• Journal of Industrial Technology
• /
• v.25 no.A
• /
• pp.81-86
• /
• 2005

A vibration monitoring system for a small class of wind turbine (W/T) is established and operated. The monitoring system consists of monolithic integrated chip accelerometer for vibration monitoring, anemometers for wind data acquisition and auxiliary sensors for atmospheric data. Using the monitoring system, vibration response of a 6kW W/T generator is investigated. Acceleration data of the W/T tower under various operation condition is acquired in real time using LabVIEW and is remotely transferred from the test site to the laboratory in school by internet. Vibration state of the tower structure is diagnosed within the operating speed range. Resonance frequency range of the test model is investigated with the wind speed data of the test site.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.1383-1385
• /
• 1996

A network based interactive magnetic resonance imaging (MRI) system has been developed using the World Wide Web. For this purpose, an HTTP server is developed on the host computer of the MRI system. Capabilities of video and audio conferencing are included for monitoring experiment. Using the developed system. MRI imaging has been successfully carried out at the Signal Processing Lab in the Kwangwoon University with the remote MRI system located at the Medical Image Research Center at the KAIST in Daejon.

• Proceedings of the KOSOMBE Conference
• /
• v.1996 no.05
• /
• pp.139-142
• /
• 1996

A remotly controlled interactive magnetic resonance imaging (MRI) has been tried using the World Wide Web. For this purpose, an HTTP server is developed on the host computer of the MRI system. Video and audio conferencing capability is also included for the experiment. Using the developed system, MRI imaging has been successfully carried out at the Signal Processing Lab in the Kwangwoon University with the remote MRI system located at the Medical Image Research Center in the KAIST in Daejon.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.3 no.3
• /
• pp.12-19
• /
• 2000

Knowledge for reflectance characteristic of interesting targets will provide us with actual application of remote sensing on agriculture. In this study, we have measured and analyzed reflectivity characteristics based on growing status from transplanting time to harvesting time. Rice paddies transplant into 3 fields at 20, May, 1999. Measurement of reflectivity characteristics were carried out with a portable spectroradiometer for frequencies from 300nm to 1100nm during the time period from 11:00 AM to 01:00 PM of clear sky and calm a day. The measurements for a day repeated 3 times(also, 3 times to each measurement)for reliable values. In result, we found that averaged reflectivity of visible range has about 2.34% - 2.55% in blue region(400nm-498nm), about 5.05% - 6.01% in green region(500nm-598nm) and about 4.21% - 5.24% in red region(600nm-698nm). It must be noted that the more rice canopy grows, the more spectral reflectivity decreases in visible region. Also, we separated infrared region into two cases - One case is increasing region with 700nm-780nm, the other is fixed region with 800nm-1100nm. Averaged reflectivity of these regions has about 22.3% - 23.0% in increasing region, about 29.4% - 33.1% in fixed region. It must be noted that more rice canopy grows, the more spectral reflectivity also increases up to 23, Aug. in infrared region. After 23, Aug, the reflectivity has a tendency toward decrease.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.6
• /
• pp.435-444
• /
• 2019

Recently, DL (Deep Learning) has been rapidly applied in various fields. In particular, classification and object recognition from images are major tasks in computer vision. Most of the DL utilizing imagery is primarily based on the CNN (Convolutional Neural Network) and improving performance of the DL model is main issue. While most CNNs are involve with images for training data, this paper aims to classify and recognize objects using DSM (Digital Surface Model), and slope and aspect information derived from the DSM instead of images. The DSM data sets used in the experiment were established by DGPF (German Society for Photogrammetry, Remote Sensing and Geoinformatics) and provided by ISPRS (International Society for Photogrammetry and Remote Sensing). The CNN-based SegNet model, that is evaluated as having excellent efficiency and performance, was used to train the data sets. In addition, this paper proposed a scheme for training data generation efficiently from the limited number of data. The results demonstrated DSM and derived data could be feasible for semantic classification with desirable accuracy using DL.

• Korean Journal of Remote Sensing
• /
• v.30 no.4
• /
• pp.417-430
• /
• 2014

We have found an error in the operational OMI HCHO columns, and corrected it by applying a background parameterization derived on a 4th order polynomial fit to the time series of monthly average OMI HCHO data. The corrected OMI HCHO agrees with this understanding as well as with the other sensors measurements and has no unrealistic trends. A new scientific approach, statistical analyses with EOF and SVD, was adapted to reanalyze the consistency of the corrected OMI HCHO with other satellite measurements of HCHO, CO, $NO_2$, and fire counts over Africa. The EOF and SVD analyses with MOPITT CO, OMI $NO_2$, SCIAMAHCY, and OMI HCHO show the overall spatial and temporal pattern consistent with those of biomass burning over these regions. However, some discrepancies were observed from OMI HCHO over northern equatorial Africa during the northern biomass burning seasons: The maximum HCHO was found further downwind from where maximum fire counts occur and the minimum was found in January when biomass burning is strongest. The statistical analysis revealed that the influence of biogenic activity on HCHO wasn't strong enough to cause the discrepancies, but it is caused by the error in OMI HCHO from using the wrong Air Mass Factor (AMF) associated with biomass burning aerosol. If the error is properly taken into consideration, the biomass burning is the strongest source of HCHO seasonality over the regions. This study suggested that the statistical tools are a very efficient method for evaluating satellite data.

• Korean Journal of Remote Sensing
• /
• v.24 no.4
• /
• pp.289-298
• /
• 2008

An algorithm was developed to retrieve both cloud optical thickness and effective particle radius considered the aerosol effect on clouds. This study apply the algorithm of Nakajima and Nakajima (1995) that is used to retrieve cloud optical thickness and effective particle radius from visible, near infrared satellite spectral measurements. To retrieve cloud properties, Look-up table (LUT) was made under different atmospheric conditions by using a radiative transfer model. Especially the vertical distribution of aerosol is based on a tropospheric aerosol profile in radiative transfer model. In the case study, we selected the extensive forest fire occurred in Russia in May 2003. The aerosol released from this fire may be transported to Korea. Cloud properties obtained from these distinct atmospheric situations are analysed in terms of their possible changes due to the interactions of the clouds with the aerosol particle plumes. Cloud properties over the East sea at this time was retrieved using new algorithm. The algorithm is applied to measurements from the MODerate Resolution Imaging Spectrometer (MODIS) onboard the Terra spacecrafts. As a result, cloud effective particle radius was decreased and cloud optical thickness was increased during aerosol event. Specially, cloud effective particle radius is hardly greater than $20{\mu}m$ when aerosol particles were present over the East Sea. Clouds developing in the aerosol event tend to have more numerous but smaller droplets.

• Korean Journal of Remote Sensing
• /
• v.34 no.6_3
• /
• pp.1427-1443
• /
• 2018

Object-based image analysis (OBIA) allows higher computation efficiency and usability of information inherent in the image, as it reduces the complexity of the image while maintaining the image properties. Superpixel methods oversegment the image with a smaller image unit than an ordinary object segment and well preserve the edges of the image. SLIC (Simple linear iterative clustering) is known for outperforming the previous superpixel methods with high image segmentation quality. Although the input parameter for SLIC, number of superpixels has considerable influence on image segmentation results, impact analysis for SLIC parameter has not been investigated enough. In this study, we performed optimal parameter analysis and evaluation of change detection for SLIC-based superpixel techniques using KOMPSAT data. Forsuperpixel generation, three superpixel methods (SLIC; SLIC0, zero parameter version of SLIC; SNIC, simple non-iterative clustering) were used with superpixel sizes in ranges of $5{\times}5$ (pixels) to $50{\times}50$ (pixels). Then, the image segmentation results were analyzed for how well they preserve the edges of the change detection reference data. Based on the optimal parameter analysis, image segmentation boundaries were obtained from difference image of the bi-temporal images. Then, DBSCAN (Density-based spatial clustering of applications with noise) was applied to cluster the superpixels to a certain size of objects for change detection. The changes of features were detected for each superpixel and compared with reference data for evaluation. From the change detection results, it proved that better change detection can be achieved even with bigger superpixel size if the superpixels were generated with high regularity of size and shape.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.2
• /
• pp.91-108
• /
• 2022

In most cases, optical images have been used as training data of DL (Deep Learning) models for object detection, recognition, identification, classification, semantic segmentation, and instance segmentation. However, properties of 3D objects in the real-world could not be fully explored with 2D images. One of the major sources of the 3D geospatial information is DSM (Digital Surface Model). In this matter, characteristic information derived from DSM would be effective to analyze 3D terrain features. Especially, man-made objects such as buildings having geometrically unique shape could be described by geometric elements that are obtained from 3D geospatial data. The background and motivation of this paper were drawn from concept of the intrinsic image that is involved in high-level visual information processing. This paper aims to extract buildings after classifying terrain features by training DL model with DSM-derived information including slope, aspect, and SRI (Shaded Relief Image). The experiments were carried out using DSM and label dataset provided by ISPRS (International Society for Photogrammetry and Remote Sensing) for CNN-based SegNet model. In particular, experiments focus on combining multi-source information to improve training performance and synergistic effect of the DL model. The results demonstrate that buildings were effectively classified and extracted by the proposed approach.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_3
• /
• pp.913-924
• /
• 2022

In this paper, we presented a database construction of undersea images for the Habitats of Ecklonia cava and Sargassum and conducted an experiment for semantic segmentation using state-of-the-art (SOTA) models such as High Resolution Network-Object Contextual Representation (HRNet-OCR) and Shifted Windows-L (Swin-L). The result showed that our segmentation models were superior to the existing experiments in terms of the 29% increased mean intersection over union (mIOU). Swin-L model produced better performance for every class. In particular, the information of the Ecklonia cava class that had small data were also appropriately extracted by Swin-L model. Target objects and the backgrounds were well distinguished owing to the Transformer backbone better than the legacy models. A bigger database under construction will ensure more accuracy improvement and can be utilized as deep learning database for undersea images.