• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.9 no.4
• /
• pp.745-754
• /
• 2019

As GeoWeb 2.0 technologies are widely used, various kinds of services that mashup spatial data and user data are being developed. In particular, various spatial information platforms such as Google Maps, OpenStreetMap, Daum Map, Naver Map, olleh Map, and VWorld based on GeoWeb 2.0 technologies support mashup service. The mashup service which is supported through the Open APIs of the platforms, provides various kinds of spatial data such as 2D map, 3D map, and aerial image. Also, application fields using the mashup service are greatly expanded. Recently, as user data for mashup have been greatly increased, there was a problem in mashup performance. However, the research on the mashup performance improvement is currently insufficient, even the research on the mashup performance comparison of various platforms has not been performed. In this paper, we perform comparative analysis of the mashup performance for large amounts of user data and spatial data using various spatial information platforms available in Korea. Specifically, we propose two performance analysis indexes of mashup time and user interaction time in order to analyze the mashup performance efficiently. Also, we implement a system for the performance analysis. Finally, from the performance analysis result, we propose a spatial information platform that can be efficiently applied to cases when user data increases greatly and user interaction occurs frequently.

• Journal of Internet Computing and Services
• /
• v.22 no.2
• /
• pp.59-68
• /
• 2021

As artificial intelligence(AI) technologies, which have made rapid growth recently, began to be applied to the marine environment such as ships, there have been active researches on the application of CNN-based models specialized for digital videos. In E-Navigation service, which is combined with various technologies to detect floating objects of clash risk to reduce human errors and prevent fires inside ships, real-time processing is of huge importance. More functions added, however, mean a need for high-performance processes, which raises prices and poses a cost burden on shipowners. This study thus set out to propose a method capable of processing information at a high rate while maintaining the accuracy by applying Quantization techniques of a deep learning model. First, videos were pre-processed fit for the detection of floating matters in the sea to ensure the efficient transmission of video data to the deep learning entry. Secondly, the quantization technique, one of lightweight techniques for a deep learning model, was applied to reduce the usage rate of memory and increase the processing speed. Finally, the proposed deep learning model to which video pre-processing and quantization were applied was applied to various embedded boards to measure its accuracy and processing speed and test its performance. The proposed method was able to reduce the usage of memory capacity four times and improve the processing speed about four to five times while maintaining the old accuracy of recognition.

• Korean Journal of Remote Sensing
• /
• v.37 no.6_3
• /
• pp.1965-1974
• /
• 2021

Satellite remote sensing approach can be actively used for forest monitoring. Especially, it is much meaningful to utilize Korea multi-purpose satellites, an independently operated satellite in Korea, for forest monitoring of Korea, Recently, several studies have been performed to exploit meaningful information from satellite remote sensed data via machine learning approaches. The forest information produced through machine learning approaches can be used to support the efficiency of traditional forest monitoring methods, such as in-situ survey or qualitative analysis of aerial image. The performance of machine learning approaches is greatly depending on the characteristics of study area and data. Thus, it is very important to survey the best model among the various machine learning models. In this study, the performance of deep neural network to classify artificial or natural forests was analyzed in Samcheok, Korea. As a result, the pixel accuracy was about 0.857. F1 scores for natural and artificial forests were about 0.917 and 0.433 respectively. The F1 score of artificial forest was low. However, we can find that the artificial and natural forest classification performance improvement of about 0.06 and 0.10 in F1 scores, compared to the results from single layered sigmoid artificial neural network. Based on these results, it is necessary to find a more appropriate model for the forest type classification by applying additional models based on a convolutional neural network.

• Korean Journal of Remote Sensing
• /
• v.37 no.2
• /
• pp.233-243
• /
• 2021

Photovoltaic panels are hazardous electronic waste that has heavy metal as one of the hazardous components. Each year, hazardous electronic waste is increasing worldwide and every heavy rainfall exposes the photovoltaic panel to become the source of heavy metal soil contamination. the development needs a monitoring technology for this hazardous exposure. this research use relationships between SAR temporal baseline and coherence of Sentinel-1 satellite to detected photovoltaic panel. Also, the photovoltaic plant detection tested using the difference between that photovoltaic panel and the other difference surface of coherence. The author tested the photovoltaic panel and its environment to calculate differences in coherence relationships. As a result of the experiment, the coherence of the photovoltaic panel, which is assumed to be a permanent scatterer, shows a bias that is biased toward a median value of 0.53 with a distribution of 0.50 to 0.65. Therefore, further research is needed to improve errors that may occur during processing. Additionally, the author found that the change detection using a temporal baseline is possible as the rate of reduction of coherence of photovoltaic panels differs from those of artificial objects such as buildings. This result could be an efficient way to continuously monitor regardless of weather conditions, which was a limitation of the existing optical satellite image-based photovoltaic panel detection research and to understand the spatial distribution in situations such as photovoltaic panel loss.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.2
• /
• pp.492-497
• /
• 2019

This study analysed the legal concept of personal information(PI), which was not differentiated from behavioral information, and established it clearly for invigorating online targeted advertising(OTA), which draw attention in big data era; by selecting Guidelines of Assessment of Data Breach Incident Factors and Guidelines of Measures for No-Identifying Personal Information based on Personal Information Protection Act(PIPA) and Enforcement Decree of the PIPA. As a result, PI was defined as any kind of information relating to (1)a living individual(not group, corporate body or things etc.); (2)makes possibly identify the individual by his or her identifiers such as name, resident registration number, image, etc. (not included if not identify the individual); and (3)including information like attribute values which makes possibly identify any specific individual, if not by itself, but combined with other information which can be actually collected and combined). Specifically, PI includes basic, proper distinguishable, sensitive and other PI. It is suggested that PI concept should be researched continually with digital technology development; the effectiveness of the Guidelines of PI Protection in OTA, the legal principles of PI protection from not only users' but business operators' perspectives and the differentiation between PI and behavioral information in OTA should be researched.

• Journal of dental hygiene science
• /
• v.18 no.6
• /
• pp.367-373
• /
• 2018

The consumption of beverages among children is rising. The purpose of this study was to examine the effect of kid's drink on dental resin-based pit and fissure sealant. Pororo, I-kicker, Sunkist kids were included in the experimental groups, and Samdasu was included in the control group. A conventional dental sealant material ($Clinpro^{TM}Sealant^{(R)}$) was selected for this study. Resin specimens (8 mm in diameter and 1 mm in thickness) were prepared according to manufacturers' instructions and the initial roughness (Ra) was then measured. The pH of all the four groups was measured using a pH meter. The specimens were individually immersed in 5 ml of the experimental solutions and stored at $37^{\circ}C$ for 72 hours. Following this, the surface roughness of the resin specimens was measured by Surftest. The concentration of residual monomer released was determined by high performance liquid chromatography (HPLC). The surface morphology of the resin specimen was evaluated before and after storage by scanning electron microscopy (SEM). Data were statistically analyzed using Kruskal-Wallis and Duncan's test. The results showed that all the children's beverages examined in this study contained citric acid. The pH of I-kicker was the lowest ($3.03{\pm}0.01$), followed by that of Sunkist kids ($3.26{\pm}0.02$) and Pororo ($3.47{\pm}0.02$). We observed an increase in the surface roughness of resin specimens after 72 h of immersion in all the beverages tested (p<0.05). There was matrix degradation after immersion, visualized on SEM image, in all the beverage groups. Bisphenol-A-glycidyl methacrylate was not detected after 72 hours, but triethylene glycol dimethacrylate levels were increased in all the beverages tested during the 72 hours by HPLC. These results suggest that intake of beverages containing acid can cause degradation of the resin-based pit and fissure sealants in children.

• Journal of Convergence for Information Technology
• /
• v.8 no.6
• /
• pp.367-374
• /
• 2018

The purpose of this research is to promote traditional markets, which are part of urban regeneration project, and to promote cultural and tourism market by applying characteristics and differentiated elements through story development through market-related resources in order to secure identity of the JeCheon Central market that lost function of the traditional market and regional aging of the traditional market. To this end, the basic survey and analysis of the Jecheon area and the current situation of the Jecheon Central Market were conducted to diagnose problems and to analyze keywords through surveys by local merchants and visitors. By drawing up measures to vitalize the Jecheon Central Market by floor and space, the Jecheon Central Market's design story is developed and applied so that it can restore the image of the local traditional market through regional and cultural elements and become a center of space and culture that can become a landmark for the region in the future. The storytelling designed for this purpose shall be linked to the spatial planning of each floor as well as the C.I. and exterior of the C.I. and the building of the Jecheon Central Market, and the identity of the Jecheon Central Market can be reestablishe.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.4
• /
• pp.859-866
• /
• 2019

Cybersecurity attack targeting a specific user is rising in number, even enterprises are trying to strengthen their cybersecurity. Network segmentation environment where public network and private network are separated could block information coming from the outside, however, it is unable to control outside information for business efficiency and productivity. Even if enterprises try to enhance security policies and introduce the network segmentation system and a solution incorporating CDR technology to remove unnecessary data contained in files, it is still exposed to security threats. Therefore, we suggest a system that uses file format conversion to transmit a secure file in the network separation environment. The secure file is converted into an image file from a document, as it reflects attack patterns of inserting malicious code into the document file. Additionally, this paper proposes a system in the environment which functions that a document file can keep information for incident response, considering forensic readiness.

• Journal of the Korea Convergence Society
• /
• v.10 no.6
• /
• pp.7-14
• /
• 2019

In this paper, we propose a Cascade Classifier face detection method using the Haar-like feature, which is complemented by the Flood Fill algorithm for lossy areas due to illumination and shadow in YCbCr color space extraction. The Cascade Classifier using Haar-like features can generate noise and loss regions due to lighting, shadow, etc. because skin color extraction using existing YCbCr color space in image only uses threshold value. In order to solve this problem, noise is removed by erosion and expansion calculation, and the loss region is estimated by using the Flood Fill algorithm to estimate the loss region. A threshold value of the YCbCr color space was further allowed for the estimated area. For the remaining loss area, the color was filled in as the average value of the additional allowed areas among the areas estimated above. We extracted faces using Haar-like Cascade Classifier. The accuracy of the proposed method is improved by about 4% and the detection rate of the proposed method is improved by about 2% than that of the Haar-like Cascade Classifier by using only the YCbCr color space.

• 재활복지
• /
• v.21 no.3
• /
• pp.129-146
• /
• 2017

In an event of an emergency, such as fire in a building, visually impaired and blind people are prone to exposed to a level of danger that is greater than that of normal people, for they cannot be aware of it quickly. Current fire detection methods such as smoke detector is very slow and unreliable because it usually uses chemical sensor based technology to detect fire particles. But by using vision sensor instead, fire can be proven to be detected much faster as we show in our experiments. Previous studies have applied various image processing and machine learning techniques to detect fire, but they usually don't work very well because these techniques require hand-crafted features that do not generalize well to various scenarios. But with the help of recent advancement in the field of deep learning, this research can be conducted to help solve this problem by using deep learning-based object detector that can detect fire using images from security camera. Deep learning based approach can learn features automatically so they can usually generalize well to various scenes. In order to ensure maximum capacity, we applied the latest technologies in the field of computer vision such as YOLO detector in order to solve this task. Considering the trade-off between recall vs. complexity, we introduced two convolutional neural networks with slightly different model's complexity to detect fire at different recall rate. Both models can detect fire at 99% average precision, but one model has 76% recall at 30 FPS while another has 61% recall at 50 FPS. We also compare our model memory consumption with each other and show our models robustness by testing on various real-world scenarios.