Journal of IKEEE (전기전자학회논문지)

Institute of Korean Electrical and Electronics Engineers (한국전기전자학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on Frame Interpolation and Nonlinear Moving Vector Estimation Using GRNN

GRNN 알고리즘을 이용한 비선형적 움직임 벡터 추정 및 프레임 보간연구

  • Lee, Seung-Joo (Dept. of Electronics Engineering, Kookmin University) ;
  • Bang, Min-Suk (Dept. of Electronics Engineering, Kookmin University) ;
  • Yun, Kee-Bang (Dept. of Electronic Engineering, Incheon National University) ;
  • Kim, Ki-Doo (Dept. of Electronics Engineering, Kookmin University)
  • Received : 2013.11.05
  • Accepted : 2013.11.25
  • Published : 2013.12.30
Download PDF
⟨ Previous Next ⟩

Abstract

Under nonlinear characteristics of frames, we propose the frame interpolation using GRNN to enhance the visual picture quality. By full search with block size of 128x128~1x1 to reduce blocky artifact and image overlay, we select the frame having block of minimum error and re-estimate the nonlinear moving vector using GRNN. We compare our scheme with forward(backward) motion compensation, bidirectional motion compensation when the object movement is large or the object image includes zoom-in and zoom-out or camera focus has changed. Experimental results show that the proposed method provides better performance in subjective image quality compared to conventional MCFI methods.

본 논문에서는 비선형적 움직임에 대하여 시각적 화질 향상에 목적을 둔 프레임 보간 기법을 제안한다. 그러므로 블록 현상과 영상의 중첩을 감소시키고자 블록 크기를 128x128부터 1x1까지 순차적으로 전역탐색을 실시하여 최소 오차값이 가장 작은 블록이 포함된 프레임을 선택하고, 비선형적인 움직임 벡터를 GRNN(General Regression Neural Network) 알고리즘을 이용하여 재 추정함으로써 프레임을 보간하는 알고리즘을 제안한다. 이러한 알고리즘의 성능 분석을 위해 프레임 반복, 단방향 움직임 보상, 양방향 움직임 보상의 기법들과 비교한다. 객체의 움직임이 크거나 카메라 초점의 이동과 줌인(zoom-in), 줌아웃(zoom-out) 효과가 들어간 대상 영상에 대하여 주관적 화질면에서 성능이 향상됨을 보인다.

Keywords

References

  1. Hyeong-chul Oh, Joo-hyun Lee, Chang-ki Min, and Je-chang Jeong, "Adaptive Frame Rate Up-Conversion Algorithm using the Neighbouring Pixel Information and Bilateral Motion Estimation," The Journal of Korea Information and Communications Society, Vol. 35, No. 9, pp. 761-770, 2010.
  2. J.H. Park, "A Study on the Frame Rate Conversion Using the Shape Based Motion Estimation and the Edge Direction Information" PhD Thesis, Chonnam National University, February 2011.
  3. G. Dane and T. Q. Nguyen, "Optimal Temporal Interpolation Filter for Motion-compensated Frame Rate Up Conversion," IEEE Transactions on Image Processing, Vol. 15, No. 4, pp. 978-991, April 2006. https://doi.org/10.1109/TIP.2005.863947
  4. Seung-hyun Choi and Seong-won Lee, "Frame Interpolation using Dominant MV," Journal of the Institute of Electronics Engineers of Korea, Vol. 46, No. 6, pp. 123-131, November 2009.
  5. Jin-soo Kim, "Frame-Adaptive Distortion Estimation for Motion Compensated Interpolated Frame," Journal of the Korea Contents Association, Vol. 12, No. 3, pp. 1-8, March 2012. https://doi.org/10.5392/JKCA.2012.12.03.001
  6. Hyeong-chul Oh, "Efficient frame rate up-conversion algorithm using adaptive threshold and motionre-estimation," MS Thesis, Hanyang University, February 2011.
  7. Min-kyu Lee and Hyun-wook Park, "Frame Interpolation using Bilateral Motion Refinement with Rotation," Journal of the Institute of Electronics Engineers of Korea, Vol. 46, No. 5, pp. 135-142, September 2009.
  8. K. A. Bugwadia, E. D. Petajan, and N. N. Puri, "Progressive-scan Rate up-conversion of 24/30 Source Materials for HDTV," IEEE Transactions on Consumer Electronics, Vol. 42, No. 3, pp. 312-321, August 1996. https://doi.org/10.1109/30.536125
  9. Iain E. Richardson, H.264 and MPEG-4 Video Compression, John Wiley & Sons Inc., 2003.
  10. Young-oh Han, "A study on motion prediction and subband coding of moving pictures using GRNN," Journal of the Korea Institute of Electronic Communication Sciences, Vol. 5, No. 3, pp. 256-261, June 2010.
  11. D. Tomandl and A. Schober, "A Modified General Regression Neural Network with New Efficient Training Algorithms as a Robust 'black box'-Tool for Data Analysis," Elsevier Neural Network., Vol. 14, No. 8, pp. 1023-1034, 2001. https://doi.org/10.1016/S0893-6080(01)00051-X
  12. C. Li and A. C. Bobik, "Blind Image Quality Assessment Using a General Regression Neural Network," IEEE Transactions on Neural Network, Vol. 22, No. 5, pp. 793-799, 2011. https://doi.org/10.1109/TNN.2011.2120620
  13. Ji-yoon Park and Chang-woo Lee, "Efficient Motion Vector Correction Method m Motion Compensated Interpolation Technique Using Bilateral Motion Estimation," The Journal of Korea Information and Communications Society, Vol. 34, No. 7, pp. 687-696, July 2009.

Cited by

  1. Implementation of Real-Time Multi-Camera Video Surveillance System with Automatic Resolution Control Using Motion Detection vol.18, pp.4, 2014, https://doi.org/10.7471/ikeee.2014.18.4.612