파노라마 영상을 이용한 특징점 탐지 알고리즘의 성능평가
Performance tests of feature detection algorithms using panoramic images
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
지리정보 분야의 영상 데이터 활용성 증대에 따라 길거리의 시각적 정보를 제공할 수 있는 파노라마 영상에대한 관심이 증가하고 있다. 여러 분야에서 길거리 영상을 참조 자료로 활용해 이동 객체에 대한 위치 정보 도출 연구가 이루어지고 있지만, 널리 활용되는 특징점 탐지 알고리즘에 대한 정량적인 성능평가는 부족한 상황이다. 이에본 연구에서는 SIFT(Scale Invariant Feature Transform), SURF(Speed Up Robust Features), ORB(Oriented FAST and Rotated Brief) 알고리즘을 대상으로, 파노라마 영상과 평면 영상의 투영법 차이, 카메라 회전에 따른특징점 정확도 변화, 주변 건물 환경 변화에 따른 특징점 정확도 평가를 수행하였고, 각 기법의 장단점과 활용을 위한 방향성을 제시하였다.
Abstract
With the increasing utilization of image database in GIS field, panoramic images which can provide visual information of streets draw attention nowadays. Although many studies in various areas have made efforts to derive location information about moving objects, quantitative analysis of the performance of commonly used feature detection algorithms is actually insufficient. Thus in this research, we carried out performance tests of the algorithms such as SIFT(Scale Invariant Feature Transform), SURF(Speed Up Robust Features), and ORB(Oriented FAST and Rotated Brief) in terms of the different projections between panoramic and planar images, accuracy variations depending on camera rotation, and the changes of street environment. The advantages and disadvantages of the methods were analysed with the implications for their utilization.
참고문헌 (15)
- Agarwal, P, "Metric localization using google street view", Proceedings of 2015 IEEE, 2015, 3111~3118
- Chen, D., "City-scale landmark identification on mobile devices", proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.737~744 (2011)
- "Google Maps API - Street View Service" (2017)[URL]
- Ji, S, "Particle filtering methods for georeferencing panoramic image sequence in complex urban scenes", ISPRS Journal of Photogrammetry and Remote Sensing, 105, pp.1~12 (2015)
- Karami, E, "Image matching using SIFT, SURF, BRIEF and ORB: Performance comparison for distorted images", Proceedings of 2015 Newfoundland Electrical and Computer Engineering Conference (NECEC) (2015)
- Lowe, D.G, "Distinctive image features from scale-invariant keypoints", International Journal of Computer Vision, 60(2), pp.91~110 (2004)
- Muja, M, "Fast approximate nearest neighbors with automatic algorithm configuration", Proceedings of 8th International Conference on Computer Vision, Theory and Applications(VISAPP), pp.331~340 (2009)
- Panchal, P.M, "A comparison of SIFT and SURF", International Journal of Innovative Research in Computer and Communication Engineering, 1(2), pp.323~327 (2013)
- Pepperell, E, "Routed roads: Probabilistic vision-based place recognition for changing conditions, split streets and varied viewpoints", The International Journal of Robotics Research, 35(9), pp.1057~1179 (2016)
- Rublee, E, "ORB: An efficient alternative to SIFT or SURF", Proceedings of 2011 IEEE International Conference on Computer Vision(ICCV), pp.2564~2571 (2011)
- Szeliski, R, "Computer Vision: Algorithms and Applications" (2010), Springer Science & Business Media
- Szeliski, R, "Creating full view panoramic image mosaics and environment maps", Proceedings of the 24th Annual Conference on Computer Graphics and Interactive Techniques, pp.251~258 (1997)
- Taneja, A, "Never get lost again: Vision based navigation using streetview images", Proceedings of Asian Conference on Computer Vision, pp.99~114 (2014)
- Wu, J, "A comparative study of SIFT and its variants", Measurement Science Review, 13(3), pp.122~131 (2013)
- Yu, L, "Monocular urban localization using street view", Proceedings of 14th International Conference on Control, Automation, Robotics and Vision (ICARCV), pp.1~6 (2016)