• Journal of Software Assessment and Valuation
• /
• v.16 no.2
• /
• pp.135-142
• /
• 2020

In this paper, we describe the iris localization method in RGB images. Most of the iris localization methods are developed for infrared images, thus an iris localization method in RGB images is required for various applications. The proposed method consists of four stages: i) detection of the candidate irises using circular Hough transform (CHT) from an input image, ii) detection of a pupil center based on deep learning, iii) determine the iris using the pupil center, and iv) correction of the iris region. The candidate irises are detected in the order of the number of intersections of the center point candidates after generating the Hough space, and the iris in the candidates is determined based on the detected pupil center. Also, the error due to distortion of the iris shape is corrected by finding a new boundary point based on the detected iris center. In experiments, the proposed method has an improved accuracy about 27.4% compared to the CHT method.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_1
• /
• pp.621-635
• /
• 2023

The assessment of structural condition is a crucial process for evaluating its usability and determining the diagnostic cycle. The currently employed manpower-based methods suffer from issues related to safety, efficiency, and objectivity. To address these concerns, research based on deep learning using images is being conducted. However, acquiring structural damage data is challenging, making it difficult to construct a substantial amount of training data, thus limiting the effectiveness of deep learning-based condition assessment. In this study, we propose a foundation model-based 2-step structural damage analysis to overcome the lack of training data in image-based structural condition assessments. We subdivided the elements of structural condition assessment into instantiation and quantification. In the quantification step, we applied a foundation model for image segmentation. Our method demonstrated a 10%-point increase in mean intersection over union compared to conventional image segmentation techniques, with a notable 40%-point improvement in the case of rebar exposure. We anticipate that our proposed approach will enhance performance in domains where acquiring training data is challenging.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.763-768
• /
• 2023

In this paper, the existing statistical method (ARIMA) and machine learning method (Informer) were developed and compared to predict the distribution volume of pharmaceuticals. It was found that a machine learning-based model is advantageous for daily data prediction, and it is effective to use ARIMA for monthly prediction and switch to Informer as the data increases. The prediction error rate (RMSE) was reduced by 26.6% compared to the previous method, and the prediction accuracy was improved by 13%, resulting in a result of 86.2%. Through this thesis, we find that there is an advantage of obtaining the best results by ensembleing statistical methods and machine learning methods. In addition, machine learning-based AI models can derive the best results through deep learning operations even in irregular situations, and after commercialization, performance is expected to improve as the amount of data increases.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.4
• /
• pp.691-708
• /
• 2022

As the technology of machine learning and deep learning became common, it began to be applied to research on anomaly(abnormal) detection of industrial control systems. In Korea, the HAI dataset was developed and published to activate artificial intelligence research for abnormal detection of industrial control systems, and an AI contest for detecting industrial control system security threats is being conducted. Most of the anomaly detection studies have been to create a learning model with improved performance through the ensemble model method, which is applied either by modifying the existing deep learning algorithm or by applying it together with other algorithms. In this study, a study was conducted to improve the performance of anomaly detection with a post-processing method that detects abnormal data and corrects the labeling results, rather than the learning algorithm and data pre-processing process. Results It was confirmed that the results were improved by about 10% or more compared to the anomaly detection performance of the existing model.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.5
• /
• pp.999-1008
• /
• 2023

Early prediction of chronic diseases such as diabetes is an important issue, and improving the accuracy of diabetes prediction is especially important. Various machine learning and deep learning-based methodologies are being introduced for diabetes prediction, but these technologies require large amounts of data for better performance than other methodologies, and the learning cost is high due to complex data models. In this study, we aim to verify the claim that DNN using the pima dataset and k-fold cross-validation reduces the efficiency of diabetes diagnosis models. Machine learning classification methods such as decision trees, SVM, random forests, logistic regression, KNN, and various ensemble techniques were used to determine which algorithm produces the best prediction results. After training and testing all classification models, the proposed system provided the best results on XGBoost classifier with ADASYN method, with accuracy of 81%, F1 coefficient of 0.81, and AUC of 0.84. Additionally, a domain adaptation method was implemented to demonstrate the versatility of the proposed system. An explainable AI approach using the LIME and SHAP frameworks was implemented to understand how the model predicts the final outcome.

• The Journal of Bigdata
• /
• v.6 no.2
• /
• pp.125-137
• /
• 2021

The number of CCTV installed in Korea is over 1.3 million, increasing by more than 15% annually. However, due to the limited budget compared to the installation demand, the infrastructure is composed of 500,000 pixel low-quality CCTV, and there is a limits on identification of objects in the video. Public CCTV has high utility in various fields such as crime prevention, traffic information collection (control), facility management, and fire prevention. Especially, since installed in high height, it works as its role in solving diverse crime and is in increasing trend. However, the current public CCTV field is operated with potential problems such as inability to identify due to environmental factors such as fog, snow, and rain, and the low-quality of collected images due to the installation of low-quality CCTV. Therefore, in this study, in order to remove the typical low-quality elements of public CCTV, the method of attenuating scattered light in the image caused by dust, water droplets, fog, etc and algorithm application method which uses deep-learning algorithm to improve input video into videos over quality over 4K are suggested.

• Annual Conference of KIPS
• /
• 2022.11a
• /
• pp.628-629
• /
• 2022

딥러닝의 높은 예측 정확도를 위해서는 많은 양의 학습 데이터가 필요하다. 그러나 실세계에서 많은 양의 레이블이 붙은 데이터를 구하는 것은 어렵고 많은 비용이 든다. 때문에 레이블이 없이도 양질의 표현 학습이 가능한 자기지도학습이 각광을 받고 있다. 와서스타인 거리는 생성모델에도 쓰이지만 의사 레이블 (pseudo label) 을 만들어 레이블이 없는 데이터들을 분류 하는데도 좋은 성능을 보이고 있다. 따라서. 본 연구는 와서스타인 거리를 자기지도학습에 접목시키는 방법을 제안한다. 실험을 통해 연구의 가능성을 보인다.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.113-118
• /
• 2023

딥러닝을 이용한 언어 모델은 다양한 분야에서 사용되고 있는데, 그 중 교육 분야에선 꾸준히 시험 문항을 자동으로 생성하려는 요구가 존재해 왔다. 그러나 빈칸 추론 문항, 그 중에서도 어구 단위 빈칸 추론 문항은 학습 및 평가 목적으로 널리 쓰이고 있지만, 이를 자동 생성하려는 연구는 상대적으로 드물다. 이에 본 연구에선 masked language modeling (MLM)을 이용한 난이도 조절이 가능한 어구 단위 빈칸 추론 문항 생성 시스템을 제안한다. 본 시스템은 정답 생성 모델의 attention 정보에 따라 지문 내 중요한 어구를 삭제해 오답을 생성하고, 동시에 어구의 삭제 비율을 조절함으로써 더 쉽거나 더 어려운 오답을 만들어낼 수 있다. 평가 결과, 제안한 시스템은 기존 접근법보다 정답과의 유사도가 최고 28.3% 낮았고, 또한 난이도 설정에 따라 쉬운 오답이 어려운 오답에 비해 유사도가 15.1% 낮아, 더 정답과 먼 뜻의 오답을 생성해내었다.

• The Journal of the Korea Contents Association
• /
• v.21 no.10
• /
• pp.48-58
• /
• 2021

For securing technology and business competences of companies that is the engine of domestic industrial growth, government-supported policy programs for the creation of commercialization results in various forms such as 『Technology Transaction Market Vitalization』 and 『Technology Finance-based R&D Commercialization Support』 have been carried out since 2014. So far, various studies on technology valuation theories and evaluation variables have been formalized by experts from various fields, and have been utilized in the field of technology commercialization. However, Their practicality has been questioned due to the existing constraint that valuation results are assessed lower than the expectation in the evaluation sector. Even considering that the evaluation results may differ depending on factors such as the corporate situation and investment environment, it is necessary to establish a reference infrastructure to secure the objectivity and reliability of the technology valuation results. In this study, we investigate the evaluation infrastructure built by each institution and examine whether the latest artificial neural networks and deep learning technologies are applicable for performing predictive simulation of technology values based on principal variables, and predicting sales estimates and qualitative evaluation scores in order to embed onto the technology valuation system.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.103-128
• /
• 2021

Recently, investors' interest and the influence of stock-related information dissemination are being considered as significant factors that explain stock returns and volume. Besides, companies that develop, distribute, or utilize innovative new technologies such as artificial intelligence have a problem that it is difficult to accurately predict a company's future stock returns and volatility due to macro-environment and market uncertainty. Market uncertainty is recognized as an obstacle to the activation and spread of artificial intelligence technology, so research is needed to mitigate this. Hence, the purpose of this study is to propose a machine learning model that predicts the volatility of a company's stock price by using the internet search volume of artificial intelligence-related technology keywords as a measure of the interest of investors. To this end, for predicting the stock market, we using the VAR(Vector Auto Regression) and deep neural network LSTM (Long Short-Term Memory). And the stock price prediction performance using keyword search volume is compared according to the technology's social acceptance stage. In addition, we also conduct the analysis of sub-technology of artificial intelligence technology to examine the change in the search volume of detailed technology keywords according to the technology acceptance stage and the effect of interest in specific technology on the stock market forecast. To this end, in this study, the words artificial intelligence, deep learning, machine learning were selected as keywords. Next, we investigated how many keywords each week appeared in online documents for five years from January 1, 2015, to December 31, 2019. The stock price and transaction volume data of KOSDAQ listed companies were also collected and used for analysis. As a result, we found that the keyword search volume for artificial intelligence technology increased as the social acceptance of artificial intelligence technology increased. In particular, starting from AlphaGo Shock, the keyword search volume for artificial intelligence itself and detailed technologies such as machine learning and deep learning appeared to increase. Also, the keyword search volume for artificial intelligence technology increases as the social acceptance stage progresses. It showed high accuracy, and it was confirmed that the acceptance stages showing the best prediction performance were different for each keyword. As a result of stock price prediction based on keyword search volume for each social acceptance stage of artificial intelligence technologies classified in this study, the awareness stage's prediction accuracy was found to be the highest. The prediction accuracy was different according to the keywords used in the stock price prediction model for each social acceptance stage. Therefore, when constructing a stock price prediction model using technology keywords, it is necessary to consider social acceptance of the technology and sub-technology classification. The results of this study provide the following implications. First, to predict the return on investment for companies based on innovative technology, it is most important to capture the recognition stage in which public interest rapidly increases in social acceptance of the technology. Second, the change in keyword search volume and the accuracy of the prediction model varies according to the social acceptance of technology should be considered in developing a Decision Support System for investment such as the big data-based Robo-advisor recently introduced by the financial sector.