• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.851-853
• /
• 2022

고령화로 인해 65세 이상 운전자가 급증하며 고령운전자의 교통사고 비율이 증가함에 따라 시급한 사회 문제로 떠오르고 있다. 이에 본 연구에서는 객체 검출, 인식 모델을 결합하고 신호등을 인식하여 Text-To-Speech(TTS)로 알리는 쿠버네티스 기반의 프레임워크를 제안한다. 객체 검출 단계에서는 YOLOv5 모델들의 성능을 비교하여 활용하였으며 객체 인식 단계에서는 C-RNN 기반의 attention-OCR 모델을 활용하였다. 이는 신호등의 내부 LED 영역이 아닌 이미지 전체를 인식하는 방식으로 오탐지 요소를 낮춰 인식률을 높였다. 결과적으로 1,628장의 테스트 데이터에서 accuracy 0.997, F1-score 0.991의 성능 평가를 얻어 제안한 프레임워크의 타당성을 입증하였다. 본 연구는 후속 연구에서 특정 도메인에 딥러닝 모델을 한정하지 않고 다양한 분야의 모델을 접목할 수 있도록 하며 고령 운전자 및 신호 위반으로 인한 교통사고 문제를 예방할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1102-1103
• /
• 2023

안전모 착용 여부를 확인하는 객체 탐지 모델을 물류 현장에서 활용하기 위해서는 안전모를 착용한 경우와 착용하지 않은 경우를 정확하게 탐지해야 한다. 하지만 학습 데이터가 안전모를 착용한 클래스와 착용하지 않은 클래스 간 불균형이 존재하는 경우 해당 데이터만으로는 태스크에 맞게 학습이됐다고 보긴 힘들다. 본 연구는 데이터 증강 기법 적용 시 임의의 데이터에 증강을 적용하는 대신 상대적으로 적은 안전모를 착용하지 않은 클래스를 포함하는 이미지에 대하여 데이터 증강 기법을 적용하였다. 여러 데이터 증강 기법 중 Rotation, Gaussian Noise, 객체를 기준으로 한 Crop을 직접 구현 및 적용하여 객체 탐지 모델인 YOLOv5의 성능을 효과적으로 높이며 더욱 강건한 모델을 개발하는 방법을 제안한다.

• Korean Journal of Artificial Intelligence
• /
• v.11 no.4
• /
• pp.29-34
• /
• 2023

The flow from developing a machine learning model to deploying it in a production environment suffers challenges. Efficient and reliable deployment is critical for realizing the true value of machine learning models. Bridging this gap between development and publication has become a pivotal concern in the machine learning community. FastAPI, a modern and fast web framework for building APIs with Python, has gained substantial popularity for its speed, ease of use, and asynchronous capabilities. This paper focused on leveraging FastAPI for deploying machine learning models, addressing the potentials associated with integration, scalability, and performance in a production setting. In this work, we explored the seamless integration of machine learning models into FastAPI applications, enabling real-time predictions and showing a possibility of scaling up for a more diverse range of use cases. We discussed the intricacies of integrating popular machine learning frameworks with FastAPI, ensuring smooth interactions between data processing, model inference, and API responses. This study focused on elucidating the integration of machine learning models into production environments using FastAPI, exploring its capabilities, features, and best practices. We delved into the potential of FastAPI in providing a robust and efficient solution for deploying machine learning systems, handling real-time predictions, managing input/output data, and ensuring optimal performance and reliability.

• International Journal of Computer Science & Network Security
• /
• v.23 no.11
• /
• pp.183-189
• /
• 2023

A stroke is a medical disease where a blood vessel in the brain ruptures, causes damage to the brain. If the flow of blood and different nutrients to the brain is intermittent, symptoms may occur. Stroke is other reason for loss of life and widespread disorder. The prevalence of stroke is high in growing countries, with ischemic stroke being the high usual category. Many of the forewarning signs of stroke can be recognized the seriousness of a stroke can be reduced. Most of the earlier stroke detections and prediction models uses image examination tools like CT (Computed Tomography) scan or MRI (Magnetic Resonance Imaging) which are costly and difficult to use for actual-time recognition. Machine learning (ML) is a part of artificial intelligence (AI) that makes software applications to gain the exact accuracy to predict the end results not having to be directly involved to get the work done. In recent times ML algorithms have gained lot of attention due to their accurate results in medical fields. Hence in this work, Stroke disease identification system by using Machine Learning algorithm is presented. The ML algorithm used in this work is Artificial Neural Network (ANN). The result analysis of presented ML algorithm is compared with different ML algorithms. The performance of the presented approach is compared to find the better algorithm for stroke identification.

• Journal of the Korean Society of Industry Convergence
• /
• v.27 no.2_1
• /
• pp.257-268
• /
• 2024

Soccer is type of sport that carries a high risk of injury. Injury is not only cause in the unlucky soccer carrier and also team performance as well as financial effects can be worse since soccer is a team-based game. The duration of recovery from a soccer injury typically relies on its type and severity. Therefore, we conduct this research in order to predict the probability of players injury type using machine learning technologies in this paper. Furthermore, we compare different machine learning models to find the best fit model. This paper utilizes various supervised classification machine learning models, including Decision Tree, Random Forest, K-Nearest Neighbors (KNN), and Naive Bayes. Moreover, based on our finding the KNN and Decision models achieved the highest accuracy rates at 70%, surpassing other models. The Random Forest model followed closely with an accuracy score of 62%. Among the evaluated models, the Naive Bayes model demonstrated the lowest accuracy at 56%. We gathered information about 54 professional soccer players who are playing in the top five European leagues based on their career history. We gathered information about 54 professional soccer players who are playing in the top five European leagues based on their career history.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.42 no.8
• /
• pp.674-682
• /
• 2014

Radio Frequency Test Set (RFTS) is essential to verify Telemetry, Command & Ranging (TC&R) RF subsystem of both Low Earth Orbit (LEO) and Geostationary Earth Orbit (GEO) satellite during Assembly Integration & Test (AI&T). The existing RFTS was specialized for each project and needed to be modified for each new satellite. The new design enables RFTS to be used in various projects. The hardware and software was designed considering this and therefore it could be directly used in other projects within a similar test period without modification or inconvenience. It will be also easily controlled, modified, and managed through the extension in modularization according to each function and the use of COTS (commercial on-the-self) and this will improve system reliability. A more reliable RF test measurement is also provided in this new RFTS by using an accurate reference clock signal.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2008.03a
• /
• pp.16-21
• /
• 2008

Aero-engine, as one kind of rotating machinery with complex structure and high rotating speed, has complicated vibration faults. Therefore, condition monitoring and fault diagnosis system is very important for airplane security. In this paper, a vibration data acquisition and intelligent fault diagnosis system is introduced. First, the vibration data acquisition part is described in detail. This part consists of hardware acquisition modules and software analysis modules which can realize real-time data acquisition and analysis, off-line data analysis, trend analysis, fault simulation and graphical result display. The acquisition vibration data are prepared for the following intelligent fault diagnosis. Secondly, two advanced artificial intelligent(AI) methods, mapping-based and rule-based, are discussed. One is artificial neural network(ANN) which is an ideal tool for aero-engine fault diagnosis and has strong ability to learn complex nonlinear functions. The other is data mining, another AI method, has advantages of discovering knowledge from massive data and automatically extracting diagnostic rules. Thirdly, lots of historical data are used for training the ANN and extracting rules by data mining. Then, real-time data are input into the trained ANN for mapping-based fault diagnosis. At the same time, extracted rules are revised by expert experience and used for rule-based fault diagnosis. From the results of the experiments, the conclusion is obvious that both the two AI methods are effective on aero-engine vibration fault diagnosis, while each of them has its individual quality. The whole system can be developed in local vibration monitoring and real-time fault diagnosis for aero-engine.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.5
• /
• pp.205-212
• /
• 2019

Scene graph is a kind of knowledge graph that represents both objects and their relationships found in a image. This paper proposes a 3D scene graph generation model for three-dimensional indoor environments. An 3D scene graph includes not only object types, their positions and attributes, but also three-dimensional spatial relationships between them, An 3D scene graph can be viewed as a prior knowledge base describing the given environment within that the agent will be deployed later. Therefore, 3D scene graphs can be used in many useful applications, such as visual question answering (VQA) and service robots. This proposed 3D scene graph generation model consists of four sub-networks: object detection network (ObjNet), attribute prediction network (AttNet), transfer network (TransNet), relationship prediction network (RelNet). Conducting several experiments with 3D simulated indoor environments provided by AI2-THOR, we confirmed that the proposed model shows high performance.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.3
• /
• pp.110-115
• /
• 2021

With the era of the 4th Industrial Revolution, education on artificial intelligence is one of the important topics. However, since existing education is aimed at knowledge, it is not suitable for developing the active problem-solving ability and AI utilization ability required by artificial intelligence education. To solve this problem, we proposes PBL-based education method in which learners learn in the process of solving the presented problem. The problem presented to the learner is a completed project. This project consists of three types: a classification model, the training data of the classification model, and the block code to be executed according to the classified result. The project works, but each component is designed to perform a low level of operation. In order to solve this problem, the learners can expect to improve their computational thinking skills by finding problems in the project through testing, finding solutions through discussion, and improving to a higher level of operation.

• Journal of Korea Entertainment Industry Association
• /
• v.15 no.4
• /
• pp.279-286
• /
• 2021

In this study, we looked at the appearance of Edutech, which is being put into the educational field after Corona 19, with the advent of the 4th industrial revolution. In the era of the 4th industrial revolution, the infrastructure, data, and service of Smart Stick that actively utilized ICT became the main pillars of smart education. In particular, smart education is being implemented through e-learning, smart learning, and edutech, and on this basis, it has become possible through the expansion and use of the Internet and computers, the dissemination of smart devices, and a software foundation using big data. Based on this, it was confirmed that Edutech is being implemented through the establishment of a quarantine safety net, a learning safety net, and a care safety net for individual learners and safe life based on artificial intelligence. Lastly, in order for edutech education using big data to become a discourse for everyone, it is necessary to consider artificial intelligence and ethics in the use and application of edutech.