• Journal of Korea Multimedia Society
• /
• v.21 no.2
• /
• pp.173-180
• /
• 2018

This paper proposes a method to improve the performance of face recognition via super-resolution method using sparse representation and deep learning from low-resolution facial images. Recently, there have been many researches on ultra-high-resolution images using deep learning techniques, but studies are still under way in real-time face recognition. In this paper, we combine the sparse representation and deep learning to generate super-resolution images to improve the performance of face recognition. We have also improved the processing speed by designing in parallel structure when applying sparse representation. Finally, experimental results show that the proposed method is superior to conventional methods on various images.

• Proceedings of the KSRS Conference
• /
• 2004.10a
• /
• pp.14-16
• /
• 2004

Although there have been wide range of studies to characterize forest stands based upon spectral information of satellite image, it was not fully understood the texture information of forest stand using high resolution data. The objective of this study is to evaluate several texture measures for characterizing forest stand structure, such as species composition, diameter at breast height(DBH), stand density, and age. High resolution IKONOS satellite imagery data were acquired in August 200 lover the forested area near Ulsan, Korea. Primary forest types were plantation pine, mixed forest, and natural deciduous forest of stand age ranging from 10 to 50 years old. Several GLCM-based texture measures were compared with forest stand characteristics. In overall, a texture measure (contrast) calculated using red band were better to differentiate species and age group than other texture measures and near infrared bands.

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.247-249
• /
• 2007

Crown parameters are important roles in tree species identification, because the canopy is the aggregate of all the crowns. However, crown measurements with spaceborne image data have remained more difficult than on aerial photographs since trees show more structural detail at higher resolutions. This recognized problem led to the initiation of the research to determine if high resolution satellite image data could be used to identify and classify single tree species. In this paper, shape parameters derived from pixel-based crown area measurements and texture features derived from GLCM parameters in QuickBird image were tested and compared for individual tree species identification. As expected, initial studies have shown that the crown parameters and the canopy texture parameters provided a differentiating method between coniferous trees and broad-leaved trees within the compartment(less than forest stand) for single extraction from spaceborne high resolution image.

• Asian Journal of Atmospheric Environment
• /
• v.11 no.4
• /
• pp.300-312
• /
• 2017

To improve the understanding of secondary organic aerosol (SOA) formation from the photo-oxidation of anthropogenic and biogenic precursors at the regional background station on Baengnyeong Island, Korea, gas phase and aerosol chemistries were investigated using the Proton Transfer Reaction Time of Flight Mass Spectrometer (PTR-ToF-MS) and the Aerodyne High Resolution Time of Flight Aerosol Mass Spectrometer (HR-ToF-AMS), respectively. HR-ToF-AMS measured fine particles ($PM_1$; diameter of particle matter less than $1{\mu}m$) at a 6-minute time resolution from February to November 2012, while PTR-ToF-MS was deployed during an intensive period from September 21 to 29, 2012. The one-minute time-resolution and high mass resolution (up to $4000m{\Delta}m^{-1}$) data from the PTR-ToF-MS provided the basis for calculations of the concentrations of anthropogenic and biogenic volatile organic compounds (BVOCs) including oxygenated VOCs (OVOCs). The dominant BVOCs from the site are isoprene (0.23 ppb), dimethyl sulphide (DMS, 0.20 ppb), and monoterpenes (0.38 ppb). Toluene (0.45 ppb) and benzene (0.32 ppb) accounted for the majority of anthropogenic VOCs (AVOCs). OVOCs including acetone (3.98 ppb), acetaldehyde (2.67 ppb), acetic acid (1.68 ppb), and formic acid (2.24 ppb) were measured. The OVOCs comprise approximately 75% of total measured VOCs, suggesting the occurrence of strong oxidation processes and/or long-range transported at the site. A strong photochemical aging and oxidation of the atmospheric pollutants were also observed in aerosol measured by HR-ToF-AMS, whereby a high $f_{44}:f_{43}$ value is shown for organic aerosols (OAs); however, relatively low $f_{44}:f_{43}$ values were observed when high concentrations of BVOCs and AVOCs were available, providing evidence of the formation of SOA from VOC precursors at the site. Overall, the results of this study revealed several different SOA formation mechanisms, and new particle formation and particle growth events were identified using the powerful tools scanning mobility particle sizer (SMPS), PTR-ToF-MS, and HR-ToF-AMS.

• Journal of Animal Reproduction and Biotechnology
• /
• v.38 no.4
• /
• pp.209-214
• /
• 2023

Background: Single nucleotide polymorphisms (SNPs) are widely used genetic markers with applications in human disease diagnostics, animal breeding, and evolutionary studies, but existing genotyping methods can be labor-intensive and costly. The aim of this study is to develop a simple and rapid method for identification of a single nucleotide change. Methods: A modified Polymerase Chain Reaction Amplification of Multiple Specific Alleles (PAMSA) and high resolution melt (HRM) analysis was performed to discriminate a bovine polymorphism in the NCAPG gene (rs109570900, 1326T > G). Results: The inclusion of tails in the primers enabled allele discrimination based on PCR product lengths, detected through agarose gel electrophoresis, successfully determining various genotypes, albeit with some time and labor intensity due to the use of relatively costly high-resolution agarose gels. Additionally, high-resolution melt (HRM) analysis with tailed primers effectively distinguished the GG genotype from the TT genotype in bovine muscle cell lines, offering a reliable way to distinguish SNP polymorphisms without the need for time-consuming AS-PCR. Conclusions: Our experiments demonstrated the importance of incorporating unique mismatched bases in the allele-specific primers to prevent cross-amplification by fragmented primers. This efficient and cost-effective method, as presented here, enables genotyping laboratories to analyze SNPs using standard real-time PCR.

• Publications of The Korean Astronomical Society
• /
• v.33 no.2
• /
• pp.21-30
• /
• 2018

Formation processes of high-mass stars have been long-standing issues in astronomy and astrophysics. This is mainly because of major difficulties in observational studies such as a smaller number of high-mass young stellar objects (YSOs), larger distances, and more complex structures in young high-mass clusters compared with nearby low-mass isolated star-forming regions (SFRs), and extremely large opacity of interstellar dust except for centimeter to submillimeter wavelengths. High resolution and high sensitivity observations with Atacama Large Millimeter/Submillimeter Array (ALMA) at millimeter/submillimeter wavelengths will overcome these observational difficulties even for statistical studies with increasing number of high-mass YSO samples. This review will summarize recent progresses in high-mass star-formation studies with ALMA such as clumps and filaments in giant molecular cloud complexes and infrared dark clouds (IRDCs), protostellar disks and outflows in dense cores, chemistry, masers, and accretion bursts in high-mass SFRs.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.4 no.3
• /
• pp.61-70
• /
• 2001

This paper presents a new method to extract semi-automatically roads from high spatial resolution satellite imagery. This method is based both on wavelet transform and on multiresolution analysis combined in the "$\grave{a}$ trous" algorithm. As an urban road network consists on different classes of streets, multiresolution processing allows to extract the streets class by class. The method was applied to a KVR-1000 image on a part of Busan Metropolitan City. The method was carried out for the road extraction of three different widths and it succeeded in extracting good fitted strips. The accuracy analysis for three types of streets was also performed. The overall accuracy in 4 pixels of width is 80.5%. The result suggests that this method can be used to update road networks in the studied urban network. In summary, the multiresolution approach based on the wavelet transform, used in this study, is regarded as one of effective methods to extract urban road network from high spatial resolution satellite images.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.23 no.3
• /
• pp.252-262
• /
• 2020

In Remote Sensing, a machine learning based SVM model is typically utilized for land cover classification. And study using neural network models is also being carried out continuously. But study using high-resolution imagery of KOMPSAT is insufficient. Therefore, the purpose of this study is to assess the accuracy of land cover classification by neural network models using high-resolution KOMPSAT-3 satellite imagery. After acquiring satellite imagery of coastal areas near Gyeongju City, training data were produced. And land cover was classified with the SVM, ANN and DNN models for the three items of water, vegetation and land. Then, the accuracy of the classification results was quantitatively assessed through error matrix: the result using DNN model showed the best with 92.0% accuracy. It is necessary to supplement the training data through future multi-temporal satellite imagery, and to carry out classifications for various items.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.11 no.4
• /
• pp.161-171
• /
• 2008

Nowdays the need of modeling a real world using computers leads to increase the importance of GIS. Moreover, the wide spread of the Internet service has brought about Web GIS Development. Especially, the tuning of Web GIS system has a crucial role in improving the performance of high resolution image Web GIS system. Thus, this study provides a variety of ways to improve the performance of high resolution image Web GIS system. We have tuned IBM AIX operation system, Oracle, ArcSDE in the database level and Tomcat, ArcIMS in the application level. What is more, we propose ways of coding which can have a positive impact on performance. As a result, the proposed methods guarantee high resolution image Web GIS system to improve performance.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.708-708
• /
• 2002

Recent developments in sensor technologies have provided remotely sensed data with very high spatial resolution. In order to fully utilize the potential of high resolution images, new image classification strategies are necessary. Unfortunately, the high resolution images increase the spectral within-field variability, and the classification accuracy of traditional methods based on pixel-based classification algorithms such as Maximum-Likelihood method may be decreased (Schiewe 2001). Recent development in Object Oriented Classification based on image segmentation algorithms can be used for the classification of forest patches on rugged terrain of Korea. The objectives of this paper are as follows. First, to compare the pros and cons of image classification methods based on pixel-based and object oriented classification algorithm for the forest patch classification. Landsat ETM＋ data and IKONOS data will be used for the classification. Second, to investigate ways to increase classification accuracy of forest patches. Supplemental data such as DTM and Forest Type Map of 1:25,000 scale are used for topographic correction and image segmentation. Third, to propose the best classification strategy for forest patch classification in terms of accuracy and data requirement. The research site for this paper is Namhansansung Provincial Park located at the eastern suburb of Seoul Metropolitan City for its diverse forest patch types and data availability. Both Landsat ETM＋ and IKONOS data are used for the classification. Preliminary results can be summarized as follows. First, topographic correction of reflectance is essential for the classification of forest patches on rugged terrain. Second, object oriented classification of IKONOS data enables higher classification accuracy compared to Landsat ETM＋ and pixel-based classification. Third, multi-stage segmentation is very useful to investigate landscape ecological aspect of forest communities of Korea.