A Comparative Analysis on Image Translation and Rotation Algorithms towards Implementations in Micro Aerial Vehicles

A comprehensive analysis on estimation of critical ego-motion parameters such as image translation and rotation using feature and intensity information method is presented. In feature approach, scale invariant feature transform is adopted, and in intensity method both frequency domain and compressive sampling (CS) techniques are considered. Comparative evaluation is performed with respect to their accuracy, time complexity and feasibility for deployments in aerial vehicles. A simplified version of frequency and CS methods has been proposed and experiments are conducted in standard image database and real time images. Results suggested that frequency domain approach performs faster in estimating motion parameters accurately. An indoor testing is also conducted on integrating the micro camera on a micro aerial vehicle to demonstrate the practicality and concept validation.

[1]  Cosmin Ancuti,et al.  CC-SIFT: Exploiting chromatic contrast forwide-baseline matching , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[2]  Laurent Moll,et al.  Real time correlation-based stereo: algorithm, implementations and applications , 1993 .

[3]  Guillermo Sapiro,et al.  Adaptive temporal compressive sensing for video , 2013, 2013 IEEE International Conference on Image Processing.

[4]  Anamitra Makur,et al.  Camera motion estimation using circulant compressive sensing matrices , 2013, 2013 9th International Conference on Information, Communications & Signal Processing.

[5]  Shree K. Nayar,et al.  Ego-motion and omnidirectional cameras , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[6]  Bernd Girod,et al.  Motion Analysis and Image Sequence Processing , 1993 .

[7]  Luís Corte-Real,et al.  Automatic Image Registration Through Image Segmentation and SIFT , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[8]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[9]  Zhihai He,et al.  Vision-based UAV flight control and obstacle avoidance , 2006, 2006 American Control Conference.

[10]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[11]  Takeo Kato,et al.  Vehicle Ego-Motion Estimation and Moving Object Detection using a Monocular Camera , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[12]  Amnon Shashua,et al.  A robust method for computing vehicle ego-motion , 2000, Proceedings of the IEEE Intelligent Vehicles Symposium 2000 (Cat. No.00TH8511).

[13]  P. Anandan,et al.  Hierarchical Model-Based Motion Estimation , 1992, ECCV.