• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.29-36
• /
• 2021

In this paper, we classify ECG signal data for mobile devices using deep learning models. To classify abnormal heartbeats with high accuracy, three factors of the deep learning model are selected, and the classification accuracy is compared according to the changes in the conditions of the factors. We apply a CNN model that can self-extract features of ECG data and compare the performance of a total of 48 combinations by combining conditions of the depth of model, optimization method, and activation functions that compose the model. Deriving the combination of conditions with the highest accuracy, we obtained the highest classification accuracy of 97.88% when we applied 19 convolutional layers, an optimization method SGD, and an activation function Mish. In this experiment, we confirmed the suitability of feature extraction and abnormal beat detection of 1-channel ECG signals using CNN.

• Journal of Korean Clinical Nursing Research
• /
• v.26 no.3
• /
• pp.285-295
• /
• 2020

Purpose: The purpose of this study was to develop and validate a education nurse specialist-led educational satisfaction scale for nurses working at comprehensive nursing care service wards. Methods: A methodological study was conducted. A total of 237 nurses working at comprehensive nursing care service wards in a general hospital participated in this study. The scale was developed through literature reviews, in-depth interview, development of preliminary items, verification of content validity, development of secondary items, evaluation of construct validity, and extraction of final items. Analysis included exploratory factor analysis, pearson's analysis, and reliability analysis using cronbach's α. Results: The education nurse specialist-led educational satisfaction scale for nurses working at comprehensive nursing care service ward consisted of 21 items. Two factors (critical thinking disposition and clinical competency) were identified which explained 63.5% of the total variance. Cronbach's α of each factors were >.95. Conclusion: The results suggested that the education nurse specialist-led educational satisfaction scale for nurses working at comprehensive nursing care service wards demonstrated acceptable validity and reliability. Items of the instrument can assess the level of satisfaction with regards to education led by education nurse specialist among nurses working at comprehensive nursing care service wards.

• Journal of Korean Society for Quality Management
• /
• v.50 no.1
• /
• pp.77-89
• /
• 2022

Purpose: In order to improve the audit quality of a company, an in-depth analysis is required to categorize the audit report in the form of a text document containing the details of the external audit. This study introduces a systematic methodology to extract keywords for each group that determines the differences between groups such as 'audit plan' and 'interim audit' using audit reports collected in the form of text documents. Methods: The first step of the proposed methodology is to preprocess the document through text mining. In the second step, the documents are classified into groups using machine learning techniques and based on this, important vocabularies that have a dominant influence on the performance of classification are extracted. In the third step, the association rules for each group's documents are found. In the last step, the final keywords for each group representing the characteristics of each group are extracted by comparing the important vocabulary for classification with the important vocabulary representing the association rules of each group. Results: This study quantitatively calculates the importance value of the vocabulary used in the audit report based on machine learning rather than the qualitative research method such as the existing literature search, expert evaluation, and Delphi technique. From the case study of this study, it was found that the extracted keywords describe the characteristics of each group well. Conclusion: This study is meaningful in that it has laid the foundation for quantitatively conducting follow-up studies related to key vocabulary in each stage of auditing.

• Journal of Industrial Convergence
• /
• v.21 no.10
• /
• pp.47-56
• /
• 2023

The purpose of this study was to explore educational awareness and meaning by applying an role-playing game (RPG) type class based on constructivism in physical education class in middle school. Regarding the research method, data collection was conducted through in-depth interviews, observations, and related documents, and data analysis was performed using case record analysis, inductive category analysis method, and case extraction process. Results of the study First, in terms of teachers' perceptions, there are both positive and negative aspects of teachers' perceptions of middle school physical education classes using the role-playing game (RPG) format. Second, in terms of educational meaning, the educational meaning of middle school physical education classes using the role-playing game (RPG) format emphasizes the need for a systematic lesson plan and sufficient support system. In conclusion, positive attitudes toward physical education as well as educational implications can be formed, and the basic foundation of the constructivist physical education class method and its preparation are thought to be of greater significance than anything else.

• Transactions on Semiconductor Engineering
• /
• v.2 no.1
• /
• pp.17-20
• /
• 2024

This paper presents a novel lightweight object detection model tailored for low-powered edge devices, addressing the limitations of traditional resource-intensive computer vision models. Our proposed detector, inspired by the Single Shot Detector (SSD), employs a compact yet robust network design. Crucially, it integrates an 'enhancer block' that significantly boosts its efficiency in detecting smaller objects. The model comprises two primary components: the Light_Block for efficient feature extraction using Depth-wise and Pointwise Convolution layers, and the Enhancer_Block for enhanced detection of tiny objects. Trained from scratch on the Udacity Annotated Dataset with image dimensions of 300x480, our model eschews the need for pre-trained classification weights. Weighing only 5.5MB with approximately 0.43M parameters, our detector achieved a mean average precision (mAP) of 27.7% and processed at 140 FPS, outperforming conventional models in both precision and efficiency. This research underscores the potential of lightweight designs in advancing object detection for edge devices without compromising accuracy.

• Journal of Advanced Navigation Technology
• /
• v.28 no.2
• /
• pp.238-245
• /
• 2024

In this paper, we propose a framework for improving the performance of semantic segmentation of agricultural multispectral image using feature fusion techniques. Most of the semantic segmentation models being studied in the field of smart farms are trained on RGB images and focus on increasing the depth and complexity of the model to improve performance. In this study, we go beyond the conventional approach and optimize and design a model with multispectral and attention mechanisms. The proposed method fuses features from multiple channels collected from a UAV along with a single RGB image to increase feature extraction performance and recognize complementary features to increase the learning effect. We study the model structure to focus on feature fusion and compare its performance with other models by experimenting with favorable channels and combinations for crop images. The experimental results show that the model combining RGB and NDVI performs better than combinations with other channels.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.7
• /
• pp.1706-1725
• /
• 2024

The unique U-shaped structure of U-Net network makes it achieve good performance in image segmentation. This network is a lightweight network with a small number of parameters for small image segmentation datasets. However, when the medical image to be segmented contains a lot of detailed information, the segmentation results cannot fully meet the actual requirements. In order to achieve higher accuracy of medical image segmentation, a novel improved U-Net network architecture called multi-scale encoder-decoder U-Net+ (MEDU-Net+) is proposed in this paper. We design the GoogLeNet for achieving more information at the encoder of the proposed MEDU-Net+, and present the multi-scale feature extraction for fusing semantic information of different scales in the encoder and decoder. Meanwhile, we also introduce the layer-by-layer skip connection to connect the information of each layer, so that there is no need to encode the last layer and return the information. The proposed MEDU-Net+ divides the unknown depth network into each part of deconvolution layer to replace the direct connection of the encoder and decoder in U-Net. In addition, a new combined loss function is proposed to extract more edge information by combining the advantages of the generalized dice and the focal loss functions. Finally, we validate our proposed MEDU-Net+ MEDU-Net+ and other classic medical image segmentation networks on three medical image datasets. The experimental results show that our proposed MEDU-Net+ has prominent superior performance compared with other medical image segmentation networks.

• Journal of the Mineralogical Society of Korea
• /
• v.28 no.1
• /
• pp.71-81
• /
• 2015

This study was carried out in order to investigate the mineralogical characteristics of the core sediments (03GHP-02 and HB13-2), obtained from the Ulleung Basin and Hupo Basin, Korea. The results on mineral compositions, clay mineral compositions, and the total contents and sequential extraction of different fractions of the phosphorus in core samples showed that those values are different in two cores and also at different depths. In both samples, mineral compositions were the same, composed mainly of quartz, microcline, albite, calcite, opal A, pyrite, and clay minerals (illite, chlorite, kaolinite, and smectite). However, the sample from Hupo Basin contains more opal A. Both samples, especially the ones from Hupo Basin contains more smectite than those reported from East Sea, indicating the influence of paleo-Hwangwei River and the Tertiary Formation of Korea Peninsula. For the samples from Uleung Basin, at 0.7-3.5 m range in depth, the low content of opal A and the low illite crystallinity index can be inferred to indicate the relatively cool climate, corresponding to the ice age. Also, the content of total phosphorus was low in those samples. It was reported that East Sea at that time was isolated from the neighboring seas due to the decrease of the sea level, and as a result, the influx of sediments was supposed to be little through the strait and rivers. For the samples from Hupo Basin, there is no significant changes in clay mineral composition and the distribution of phosphorus with increasing depth. This little change can be interpreted to indicate that the sediments comprising the core might be deposited in a relatively short period of time or deposited in sedimentary environment in which there's no significant changes in sediment supplies. The values of crystallinity index of clay minerals are high in those samples, indicating that it was relatively warm during that time. Although the increase of fluctuation pattern can be observed, showing that the climate of this period often changed, it is supposed that it was generally warm.

• Korean Journal of Ecology and Environment
• /
• v.44 no.1
• /
• pp.31-41
• /
• 2011

To optimize the natural chemical agents against nuisance phytoplankton, we examined algal removal activity (ABA) of Plant-Mineral Composite (PMC), which already developed by our teams (Kim et al., 2010), on various conditions. The PMC are consisted of extracted-mixtures with indigenous plants (Camellia sinensis, Quercusacutissima and Castanea crenata) and minerals (Loess, Quartz porphyry, and natural zeolite), and characterized by coagulation and floating of low-density suspended solids. A simple extraction process was adopted, such as drying and grinding of raw material, water-extraction by high temperature-sonication and filtering. All tests were performed in 3 L plastic chambers varying conditions; six different concentrations ($0{\sim}1.0\;mL\;L^{-1}$), six light intensities ($8{\sim}1,400\;{\mu}mol\;m^{-2}s^{-1}$), three temperatures ($10{\sim}30^{\circ}C$), four pHs (7~10), five water depths (10~50 cm), and three different waters dominated by cyanobacteria, diatom, and green algae, respectively. Results indicate that the highest ABA of PMC was seen at $0.05\;mL\;L^{-1}$ in treatment concentrations, where showed a reduction of more than 80% of control phytoplankton biomass, while $1,400\;{\mu}mol\;m^{-2}s^{-1}$ in light intensity (>90%), $20{\sim}30^{\circ}C$ temperature (>60%), 7~9 in pH (>90%), below 50 cm in water depth (>90%), and cyanobacterial dominating waters (>80%), respectively. Over the test, ABA of PMC were more obvious on the algal biomass (chlorophyll-${\alpha}$) than suspended solids, suggesting a selectivity of PMC to particle size or natures. These results suggest that PMC agents can play an important role as natural agents to remove the nuisant algal aggregates or seston of eutrophic lake, where occur cyanobacterial bloom in a shallow shore of lake during warm season.

• Journal of Soil and Groundwater Environment
• /
• v.10 no.2
• /
• pp.35-43
• /
• 2005

Bioslurping combines the three remedial approaches of bioventing, vacuum-enhanced free-product recovery, and soil vapor extraction. Bioslurping is less effective in tight (low-permeability) soils. The greatest limitation to air permeability is excessive soil moisture. Optimum soil moisture is very soil-specific. Too much moisture can reduce air permeability of the soil and decrease its oxygen transfer capability. Too little moisture will inhibit microbial activity. So Modified Fenton reaction as chemical treatment which can overcome the weakness of Bioslurping was experimented for simultaneous treatment. Although the diesel removal efficiency of SVE process increased in proportion to applied vacuum pressure, SVE process was difficulty to remediation quickly semi- or non-volatile compounds absorbed soil strongly. And SVE process had variation of efficiency with distance from the extraction well and depth a air flow form of hemisphere centering around the well. Below 0.1 % hydrogen peroxide shows the potential of using hydrogen peroxide as oxygen source but the co-oxidation of chemical and biological treatment was impossible because of the low efficiency of Modified Fenton reaction at 0.1 % (wt) hydrogen peroxide. NTA was more efficiency than EDTA as chelating agent and diesel removal efficiency of Modified Fenton reaction increased in proportion to hydrogen peroxide concentration. Hexadecane as typical aliphatic compound was removed less than Toluene as aromatic compound because of its structural stability in Modified Fenton reaction. What minimum 10% hydrogen peroxide concentration has good remediation efficiency of diesel contaminated groundwater may show the potential use of Modified Fenton reaction after bioslurping treatment.