• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.173-178
• /
• 2022

Recently, a recommendation system using deep learning in social network services has been actively studied. However, in the case of a recommendation system using deep learning, the cold start problem and the increased learning time due to the complex computation exist as the disadvantage. In this paper, the user-tailored exercise routine recommendation algorithm is proposed using the user's metadata. Metadata (the user's height, weight, sex, etc.) set as the input of the model is applied to the designed model in the proposed algorithms. The exercise recommendation system model proposed in this paper is designed based on the neural collaborative filtering (NCF) algorithm using multi-layer perceptron and matrix factorization algorithm. The learning proceeds with proposed model by receiving user metadata and exercise information. The model where learning is completed provides recommendation score to the user when a specific exercise is set as the input of the model. As a result of the experiment, the proposed exercise recommendation system model showed 10% improvement in recommended performance and 50% reduction in learning time compared to the existing NCF model.

• Journal of the Korea Convergence Society
• /
• v.10 no.11
• /
• pp.1-7
• /
• 2019

Recently, various approaches to detect network attacks using machine learning have been studied and are being applied to detect new attacks and to increase precision. However, the machine learning method is dependent on feature extraction and takes a long time and complexity. It also has limitation of performace due to learning data imbalance. In this study, we propose a method to solve the degradation of classification performance due to imbalance of learning data among the limit points of detection system. To do this, we generate data using Generative Adversarial Networks (GANs) and propose a classification method using Convolutional Neural Networks (CNNs). Through this approach, we can confirm that the accuracy is improved when applied to the NSL-KDD and UNSW-NB15 datasets.

• Journal of Convergence for Information Technology
• /
• v.10 no.5
• /
• pp.111-116
• /
• 2020

Recently, many safety accidents involving children shuttle buses take place. Without a teacher for help, a safety accident occurs when the driver can't see a child who is getting off in the blind spot of both frontside and backside. A deep learning-based smart mirror allows user authentication and provides various services. Especially, It can be a role of helper for children, and prevent accidents that can occur when drivers or assistant teachers do not see them. User authentication is carried out with children's face registered in advance. Safety accidents can be prevented by an approximate sensor and a camera in frontside and backside of the bus. This study suggests a way of checking out whether children are missed in the process of getting in and out of the bus, designs a system that reduce blind spots in the front and back of the vehicle, and builds a safety system that provide various services using GPS.

• Journal of Internet Computing and Services
• /
• v.21 no.5
• /
• pp.109-118
• /
• 2020

Recently, people are spending a lot of time inside their homes because of various diseases. It is difficult to ask others for help in the case of a single-person household that is injured in the house or infected with a disease and needs help from others. In this study, an algorithm is proposed to detect emergency event, which are situations in which single-person households need help from others, such as injuries or disease infections, in their homes. It proposes vision pattern detection algorithms using home CCTVs, audio pattern detection algorithms using artificial intelligence speakers, activity pattern detection algorithms using acceleration sensors in smartphones, and dust pattern detection algorithms using air purifiers. However, if it is difficult to use due to security issues of home CCTVs, it proposes a fusion method combining audio, activity and dust pattern sensors. Each algorithm collected data through YouTube and experiments to measure accuracy.

• Journal of Convergence for Information Technology
• /
• v.10 no.8
• /
• pp.23-34
• /
• 2020

Recently, the artificial intelligence deep learning field has been hard to commercialize due to the high computing power and the price problem of computing resources. In this paper, we apply a double pruning techniques to evaluate the performance of the in-depth neural network and various datasets. Double pruning combines basic Network-slimming and Parameter-prunning. Our proposed technique has the advantage of reducing the parameters that are not important to the existing learning and improving the speed without compromising the learning accuracy. After training various datasets, the pruning ratio was increased to reduce the size of the model.We confirmed that MobileNet-V3 showed the highest performance as a result of NetScore performance analysis. We confirmed that the performance after pruning was the highest in MobileNet-V3 consisting of depthwise seperable convolution neural networks in the Cifar 10 dataset, and VGGNet and ResNet in traditional convolutional neural networks also increased significantly.

• Journal of Convergence for Information Technology
• /
• v.9 no.12
• /
• pp.17-23
• /
• 2019

In this paper, I propose a system development methodology that accepts images taken by camera attached to drone in real time while controlling mini drone and recognize and confirm the face of certain person. For the development of this system, OpenCV, Python related libraries and the drone SDK are used. To increase face recognition ratio of certain person from real-time drone images, it uses Deep Learning-based facial recognition algorithm and uses the principle of Triples in particular. To check the performance of the system, the results of 30 experiments for face recognition based on the author's face showed a recognition rate of about 95% or higher. It is believed that research results of this paper can be used to quickly find specific person through drone at tourist sites and festival venues.

• The Journal of Engineering Geology
• /
• v.32 no.4
• /
• pp.697-723
• /
• 2022

Data-driven models to predict groundwater levels 30 days in advance were developed for 12 groundwater monitoring stations in the middle-Jeju watershed, Jeju Island. Stacked long short-term memory (stacked-LSTM), a deep learning technique suitable for time series forecasting, was used for model development. Daily time series data from 2001 to 2022 for precipitation, groundwater usage amount, and groundwater level were considered. Various models were proposed that used different combinations of the input data types and varying lengths of previous time series data for each input variable. A general procedure for deep-learning-based model development is suggested based on consideration of the comparative validation results of the tested models. A model using precipitation, groundwater usage amount, and previous groundwater level data as input variables outperformed any model neglecting one or more of these data categories. Using extended sequences of these past data improved the predictions, possibly owing to the long delay time between precipitation and groundwater recharge, which results from the deep groundwater level in Jeju Island. However, limiting the range of considered groundwater usage data that significantly affected the groundwater level fluctuation (rather than using all the groundwater usage data) improved the performance of the predictive model. The developed models can predict the future groundwater level based on the current amount of precipitation and groundwater use. Therefore, the models provide information on the soundness of the aquifer system, which will help to prepare management plans to maintain appropriate groundwater quantities.

• Journal of the Korean Society of Industry Convergence
• /
• v.25 no.4_2
• /
• pp.687-697
• /
• 2022

An automated system is needed for the effectiveness of non-destructive testing. In order to utilize the radiographic testing data accumulated in the film, the types of welding defects were classified into 9 and the shape of defects were analyzed. Data was preprocessed to use deep learning with high performance in image classification, and a combination of one-stage/two-stage method and convolutional neural networks/Transformer backbone was compared to confirm a model suitable for welding defect detection. The combination of two-stage, which can learn step-by-step, and deep-layered CNN backbone, showed the best performance with mean average precision 0.868.

• Journal of the Korean Society of Industry Convergence
• /
• v.25 no.4_2
• /
• pp.573-586
• /
• 2022

Corrosion can cause dangerous and expensive damage and failures of ship hulls and equipment. Therefore, it is necessary to maintain the vessel by periodic corrosion inspections. During visual inspection, many corrosion locations are inaccessible for many reasons, especially safety's point of view. Including subjective decisions of inspectors is one of the issues of visual inspection. Automation of visual inspection is tried by many pieces of research. In this study, we propose image preprocessing methods by image patch segmentation and thresholding. YOLOv5 was used as an object detection model after the image preprocessing. Finally, it was evaluated that corrosion detection performance using the proposed method was improved in terms of mean average precision.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.738-740
• /
• 2021

산업통상자원부에서 제공하는 KOTRA 무역 데이터는 해당 품목과 해당 국가에 대하여 GDP, 관세율, 비즈니스 점수, 과/차년도 수출금액 등을 제공한다. 그러나 무역 수출품목은 수 없이 많을 뿐더러 그에 따른 대량의 데이터를 매년 인간의 분석을 통해 유의미한 결과를 이끌어내는 것은 상당히 큰 시간과 비용을 요구한다. 따라서 이번 연구에선 대량의 데이터를 학습하여 단기간에 저비용으로 결과를 예측할 수 있는 심층신경망 모델을 구현해 보았다.