• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.33 no.1
• /
• pp.63-70
• /
• 2015

An Image fusion, or Pansharpening is a methodology of increasing the spatial resolution of image with low-spatial resolution using high-spatial resolution images. In this paper, we have performed an image fusion of hyperspectral imagery by using panchromatic image with high-spatial resolution, multispectral and hyperspectral images with low-spatial resolution, which had been acquired by ALI and Hyperion of EO-1 satellite sensors. The study has been mainly focused on evaluating performance of fusion process following to the image fusion methodology of the block association, which had applied to ALI and Hyperion dataset by considering spectral characteristics between multispectral and hyperspectral images. The results from experiments have been identified that the proposed algorithm efficiently improved the spatial resolution and minimized spectral distortion comparing with results from a fusion of the only panchromatic and hyperspectral images and the existing block-based fusion method. Through the study in a proposed algorithm, we could concluded in that those applications of airborne hyperspectral sensors and various hyperspectral satellite sensors will be launched at future by enlarge its usages.

• Convergence Security Journal
• /
• v.20 no.3
• /
• pp.71-78
• /
• 2020

Bitcoin-based blockchain technology provides an infrastructure that enables anonymous smart contracts, low-cost remittances, and online payments. However, the block-chain technology that implements the bitcoin has scalability constraints in tradeoffs between throughput and latency. To solve these problems, the Byzantine fault tolerant block-chain technique has been proposed. This technique improves throughput without increasing latency by selecting a leader and constructing many microblocks that do not contain proofs of work within the existing block by the leader. However, this technique may be less secure than existing techniques in selecting the reader.

• Annual Conference of KIPS
• /
• 2022.11a
• /
• pp.69-71
• /
• 2022

비트코인은 현재 가장 많이 사용하는 전자 화폐 거래 시스템으로, Proof-of-Work를 합의 알고리즘으로 사용한다. 이때 비트코인에 대한 다양한 공격기법들이 연구되었으며 대표적으로 Selfish Mining과 블록 보류 공격이 존재한다. 본 논문에서는 블록 보류 공격에 대한 방지 기법의 동향에 대해서 조사하였으며, 대표적으로 딥러닝, 평판 기반 블록체인, 게임 전략을 통한 방지 기법이 존재하였다.

• Annual Conference of KIPS
• /
• 2022.11a
• /
• pp.261-262
• /
• 2022

최근 몇 년간 블록체인 기술은 빠르게 발전하여 많은 산업에 영향을 끼쳤다. 하지만 동시에 비트코인, 이더리움 등 블록체인 네트워크 내에서 많은 금융 범죄들이 발생하고 있다. 최근에는 이와 같은 비정상적인 활동을 탐지하기 위해 많은 연구가 진행되고 있다. 본 논문에서는 딥러닝을 기반으로 블록체인 네트워크 내의 악성 노드를 탐지하는 기법에 대해 살펴본다. 대부분의 연구가 높은 정확도를 달성하였으며, 악성 노드뿐만 아니라 악성 노드가 수행한 사기 트랜잭션을 탐지하는 연구도 진행되고 있었다.

• Journal of the Korea Convergence Society
• /
• v.9 no.11
• /
• pp.75-81
• /
• 2018

Anonymity and privacy issues are becoming important as all transactions in the blockchain are open to users. Public blockchains appear to guarantee anonymity by using public-key addresses on behalf of users, but they can weaken anonymity by tracking with various analytic techniques based on transaction graph. In this paper, we propose a scheme to protect anonymity and privacy by converging various security techniques such as k-anonymity, mixing, blind signature, multi-phase processing, random selection, and zero-knowledge proof techniques with incentive mechanism and contributor participation. Through performance analysis, our proposed scheme shows that it is difficult to invade privacy and anonymity through collusion attacks if the number of contributors is larger than that of conspirators.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.06a
• /
• pp.32-34
• /
• 2019

최근 딥러닝 기반의 다양한 물체 검출 알고리즘이 제안되어 높은 성능을 보이고 있다. 본 논문은 이러한 딥러닝 기반 물체 검출의 성능을 향상시키기 위해 입력영상에서 추출된 특징 지도를 강화하는 비지역적 특징 융합과, 이를 이용한 물체 검출 기법을 제안한다. 제안 기법은 입력영상에서 CNN 을 통해 추출한 특징 지도를 비지역적 특징 강화 블록을 통해 강화한다. 해당 블록 내에서 입력된 특징 지도는 먼저 여러 리셉티브 필드를 갖는 특징 지도로 분기된다. 그리고 분기된 특징 지도들은 비지역적 특징 융합 모듈에 의해 융합되어 강화된다. 이러한 과정을 통해 강화된 특징 지도는 비지역적 문맥 정보가 강화된 특성을 가지며, 해당 특징 지도를 이용하여 최종적으로 물체 검출을 수행한다. Pascal VOC 공인 데이터세트를 통한 실험 결과, 제안 기법은 기존 비교 기법 대비 향상된 검출 성능을 보인다.

• Journal of Convergence for Information Technology
• /
• v.8 no.3
• /
• pp.85-90
• /
• 2018

Blockchain technology provides distributed trust structure; with this, we can implement a system that cannot be forged and make Smart Contract possible. With blockchain technology emerging as next generation security technology, there have been studies on authentication and security services that ensure integrity. Although Internet-based services have been going with user authentication with password, the information can be stolen through a client and a network and the server is exposed to hacking. For the reason, we suggest blockchain technology and OTP based authentication mechanism to ensure integrity. In particular, the Two-Factor Authentication is able to ensure secure authentication by combining OTP authentication and biometric authentication without using password. As the suggested authentication applies multiple hash functions and generates transactions to be placed in blocks in order for biometric information not to be identified, it is protected from server attacks by being separate from the server.

• Convergence Security Journal
• /
• v.19 no.3
• /
• pp.13-19
• /
• 2019

MANET has many security vulnerabilities because it consists of only mobile nodes using wireless. In particular, it is a very important factor determining network performance that excludes the participation of malicious nodes through accurate reliability measurements and authentication of nodes participating in the network. In this paper, we proposed a technique applied with blockchain technology in order to prevent forgery of authentication information for nodes participating in the network. And, an area-based hierarchical structure was applied to increase the efficiency of authentication for nodes and apply the optimal technique of block generation and exchange protocol. In addition, four data payloads were added to the block header in order to add authentication information for nodes in block. To improve the reliability by applying the blockchain technique to the hop-by-hop data transfer method between mobile nodes, blockchain exchange protocol through transaction creation, block packaging and verification processes were implemented. We performed the comparative experiment with the existing methods to evaluate the performance of the proposed method and confirmed the excellent performance by the experiment results.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.37-40
• /
• 2020

최근 블록체인이 활성화되면서 블록체인 시장 및 블록체인 모델의 활용도가 늘어나고 있다. 그중 하이퍼레저 패브릭은 프라이빗 블록체인의 대표적인 플랫폼이다. 하이퍼레저 패브릭에서 클라이언트는 트랜잭션을 보증 피어에게 전송할 때 사전에 정해진 보증 피어에게만 전송한다. 이는 트랜잭션의 실행 비용 및 보증 피어의 성능을 고려하지 않아 보증 시간을 증가시키는 문제가 발생한다. 본 논문은 이 문제를 해결하기 위해 트랜잭션의 실행 비용에 따라 효율적으로 보증 피어를 라우팅하는 기법을 제안한다.

• Journal of Korea Multimedia Society
• /
• v.16 no.11
• /
• pp.1250-1260
• /
• 2013

The contrast enhancement techniques based on Laplacian pyramid image fusion have a benefit that they can faithfully describe the image information because they combine the multiple resource images by selecting the desired pixel in each image. However, they also have some problem that the output image may contain noise, because the methods evaluate the visual information on the basis of each pixel. In this paper, an improved contrast enhancement method, which effectively suppresses the noise, using image fusion is proposed. The proposed method combines the resource images by making Laplacian pyramids generated from weight maps, which are produced by measuring the difference between the block-based local well exposedness and local homogeneity for each resource image. We showed the proposed method could produce less noisy images compared to the conventional techniques in the test for various images.