• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.6
• /
• pp.1530-1544
• /
• 2023

Data monitoring is an important foundation of modern science. In most cases, the monitoring data is time-series data, which has high application value. The deep learning algorithm has a strong nonlinear fitting capability, which enables the recognition of time series by capturing anomalous information in time series. At present, the research of time series recognition based on deep learning is especially important for data monitoring. Deep learning algorithms require a large amount of data for training. However, abnormal sample is a small sample in time series, which means the number of abnormal time series can seriously affect the accuracy of recognition algorithm because of class imbalance. In order to increase the number of abnormal sample, a data augmentation method called GANBATS (GAN-based Bi-LSTM and Attention for Time Series) is proposed. In GANBATS, Bi-LSTM is introduced to extract the timing features and then transfer features to the generator network of GANBATS.GANBATS also modifies the discriminator network by adding an attention mechanism to achieve global attention for time series. At the end of discriminator, GANBATS is adding averagepooling layer, which merges temporal features to boost the operational efficiency. In this paper, four time series datasets and five data augmentation algorithms are used for comparison experiments. The generated data are measured by PRD(Percent Root Mean Square Difference) and DTW(Dynamic Time Warping). The experimental results show that GANBATS reduces up to 26.22 in PRD metric and 9.45 in DTW metric. In addition, this paper uses different algorithms to reconstruct the datasets and compare them by classification accuracy. The classification accuracy is improved by 6.44%-12.96% on four time series datasets.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.189-192
• /
• 2021

This study applies a deep learning-based long short-term memory(LSTM) model to track tracking technology. In the case of existing track tracking technology, the weight of constant velocity, constant acceleration, stiff turn, and circular(3D) flight is automatically changed when tracking track in real time using LMIPDA based on Kalman filter according to flight characteristics of an aircraft such as constant velocity, constant acceleration, stiff turn, and circular(3D) flight. In this process, it is necessary to improve performance of changing flight characteristic weight, because changing flight characteristics such as stiff turn flight during constant velocity flight could incur the loss of track and decreasing of the tracking performance. This study is for improving track tracking performance by predicting the change of flight characteristics in advance and changing flight characteristic weigh rapidly. To get this result, this study makes deep learning-based Long Short-Term Memory(LSTM) model study the plot and target of simulator applied with radar error model, and compares the flight tracking results of using Kalman filter with those of deep learning-based Long Short-Term memory(LSTM) model.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.169-171
• /
• 2021

Image processing through cameras such as self-driving, CCTV, mobile phone security, and parking facilities is being used to solve many real-life problems. Simple classification is solved through image processing, but it is difficult to find images or in-image features of complexly mixed objects. To solve this feature point, we utilize deep learning techniques in classification, detection, and segmentation of image data so that we can think and judge closely. Of course, the results are better than just image processing, but we confirm that the results judged by the method of image segmentation using deep learning have deviations from the real object. In this paper, we study how to perform accuracy improvement through simple image processing just before outputting the output of deep learning image segmentation to increase the precision of image segmentation.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.113-119
• /
• 2022

In software fault prediction, a multi classification model that predicts the fault severity category of a module can be much more useful than a binary classification model that simply predicts the presence or absence of faults. A small number of severity-based fault prediction models have been proposed, but no classifier using deep learning techniques has been proposed. In this paper, we construct MLP models with 3 or 5 hidden layers, and they have a structure with a fixed or variable number of hidden layer nodes. As a result of the model evaluation experiment, MLP-based deep learning models shows significantly better performance in both Accuracy and AUC than MLPs, which showed the best performance among models that did not use deep learning. In particular, the model structure with 3 hidden layers, 32 batch size, and 64 nodes shows the best performance.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.611-616
• /
• 2023

ECG(electrocardiogram) is a test used to measure the rate and regularity of heartbeats, as well as the size and position of the chambers, the presence of any damage to the heart, and the cause of all heart diseases can be found. Because the ECG signal obtained using the ECG-KIT includes noise in the ECG signal, noise must be removed from the ECG signal to apply to the deep learning. In this paper, the noise of the ECG signal was removed using the digital IIR Butterworth low-pass filter. When the performance evaluation of the three activation functions, sigmoid(), ReLU(), and tanh() functions, was compared using the deep learning model of LSTM, it was confirmed that the activation function with the smallest error was the tanh() function. Also, When the performance evaluation and elapsed time were compared for LSTM and GRU models, it was confirmed that the GRU model was superior to the LSTM model.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.370-376
• /
• 2023

In the underwater image, it is not clear to distinguish the shape of the target due to underwater noise and low resolution. In addition, as an input of deep learning, underwater images require pre-processing and segmentation must be preceded. Even after pre-processing, the target is not clear, and the performance of detection and identification by deep learning may not be high. Therefore, it is necessary to distinguish and clarify the target. In this study, the importance of target shadows is confirmed in underwater images, object detection and target area acquisition by shadows, and data containing only the shape of targets and shadows without underwater background are generated. We present the process of converting the shadow image into a 3-mode image in which the target is white, the shadow is black, and the background is gray. Through this, it is possible to provide an image that is clearly pre-processed and easily discriminated as an input of deep learning. In addition, if the image processing code using Open Source Computer Vision (OpenCV)Library was used for processing, the processing speed was also suitable for real-time processing.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_3
• /
• pp.1707-1720
• /
• 2023

With the increase in satellite time series data, the utility of remote sensing data is growing. In the analysis of time series data, the relative positional accuracy between images has a significant impact on the results, making image registration essential for correction. In recent years, research on image registration has been increasing by applying deep learning, which outperforms existing image registration algorithms. To train deep learning-based registration models, a large number of image pairs are required. Additionally, creating a correlation map between the data of existing deep learning models and applying additional computations to extract registration points is inefficient. To overcome these drawbacks, this study developed a data augmentation technique for training image registration models and applied it to OffsetNet, a registration model that predicts the offset amount itself, to perform image registration for KOMSAT-2, -3, and -3A. The results of the model training showed that OffsetNet accurately predicted the offset amount for the test data, enabling effective registration of the master and slave images.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.21-30
• /
• 2024

This research aims to design a system capable of generating real web pages based on deep learning and big data, in three stages. First, a classification system was established based on the industry type and functionality of e-commerce websites. Second, the types of components of web pages were systematically categorized. Third, the entire web page auto-generation system, applicable for deep learning, was designed. By re-engineering the deep learning model, which was trained with actual industrial data, to analyze and automatically generate existing websites, a directly usable solution for the field was proposed. This research is expected to contribute technically and policy-wise to the field of generative AI-based complete website creation and industrial sectors.

• Journal of the Korean Society of Radiology
• /
• v.81 no.6
• /
• pp.1290-1304
• /
• 2020

Medical image analyses have been widely used to differentiate normal and abnormal cases, detect lesions, segment organs, etc. Recently, owing to many breakthroughs in artificial intelligence techniques, medical image analyses based on deep learning have been actively studied. However, sufficient medical data are difficult to obtain, and data imbalance between classes hinder the improvement of deep learning performance. To resolve these issues, various studies have been performed, and data augmentation has been found to be a solution. In this review, we introduce data augmentation techniques, including image processing, such as rotation, shift, and intensity variation methods, generative adversarial network-based method, and image property mixing methods. Subsequently, we examine various deep learning studies based on data augmentation techniques. Finally, we discuss the necessity and future directions of data augmentation.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.2
• /
• pp.453-458
• /
• 2024

Optical Character Recognition is the technology that recognizes text in images and converts them into digital format. Deep learning-based OCR is being used in many industries with large quantities of recorded data due to its high recognition performance. To improve medical services, deep learning-based OCR was actively introduced by the medical industry. In this paper, we discussed trends in OCR engines and medical OCR and provided a roadmap for development of medical OCR. By using natural language processing on detected text data, current medical OCR has improved its recognition performance. However, there are limits to the recognition performance, especially for non-standard handwriting and modified text. To develop advanced medical OCR, databaseization of medical data, image pre-processing, and natural language processing are necessary.