• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.4
• /
• pp.771-780
• /
• 2024

In this study, we implemented a real-time pest detection and prediction system for a strawberry farm using a computer vision model based on the YOLOv5 architecture and an Isolation Forest Classifier. The model performance evaluation showed that the YOLOv5 model achieved a mean average precision (mAP 0.5) of 78.7%, an accuracy of 92.8%, a recall of 90.0%, and an F1-score of 76%, indicating high predictive performance. This system was designed to be applicable not only to strawberry farms but also to other crops and various environments. Based on data collected from a tomato farm, a new AI model was trained, resulting in a prediction accuracy of over 85% for major diseases such as late blight and yellow leaf curl virus. Compared to the previous model, this represented an improvement of more than 10% in prediction accuracy.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.9
• /
• pp.317-322
• /
• 2022

The rapid development of information technology is also bringing about many changes in the medical environment. In particular, it is leading the rapid change of medical image information systems using big data and artificial intelligence (AI). The prescription delivery system (OCS), which consists of an electronic medical record (EMR) and a medical image storage and transmission system (PACS), has rapidly changed the medical environment from analog to digital. When combined with multiple solutions, PACS represents a new direction for advancement in security, interoperability, efficiency and automation. Among them, the combination with artificial intelligence (AI) using big data that can improve the quality of images is actively progressing. In particular, AI PACS, a system that can assist in reading medical images using deep learning technology, was developed in cooperation with universities and industries and is being used in hospitals. As such, in line with the rapid changes in the medical image information system in the medical environment, structural changes in the medical market and changes in medical policies to cope with them are also necessary. On the other hand, medical image information is based on a digital medical image transmission device (DICOM) format method, and is divided into a tomographic volume image, a volume image, and a cross-sectional image, a two-dimensional image, according to a generation method. In addition, recently, many medical institutions are rushing to introduce the next-generation integrated medical information system by promoting smart hospital services. The next-generation integrated medical information system is built as a solution that integrates EMR, electronic consent, big data, AI, precision medicine, and interworking with external institutions. It aims to realize research. Korea's medical image information system is at a world-class level thanks to advanced IT technology and government policies. In particular, the PACS solution is the only field exporting medical information technology to the world. In this study, along with the analysis of the medical image information system using big data, the current trend was grasped based on the historical background of the introduction of the medical image information system in Korea, and the future development direction was predicted. In the future, based on DICOM big data accumulated over 20 years, we plan to conduct research that can increase the image read rate by using AI and deep learning algorithms.

• Journal of Korea Society of Industrial Information Systems
• /
• v.26 no.1
• /
• pp.31-40
• /
• 2021

The important goal of the unmanned vehicle technology is on controlling the direction and speed of the vehicle with information acquired from various sensors, without the intervention of the driver, until the vehicle reaches to its destination. In this paper, our focus is on developing an unmanned conveyance system by exploiting low-cost sensing technology for indoor factories or warehouses, where the moving range of the vehicle is limited. To this end, we propose an architecture of a scalable automated conveyance system. Our proposed system includes a number of unmanned conveyance vehicles, and the efficient control mechanism of the vehicles without neither conflicts nor deadlock between the vehicles being simultaneously moved. By implementing the real prototype of the system, we successfully verify the efficiency and functionality of the proposed system.

• Journal of Software Assessment and Valuation
• /
• v.16 no.1
• /
• pp.45-53
• /
• 2020

In this paper, we describe a new method for acquiring calibration data using a user interaction process, which occurs continuously during web browsing in gaze estimation, and for performing calibration naturally while estimating the user's gaze. The proposed non-intrusive calibration is a tuning process over the pre-trained gaze estimation model to adapt to a new user using the obtained data. To achieve this, a generalized CNN model for estimating gaze is trained, then the non-intrusive calibration is employed to adapt quickly to new users through online learning. In experiments, the gaze estimation model is calibrated with a combination of various user interactions to compare the performance, and improved accuracy is achieved compared to existing methods.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.407-411
• /
• 2022

최근 COVID-19로 인해 대중의 의학 분야 관심이 증가하고 있다. 대부분의 의학문서는 전문용어인 의학용어로 구성되어 있어 대중이 이를 보고 이해하기에 어려움이 있다. 의학용어를 쉬운 뜻으로 풀이하는 모델을 이용한다면 대중이 의학 문서를 쉽게 이해할 수 있을 것이다. 이런 문제를 완화하기 위해서 본 논문에서는 Transformer 기반 번역 모델을 이용한 의학용어 탐지 및 해석 모델을 제안한다. 번역 모델에 적용하기 위해 병렬말뭉치가 필요하다. 본 논문에서는 다음과 같은 방법으로 병렬말뭉치를 구축한다: 1) 의학용어 사전을 구축한다. 2) 의학 드라마의 자막으로부터 의학용어를 찾아서 그 뜻풀이로 대체한다. 3) 원자막과 뜻풀이가 포함된 자막을 나란히 배열한다. 구축된 병렬말뭉치를 이용해서 Transformer 번역모델에 적용하여 전문용어를 찾아서 해석하는 모델을 구축한다. 각 문장은 음절 단위로 나뉘어 사전학습 된 KoCharELECTRA를 이용해서 임베딩한다. 제안된 모델은 약 69.3%의 어절단위 BLEU 점수를 보였다. 제안된 의학용어 해석기를 통해 대중이 의학문서를 좀 더 쉽게 접근할 수 있을 것이다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.4
• /
• pp.47-52
• /
• 2023

With the growing data size and the increased computing load in machine learning, energy-efficient resource planning in IoT systems is becoming increasingly important. In this paper, we suggest a new resource planning policy for real-time workloads that can be fluctuated over time in IoT systems. To handle such situations, we categorize real-time tasks into fixed tasks and variable tasks, and optimize the resource planning for various workload conditions. Based on this, we initiate the IoT system with the configuration for the fixed tasks, and when variable tasks are activated, we update the resource planning promptly for the situation. Simulation experiments show that the proposed policy saves the processor and memory energy significantly.

• Smart Media Journal
• /
• v.13 no.10
• /
• pp.59-66
• /
• 2024

This study utilizes data collected by the Rural Development Administration from smart farm sites to identify key variables affecting cucumber shipment and proposes the most accurate prediction model through comparative analysis of various forecasting models. The dataset includes daily weather conditions, cultivation environments, and management activities from 36 different crop seasons. The predictive models used in this study include Multiple Regression, ARIMA(Auto Regressive Integrated Moving Average), LSTM(Long Short-Term Memory), and SARIMA(Seasonal Auto Regressive Integrated Moving Average). Model performance was evaluated using RMSE and MAE, with SARIMA demonstrating the best results. By optimizing the hyperparameters, SARIMA's prediction accuracy improved significantly, effectively capturing the strong seasonality in cucumber shipments.

• Annual Conference of KIPS
• /
• 2024.10a
• /
• pp.734-737
• /
• 2024

Large Language Model(LLM) 사용이 증가하면서, LLM의 성별 및 인종에 대한 편향성은 사회적 불평등을 심화시킬 수 있는 중요한 문제로 대두되고 있다. 이에 LLM의 편향을 정확하고 신뢰성 있게 측정하는 도구가 필요하다. 본 논문은 LLM의 편향을 평가하는 방법론을 워드 임베딩 분석과 텍스트 생성 편향 분석으로 나누어 검토한다. 워드 임베딩 분석 방법은 단어 벡터 간 거리를 측정해 편향을 정량적으로 평가하는 방식으로, 간호사나 군인과 같은 단어들이 성별이나 인종과 같은 특정 집단과 얼마나 가깝게 매핑되는지를 분석하는 방식이다. 그러나 이 방법은 단어의 문맥적 의미 변화를 충분히 반영하지 못하는 한계가 있다. 반면, 텍스트 생성 편향 분석 방법은 LLM이 실제로 생성한 텍스트에서 나타나는 편향을 직접 평가하는 방식이다. 이를 위해 연구자는 성별이나 인종과 관련된 편향이 드러날 수 있는 문장들로 데이터셋을 구성하고, LLM이 이를 어떻게 처리하는지 분석한다. 이방법은 문맥을 반영해 모델이 생성한 텍스트에서 편향을 평가할 수 있다는 장점이 있지만, 연구자가 데이터셋을 구축하는 과정에서 주관적 판단이나 편향이 개입될 가능성이 있으며, 평가할 수 있는 시나리오가 제한적이라는 한계가 있다. 본 논문은 이러한 한계를 극복하기 위한 향후 연구로, 합성 데이터를 활용하여 데이터셋을 구축하고, 이를 통해 텍스트 생성 편향을 분석하는 방법을 제안한다. 합성 데이터는 다양한 시나리오를 기반으로 무한히 생성할 수 있어, 특정 시나리오에 제한되지 않고 LLM의 편향을 폭넓게 평가할 수 있다. 또한 연구자의 개입을 줄여 데이터셋 구축 시 발생할 수 있는 편향을 최소화하고, 더 공정하고 신뢰성 있는 평가를 가능하게 한다. 이에 따라 합성 데이터를 이용한 텍스트 생성 편향 분석 방법은 LLM의 성별 및 인종 편향을 보다 객관적으로 평가하는 도구로서 중요한 역할을 할 것으로 기대한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.6
• /
• pp.855-860
• /
• 2018

The IT technologies applying to IoT(Internet of Things), Big Data, and AI(Artificial Intelligence) are needed in the era of 4th industrial revolution. So, the IT convergence courses of computer science major which will be required in the companies in order to prepare the crises of 4th industry revolution are necessary. And, one approach to cope with this problem is the training of IT convergence man power based on NCS(National Competency Standard) education. In this paper, we propose and analyze the NCS education courses for computer science major in order to teach the students who are needed in the Korean domestic companies preparing the 4th industrial revolution. The skills and applications of Chatbot, Blockchain, and CPS(Cyber Physical System) for the post mobile and post Internet technologies are included in the proposed courses.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.11
• /
• pp.427-432
• /
• 2019

In this paper, we propose three approaches to modeling Android malware. The first method involves human security experts for meticulously selecting feature sets. With the second approach, we choose 300 features with the highest importance among the top 99% features in terms of occurrence rate. The third approach is to combine multiple models and identify malware through weighted voting. In addition, we applied a novel method of eliminating permission information which used to be regarded as a critical factor for distinguishing malware. With our carefully generated feature sets and the weighted voting by the ensemble algorithm, we were able to reach the highest malware detection accuracy of 97.8%. We also verified that discarding the permission information lead to the improvement in terms of false positive and false negative rates.