acquiring alarm data from real NPP and simulation facility for data augmentation, and noisy data is reproduced by adding Gaussian noise. Using DAEs with 3, 5, 7, and 9 layers, features are extracted for each model and classified into neural networks. Finally, the results obtained from each DAE are classified by majority voting. Also, through comparison with other methods, the accuracy and the false alarm rate are compared, and the excellence of the proposed method is confirmed.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1273-1283
• /
• 2022

Recently, forest fires have frequently occurred due to climate change, leading to human and property damage every year. The forest fire monitoring technique using remote sensing can obtain quick and large-scale information of fire-damaged areas. In this study, the Gangneung and Donghae forest fires that occurred in March 2022 were analyzed using the spectral band of Sentinel-2, the normalized difference vegetation index (NDVI), and the normalized difference water index (NDWI) to classify the affected areas of forest fires. The U-net based convolutional neural networks (CNNs) model was simulated for the fire-damaged areas. The accuracy of forest fire classification in Donghae and Gangneung classification was high at 97.3% (f₁=0.486, IoU=0.946). The same model used in Donghae and Gangneung was applied to Uljin and Samcheok areas to get rid of the possibility of overfitting often happen in machine learning. As a result, the portion of overlap with the forest fire damage area reported by the National Institute of Forest Science (NIFoS) was 74.4%, confirming a high level of accuracy even considering the uncertainty of the model. This study suggests that it is possible to quantitatively evaluate the classification of forest fire-damaged area using a spectral band and indices similar to that of the Compact Advanced Satellite 500 (CAS500-4) in the Sentinel-2.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.12
• /
• pp.59-68
• /
• 2022

This paper presents a method for 1:1 verification by comparing the similarity between the given real product image and the drawing image. The proposed method combines two existing CNN-based deep learning models to construct a Siamese Network. After extracting the feature vector of the image through the FC (Fully Connected) Layer of each network and comparing the similarity, if the real product image and the drawing image (front view, left and right side view, top view, etc) are the same product, the similarity is set to 1 for learning and, if it is a different product, the similarity is set to 0. The test (inference) model is a deep learning model that queries the real product image and the drawing image in pairs to determine whether the pair is the same product or not. In the proposed model, through a comparison of the similarity between the real product image and the drawing image, if the similarity is greater than or equal to a threshold value (Threshold: 0.5), it is determined that the product is the same, and if it is less than or equal to, it is determined that the product is a different product. The proposed model showed an accuracy of about 71.8% for a query to a product (positive: positive) with the same drawing as the real product, and an accuracy of about 83.1% for a query to a different product (positive: negative). In the future, we plan to conduct a study to improve the matching accuracy between the real product image and the drawing image by combining the parameter optimization study with the proposed model and adding processes such as data purification.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.1
• /
• pp.163-169
• /
• 2020

Recently, research on automation and unmanned operation of machines in the industrial field has been conducted with the advent of AI, Big data, and the IoT, which are the core technologies of the Fourth Industrial Revolution. The machines for these automation processes are controlled based on the data collected from the sensors attached to them, and further, the processes are managed. Conventionally, the abnormalities of sensors are periodically checked and managed. However, due to various environmental factors and situations in the industrial field, there are cases where the inspection due to the failure is not missed or failures are not detected to prevent damage due to sensor failure. In addition, even if a failure occurs, it is not immediately detected, which worsens the process loss. Therefore, in order to prevent damage caused by such a sudden sensor failure, it is necessary to identify the failure of the sensor in an embedded system in real-time and to diagnose the failure and determine the type for a quick response. In this paper, a deep neural network-based fault diagnosis system is designed and implemented using Raspberry Pi to classify typical sensor fault types such as erratic fault, hard-over fault, spike fault, and stuck fault. In order to diagnose sensor failure, the network is constructed using Google's proposed Inverted residual block structure of MobilieNetV2. The proposed scheme reduces memory usage and improves the performance of the conventional CNN technique to classify sensor faults.