• Spatial Information Research
• /
• v.19 no.2
• /
• pp.47-56
• /
• 2011

Typical tree-based spatial index structures are divided into a data-partitioning index structure such as R-Tree and a space-partitioning index structure such as KD-Tree. In recent years, researches on hybrid index structures combining advantages of these index structures have been performed extensively. However, because the split boundary extension of the node to which a new spatial object is inserted may extend split boundaries of other neighbor nodes in existing researches, overlaps between nodes are increased and the query processing cost is raised. In this paper, we propose a hybrid index structure, called SQR-Tree that can support efficient processing of spatial queries to solve these problems. SQR-Tree is a combination of SQ-Tree(Spatial Quad- Tree) which is an extended Quad-Tree to process non-size spatial objects and R-Tree which actually stores spatial objects associated with each leaf node of SQ-Tree. Because each SQR-Tree node has an MBR containing sub-nodes, the split boundary of a node will be extended independently and overlaps between nodes can be reduced. In addition, a spatial object is inserted into R-Tree in each split data space and SQ-Tree is used to identify each split data space. Since only R-Trees of SQR-Tree in the query area are accessed to process a spatial query, query processing cost can be reduced. Finally, we proved superiority of SQR-Tree through experiments.

• Journal of Broadcast Engineering
• /
• v.24 no.3
• /
• pp.495-505
• /
• 2019

Typical algorithms of deep learning include CNN(Convolutional Neural Networks), which are mainly used for image recognition, and RNN(Recurrent Neural Networks), which are used mainly for speech recognition and natural language processing. Among them, CNN is able to learn from filters that generate feature maps with algorithms that automatically learn features from data, making it mainstream with excellent performance in image recognition. Since then, various algorithms such as R-CNN and others have appeared in object detection to improve performance of CNN, and algorithms such as YOLO(You Only Look Once) and SSD(Single Shot Multi-box Detector) have been proposed recently. However, since these deep learning-based detection algorithms determine the success of the detection in the still images, stable object tracking and detection in the video requires separate tracking capabilities. Therefore, this paper proposes a method of combining Kalman filters into deep learning-based detection networks for improved object tracking and detection performance in the video. The detection network used YOLO v2, which is capable of real-time processing, and the proposed method resulted in 7.7% IoU performance improvement over the existing YOLO v2 network and 20 fps processing speed in FHD images.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.38 no.6
• /
• pp.623-633
• /
• 2020

As the time spent in indoor space has increased, the demand for services targeting indoor spaces also continues to increase. To provide indoor spatial information services, the construction of indoor spatial information should be done first. In the study, a method of generation IndoorGML, which is the international standard data format for Indoor space, from existing BIM data. The characteristics of IFC objects were investigated, and objects that need to be converted to IndoorGML were selected and classified into objects that restrict the expression of Indoor space and internal passages. Using the proposed method, a part of data set provided by the BIMserver github and the IFC model of the 21st Century Building in University of Seoul were used to perform experiments to generate PrimalSpaceFeatures of IndoorGML. As a result of the experiments, the geometric information of IFC objects was represented completely as IndoorGML, and it was shown that NavigableBoundary, one of major features of PrimalSpaceFeatures in IndoorGML, was accurately generated. In the future, the proposed method will improve to generate various types of objects such as IfcStair, and additional method for automatically generating MultiLayeredGraph of IndoorGML using PrimalSpaceFeatures should be developed to be sure of completeness of IndoorGML.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.1
• /
• pp.41-49
• /
• 2023

A convolutional neural network (CNN) is a representative algorithm for implementing artificial neural networks. CNNs have improved on the issues of rapid increase in calculation amount and low object classification rates, which are associated with a conventional multi-layered fully-connected neural network (FNN). However, because of the rapid development of IT devices, the maximum resolution of images captured by current smartphone and tablet cameras has reached 108 million pixels (MP). Specifically, a traditional CNN algorithm requires a significant cost and time to learn and process simple, high-resolution images. Therefore, this study proposes an improved CNN algorithm for implementing an object classification learning model for simple, high-resolution images. The proposed method alters the adjacency matrix value of the pooling layer's max pooling operation for the CNN algorithm to reduce the high-resolution image learning model's creation time. This study implemented a learning model capable of processing 4, 8, and 12 MP high-resolution images for each altered matrix value. The performance evaluation result showed that the creation time of the learning model implemented with the proposed algorithm decreased by 36.26% for 12 MP images. Compared to the conventional model, the proposed learning model's object recognition accuracy and loss rate were less than 1%, which is within the acceptable error range. Practical verification is necessary through future studies by implementing a learning model with more varied image types and a larger amount of image data than those used in this study.

• Korean Journal of Remote Sensing
• /
• v.39 no.2
• /
• pp.157-168
• /
• 2023

Image segmentation and supervised classification techniques are widely used to monitor the ground surface using high-resolution remote sensing images. In order to classify various objects, a process of defining a class corresponding to each object and selecting samples belonging to each class is required. Existing methods for extracting class samples should select a sufficient number of samples having similar intensity characteristics for each class. This process depends on the user's visual identification and takes a lot of time. Representative samples of the class extracted are likely to vary depending on the user, and as a result, the classification performance is greatly affected by the class sample extraction result. In this study, we propose an image classification technique that minimizes user intervention when extracting class samples by applying the histogram back-projection technique and has consistent intensity characteristics of samples belonging to classes. The proposed classification technique using histogram back-projection showed improved classification accuracy in both the experiment using hue subchannels of the hue saturation value transformed image from Compact Advanced Satellite 500-1 imagery and the experiment using the original image compared to the technique that did not use histogram back-projection.

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06b
• /
• pp.395-397
• /
• 2011

RT-eCos3.0은 대표적 분산 실시간 객체 모델인 TMO(Time-triggered Message-triggered Object)의 실행을 제공하기 위하여 공개소스 eCos3.0 기반으로 개발된 초경량 경성 실시간 임베디드 운영체제이다. RT-eCos3.0에서는 그간 스레드의 최장 수행 시간 입력이 필요 없는 EDF 및 FIFO 스케줄러를 지원하여 왔다. 본 논문에서는 TMO의 시간 구동 스레드와 메시지 구동 스레드의 스레드 등록 시 최장 수행 시간을 입력 받아 이를 기반으로 마감시간까지의 수행시간 대비 잔여시간을 이용하는 LLF (Least Laxity First) 스케줄러를 클럭 인터럽트 핸들러 내에 구현하고 각 스레드로 하여금 스케줄링 정책을 선택할 수 있도록 구현하였다.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.547-551
• /
• 2009

In recent years, a film restoration has gained increasing attention by many researchers, to emergence of variety multimedia and to importance of video preservation. Blotch is the most frequent degradation in old film. This paper presents a blotch detection method using color and shape feature. The proposed method is two major modules: a SROD detector using impulsive feature and NN-based detector using shape feature. To assess the validity of the proposed method, the experiments have been performed on several old films.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.642-644
• /
• 2003

본 연구에서는 LAN 기반 환경에서 반물입 혹은 몰입형 가상현실 장비의 대표적인 형태인 수평형, 수직형, 정방형 시스템들을 연동하여 복합 공유가상환경을 구축하였다. 또한 간단한 인터랙션을 포함하는 어플리케이션을 개발하여 공유가상환경 내에 위치한 사용자가 서로 다른 형태의 인터랙션 장비를 이용하여 가상객체와 실시간 상호작용 할 수 있도록 하고, 그 결과를 가상환경 내에 있는 사용자가 공유할 수 있도록 하였다. 기존의 연구가 동기종간의 공유가상환경을 구축하거나 이기종간의 단방향 인터랙션을 지원하는데 비해 본 논문에서는 다기종간의 공유가상환경 구축 및 실시간 상호인터랙션 방법을 제안하고 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10c
• /
• pp.56-61
• /
• 2006

최근 공간 네트워크 데이터베이스를 위한 질의처리 알고리즘에 관한 연구가 많이 진행되어 왔다. 그러나 현재 좌표-기반 질의에 대한 연구는 활발히 진행중인 반면, 경로-기반 질의에 대한 연구는 매우 미흡한 실정이다. 공간 네트워크 데이터베이스에서는 이동객체가 공간 네트워크상에서만 이동하기 때문에 경로-기반 질의의 유용성이 매우 증대되므로, 경로-기반 질의에 대한 효율적인 질의처리 알고리즘 연구가 필수적이다. 따라서 본 논문에서는 경로-기반 질의의 대표적인 방법인 In-Route Nearest Neighbor 질의처리 알고리즘을 분석하여 기존 연구에서 고려하지 않은 시간 및 공간제약을 고려한 경로-기반 질의처리 알고리즘을 설계한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10b
• /
• pp.1335-1338
• /
• 2001

다자간 그룹통신은 화상회의 등에서 매우 유용한 기술이다. 그러나, 개발의 어려움으로 인해 사용되지 못하고 있다. 현재는 예전과 달리, 그룹통신에 적응될 수 있는 많은 기술이 나왔는데 그 대표적인 것이 신뢰성을 제공하는 멀티캐스트 기술이다. 신뢰적 멀티캐스트 기술은 일-대-다 또는 다-대-다 전송을 지원하는 IP 멀티캐스트에 신뢰성을 제공하는 기술이다. 본 논문에서는 신뢰적인 멀티캐스트 전송기술을 이용한 프레임워크 형태의 API를 객체 지향적인 기법으로 설계한다.