• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.2
• /
• pp.78.1-78.1
• /
• 2019

There are a huge number of faint objects that have not been observed due to the lack of large and deep surveys. In this study, we demonstrate that a deep learning approach can produce a better quality deep image from a single pass imaging so that could be an alternative of conventional image stacking technique or the expensive large and deep surveys. Using data from the Sloan Digital Sky Survey (SDSS) stripe 82 which provide repeatedly scanned imaging data, a training data set is constructed: g-, r-, and i-band images of single pass data as an input and r-band co-added image as a target. Out of 151 SDSS fields that have been repeatedly scanned 34 times, 120 fields were used for training and 31 fields for validation. The size of a frame selected for the training is 1k by 1k pixel scale. To avoid possible problems caused by the small number of training sets, frames are randomly selected within that field each iteration of training. Every 5000 iterations of training, the performance were evaluated with RMSE, peak signal-to-noise ratio which is given on logarithmic scale, structural symmetry index (SSIM) and difference in SSIM. We continued the training until a GAN model with the best performance is found. We apply the best GAN-model to NGC0941 located in SDSS stripe 82. By comparing the radial surface brightness and photometry error of images, we found the possibility that this technique could generate a deep image with statistics close to the stacked image from a single-pass image.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.1055-1065
• /
• 2023

Recently, various studies using deep reinforcement learning (deep RL) technology have been conducted to solve complex problems using big data collected at industrial internet of things. Deep RL uses reinforcement learning"s trial-and-error algorithms and cumulative compensation functions to generate and learn its own data and quickly explore neural network structures and parameter decisions. However, studies so far have shown that the larger the size of the learning data is, the higher are the memory usage and search time, and the lower is the accuracy. In this study, model-agnostic learning for efficient federated deep RL was utilized to solve privacy invasion by increasing robustness as 55.9% and achieve 97.8% accuracy, an improvement of 5.5% compared with the comparative optimization-based meta learning models, and to reduce the delay time by 28.9% on average.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_2
• /
• pp.837-847
• /
• 2023

In this study, we explored a method for assessing the extent of damage caused by chemical substances at an accident site through the use of a vegetation index. Data collection involved the deployment of two different drone types, and the damaged area was determined using photogrammetry technology from the 3D point cloud data. To create a vegetation index image, we utilized spectral band data from a multi-spectral sensor to generate an orthoimage. Subsequently, we conducted statistical analyses of the accident site with respect to the damaged area using a predefined threshold value. The Kappa values for the vegetation index, based on the near-infrared band and the green band, were found to be 0.79 and 0.76, respectively. These results suggest that the vegetation index-based approach for analyzing damage areas can be effectively applied in investigations of chemical accidents.

• Proceedings of the National Institute of Ecology of the Republic of Korea
• /
• v.4 no.4
• /
• pp.159-176
• /
• 2023

The conservation of the raccoon dog (Nyctereutes procyonoides) in South Korea requires the protection and preservation of natural habitats while additionally ensuring coexistence with human activities. Applying habitat map modeling techniques provides information regarding the distributional patterns of raccoon dogs and assists in the development of future conservation strategies. The purpose of this study is to generate potential habitat distribution maps for the raccoon dog in South Korea using geospatial technology-based models. These models include the frequency ratio (FR) as a bivariate statistical approach, the group method of data handling (GMDH) as a machine learning algorithm, and convolutional neural network (CNN) and long short-term memory (LSTM) as deep learning algorithms. Moreover, the imperialist competitive algorithm (ICA) is used to fine-tune the hyperparameters of the machine learning and deep learning models. Moreover, there are 14 habitat characteristics used for developing the models: elevation, slope, valley depth, topographic wetness index, terrain roughness index, slope height, surface area, slope length and steepness factor (LS factor), normalized difference vegetation index, normalized difference water index, distance to drainage, distance to roads, drainage density, and morphometric features. The accuracy of prediction is evaluated using the area under the receiver operating characteristic curve. The results indicate comparable performances of all models. However, the CNN demonstrates superior capacity for prediction, achieving accuracies of 76.3% and 75.7% for the training and validation processes, respectively. The maps of potential habitat distribution are generated for five different levels of potentiality: very low, low, moderate, high, and very high.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.6
• /
• pp.101-112
• /
• 2024

In this paper, we present EDF (Event Data Factory), an interactive tool designed to assist event log generation for process mining. EDF integrates various data connectors to improve its capability to assist users in connecting to diverse data sources. Our tool employs low-code/no-code technology, along with graph-based visualization, to help non-expert users understand process flow and enhance the user experience. By utilizing metadata information, EDF allows users to efficiently generate an event log containing case, activity, and timestamp attributes. Through log quality metrics, our tool enables users to assess the generated event log quality. We implement EDF under a cloud-based architecture and run a performance evaluation. Our case study and results demonstrate the usability and applicability of EDF. Finally, an observational study confirms that EDF is easy to use and beneficial, expanding small and medium-sized enterprises' (SMEs) access to process mining applications.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.77-84
• /
• 2024

In this paper, we propose a method for accurately and rapidly detecting defects in wire harnesses by utilizing computer vision to calculate six crucial measurement values: the length of crimped terminals, the dimensions (width) of terminal ends, and the width of crimped sections (wire and core portions). We employ Harris corner detection to locate object positions from two types of data. Additionally, we generate reference points for extracting measurement values by utilizing features specific to each measurement area and exploiting the contrast in shading between the background and objects, thus reflecting the slope of each sample. Subsequently, we introduce a method using the Euclidean distance and correction coefficients to predict values, allowing for the prediction of measurements regardless of changes in the wire's position. We achieve high accuracy for each measurement type, 99.1%, 98.7%, 92.6%, 92.5%, 99.9%, and 99.7%, achieving outstanding overall average accuracy of 97% across all measurements. This inspection method not only addresses the limitations of conventional visual inspections but also yields excellent results with a small amount of data. Moreover, relying solely on image processing, it is expected to be more cost-effective and applicable with less data compared to deep learning methods.

• Journal of Advanced Navigation Technology
• /
• v.28 no.5
• /
• pp.738-744
• /
• 2024

The goal of this system is to effectively summarize and visualize important DICOM image data in the medical field. Using React and Node.js, the system collects and parses DICOM images, extracting critical medical information in the process. It then employs a large language model (LLM) to generate automatic summary reports, providing users with personalized medical information. This approach enhances accessibility to medical data and leverages web technologies to process large-scale data quickly and reliably. The system also aims to improve communication between patients and doctors, enhancing the quality of care and enabling medical staff to make faster, more accurate decisions. Additionally, it seeks to improve patients' medical experiences and overall satisfaction. Ultimately, the system aims to improve the quality of healthcare services.

• The KIPS Transactions:PartD
• /
• v.14D no.7
• /
• pp.733-742
• /
• 2007

A data stream is a massive unbounded sequence of data elements continuously generated at a rapid rate. Due to this reason, memory usage for data stream analysis should be confined finitely although new data elements are continuously generated in a data stream. To satisfy this requirement, data stream processing sacrifices the correctness of its analysis result by allowing some errors. The old distribution statistics are diminished by a predefined decay rate as time goes by, so that the effect of the obsolete information on the current result of clustering can be eliminated without maintaining any data element physically. This paper proposes a grid based clustering algorithm for a data stream. Given a set of initial grid cells, the dense range of a grid cell is recursively partitioned into a smaller cell based on the distribution statistics of data elements by a top down manner until the smallest cell, called a unit cell, is identified. Since only the distribution statistics of data elements are maintained by dynamically partitioned grid cells, the clusters of a data stream can be effectively found without maintaining the data elements physically. Furthermore, the memory usage of the proposed algorithm is adjusted adaptively to the size of confined memory space by flexibly resizing the size of a unit cell. As a result, the confined memory space can be fully utilized to generate the result of clustering as accurately as possible. The proposed algorithm is analyzed by a series of experiments to identify its various characteristics

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.42 no.4 s.304
• /
• pp.33-42
• /
• 2005

This paper proposes a classification method for gene expression data, using membership function and neural network. The gene expression is a process to produce mRNA and protains which generate a living body, and the gene expression data is important to find out the functions and correlations of genes. Such gene expression data can be obtained from DNA 칩 massively and quickly. However, thousands of gene expression data may not be useful until it is well organized. Therefore a classification method is necessary to find the characteristics of gene data acquired from the gene expression. In the proposed method, a set of gene data is extracted according to the fisher's criterion, because we assume that selected gene data is the well-classified data sample. However, the selected gene data does not guarantee well-classified data sample and we calculate feature values using membership function to reduce the influence of outliers in gene data. Feature vectors estimated from the selected feature values are used to train back propagation neural network. The experimental results show that the clustering performance of the proposed method has been improved compared to other existing methods in various gene expression data.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.15 no.1
• /
• pp.36-44
• /
• 2022

Reversible data hiding techniques have been developed to hide confidential data in the image by shifting the histogram of the image. These techniques have a weakness in which the security of hidden confidential data is weak. In this paper, to solve this drawback, we propose a technique of triple encrypting confidential data using pixel value information and hiding it in the cover image. When confidential data is triple encrypted using the proposed technique and hidden in the cover image to generate a stego-image, since encryption based on pixel information is performed three times, the security of confidential data hidden by triple encryption is greatly improved. In the experiment to measure the performance of the proposed technique, even if the triple-encrypted confidential data was extracted from the stego-image, the original confidential data could not be extracted without the encryption keys. And since the image quality of the stego-image is 48.39dB or higher, it was not possible to recognize whether confidential data was hidden in the stego-image, and more than 30,487 bits of confidential data were hidden in the stego-image. The proposed technique can extract the original confidential data from the triple-encrypted confidential data hidden in the stego-image without loss, and can restore the original cover image from the stego-image without distortion. Therefore, the proposed technique can be effectively used in applications such as military, medical, digital library, where security is important and it is necessary to completely restore the original cover image.