﻿ 基于提升小波的光流估计算法 Optical Flow Estimation Algorithm Based on Shift Wavelet Transform

Geomatics Science and Technology
Vol.05 No.02(2017), Article ID:20288,7 pages
10.12677/GST.2017.52006

Optical Flow Estimation Algorithm Based on Shift Wavelet Transform

Lichuan Geng1,2, Zexun Geng1,2

1School of Urban-Rural Planning and Landscape Architecture, Xuchang University, Xuchang Henan

2Collaborative Innovation Center for UAVRS, Xuchang University, Xuchang Henan

Received: Apr. 6th, 2017; accepted: Apr. 25th, 2017; published: Apr. 28th, 2017

ABSTRACT

In order to improve the estimation precision of displacement among low resolution images, a wavelet based optical flow estimation algorithm is promoted. We calculated the image optical flow at different scales to improve the registration accuracy of optical flow estimation algorithm with large displacement. The results show that this improved algorithm achieves high precision pixel displacement estimation.

Keywords:Wavelet Transform, Motion Estimation, Optical Flow, Multi-Resolution

1许昌学院，城乡规划与园林学院，河南 许昌

2许昌学院，无人机低空遥感技术协同创新中心，河南 许昌

1. 引言

2. 传统的光流估算方法

(1)

(2)

(3)

(4)

3. 提升小波原理

4. 基于提升小波的光流估计

5. 实验及分析

6. 结论

1) 文中结合实际应用要求，引入多分辨率技术，改进了光流估计算法对于大位移的不能准确性估计，扩展了光流估计的应用场合，同时使光流估计的鲁棒性有了一定的提高。

2) 改进的光流算法虽在文中取得了较好的应用效果，但随着应用场合的不同，对该算法的抗噪声和快速的计算能力提出了更高的要求。

3) 在今后的研究中将继续围绕光流算法的实时性和良好的鲁棒性进行深入探索，使其能在特殊的硬件支持下实现实时动态检测，并将其研究成果扩展到其他研究领域。

Figure 1. Single-level image wavelet transform

Figure 2. Schematic diagram of multi-resolution image

Figure 3. Flow of our algorithm

(a) 第一帧图像(b) 第二帧图像

Figure 4. Experimental images

Figure 5. The db4 shift wavelet optical flow vector

Figure 6. The db6 shift wavelet optical flow vector

Figure 7. The L-K algorithm optical flow vector

Optical Flow Estimation Algorithm Based on Shift Wavelet Transform[J]. 测绘科学技术, 2017, 05(02): 40-46. http://dx.doi.org/10.12677/GST.2017.52006

1. 1. 解凯, 郭恒业, 张田文. 图像Mosaics技术综述[J]. 电子学报, 2004, 32(4): 630-634.

2. 2. Ng, M.K. and Bose, N.K. (2003) Mathematical Analysis of Super-Resolution Methodology. IEEE Signal Processing Magazine, 20, 62.

3. 3. Irani, M. and Peleg, S. (1991) Improving Resolution by Image Registration. CVGIP: Graphical Models and Image, 53, 231.

4. 4. Huang, Y., Ma, S. and Shen, C. (2003) Predictive Line Search: An Efficient Motion Estimation Algorithm for MPEG-4 Encoding Systems on Multimedia Processors. IEEE Trans on Circuits and Systems for Video Technology, 13, 111.

5. 5. Hornb, S.B. (1981) Determining Optical Flow. Artificial Intelligence, 17, 185.

6. 6. Horn, B.K.P. and Schunck, B.G. (1981) Determining Optical Flow. Artificial Intelligence, 17, 185-203.

7. 7. Lucas, B.D. and Kanade, T. (1981) An Iterative Image Registration Technique with an Application to Stereo Vision. International Joint Conference on Artificial Intelligence.

8. 8. Fleet, D.J. and Jepson, A.D. (1990) Computation of Component Image Velocity from Local Phase Information. International Journal of Computer Vision, 5, 77-104.

9. 9. Weber, J. and Malik, J. (1994) Robust Computation of the Optical Flow in Multi-Scale Differential Framework. International Journal of Computer Vision, 2, 5-19.

10. 10. Bernard, C. (1998) Wavelets and Ill-Posed Problems: Optical Flow and Scattered Data Interpolation. Ecole Polytechnique, France.

11. 11. 董长虹, 高志, 余啸海. Matlab小波分析工具箱原理与应用[M]. 北京: 国防工业出版社, 2004: 108-127.

12. 12. Sweldens, W. (1996) The Lifting Scheme: A Custom-Design Construction of Biorthogonal Wavelets. Application Computer Harmonic Analysis, 3, 186-200.

13. 13. Sweldens, W. (1998) The Lifting Scheme: A Construction of Second Generation Wavelets. SIAM Journal on Mathematical Analysis, 29, 511-546.

14. 14. 郑世友, 费树岷, 龙飞. 基于小波提升框架的图像序列中运动目标检测算法[J]. 中国图象图形学报, 2005, 10(5): 596-602.

15. 15. 徐韶华, 李红. 基于小波提升框架及小波能量的红外弱目标检测方法[J]. 红外技术, 2006, 28(11): 669-672.