• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.269-270
• /
• 2008

In this paper, we discuss the design of an urban drainpipe system based on Geographic Information System (GIS). First we introduce the reasons why we establish this system, and then analyzed the construction, database architecture, network architecture of the system, at last we give a develop direction in the future.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 2004.11a
• /
• pp.193-193
• /
• 2004

• Proceedings of the Korean Magnestics Society Conference
• /
• 2003.12a
• /
• pp.56-56
• /
• 2003

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1737-1741
• /
• 2022

Active fire monitoring using high-resolution drone images and deep learning technologies is now an initial stage and requires various approaches for research and development. This letter examined the detection of active fire objects using You Look Only Once Version 7 (YOLOv7), a state-of-the-art (SOTA) model that has rarely been used in fire detection with drone images. Our experiments showed a better performance than the previous works in terms of multiple quantitative measures. The proposed method can be applied to continuous monitoring of wide areas, with an integration of additional development of new technologies.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.3E
• /
• pp.91-97
• /
• 2004

In this paper, we propose a classification system of TV program scenes based on audio information. The system classifies the video scene into six categories of commercials, basketball games, football games, news reports, weather forecasts and music videos. Two type of audio feature set are extracted from each audio frame-timbral features and coefficient domain features which result in 58-dimensional feature vector. In order to reduce the computational complexity of the system, 58-dimensional feature set is further optimized to yield l0-dimensional features through Sequential Forward Selection (SFS) method. This down-sized feature set is finally used to train and classify the given TV program scenes using κ -NN, Gaussian pattern matching algorithm. The classification result of 91.6% reported here shows the promising performance of the video scene classification based on the audio information. Finally, the system stability problem corresponding to different query length is investigated.

• Journal of Environmental Science International
• /
• v.18 no.8
• /
• pp.899-910
• /
• 2009

We classified wind sectors according to the wind features in South Korea. In order to get the information of wind speed and wind direction, we used and improved on the atmospheric numerical model. We made use of detailed topographical data such as terrain height data of an interval of 3 seconds and landuse data produced at ministry of environment, Republic of Korea. The result of simulated wind field was improved. We carried out the cluster analysis to classify the wind sectors using the K-means clustering. South Korea was classified as 8 wind sectors to the annual wind field.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_2
• /
• pp.859-873
• /
• 2023

Wildfires cause a lot of environmental and economic damage to the Earth over time. Various experiments have examined the harmful effects of wildfires. Also, studies for detecting wildfires and pollutant emissions using satellite remote sensing have been conducted for many years. The wildfire product for the Geostationary Environmental Monitoring Spectrometer (GEMS), Korea's first environmental satellite sensor, has not been provided yet. In this study, a false-color composite for better expression of wildfire smoke was created from GEMS and used in a U-Net model for wildfire detection. Then, a classification model was constructed to distinguish yellow dust from the wildfire smoke candidate pixels. The proposed method can contribute to disaster monitoring using GEMS images.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_2
• /
• pp.875-883
• /
• 2023

In this letter, we used PlanetScope imagery to conduct experiments on the color characteristics for oil type classification of marine oil spills through Red-Green-Blue (RGB) histogram analysis. The histograms of marine oil spills can be divided into three categories (dark black tones, light silver tones, and light rainbow tones) according to the distribution of pixel values in each band. Thick oil layers with dark black tones can be classified as heavy oil, while thin oil layers with light silver and rainbow tones can be classified as light oil. As more images are analyzed in the future, these oil spill detection and classification methods will become more generalized and reliable.

• Proceedings of the KIEE Conference
• /
• 2006.07d
• /
• pp.1852-1853
• /
• 2006

This paper describes the development of temperature controller using program control method for small heating system. This system consists of three parts; sensing part, control part that includes the PID control algorithm, and actuating part. We introduce a zero-crossing control method of TRIAC and firmware technique using single chip micro-controller(ATmega8535) in the control part.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.535-537
• /
• 2005

본 논문에서는 시소러스 상에서 개념간 세분화를 위한 의미적 기준인 개념 패싯 (Conceptual Facet)과 관계 패싯 (Relational Facet) 그룹을 사용하는데, 패싯이란 공통의 특성을 갖는 개념들을 함께 그룹화하고 용어간 관계를 구조화하기 위한 장치이다. 개념 패싯은 용어가 갖는 대표적인 의미속성, 범주를 의미하며, 용어 각각을 구별되는 의미장 (Semantic Field)에 분류하도록 한다. 관계 패싯은 상$\cdot$하위 개념 간 의미 관계를 표현한 메타 개념이다. 본 시소러스는 여러 전문가들의 다양한 관점을 반영하도록 설계되었다. 관점이란 주관적이며, 임의적이어서 개별 개념에 내재된 자질 또는 속성과는 구별되는 독립적 속성이다. 개념 패싯, 관계 패싯의 도입은 계층 관계, 동등 관계, 범주 관계 등과 더불어 용어 간 관계를 보다 구체적으로 명시함으로써 최종 사용자에게 검색의 효율성과 정확성을 제공할 수 있다.