• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.38 no.1
• /
• pp.46-57
• /
• 2001

To access the multimedia data, such as video data with temporal properties, the content-based image retrieval technique is required. Moreover, one of the basic techniques for content-based image retrieval is an extraction of representative key-frames. Not only did we implement this method, but also by analyzing the video data, we have proven the proposed method to be both effective and accurate. In addition, this method is expected to solve the real world problem of building video databases, as it is very useful in building an index.

• Journal of IKEEE
• /
• v.7 no.2 s.13
• /
• pp.210-217
• /
• 2003

A novel characteristic value extraction method based on mathematical morphology is proposed. Morphological spatial frequency defined by morphological pattern distribution function is introduced and applied to define a new feature called ‘average height.' The average height is used to define a characteristic value which is to be used to generate an index key value for content-based image retrieval. Superiority of the method was proved for various images by experiment. Furthermore the fact that the proposed method does not need threshold to obtain binary image provides its applicability to content-based image retrieval.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.10
• /
• pp.2788-2808
• /
• 2023

Brain tumors are one of the most threatening malignancies for humans. Misdiagnosis of brain tumors can result in false medical intervention, which ultimately reduces a patient's chance of survival. Manual identification and segmentation of brain tumors from Magnetic Resonance Imaging (MRI) scans can be difficult and error-prone because of the great range of tumor tissues that exist in various individuals and the similarity of normal tissues. To overcome this limitation, the Amended Convolutional Neural Network (ACNN) model has been introduced, a unique combination of three techniques that have not been previously explored for brain tumor detection. The three techniques integrated into the ACNN model are image tissue preprocessing using the Kalman Bucy Smoothing Filter to remove noisy pixels from the input, image tissue segmentation using the Isotonic Regressive Image Tissue Segmentation Process, and feature extraction using the Marr Wavelet Transformation. The extracted features are compared with the testing features using a sigmoid activation function in the output layer. The experimental findings show that the suggested model outperforms existing techniques concerning accuracy, precision, sensitivity, dice score, Jaccard index, specificity, Positive Predictive Value, Hausdorff distance, recall, and F1 score. The proposed ACNN model achieved a maximum accuracy of 98.8%, which is higher than other existing models, according to the experimental results.

• The KIPS Transactions:PartB
• /
• v.16B no.1
• /
• pp.71-78
• /
• 2009

For more effective index extraction and index weight determination, studies of extracting indices are carried out by using document content as well as structure. However, most of studies are concentrating in calculating the importance of context rather than that of XML tag. These conventional studies determine its importance from the aspect of common sense rather than verifying that through an objective experiment. This paper, for the automatic indexing by using the tag information of XML document that has taken its place as the standard for web document management, classifies major tags of constructing a paper according to its importance and calculates the term weight extracted from the tag of low weight. By using the weight obtained, this paper proposes a method of calculating the final weight while updating the term weight extracted from the tag of high weight. In order to determine more objective weight, this paper tests the tag that user considers as important and reflects it in calculating the weight by classifying its importance according to the result. Then by comparing with the search performance while using the index weight calculated by applying a method of determining existing tag importance, it verifies effectiveness of the index weight calculated by applying the method proposed in this paper.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.3
• /
• pp.123-138
• /
• 2023

The Fourth Industrial Revolution and sensor technology have led to increased utilization of sensor data. In our modern society, data complexity is rising, and the extraction of valuable information has become crucial with the rapid changes in information technology (IT). Recurrent neural networks (RNN) and long short-term memory (LSTM) models have shown remarkable performance in natural language processing (NLP) and time series prediction. Consequently, there is a strong expectation that models excelling in NLP will also excel in time series prediction. However, current research on Transformer models for time series prediction remains limited. Traditional RNN and LSTM models have demonstrated superior performance compared to Transformers in big data analysis. Nevertheless, with continuous advancements in Transformer models, such as GPT-2 (Generative Pre-trained Transformer 2) and ProphetNet, they have gained attention in the field of time series prediction. This study aims to evaluate the classification performance and interval prediction of remaining useful life (RUL) using an advanced Transformer model. The performance of each model will be utilized to establish a health index (HI) for cutting blades, enabling real-time monitoring of machine health. The results are expected to provide valuable insights for machine monitoring, evaluation, and management, confirming the effectiveness of advanced Transformer models in time series analysis when applied in industrial settings.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.111-136
• /
• 2018

In this paper, we propose a methodology to extract answer information about queries from various types of unstructured documents collected from multi-sources existing on web in order to expand knowledge base. The proposed methodology is divided into the following steps. 1) Collect relevant documents from Wikipedia, Naver encyclopedia, and Naver news sources for "subject-predicate" separated queries and classify the proper documents. 2) Determine whether the sentence is suitable for extracting information and derive the confidence. 3) Based on the predicate feature, extract the information in the proper sentence and derive the overall confidence of the information extraction result. In order to evaluate the performance of the information extraction system, we selected 400 queries from the artificial intelligence speaker of SK-Telecom. Compared with the baseline model, it is confirmed that it shows higher performance index than the existing model. The contribution of this study is that we develop a sequence tagging model based on bi-directional LSTM-CRF using the predicate feature of the query, with this we developed a robust model that can maintain high recall performance even in various types of unstructured documents collected from multiple sources. The problem of information extraction for knowledge base extension should take into account heterogeneous characteristics of source-specific document types. The proposed methodology proved to extract information effectively from various types of unstructured documents compared to the baseline model. There is a limitation in previous research that the performance is poor when extracting information about the document type that is different from the training data. In addition, this study can prevent unnecessary information extraction attempts from the documents that do not include the answer information through the process for predicting the suitability of information extraction of documents and sentences before the information extraction step. It is meaningful that we provided a method that precision performance can be maintained even in actual web environment. The information extraction problem for the knowledge base expansion has the characteristic that it can not guarantee whether the document includes the correct answer because it is aimed at the unstructured document existing in the real web. When the question answering is performed on a real web, previous machine reading comprehension studies has a limitation that it shows a low level of precision because it frequently attempts to extract an answer even in a document in which there is no correct answer. The policy that predicts the suitability of document and sentence information extraction is meaningful in that it contributes to maintaining the performance of information extraction even in real web environment. The limitations of this study and future research directions are as follows. First, it is a problem related to data preprocessing. In this study, the unit of knowledge extraction is classified through the morphological analysis based on the open source Konlpy python package, and the information extraction result can be improperly performed because morphological analysis is not performed properly. To enhance the performance of information extraction results, it is necessary to develop an advanced morpheme analyzer. Second, it is a problem of entity ambiguity. The information extraction system of this study can not distinguish the same name that has different intention. If several people with the same name appear in the news, the system may not extract information about the intended query. In future research, it is necessary to take measures to identify the person with the same name. Third, it is a problem of evaluation query data. In this study, we selected 400 of user queries collected from SK Telecom 's interactive artificial intelligent speaker to evaluate the performance of the information extraction system. n this study, we developed evaluation data set using 800 documents (400 questions * 7 articles per question (1 Wikipedia, 3 Naver encyclopedia, 3 Naver news) by judging whether a correct answer is included or not. To ensure the external validity of the study, it is desirable to use more queries to determine the performance of the system. This is a costly activity that must be done manually. Future research needs to evaluate the system for more queries. It is also necessary to develop a Korean benchmark data set of information extraction system for queries from multi-source web documents to build an environment that can evaluate the results more objectively.

• Economic and Environmental Geology
• /
• v.34 no.3
• /
• pp.283-299
• /
• 2001

The middle cover soils and clay liners collected from the Kumheung landfill in Gongiu City were analysed for As, Ba, Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, Sr, Ti and Zn concentrations using 0.] N HCl digestion and total/sequential extraction experiments followed by ICP-AES determination. The uncontaminated soil and sediment samples were also analyzed for the comparison. The results of sequential extraction showed that Cu was dominant in the oxidizable fraction, and As, Ni, Sr, Ba, and Mn were in the exchangeable fraction. Zinc and Mn occurred mostly in association with reducible, residual and carbonate fractions. Most of Cd and Pb were bound to the reducible and oxidizable fractions. The main carrier of Co, Cr, Fe and 11 was the residual fraction and another important carrier was the reducible fraction. The percentage of the metals of organically-bound form in the middle cover soils and clay liner was in the order of Cu(48%) > Ti(42%) > Pb(27%) > As(25%) > Cd(20%). As deduced from sequential extraction analysis, potential order of metal mobility in the middle cover soils and clay liner from the landfill was proposed: Cd > Sr > As > Ni > Mn > Ba > Cu > Pb > Zn » Co > 11 > Fe > Cr. Based on the 'geoaccumulation index' and the 'enrichment factor' normalized to A], the level of contamination of Cu, Ni and C1' was significant in the samples from Kumheung landfill and surrounding farmland. Their enrichments were attributed partly to anthropogenic pollutions.

• Journal of the East Asian Society of Dietary Life
• /
• v.14 no.3
• /
• pp.243-250
• /
• 2004

The effect of hemicellulose extracted from Shiitake mushroom(Lentinus edodes) on the level of blood sugar and cholesterol in the diabetes-induced rat by streptozotocin(STZ) was investigated. The yield of hemicellulose by extraction process of 5% salt extraction, preparation of alcohol insoluble substance, IN KOH extraction, acid precipitation(pH 3.0), and dialysis was 9.24%. The experimental plots divided to 1% cellulose group(control), 0.5% hemicellulose group(H-l) and 1% hemicellulose group(H-2). The groups were fed for 6 weeks, then continuously fed for 1 week after induction of diabetes by STZ. Feed intakes, weight gain and feed efficiency of the each groups were not significantly different, while water intakes and liver weight of H-2 group were lower than those of control and H-l group. Weight of liver in the H-2 group was significantly lower than those of control and H-l groups. The amounts of feces were 0.32 g/day in the control group, 0.43∼0.44 g/day in the H-l and H-2 groups, while the amounts of urine were 15.28 mL/day in the control group, 10.83∼11.20 mL/day in the H-l and H-2 groups. The content of blood glucose before diabetes induction(fed for 3∼5 weeks) was 111.2-132.6 mg/dL in the control group, not significantly different from others; After diabetes induction, however, the contents were 212.8 mg/dL in the control group, 140.0-144.0 mg/dL in the H-l and H-2 groups, which showed significant difference. Urine glucose contents of H-2 group before and after diabetes induction were lower than those of control and H-l groups. There was no significant difference in the content of neutral lipid between each groups. Total cholesterol contents were 101.6 mg/dL in the control group, 56.∼64.0 mg/dL in the hemicellulose groups. HDL-cholesterol content and atherogenic index of hemicellulose groups were lower than those of control group, respectively. In conclusion, the hemicellulose extracted from Shiitake mushroom represented improving and preventing effects for diabetes.

• Korean Journal of Soil Science and Fertilizer
• /
• v.47 no.6
• /
• pp.443-450
• /
• 2014

To determine Pb species in soils following the immobilization process, sequential extraction has been used despite the possibility of overestimating Pb species from unintended reactions during chemical extraction. Meanwhile, the application of extended X-ray absorption fine structure (EXAFS) has been shown to provide a more precise result than chemical extraction. In this study, the immobilization of Pb in contaminated soils treated with liming materials such as oyster shell (OS) or eggshell (ES) was evaluated with thermodynamic modelling and EXAFS analysis. Thermodynamic modelling by visual MINTEQ predicted the precipitation of $Pb(OH)_2$ in OS and ES treated soils. In particular, the values of saturation index (SI) for $Pb(OH)_2$ in OS (SI=0.286) and ES (SI=0.453) treated soils were greater than in the control soil (SI=0.281). Linear combination fitting (LCF) analysis confirmed the presence of $C_{12}H_{10}O_{14}Pb_3$ (lead citrate, 44.7%) by citric acid from plant root, Pb-gibbsite (Pb adsorbed gibbsite, 26.4%), and Pb-kaolinite (Pb adsorbed kaolinite, 20.3%) in the control soil. On the other hand, $Pb(OH)_2$ (16.8%), Pb-gibbsite (39.3%), and Pb-kaolinite (25.6%) were observed in the OS treated soil and $Pb(OH)_2$ (55.2%) and Pb-gibbsite (33.8%) were also confirmed in the ES treated soil. Our results indicate that the treatment with OS and ES immobilizes Pb by adsorption of Pb onto the soil minerals as a result of the increase in soil negative charge and the formation of stable $Pb(OH)_2$ under high pH condition of soils.

• Korean Journal of Optics and Photonics
• /
• v.20 no.3
• /
• pp.182-188
• /
• 2009

The surface plasmon resonance ellipsometer (SPRE), using a multiple air injection system with an extraction of air system, has been proposed and developed to minimize measurement error of signals due to diffusion of reagent into running buffer. Since the diffusion of reagent into running buffer affects the refractive index of the running buffer by changing the concentration, characteristics of binding between various bio-molecules don't appear clearly in measurement results. The diffusion between running buffer and reagent can be blocked by using an air bubble injection system. An extraction of air system is used to remove the noise signal due to unnecessary air bubbles flowing in a channel. Reliability of measurement results has been improved by using the valve system.