융합정보논문지 (Journal of Convergence for Information Technology)

중소기업융합학회 (Convergence Society for SMB)
권호 표지
DOI QR코드

DOI QR Code

방사 투영 프로파일을 이용한 회전각 추정 방법

Rotation Angle Estimation Method using Radial Projection Profile

  • 최민석 (삼육대학교 지능정보융합학부)
  • Choi, Minseok (Division of AI Informatics, Sahmyook University)
  • 투고 : 2021.08.24
  • 심사 : 2021.10.20
  • 발행 : 2021.10.28
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 영상 인식 환경에서 영상 정렬에 필요한 회전각 추정 방법 중 낮은 사양의 임베디드 기반 환경에 적용 가능한 방법을 제안하고 기존의 복소 모멘트를 이용하는 방법과 비교하였다. 제안된 방법은 영상을 극좌표로 변환한 후 거리축 방향으로 투영된 1차원 프로파일의 유사도 매칭을 통하여 회전각을 추정한다. 추가로 연산을 더 단순화시킨 투영 프로파일의 벡터합을 이용하는 방법을 선택할 수도 있다. 이진 패턴 영상과 흑백 명암영상을 대상으로 진행한 실험을 통하여 제안된 방법의 추정 오차가 기존의 복소 모멘트를 이용하는 방법과 큰 차이가 없으며 보다 적은 연산과 낮은 시스템 자원이 요구됨을 보였다. 추후 확장을 위하여 흑백 명암영상에서 회전 중심을 일치시키는 방법에 관한 연구가 필요할 것이다.

In this paper, we studied the rotation angle estimation methods required for image alignment in an image recognition environment. In particular, a rotation angle estimation method applicable to a low specification embedded-based environment was proposed and compared with the existing method using complex moment. The proposed method estimates the rotation angle through similarity mathcing of the 1D projection profile along the radial axis after converting an image into polar coordinates. In addition, it is also possible to select a method of using vector sum of the projection profile, which more simplifies the calculation. Through experiments conducted on binary pattern images and gray-scale images, it was shown that the estimation error of the proposed method is not significantly different from that of complex moment-based method and requires less computation and system resources. For future expansion, a study on how to match the rotation center in gray-scale images will be needed.

키워드

참고문헌

  1. L. G. Brown. (1992). A survey of image registration techniques. ACM Computing Surveys, 24(4), 325-376. DOI : 10.1145/146370.146374
  2. B. Zitova & J. Flusser. (2003). Image registration methods: a servey. Image and Vision Computing, 21(11), 977-100. DOI : 10.1016/S0262-8856(03)00137-9
  3. M. Goljan. (2018, Jan). Blind Detection of Image Rotation and Angle Estimation. Electronic Imaging, Media Watermarking, Security, and Forensics 2018 (pp. 158-1-158-10). Burlingame : Society for Imaging Science and Technology.
  4. E. Dicke, A. Wacher & W. Nahm. (2017). Estimation of the interpolation error of a three-step rotation algorithm using recorded images with rotated test pattern as ground truth. Current Directions in Biomedical Engineering, 3(2), 555-558. DOI : 10.1515/cdbme-2017-0186
  5. C. G. Cao & Q. Quyang. (2019). 2D Rotation-Angle Measurement Utilizing Least Iterative Region Segmenttion. SENSORS, 19(7), 1634. DOI : 10.3390/s19071634
  6. R. J. Prokop & A. P. Reeves. (1992). A survey of moment-based techniques for unoccluded object representation and recognition. Graphical Models and Image Processing, 54(5), 438-460. DOI : 10.1016/1049-9652(92)90027-U
  7. W. H. Tsai & S. L. Chou. (1991). Detection of generalized principal axes in rotationally symmetric shapes. Pattern Recognition, 24(2), 95-104. DOI : 10.1016/0031-3203(91)90080-O
  8. S. Khotanzad & Y. H. Hong. (1990). Invariant image recognition by Zernike moments. IEEE Transactions on Pattern Analysis and Machine Intelligence, 12(5), 489-498. DOI : 10.1109/34.55109
  9. W. Y. Kim & Y. S. Kim. (1999). Robust Rotation Angle Estimator. IEEE Transactions on Pattern Analysis and Machine Intelligence, 21(8), 768-773. DOI : 10.1109/34.784290
  10. J. M. Lee & W. Y. Kim. (2012). A Comparative Study of Rotation Angle Estimation Methods Based on Complex Moments. IEICE Transactions on Information and Systems, E95.D(1), 1485-1493. DOI : 10.1587/transinf.E95.D.1485
  11. J. Revaud, G. Lavoue & A. Baskurt. (2008). Improving Zernike Moments Comparison for Optimal Similarity and Rotation Angle Retrieval, IEEE Transactions on Pattern Analysis and Machine Intelligence, 31(4), 627-633. DOI : 10.1109/TPAMI.2008.115
  12. S. Nagashima, K. Ito, H. Ishii & K. Kobayashi. (2007, August). A high-accuracy rotation estimation algorithm based on 1D phase-only correlation. ICIAR 2007, Lecture Notes in Computer Science, 4633, 210-221. DOI : 10.1007/978-3-540-74260-9_19