• The KIPS Transactions:PartB
• /
• v.15B no.6
• /
• pp.613-620
• /
• 2008

Noun extraction plays an important part in the fields of information retrieval, text summarization, and so on. In this paper, we present a Korean base-noun extraction system and apply it to text summarization to deal with a huge amount of text effectively. The base-noun is an atomic noun but not a compound noun and we use tow techniques, filtering and segmenting. The filtering technique is used for removing non-nominal words from text before extracting base-nouns and the segmenting technique is employed for separating a particle from a nominal and for dividing a compound noun into base-nouns. We have shown that both of the recall and the precision of the proposed system are about 89% on the average under experimental conditions of ETRI corpus. The proposed system has applied to Korean text summarization system and is shown satisfactory results.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.11
• /
• pp.4473-4482
• /
• 2010

The examination of pulse, which is a typical palpation technique in the oriental medicine, has been used conventional analog system for discrimination of 28 pulses. However, the clipping phenomenon in the pulses, which used same feature extraction technique with ECG signals, has been occurred in analog system due to feature extraction method and over amplification from the input signals. It caused inaccurate to analyze the pulse signals. In this paper, we propose a digital filter design technique based on Prony's method for signal modeling and C-spline interpolation for feature extraction from pulse signal to compensate analog pulse detection system. In addition, we suggest a compensated electronic pulse detection system comprising new pulse analyzing algorithm and shape analysis technique for pulses, which were difficult to use in analog system. The feasibility for new proposed system has been confirmed comparing output signals between electronic pulse detection system having proposed filter design techniques with pulse analyzing algorithm and conventional analog system.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.14 no.6
• /
• pp.7-14
• /
• 2018

Salient information extraction from multi-view videos is a very challenging area because of inter-view, intra-view correlations, and computational complexity. There are several techniques developed for keyframes extraction from multi-view videos with very high computational complexities. In this paper, we present a keyframes extraction approach from multi-view videos using entropy and complexity information present inside frame. In first step, we extract representative shots of the whole video from each view based on structural similarity index measurement (SSIM) difference value between frames. In second step, entropy and complexity scores for all frames of shots in different views are computed. Finally, the frames with highest entropy and complexity scores are considered as keyframes. The proposed system is subjectively evaluated on available office benchmark dataset and the results are convenient in terms of accuracy and time complexity.

• Korean Journal of Food Science and Technology
• /
• v.41 no.3
• /
• pp.284-291
• /
• 2009

This study was conducted to characterize the enhanced antioxidant and anticancer activities of Berberis koreana bark following a low temperature and high pressure extraction process. The results indicate that the B. koreana bark extracted as described showed a 93% increase in DPPH radical scavenging activity. Inhibition activity of xanthine oxidase was highest by this extraction process. In addition the growth of human lung cancer cells (A549), human stomach cancer cells (AGS), human breast cancer cells (MCF-7) and human liver cancer cells (Hep3B) were inhibited by 70.8%, 86.2%, 84.3% and 62.5% respectively. These data indicate that this low temperature and high pressure extraction technique results in the efficient extraction of bioactive compounds from rigid plant materials. This process could also be combined with other techniques to improve extraction yields and identify new biologically active substances from relatively hard plants.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.4
• /
• pp.41-51
• /
• 2023

In this paper, we propose a technique of multi-time window feature extraction for anger detection in gait data. In the previous gait-based emotion recognition methods, the pedestrian's stride, time taken for one stride, walking speed, and forward tilt angles of the neck and thorax are calculated. Then, minimum, mean, and maximum values are calculated for the entire interval to use them as features. However, each feature does not always change uniformly over the entire interval but sometimes changes locally. Therefore, we propose a multi-time window feature extraction technique that can extract both global and local features, from long-term to short-term. In addition, we also propose an ensemble model that consists of multiple classifiers. Each classifier is trained with features extracted from different multi-time windows. To verify the effectiveness of the proposed feature extraction technique and ensemble model, a public three-dimensional gait dataset was used. The simulation results demonstrate that the proposed ensemble model achieves the best performance compared to machine learning models trained with existing feature extraction techniques for four performance evaluation metrics.

• Korean Journal of Remote Sensing
• /
• v.21 no.4
• /
• pp.341-369
• /
• 2005

Hyperspectral images have emerged as a new and promising remote sensing data that can overcome the limitations of existing optical image data. This study was designed to provide a comprehensive review on definition, data processing methods, and applications of hyperspectral data. Various types of airborne, spaceborne, and field hyperspectral image sensors were surveyed from the available literatures and internet search. To understand the current status of hyperspectral remote sensing technology and research development, we collected several hundreds research papers from international journals (IEEE Transactions on Geoscience and Remote Sensing, International Journal of Remote Sensing, Remote Sensing of Environment and AVIRIS Workshop Proceedings), and categorized them by sensor types, data processing techniques, and applications. Although several hyperspectral sensors have been developing, AVIRIS has been a primary data source that the most hyperspectral remote sensing researches were relied on. Since hyperspectral data have very large data volume with many spectral bands, several data processing techniques that are particularly oriented to hyperspectral data have been developed. Although atmospheric correction, spectral mixture analysis, and spectral feature extraction are among those processing techniques, they are still in experimental stage and need further refinement until the fully operational adaptation. Geology and mineral exploration were major application in early stage of hyperspectral sensing because of the distinct spectral features of rock and minerals that could be easily observed with hyperspectral data. The applications of hyperspectral sensing have been expanding to vegetation, water resources, and military areas where the multispectral sensing was not very effective to extract necessary information.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.5
• /
• pp.538-547
• /
• 2001

Currently, 2D real-time image coding techniques had great developments and many related products were commercially developed. However, these techniques lack the capability of handling 3D actuality, occurred by the advent of virtual reality, because they handle only the temporal transmission for 2D image. Besides, many 3D virtual reality researches have been studied in computer graphics. Since the graphical researches were limited to the application of artificial models, the 3D actuality for real scene images could not be managed also. Therefore, a new 3D model extraction method based on stereo vision, that can deal with real scene virtual reality, is proposed in this paper. The proposed method adapted a compact genetic algorithm using population-based incremental learning (PBIL) to matching environments, in order to reduce memory consumption and computational time of conventional genetic algorithms. Since the PBIL used a probability vector and competitive learning, the matching algorithm became simple and the computation load was considerably reduced. Moreover, the matching quality was superior than conventional methods. Even if the characteristics of images are changed, stable outputs were obtained without the modification of the matching algorithm.

• Journal of Periodontal and Implant Science
• /
• v.38 no.4
• /
• pp.729-736
• /
• 2008

Purpose: Anterior ridge defect after tooth extraction results in unfavorable appearance. Ridge augmentation procedures should be preceded by careful surgical-prosthetic treatment planning, and various techniques can be used in anterior ridge augmentation. Materials and Methods: Three patients showed deformed ridges after tooth extraction. Three different techniques ; onlay-interpositional connective tissue graft; bovine hydroxyapatite graft with free connective tissue graft; bovine hydroxyapatite graft with resorbable collagen membrane following free connective tissue graft; were used for anterior ridge augmentation. Result: Soft tissue graft can be used in small amount of ridge defect, hard tissue graft combined with soft tissue graft can be used in large amount of ridge defect. After ridge augmentation, about three months of healing period, augmented tissue was stabilized. The final restoration was initiated after this healing period, and the tissue form was maintained stable. Conclusion: Careful diagnosis and surgical-prosthetic treatment planning with joint consultation prior to surgery should be performed in order to attain an optimal esthetic results.

• Journal of the Korean Society of Tobacco Science
• /
• v.30 no.2
• /
• pp.117-124
• /
• 2008

A solid phase microextraction (SPME) method in combination with gas chromatography/mass spectrometric techniques was used for the extraction and quantification of 12 selected agrochemical residues in tobacco. The parameters such as the type of SPME fiber, adsorption/desorption time and the extraction temperature affecting the precision and accuracy of the SPME method were investigated and optimized. Among three types of fibers investigated, polyacrylate (PA), polydimethylsiloxane (PDMS) and polydimethylsiloxane-divinylbenzene (PDMS-DVB), PDMS fiber was selected for the extractions of the agrochemicals. The SPME device was automated and on-line coupled to a gas chromatograph with a mass spectrometer. Mass spectrometry (MS) was used and two different instruments, a quadrupole MS and triple quadrupole MS-MS mode, were compared. The performances of the two GC-MS instruments were comparable in terms of linearity (in the range of 0.01$\sim$0.5 $\mu$g/mL) and sensitivity (limits of detection were in the low ng/mL range). The triple quadrupole MS-MS instrument gave better precision than that of quadrupole MS system, but generally the relative standard deviations for replicates were acceptable for both instruments (< 15%). The LODs was fully satisfied the requirements of the CORESTA GRL. Recoveries of 12 selected agrochemicals in tobacco yielded more than 80% and reproducibility was found to be better than 10% RSD so that SPME procedure could be applied to the quantitative analysis of agrochemical residues in tobacco.

• The Journal of the Society of Korean Medicine Diagnostics
• /
• v.17 no.2
• /
• pp.77-89
• /
• 2013

Objectives Tongue diagnosis is an important diagnostic method in traditional Eastern medicine, and it has a high potential to be used in the future healthcare because of easy, quick, and non-contact measuring features. Recently, research and development efforts on computerized tongue diagnosis systems (CTDS) have been active that led to the technical advancements in the field of photographing techniques, image extraction and classification algorithms. In this study, we analyzed the trends in the CTDS patents. Using the WIPS search engine (www.wipsglobal.com), quantitative and qualitative patent analyses were performed through Korea, China, Japan, U.S.A and Europe. Methods For a systematic search and data analysis, we defined patent categories based on the application area and technical details. By applying thus-obtained categorical key words, we obtained 360 relevant patents on photographing techniques, image extraction and classification algorithms for the purpose of diagnosis or security. Results As a result, companies related to image acquisition, medical imaging and mobile devices and research groups of universities in East Asia were major patent applicants. In all the five countries, the number of patents have been increasing since 1980. In particular, technology related to color correction and image segmentation were most actively patented categories, and expected to continue a high application rate.