• Annual Conference of KIPS
• /
• 2022.05a
• /
• pp.655-657
• /
• 2022

본 논문은 유튜브에 업로드 된 운동 영상을 시청하는 사람의 얼굴 영역을 YoloV3을 이용하여 얼굴 영상에서 눈 및 입술영역을 검출하는 방법을 연구하여, YoloV3은 딥 러닝을 이용한 물체 검출 방법으로 기존의 특징 기반 방법에 비해 성능이 우수한 것으로 알려져 있다. 본 논문에서는 영상을 다차원적으로 분리하고 클래스 확률(Class Probability)을 적용하여 하나의 회귀 문제로 접근한다. 영상의 1 frame을 입력 이미지로 CNN을 통해 텐서(Tensor)의 그리드로 나누고, 각 구간에 따라 객체인 경계 박스와 클래스 확률을 생성해 해당 구역의 눈과 입을 검출한다. 검출된 이미지 감성 분석을 통해, 운동 영상 중 하이라이트 부분을 자동으로 선별하는 시스템을 설계하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.394-394
• /
• 2022

최근 스마트 수도미터의 보급을 통해 수용가구별 물 사용 자료를 수집할 수 있다. 이런 수용가구별 물 사용 패턴은 주말, 날씨 등 다양한 요인으로 인해 비선형적 특성을 가지고 있다. 그로인해 전통적인 시계열 예측 모형인 ARIMA 모형으로 적용하기 어렵다. 따라서 본 연구에서는 딥러닝 기반의 LSTM 모형을 통해 수용가구별 물 소비량 예측 모형을 개발하였다. 이 모형은 비선형적인 물 소비 패턴을 학습하기 위해 다양한 변수를 고려하였다. 서로 다른 종류의 4개 type (A : 단독주택, B: 아파트, C: 음식점, D : 초등학교)의 수용가구에 대한 ARIMA 모형과 LSTM 모형을 개발하였고, 학습에 사용되지 않은 새로운 데이터를 적용하여 정량적으로 예측성능을 비교했다. 그 결과, 모든 수용가구에서 LSTM 모형이 ARIMA 모형보다 성능이 우수하였다 (상관계수 : 평균89% | RMSE : 평균 5.60m³). 따라서 본 연구에서 제안한 모형은 수용가구별 물 사용량을 예측하는데 높은 활용도를 보일 것으로 기대된다.

• Annual Conference of KIPS
• /
• 2023.05a
• /
• pp.131-133
• /
• 2023

악성코드를 유포할 때 프로그램 코드만으로 악성코드의 유무를 확인할 수 없도록 조치하여 분석을 지연시키는 방식을 사용하는 방향으로 발전하고 있다. 악성코드를 실행하지 않고 코드와 구조만으로 분석하는 정적 분석으로는 악성코드를 판별할 수 없어 코드를 직접 실행해 분석하는 동적 분석을 이용해야 한다. 본 논문에서는 난독화된 비정상적인 코드를 직접 실행한 동적 분석데이터와 일반적이지 않은 섹션들의 정보를 추출한 정적 분석데이터를 이용해 동적-정적 분석 데이터와 딥러닝 모델을 통해 난독화 및 패킹된 악성코드를 탐지하는 기법을 제안한다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.193-198
• /
• 2023

We study a learner-customized lecture recommendation project using deep learning. Recommendation systems can be easily found on the web and apps, and examples using this feature include recommending feature videos by clicking users and advertising items in areas of interest to users on SNS. In this study, the sentence similarity Word2Vec was mainly used to filter twice, and the course was recommended through the Surprise library. With this system, it provides users with the desired classification of course data conveniently and conveniently. Surprise Library is a Python scikit-learn-based library that is conveniently used in recommendation systems. By analyzing the data, the system is implemented at a high speed, and deeper learning is used to implement more precise results through course steps. When a user enters a keyword of interest, similarity between the keyword and the course title is executed, and similarity with the extracted video data and voice text is executed, and the highest ranking video data is recommended through the Surprise Library.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.3
• /
• pp.19-26
• /
• 2020

Steel production requires high-quality surfaces with minimal defects. Therefore, the detection algorithms for the surface defects of steel strip should have good generalization performance. To meet the growing demand for high-quality products, the use of intelligent visual inspection systems is becoming essential in production lines. In this paper, we proposed a ShuffleDefectNet defect detection system based on deep learning. The proposed defect detection system exceeds state-of-the-art performance for defect detection on the Northeastern University (NEU) dataset obtaining a mean average accuracy of 99.75%. We train the best performing detection with different amounts of training data and observe the performance of detection. We notice that accuracy and speed improve significantly when use the overall architecture of ShuffleDefectNet.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.334-337
• /
• 2019

In this study, we built the data collection system to learn user's habit data by deep learning and to create an indoor environment according to the situation. The system consists of a data collection server and several sensor nodes, which creates the environment according to the data collected. We used Google Inception v3 network to analyze the photographs and hand-designed second DNN (Deep Neural Network) to infer behaviors. As a result of the DNN learning, we gained 98.4% of Testing Accuracy. Through this results, we were be able to prove that DNN is capable of extrapolating the situation.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.355-356
• /
• 2021

본 논문에서는 Python 3의 Keras 모듈을 이용하여 특정 자동차에 대한 최적의 판매자권장소비자가격(MSRP)을 예측하는 시스템을 제안한다. 이 시스템은 2004년에 미국에서 시판된 428종류의 자동차에 대한 정보를 제조사, 차종, 생산지, 엔진 크기, 실린더 수, 시내 주행 시 연비, 고속도로 주행 시 연비, 마력, 차체 무게, 차체 길이의 독립변수를 사용하여 자체적으로 딥러닝한 회귀모델을 통해 특정 지표가 주어진 차량에 대해 종속변수인 판매자권장소비자가격을 예측한다. Optimizer를 adam으로, 학습률을 0.005으로 설정한 경우의 검증 MAE 값이 3842.98로 가장 낮게 산출되었고, 해당 모델의 결과는 예측값과 실제값의 오차율이 ±15% 정도 내외로 예측된 표본의 비율이 약 80.14%로 측정되었다. 위 모델은 향후 신차 가격 결정 및 중고차 시장에서 구매, 판매 결정을 돕는 등 특정 시장 내에서 다양한 자동차의 가치를 판단하기에 유용할 것으로 전망된다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.5
• /
• pp.83-99
• /
• 2021

With the advent of the fourth industrial revolution, studies on driving management and driving strategies of autonomous vehicles are emerging. While obtaining microscopic traffic data on vehicles is essential for such research, we also see that conventional traffic data collection methods cannot collect the driving behavior of individual vehicles. In this study, UAV videos were used to collect traffic data from the viewpoint of the aerial base that is microscopic. To overcome the limitations of the related research in the literature, the micro-traffic data were estimated using the multiple object tracking of deep learning and an image registration technique. As a result, the speed obtained error rates of MAE 3.49 km/h, RMSE 4.43 km/h, and MAPE 5.18 km/h, and the traffic obtained a precision of 98.07% and a recall of 97.86%.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.45-50
• /
• 2023

As the ownership of pets has steadily increased in recent years, the need for an effective pet management system has grown. In this study, we propose a pet management system with a deep learning-based emotion recognition SNS. The system detects emotions through pet facial expressions using a convolutional neural network (CNN) and shares them with a user community through SNS. Through SNS, pet owners can connect with other users, share their experiences, and receive support and advice for pet management. Additionally, the system provides comprehensive pet management, including tracking pet health and vaccination and reservation reminders. Furthermore, we added a function to manage and share pet walking records so that pet owners can share their walking experiences with other users. This study demonstrates the potential of utilizing AI technology to improve pet management systems and enhance the well-being of pets and their owners.

• Journal of Digital Convergence
• /
• v.18 no.11
• /
• pp.267-272
• /
• 2020

Macroeconomics are one of the indicators that are preceded and analyzed when analyzing stocks because it shows the movement of a country's economy as a whole. The overall economic situation at the national level, such as national income, inflation, unemployment, exchange rates, currency, interest rates, and balance of payments, has a great affect on the stock market, and economic indicators are actually correlated with stock prices. It is the main source of data for analysts to watch with interest and to determine buy and sell considering the impact on individual stock prices. Therefore, economic indicators that impact on the stock price are analyzed as leading indicators, and the stock price prediction is predicted through deep learning-based prediction, after that the actual stock price is compared. If you decide to buy or sell stocks by analysis of stock prediction, then stocks can be investments, not gambling. Therefore, this research was conducted to enable automated stock trading by using macro-indicators and deep learning algorithms in artificial intelligence.