• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1505-1514
• /
• 2022

Image matching is a crucial preprocessing step for effective utilization of multi-temporal and multi-sensor very high resolution (VHR) satellite images. Deep learning (DL) method which is attracting widespread interest has proven to be an efficient approach to measure the similarity between image pairs in quick and accurate manner by extracting complex and detailed features from satellite images. However, Image matching of VHR satellite images remains challenging due to limitations of DL models in which the results are depending on the quantity and quality of training dataset, as well as the difficulty of creating training dataset with VHR satellite images. Therefore, this study examines the feasibility of DL-based method in matching pair extraction which is the most time-consuming process during image registration. This paper also aims to analyze factors that affect the accuracy based on the configuration of training dataset, when developing training dataset from existing multi-sensor VHR image database with bias for DL-based image matching. For this purpose, the generated training dataset were composed of correct matching pairs and incorrect matching pairs by assigning true and false labels to image pairs extracted using a grid-based Scale Invariant Feature Transform (SIFT) algorithm for a total of 12 multi-temporal and multi-sensor VHR images. The Siamese convolutional neural network (SCNN), proposed for matching pair extraction on constructed training dataset, proceeds with model learning and measures similarities by passing two images in parallel to the two identical convolutional neural network structures. The results from this study confirm that data acquired from VHR satellite image database can be used as DL training dataset and indicate the potential to improve efficiency of the matching process by appropriate configuration of multi-sensor images. DL-based image matching techniques using multi-sensor VHR satellite images are expected to replace existing manual-based feature extraction methods based on its stable performance, thus further develop into an integrated DL-based image registration framework.

• 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.

• The Journal of Art Theory & Practice
• /
• no.15
• /
• pp.31-49
• /
• 2013

This paper delves into the recent 'paintings' of African-American artist David Hammons, which combine rubbish-like plastic wraps with the abstract-expressionist style paintings. In straddling between rubbish and art object, his works tend to blur the boundary drawn between two opposite categories in value, art and garbage, provoking the sophisticated taste of Upper-East-side white community in Manhattan, New York. Choosing the venue of his exhibition at a commercial gallery, Hammons's creative efforts is also a critique of what can be seen as the dominance of abstract expressionism and white elitism in American art history. The artist is known for his use of unconventional materials in art making such as black hair, barbecue bones, and elephant droppings, ones that are often associated with African-American experiences in all different levels. Since his debut in the art scene in the 1970s, Hammons has pursued the view of art-making as a medium for provoking contentious issues of racial relations in the States. On the other hand, the reception of Hammons's work as African-American art can be potentially quite limiting, overlooking as it does multi-faceted meanings of his art practice. His unconventional approach to art often took him outside art galleries and museums, where he was seen using a variety of common materials for site-specific installations and performances. Staged in different parts of Manhattan, these acts of art making traverse seemingly opposite communities and cultures, often blurring their boundaries. Hammons's artistic practice can label him what Abdul Jan Mohamed calls "specular border intellectual", revealing as it does the symbiosis of binary oppositions that is basic to the experience of communnal living.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.11 no.2
• /
• pp.127-133
• /
• 2011

This study proposed the PHB mapping model for combining services with DiffServ in the MPLS network capable of high-speed switching, and the effective differential bandwidth service model in the MPLS network on the basis of the proposed mapping model. The result of simulation on the service model using PRI showed that the transmission rate was achieved as much as the assigned bandwidth regardless of the increase of traffic as far as the EF PHB did not exceed the assigned bandwidth at the peak rate when the overall traffic increased.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.41 no.5
• /
• pp.133-142
• /
• 2004

Recently, the growth of Internet and a variety of multimedia services through Internet increasingly demands high-speed packet transmission, the new routing function, and QoS guarantee on conventional routers. Thus, a new switching mechanical called the MPLS(Multi-Protocol Label Switching), was proposed by IETF(Internet Engineering Task Force) as a solution to meet these demands. In addition the deployment of MPLS network supporting Differentiated Services is required. In this paper, we propose the architecture of the MPLS switch supporting Differentiated Services in the MPLS-based network. The traffic conditioner consists of a classifier, a meter, and a marker. The VOQ-PHB module, which combines input Queue with each PHB queue, is implemented to utilize the resources efficiently. It employs the Priority-iSLIP scheduling algorithm to support high-speed switching. We have designed and verified the new and fast hardware architecture of VOQ-PHB and the traffic conditioner for QoS and high-speed switching using NS-2 simulator. In addition, the proposed architecture is modeled in VHDL, synthesized and verified by the VSS analyzer from SYNOPSYS. Finally, to justify the validity of the hardware architecture, the proposed architecture is placed and routed using Apollo tool.

• The Journal of the Korea Contents Association
• /
• v.11 no.7
• /
• pp.70-79
• /
• 2011

The multi-path routing schemes that assigns labels to sensor nodes for the reliability of data transmission and the accuracy of an aggregation query over the sensor networks where data transfer is prone to defect have been proposed. However, the existing schemes have high costs for reassigning labels to nodes when the network topology is changed. In this paper, we propose a novel routing method that avoids duplicated data and reduces the update cost of a sensor node. In order to show the superiority of the proposed scheme, we compare it with the existing scheme through the various experiments. Our experimental results show that our proposed method reduces about 95% the amount of the transmitted data for restoration to node failure and about 220% the amount of the transmitted data for query processing over the existing method on average.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1846-1850
• /
• 2005

This paper presents a constrained-based routing (CBR) algorithm called, Dynamic Possible Path per Link (D-PPL) routing algorithm, for MultiProtocol Label Switching (MPLS) networks. In MPLS on-line routing, future traffics are unknown and network resource is limited. Therefore many routing algorithms such as Minimum Hop Algorithm (MHA), Widest Shortest Path (WSP), Dynamic Link Weight (DLW), Minimum Interference Routing Algorithm (MIRA), Profiled-Based Routing (PBR), Possible Path per Link (PPL) and Residual bandwidth integrated - Possible Path per Link (R-PPL) are proposed in order to improve network throughput and reduce rejection probability. MIRA is the first algorithm that introduces interference level avoidance between source-destination node pairs by integrating topology information or address of source-destination node pairs into the routing calculation. From its results, MIRA improves lower rejection probability performance. Nevertheless, MIRA suffer from its high routing complexity which could be considered as NP-Complete problem. In PBR, complexity of on-line routing is reduced comparing to those of MIRA, because link weights are off-line calculated by statistical profile of history traffics. However, because of dynamic of traffic nature, PBR maybe unsuitable for MPLS on-line routing. Also, both PPL and R-PPL routing algorithm we formerly proposed, are algorithms that achieve reduction of interference level among source-destination node pairs, rejection probability and routing complexity. Again, those previously proposed algorithms do not take into account the dynamic nature of traffic load. In fact, future traffics are unknown, but, amount of previous traffic over link can be measured. Therefore, this is the motivation of our proposed algorithm, the D-PPL. The D-PPL algorithm is improved based on the R-PPL routing algorithm by integrating traffic-per-link parameters. The parameters are periodically updated and are dynamically changed depended on current incoming traffic. The D-PPL tries to reserve residual bandwidth to service future request by avoid routing through those high traffic-per-link parameters. We have developed extensive MATLAB simulator to evaluate performance of the D-PPL. From simulation results, the D-PPL improves performance of MPLS on-line routing in terms of rejection probability and total throughput.

• Journal of Korea Society of Industrial Information Systems
• /
• v.19 no.2
• /
• pp.85-91
• /
• 2014

The objective of the research is to automatically recognize railway objects from MLS data in which 9 key objects including terrain, track, bed, vegetation, platform, barrier, posts, attachments, powerlines are targeted. The proposed method can be divided into two main sub-steps. First, multi-scale contextual features are extracted to take the advantage of characterizing objects of interest from different geometric levels such as point, line, volumetric and vertical profile. Second, by considering contextual interactions amongst object labels, a contextual classifier is utilized to make a prediction with local coherence. In here, the Conditional Random Field (CRF) is used to incorporate the object context. By maximizing the object label agreement in the local neighborhood, CRF model could compensate the local inconsistency prediction resulting from other local classifiers. The performance of proposed method was evaluated based on the analysis of commission and omission error and shows promising results for the practical use.

• Journal of Communications and Networks
• /
• v.9 no.4
• /
• pp.511-523
• /
• 2007

In broadband convergence network(BcN), heterogeneous broadband wired & wireless subnetworks and various terminal equipments will be interconnected. In order to provide end-to-end realtime multimedia services on such heterogeneous networking environment, as a result, two major problems should be resolved: i) Multimedia session establishment & negotiation that adjusts the differences in the capability of multimedia data processing at the end terminal nodes, ii) quality of service(QoS)-guaranteed connection establishment or resource reservation with connection admission control(CAC) in each heterogeneous subnetworks along the path. The session layer signaling(e.g., SIP/SDP) should be extended for QoS negotiation, and must be tightly cooperating with network layer signaling or resource reservation with CAC function. In this paper we propose a session and connection management architecture for the QoS-guaranteed realtime multimedia service provisioning on BcN, with Q-SIP/SDP, resource reservation protocol with traffic engineering extension, and CAC functions. The detailed interaction scenario and related algorithms for QoS-guaranteed realtime multimedia session, resource reservation and connection establishment are explained and analyzed. From the experimental implementation of the proposed scheme on a small scale BcN testbed, we verified that the proposed architecture is applicable for the realtime multimedia service provisioning. We analyze the network performance and QoS parameters in detail.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.42 no.9 s.339
• /
• pp.71-84
• /
• 2005

A 'VPN over Intemet' has the benefit of being cost-effective and flexible. However, it has difficulties providing sufficient QoS and adequate transmission capacity for high bandwidth service. Given the increasing demand for high bandwidth Internet and QoS assurances in a 'VPN over Internet', IP/GMPLS based on a control plane combined with a high bandwidth, DWDM optical network is seen as a very favorable anroach for realizing the future 'OVPN over IP/GMPLS over DWDM'. Within this architecture, providing QoS guaranteed multimedia services with a differentiated QoS guaranteed protocol framework with QoS recovery is one of the key issues to implement. Therefore, in this paper, we suggest the entire interoperability procedure of control protocols based on differentiated optical QoS service (DOQOS) classes for providing end-to-end QoS in an 'OVPN over IP/GMPLS over DWDM' architecture.