• Journal of Industrial Technology
• /
• v.39 no.1
• /
• pp.27-31
• /
• 2019

Semiconductor logistics systems are facing difficulties in increasing production as production processes become more complicated due to the upgrading of fine processes. Therefore, the purpose of the research is to design predictive models that can predict traffic during the pre-planning stage, identify the risk zones that occur during the production process, and prevent them in advance. As a solution, we build FABs using automode simulation to collect data. Then, the traffic prediction model of the areas of interest is constructed using deep learning techniques (keras - multistory conceptron structure). The design of the predictive model gave an estimate of the traffic in the area of interest with an accuracy of about 87%. The expected effect can be used as an indicator for making decisions by proactively identifying congestion risk areas during the Fab Design or Factory Expansion Planning stage, as the maximum traffic per section is predicted.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.16 no.4
• /
• pp.1-7
• /
• 2020

Over the past few years, Wi-Fi signal based indoor positioning system (IPS) has been researched extensively because of its low expenses of infrastructure deployment. There are two major aspects of location-related information contained in Wi-Fi signals. One is channel state information (CSI), and one is received signal strength indicator (RSSI). Compared to the RSSI, the CSI has been widely utilized because it is able to reveal fine-grained information related to locations. However, the conventional IPS that employs a single access point (AP) does not exhibit decent performance especially in the environment of non-line-of-sight (NLOS) situations due to the reliability degeneration of signals caused by multipath fading effect. In order to address this problem, in this paper, we propose a novel method that utilizes multiple APs instead of a single AP to enhance the robustness of the IPS. In our proposed method, a hybrid neural network is applied to the CSIs collected from multiple APs. By relying more on the fingerprint constructed by the CSI collected from an AP that is less affected by the NLOS, we find that the performance of the IPS is significantly improved.

• Korean Journal of Materials Research
• /
• v.31 no.12
• /
• pp.719-726
• /
• 2021

Through the process of chemical vapor deposition, Tungsten Hexafluoride (WF₆) is widely used by the semiconductor industry to form tungsten films. Tungsten Hexafluoride (WF₆) is produced through manufacturing processes such as pulverization, wet smelting, calcination and reduction of tungsten ores. The manufacturing process of Tungsten Hexafluoride (WF₆) is required thorough quality control to improve productivity. In this paper, a real-time detection system for oxidation defects that occur in the manufacturing process of Tungsten Hexafluoride (WF₆) is proposed. The proposed system is implemented by applying YOLOv5 based on Convolutional Neural Network (CNN); it is expected to enable more stable management than existing management, which relies on skilled workers. The implementation method of the proposed system and the results of performance comparison are presented to prove the feasibility of the method for improving the efficiency of the WF₆ manufacturing process in this paper. The proposed system applying YOLOv5s, which is the most suitable material in the actual production environment, demonstrates high accuracy (mAP@0.5 99.4 %) and real-time detection speed (FPS 46).

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.229-229
• /
• 2023

강화학습은 에이전트(agent)가 주어진 환경(environment)과의 상호작용을 통해서 상태(state)를 변화시켜가며 최대의 보상(reward)을 얻을 수 있도록 최적의 행동(action)을 학습하는 기계학습법을 의미한다. 최근 알파고와 같은 게임뿐만 아니라 자율주행 자동차, 로봇 제어 등 다양한 분야에서 널리 사용되고 있다. 상수도관망 분야의 경우에도 펌프 운영, 밸브 운영, 센서 최적 위치 선정 등 여러 문제에 적용되었으나, 설계에 강화학습을 적용한 연구는 없었다. 설계의 경우, 관망의 크기가 커짐에 따라 알고리즘의 탐색 공간의 크기가 증가하여 기존의 최적화 알고리즘을 이용하는 것에는 한계가 존재한다. 따라서 본 연구는 강화학습을 이용하여 상수도관망의 구성요소와 환경요인 간의 복잡한 상호작용을 고려하는 설계 방법론을 제안한다. 모델의 에이전트를 딥 강화학습(Deep Reinforcement Learning)으로 구성하여, 상태 및 행동 공간이 커 발생하는 고차원성 문제를 해결하였다. 또한, 해당 모델의 상태 및 보상으로 절점에서의 압력 및 수요량과 설계비용을 고려하여 적절한 수량과 수압의 용수 공급이 가능한 경제적인 관망을 설계하도록 하였다. 모델의 행동은 실제로 공학자가 설계하듯이 절점마다 하나씩 차례대로 다른 절점과의 연결 여부를 결정하는 것으로, 이를 통해 관망의 레이아웃(layout)과 관경을 결정한다. 본 연구에서 제안한 방법론을 규모가 큰 그리드 네트워크에 적용하여 모델을 검증하였으며, 고려해야 할 변수의 개수가 많음에도 불구하고 목적에 부합하는 관망을 설계할 수 있었다. 모델 학습과정 동안 에피소드의 평균 길이와 보상의 크기 등의 변화를 비교하여, 제안한 모델의 학습 능력을 평가 및 보완하였다. 향후 강화학습 모델을 통해 신뢰성(reliability) 또는 탄력성(resilience)과 같은 시스템의 성능까지 고려한 설계가 가능할 것으로 기대한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.108-111
• /
• 2022

Applying deep learning to machine vision systems for defect detection of products requires vast amounts of training data about various defect cases. However, since data imbalance occurs according to the type of defect in the actual manufacturing industry, it takes a lot of time to collect product images enough to generalize defect cases. In this paper, we apply a Siamese neural network that can be learned with even a small amount of data to product defect detection, and modify the image pairing method and contrastive loss function by properties the situation of product defect image data. We indirectly evaluated the embedding performance of Siamese neural networks using AUC-ROC, and it showed good performance when the images only paired among same products, not paired among defective products, and learned with exponential contrastive loss.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.2
• /
• pp.24-29
• /
• 2023

In this paper, we propose a real-time bus arrival notification system using deep learning to guarantee movement rights for the visually impaired. In modern society, by using location information of public transportation, users can quickly obtain information about public transportation and use public transportation easily. However, since the existing public transportation information system is a visual system, the visually impaired cannot use it. In Korea, various laws have been amended since the 'Act on the Promotion of Transportation for the Vulnerable' was enacted in June 2012 as the Act on the Movement Rights of the Blind, but the visually impaired are experiencing inconvenience in using public transportation. In particular, from the standpoint of the visually impaired, it is impossible to determine whether the bus is coming soon, is coming now, or has already arrived with the current system. In this paper, we use deep learning technology to learn bus numbers and identify upcoming bus numbers. Finally, we propose a method to notify the visually impaired by voice that the bus is coming by using TTS technology.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.221-224
• /
• 2019

딥러닝을 이용한 Neural Machine Translation(NMT)의 등장으로 기계번역 분야에서 기존의 규칙 기반,통계기반 방식을 압도하는 좋은 성능을 보이고 있다. 본 논문은 기계번역 모델도 중요하지만 무엇보다 중요한 것은 고품질의 학습데이터를 구성하는 일과 전처리라고 판단하여 이에 관련된 다양한 실험을 진행하였다. 인공신경망 기계번역 시스템의 학습데이터 즉 병렬 코퍼스를 구축할 때 양질의 데이터를 확보하는 것이 무엇보다 중요하다. 그러나 양질의 데이터를 구하는 일은 저작권 확보의 문제, 병렬 말뭉치 구축의 어려움, 노이즈 등을 이유로 쉽지 않은 상황이다. 본 논문은 고품질의 학습데이터를 구축하기 위하여 병렬 코퍼스 필터링 기법을 제시한다. 병렬 코퍼스 필터링이란 정제와 다르게 학습 데이터에 부합하지 않다고 판단되며 소스, 타겟 쌍을 함께 삭제 시켜 버린다. 또한 기계번역에서 무엇보다 중요한 단계는 바로 Subword Tokenization 단계이다. 본 논문은 다양한 실험을 통하여 한-영 기계번역에서 가장 높은 성능을 보이는 Subword Tokenization 방법론을 제시한다. 오픈 된 한-영 병렬 말뭉치로 실험을 진행한 결과 병렬 코퍼스 필터링을 진행한 데이터로 만든 모델이 더 좋은 BLEU 점수를 보였으며 본 논문에서 제안하는 형태소 분석 단위 분리를 진행 후 Unigram이 반영된 SentencePiece 모델로 Subword Tokenization를 진행 하였을 시 가장 좋은 성능을 보였다.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.57 no.4
• /
• pp.479-488
• /
• 2024

In recent years, changes in the fishing ground environment have led to reduced catches by fishermen at traditional fishing spots and increased operational costs related to vessel exploration, fuel, and labor. In this study, we developed a deep learning model to classify the fishing activities of drift gillnet fishing boats using AIS (automatic identification system) trajectory data. The proposed model integrates long short-term memory and 1-dimensional convolutional neural network layers to effectively distinguish between fishing (throwing and hauling) and non-fishing operations. Training on a dataset derived from AIS and validation against a subset of CCTV footage, the model achieved high accuracy, with a classification accuracy of 90% for fishing events. These results show that the model can be used effectively to monitor and manage fishing activities in coastal waters in real time.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.8
• /
• pp.171-180
• /
• 2020

Smart Factory consists of digital automation solutions throughout the production process, including design, development, manufacturing and distribution, and it is an intelligent factory that installs IoT in its internal facilities and machines to collect process data in real time and analyze them so that it can control itself. The smart factory's equipment works in a physical combination of numerous hardware, rather than a virtual character being driven by a single object, such as a game. In other words, for a specific common goal, multiple devices must perform individual actions simultaneously. By taking advantage of the smart factory, which can collect process data in real time, if reinforcement learning is used instead of general machine learning, behavior control can be performed without the required training data. However, in the real world, it is impossible to learn more than tens of millions of iterations due to physical wear and time. Thus, this paper uses simulators to develop grid sortation systems focusing on transport facilities, one of the complex environments in smart factory field, and design cooperative multi-agent-based reinforcement learning to demonstrate efficient behavior control.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.991-1005
• /
• 2022

Satellite imaging is an effective supplementary data source for detecting and verifying nuclear activity. It is also highly beneficial in regions with limited access and information, such as nuclear installations. Time series analysis, in particular, can identify the process of preparing for the conduction of a nuclear experiment, such as relocating equipment or changing facilities. Differences in the semantic segmentation findings of time series photos were employed in this work to detect changes in meaningful items connected to nuclear activity. Building, road, and small object datasets made of KOMPSAT 3/3A photos given by AIHub were used to train deep learning models such as U-Net, PSPNet, and Attention U-Net. To pick relevant models for targets, many model parameters were adjusted. The final change detection was carried out by including object information into the first change detection, which was obtained as the difference in semantic segmentation findings. The experiment findings demonstrated that the suggested approach could effectively identify altered pixels. Although the suggested approach is dependent on the accuracy of semantic segmentation findings, it is envisaged that as the dataset for the region of interest grows in the future, so will the relevant scope of the proposed method.