• Korean Journal of Remote Sensing
• /
• v.38 no.5_3
• /
• pp.819-834
• /
• 2022

Artificial Intelligence (AI) techniques have been effectively used for image classification, object detection, and image segmentation. Along with the recent advancement of computing power, deep learning models can build deeper and thicker networks and achieve better performance by creating more appropriate feature maps based on effective activation functions and optimizer algorithms. This review paper examined technical and academic trends of Convolutional Neural Network (CNN) and Transformer models that are emerging techniques in remote sensing and suggested their utilization strategies and development directions. A timely supply of satellite images and real-time processing for deep learning to cope with disaster monitoring will be required for future work. In addition, a big data platform dedicated to satellite images should be developed and integrated with drone and Closed-circuit Television (CCTV) images.

• Spatial Information Research
• /
• v.22 no.5
• /
• pp.27-41
• /
• 2014

Recently, BIM(Building Information Modeling) has been applied to the infrastructure such as road and bridge, and information about the outside environment of buildings is needed for maintaining and managing the large urban facilities. In addition, the convergence between spatial information and Big-data has a large potentiality, in respect that considerable profits and developments in other application problems such as various simulations and urban plans, national land security, may be brought about on the basis of the interoperability of information between BIM and GIS. Therefore, this study attempted to suggest the development direction of a model integrating building for spatial information analysis and city on the subject by comparing and analyzing difference between information system and shape expression of IFC, CityGML and LandXML to efficiently link information between IFC as a standard model of BIM and CityGML as a standard model in the GIS sector and to prepare a basic fusion strategy and a method of utilization between BIM and GIS. The result of the study are as follow. Firstly, contents and structure of IFC, CityGML and LandXML are compared and analyzed. Secondly, the development direction of iBUM(Integrated Building and Urban Model) suggested, which is based on convergence technology for analysis of space information. Finally, a strategy and method of the BIM and GIS are proposed in the iBUM environment.

• The Journal of Society for e-Business Studies
• /
• v.21 no.3
• /
• pp.15-28
• /
• 2016

This paper proposes a system recommending spatial information what user wants with collecting and analyzing tweets around the user's location by using the GPS information acquired in mobile. This system has built an emotion dictionary and then derive the recommendation score of morphological analyzed tweets to provide not just simple information but recommendation through the emotion analysis information. The system also calculates distance between the recommended tweets and user's latitude-longitude coordinates and the results showed the close order. This paper evaluates the result of the emotion analysis in a total of 10 areas with two keyword 'Restaurants' and 'Performance.' In the result, the number of tweets containing the words positive or negative are 122 of the total 210. In addition, 65 tweets classified as positive or negative by analyzing emotions after a morphological analysis and only 46 tweets contained the meaning of the positive or negative actually. This result shows the system detected tweets containing the emotional element with recall of 38% and performed emotion analysis with precision of 71%.

• Korean Journal of Remote Sensing
• /
• v.36 no.3
• /
• pp.487-501
• /
• 2020

As a viable option for retrieval of LST (Land Surface Temperature), this paper presents a DNN (Deep Neural Network) based approach using 148 Landsat 8 images for South Korea. Because the brightness temperature and emissivity for the band 10 (approx. 11-㎛ wavelength) of Landsat 8 are derived by combining physics-based equations and empirical coefficients, they include uncertainties according to regional conditions such as meteorology, climate, topography, and vegetation. To overcome this, we used several land surface variables such as NDVI (Normalized Difference Vegetation Index), land cover types, topographic factors (elevation, slope, aspect, and ruggedness) as well as the T₀ calculated from the brightness temperature and emissivity. We optimized four seasonal DNN models using the input variables and in-situ observations from ASOS (Automated Synoptic Observing System) to retrieve the LST, which is an advanced approach when compared with the existing method of the bias correction using a linear equation. The validation statistics from the 1,728 matchups during 2013-2019 showed a good performance of the CC=0.910~0.917 and RMSE=3.245~3.365℃, especially for spring and fall. Also, our DNN models produced a stable LST for all types of land cover. A future work using big data from Landsat 5/7/8 with additional land surface variables will be necessary for a more reliable retrieval of LST for high-resolution satellite images.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.5
• /
• pp.49-63
• /
• 2019

Crowdedness management at subway platforms is essential to improve services, including the prevention of train delays and ensuring passenger safety. Establishing effective crowdedness mitigation measures for platforms requires accurate estimation of the congestion level. There are temporal and spatial constraints since crowdedness on subway platforms is assessed at certain locations every 1-2 years by hand counting. However, smart cards generate real-time big data 24 hours a day and could be used in estimating congestion. This study proposes a model based on data from transit cards to estimate crowdedness dynamically. Crowdedness was defined as demand, which can be translated into passengers dynamically moving along a subway network. The trajectory of an individual passenger can be identified through this model. Passenger flow that concentrates or disperses at a platform is also calculated every minute. Lastly, the platform congestion level is estimated based on effective waiting areas for each platform structure.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.18 no.4
• /
• pp.119-130
• /
• 2015

In recent years, the attention on the applications of the national or international standards to water-related information technology in practice has more increased. In particular, as the demand on spatial information technology including content development, receiving, disposal and distribution has increased, the importance of standardization has been also emphasized. However, compared with attention and demand on standardization in spatial information technology, studies on development and application of standardization are still at the initial stage. Accordingly, this study attempted to investigate the trends of international standards developed and established by International Organization for Standardization(ISO) and Open Geospatial Consortium(OGC), and to derive the adaptable integrated management standard technology for water hazard information platform. For this, we investigated big data, NoSQL, and cloud technology for the observed data provision. Furthermore, OGC WxS standard technologies for spatial information web service and ISO standards for spatial information format were investigated. Based on these investigations, we examined the strategies and plans to apply and establish the standardization for information provision.

• 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.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.23 no.3
• /
• pp.100-119
• /
• 2020

During recent years, the heat environment and particulate matter (PM₁₀) have become serious environmental problems, as increases in heat waves due to rising global temperature interact with weakening atmospheric wind speeds. There exist urban heat islands and urban pollution islands with higher temperatures and air pollution concentrations than other areas. However, few studies have examined these issues together because of a lack of micro-scale data, which can be constructed from spatial data. Today, with the help of satellite images and big data collected by private telecommunication companies, detailed spatial distribution analyses are possible. Therefore, this study aimed to examine the spatial distribution patterns of urban heat islands and urban pollution islands within Busan Metropolitan City and to compare the distributions of the two phenomena. In this study, the land surface temperature of Landsat 8 satellite images, air temperature and particulate matter concentration data derived from a private automated meteorological observation system were gridded in 30m × 30m units, and spatial analysis was performed. Analysis showed that simultaneous zones of urban heat islands and urban pollution islands included some vulnerable residential areas and industrial areas. The political migration areas such as Seo-dong and Bansong-dong, representative vulnerable residential areas in Busan, were included in the co-occurring areas. The areas have a high density of buildings and poor ventilation, most of whose residents are vulnerable to heat waves and air pollution; thus, these areas must be considered first when establishing related policies. In the industrial areas included in the co-occurring areas, concrete or asphalt concrete-based impervious surfaces accounted for an absolute majority, and not only was the proportion of vegetation insufficient, there was also considerable vehicular traffic. A hot-spot analysis examining the reliability of the analysis confirmed that more than 99.96% of the regions corresponded to hot-spot areas at a 99% confidence level.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.2
• /
• pp.73-79
• /
• 2014

Visible light communication is a communication method using an LED having a high-speed ON / OFF rate data to be transmitted, it is used as a wireless high speed data transmission. VLC system evolves, the problem of basic research is intended to improve performance and ensure reliability of the communication. The nature of the visible light communication, communication is performed only in the reach of the light, which indicates a big difference by the emission angle of the light. Therefore, in this paper, in the case of indoor environments generally, with the attributes of the multiple LED is equipped, was applied to a MIMO (Multiple Input Multiple Output) communication system. and analyzed SNR performance and total power can be obtained on the reception side by changing the emission angle of the transmitter. As a result of the simulation was run against this, it was confirmed that there is a significant impact on the performance of BER and SNR performance by the emission angle of the transmitter.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.15-30
• /
• 2023

Recently, various services using artificial intelligence(AI) are emerging in the media field as well However, most of the video editing, which involves finding an editing point and attaching the video, is carried out in a passive manner, requiring a lot of time and human resources. Therefore, this study proposes a methodology that can detect the edit points of video according to whether person in video are spoken by using Video Swin Transformer. First, facial keypoints are detected through face alignment. To this end, the proposed structure first detects facial keypoints through face alignment. Through this process, the temporal and spatial changes of the face are reflected from the input video data. And, through the Video Swin Transformer-based model proposed in this study, the behavior of the person in the video is classified. Specifically, after combining the feature map generated through Video Swin Transformer from video data and the facial keypoints detected through Face Alignment, utterance is classified through convolution layers. In conclusion, the performance of the image editing point detection model using facial keypoints proposed in this paper improved from 87.46% to 89.17% compared to the model without facial keypoints.