DCT 계수와 에지를 이용한 자막의 강건한 분할
Robust Caption Segmentation Using DCT Coefficients and Edges
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
디지털 비디오 데이터의 효과인 색인과 검색을 해서는 비디오 상 안에 존재하는 자막을 활용하는 것이 매우 요하다. 본 논문에서는 DCT(discrete cosine transform) 계수와 에지를 이용하여 입력 상 안에 포함된 자막 을 강건하게 검출하는 방법을 제안한다. 제안된 방법에서는 먼 입력 상으로부터 자막이 존재할 것이라고 상되 는 후보 자막 역을 수평과 수직 방향의 고주 특징을 이용하여 검출한다. 그런 다음, 후보 자막 역에서 자막을 형성하는 역들만을 에지의 군집화를 통해 추출한다. 그리고 앞에서 추출한 자막 역에서 배경을 제외하고 문자만 을 추출한다. 마지막으로, 추출된 자막 역이 비 자막 역을 포함하지 않고 실제 자막 역만을 포함하고 있는지 를 신경망을 이용하여 최종으로 검증한다. 실험에서는 본 논문에서 제안한 자막 역 검출 알고리즘이 서로 다른 상황에서 자막을 강건하게 추출한다는 것을 여러 가지의 실험을 통해 보여다.
Abstract
Utilizing captions is very important for effective video indexing and retrieval in digital video data. In this paper, we propose a method of robustly detecting captions using DCT(discrete cosine transform) coefficients and edges in input images. The suggested method first detects candidate caption regions by using horizontal and vertical high-frequency DCT coefficients. The method then obtains character regions containing captions from candidate regions through edge clustering. Subsequently, we separate captions from background in the obtained candidate region. Finally, we determine actual caption regions among candidate regions using an artificial neural network. Experimental results show that the proposed method can robustly extract caption regions from different input images.
참고문헌 (17)
- G.-D Kim, "SMART : High Performance Internet Server for HD Video On Demand", The Journal of Korean Institute of Next Generation Computing, 2(2), pp.7~15 (2006)
- 황우민, "Negotiated Seamless Video Streaming Service with GPS and Digital Map", 한국차세대컴퓨팅학회 논문지, 6(3), pp.20~28 (2010), 한국차세대컴퓨팅학회
- D. Márquez, "A model-based iterative method for caption extraction in compressed MPEG video", Lecture Notes in Computer Science, 4816, pp.91~94 (2007)
- X. Huang, "A novel video text extraction approach based on Log-Gabor filters", Proceedings of the International Congress on Image and Signal Processing (CISP), 1, pp.474~478 (2011)
- J.-W. Wu, "A discrete particle swarm optimization algorithm for domain independent linear text segmentation", Proceedings of the IEEE International Conference on granular computing (GRC), pp.519~524 (2010)
- M. R. Lyu, "A comprehensive method for multilingual video text detection, localization, and extraction", IEEE Transactions on Circuits and Systems for Video Technology, 15(2), pp.243~255 (2005)
- X. Qian, "Text detection, localization, and tracking in compressed video", Signal Processing: Image Communication, 22(9), pp.752~768 (2007)
- W. Kim, "A new approach for overlay text detection and extraction from complex video scene", IEEE Transactions on Image Processing, 18(2), pp.401~411 (2009)
- P. P. Roy, "Multi-oriented touching text character segmentation in graphical documents using dynamic programming", Pattern Recognition, 45(5), pp.1972~1983 (2012)
- 아드리안토 테드조쿠수모, "A Two-step Character Binarization Method Using Standard Deviationin Local Area", 한국차세대컴퓨팅학회 논문지, 9(4), pp.24~34 (2013), 한국차세대컴퓨팅학회
- S. A. Angadi, "A robust segmentation technique for line, word and character extraction from Kannada text in low resolution display board images", Proceedings of the International Conference on Signal and Image Processing (ICSIP), pp.42~49 (2014)
- X. Huang, "A method of text segmentation from scanned image with complex background", Proceedings of the International Conference on Management and Service Science (MASS), pp.1~4 (2009)
- D. Q. Zhang, "General and domain-specific techniques for detecting and recognizing superimposed text in video", Proceedings of the International Conference on Image Processing (ICIP), 1, pp.I-593~I-596 (2002)
- K.-J. Ahn, "Determining intensity value of characters and backgrounds on caption", Proceedings of the Conference on Korea Society of Computer and Information, 18(2), pp.125~128 (2010)
- R. C. Gonzalez, "Digital Image Processing, Third Edition" (2007), Prentice Hall
- A.-H. Cho, "The extraction of character from an English name card by using smearing method and contour tracking algorithm", Proceedings of the Korea Intelligent Information System Society Conference, pp.410~413 (2002)
- J. Song, "A robust statistic method for classifying color polarity of video text", Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 3, pp.385~388 (2003)