• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2011년도 제43차 동계학술발표논문집 19권1호
• /
• pp.55-58
• /
• 2011

This paper proposes a method of occluded number recognition by matching interest points. Interest points of input pattern are found via SURF features extracting and matched to interest points of clusters in database following three steps: SURF matching, coordinate matching and SURF matching on coordinate matched points. Then the satisfied interest points are counted to compute matching rate of each cluster. The input pattern will be assigned to cluster having highest matching rate. We have experimented our method to different numerical fonts and got encouraging results.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 추계학술대회 논문집 학회본부 D
• /
• pp.747-749
• /
• 2000

In this paper, we proposed color stereo matching algorithm using dynamic programming. The conventional gray stereo matching algorithms show blur at depth discontinuities and non-existence of matching pixel in occlusion lesions. Also it accompanies matching error by lack of matching information in the untextured region. This paper defines new cost function makes up for the problems happening in conventional gray stereo matching algorithm. New cost function contain the following properties. I) Edge points are corresponded to edge points. ii) Non-edge points are corresponded to non-edge points. iii) In case of exiting the amount of edges, the cost function has some weight in proportion to path distance. Proposed algorithm was applied in various images obtained by parallel camera model. As the result, proposed algorithm showed improved performance in the aspect of matching error and processing in the occlusion region compared to conventional gray stereo matching algorithms.

• 한국측량학회:학술대회논문집
• /
• 한국측량학회 2004년도 춘계학술발표회논문집
• /
• pp.171-176
• /
• 2004

Image matching is fundamental process in photogrammetry and computer vision to identify and to measure corresponding features on the multiple images. Uniqueness of the matching entities and robustness of the algorithm are the key issues that have influence on quality of the matching result. The optimal solution could be obtained by utilizing appropriate matching entities in the first place. In this study, candidate matching points were extracted by interest operator, and an area-based matching method was applied with characteristics of the gray value distribution as the matching entities. The characteristic information is based on the concept of "intrinsic image" (or parameter image). The information was utilized as additional and/or complementary matching entities. Matching on interest points with the characteristic information resulted in high quality of matching because matching windows were created with surrounding pixels of the interest points that contain distinct and unique features. The experiment shows that matching quality and reliability increase by exploiting interest operator, and the characteristic information has potential to be matching entity.

• Journal of Information Processing Systems
• /
• 제15권1호
• /
• pp.22-33
• /
• 2019

This paper introduces an efficient fingerprint matching method based on multiple reference minutiae points. First, we attempt to effectively align two fingerprints by employing multiple reference minutiae points. However, the corresponding minutiae points between two fingerprints are ambiguous since a minutia of one fingerprint can be a match to any minutia of the other fingerprint. Therefore, we introduce a novel method based on linear classification concept to establish minutiae correspondences between two fingerprints. Each minutiae correspondence represents a possible alignment. For each possible alignment, a matching score is computed using minutiae and ridge orientation features and the maximum score is then selected to represent the similarity of the two fingerprints. The proposed method is evaluated using fingerprint databases, FVC2002 and FVC2004. In addition, we compare our approach with two existing methods and find that our approach outperforms them in term of matching accuracy, especially in the case of non-linear distorted fingerprints. Furthermore, the experiments show that our method provides additional advantages in low quality fingerprint images such as inaccurate position, missing minutiae, and spurious extracted minutiae.

• 한국측량학회지
• /
• 제38권2호
• /
• pp.165-172
• /
• 2020

There is growing interest in Vision-based video image matching owing to the constantly developing technology of unmanned-based systems. The purpose of this paper is the development of a fast and effective matching technique for the UAV oblique video image. We first extracted initial matching points using NCC (Normalized Cross-Correlation) algorithm and improved the computational efficiency of NCC algorithm using integral image. Furthermore, we developed a triangulation-based outlier removal algorithm to extract more robust matching points among the initial matching points. In order to evaluate the performance of the propose method, our method was quantitatively compared with existing image matching approaches. Experimental results demonstrated that the proposed method can process 2.57 frames per second for video image matching and is up to 4 times faster than existing methods. The proposed method therefore has a good potential for the various video-based applications that requires image matching as a pre-processing.

• 한국측량학회지
• /
• 제28권4호
• /
• pp.463-469
• /
• 2010

화성의 고도정보를 제공하는 MOLA 센서는 화성전역에 대한 데이터를 제공하지 못하므로, 수치표고모형을 만들기 위해서는 MOC-NA영상을 이용한 영상정합이 수행되어야만 한다. 그러나 특색(feature)이 적고 명암대비가 낮은 화성영상의 특성상, 자동 영상정합은 어려운 실정이다. 본 논문은 MOC-NA 영상에 대하여 영역기준 영상정합에 기반한 반 자동 영상정합의 알고리즘을 다룬다. 공액점을 나타내는 시드(seed)포인트 들이 수동으로 스테레오 영상에 추가되고 이를 바탕으로 특징점들이 자동으로 삽입된다. 각 영상의 특징점들은 서로의 초기 공액점으로 사용되며, 영역기준 영상정합으로 정제된다. 영상정합의 과정 중 정합에 실패한 점들은 초기 공액점의 위치를 정합에 성공한 주변의 여섯 점들을 이용하여 재 계산한 후 정제된다. 타깃영상과 검색영상의 역할을 바꾸어 수행한 영상정합의 질적 평가 결과, 97.3%의 점들이 한 화소 이하의 절대거리를 나타내었다.

• 전자공학회논문지B
• /
• 제29B권8호
• /
• pp.44-53
• /
• 1992

In this paper, a matching method is proposed to improve the correct matching rate in stereo correspondence matching in which the fingerprint of zero-crossing points on the scale-space is used as the robust matching feature. The dynamic programming, which is appropriate for the fingerprint feature, is introduced for correspondence matching. We also improve the matching rate by using the post-processing for correcting mismatched points. In simulation, we apply the proposed algorithm to the synthetic and real images and obtain good matching results.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.1467-1469
• /
• 2003

Image matching is fundamental process to identify conjugate points on the stereo images. However, standard methods or general solutions for matching problem have not been found yet, in spite of long history. Quality of the matching basically depends on uniqueness of the matching entity and robustness of the algorithm. In this study, conjugate points were extracted by implementing interest operator, then area based matching method was applied to the topographical characteristics of the gray value as the matching entities. The matching entities were utilized based on the concept of the intrinsic image.

• 한국측량학회지
• /
• 제14권1호
• /
• pp.9-16
• /
• 1996

본 연구에서는 상호표정과 사진기준점측량을 위한 새로운 매칭방법을 제안하였다. 이 방법은 수작업에 의한 점이사와 관측을 대신하여 디지탈 항공사진에서 연결점과 접합점으로 사용할 수 있는 공액점들을 매칭한다. 제안된 매칭방법에는 다음과 같은 세 가지의 독특한 단계가 포함되어 있다. 첫째 단계는 interest point를 찾고 이들을 매칭후보점으로 사용한다. 둘째 단계는 중복찰영된 서상에서 표준상관계수법에 의해 양방향으로(왼쪽에서 오른쪽으로 오른쪽에서 왼쪽으로) 매칭을 실시한다. 세째 단계는 양매칭방향에서 일치성이 있는 쌍들을 선택한다. 제안된 매칭방법에 기초한 컴퓨터 프로그램을 개발하고, 전면지상기준점측량을 실시한 디지탈 항공사진을 이용하여 제안된 방법을 사용하였을 때의 신뢰성과 위치정확도를 조사하기 위한 시험을 하였다. 시험 결과 제안된 매칭방법은 다른 매칭방법보다 불량하게 매칭된 쌍들을 제거하는 데 더 효과적이었으며 , 따라서 이 방법을 상호표정 및 사진기준점측량에 사용할 경우 매우 매우 신뢰성이 높다.

• 제어로봇시스템학회논문지
• /
• 제17권3호
• /
• pp.248-253
• /
• 2011

This paper proposes the robust matching algorithm for optical images obtained by WSI(White-light Scanning Interferometer) machine. The matching algorithms are divided by two part according to the matching points: algorithm whether the matching points between two images exist or not. Also, after matching the images, we propose the algorithm to smooth the matched image. Finally, we show the effectiveness and feasibility of the proposed method through some experiments.