In order to overcome the shortcomings of insufficient operation accuracy and sensitivity in the operation of underwater machines, it is necessary to improve the positioning accuracy of the multi-camera for the underwater robots, which requires an accurate and rapid registration method for underwater images. In this paper, a novel underwater image registration method is proposed by using SURF (Speeded Up Robust Features) feature detection and PCET (Polar Complex Exponential Transform) moments for feature description and matching with fast calculation speed and better registration accuracy. SURF feature detection algorithm is used to extract the feature points firstly. Then, the region descriptor is constructed using PCET moments to represent the local feature of image patch for rough matching. Finally, the RANSAC (RANdom SAmple Consensus) algorithm is applied to remove mismatches for obtaining the correct transformation parameters. The experimental results confirm that the proposed method can achieve an accurate registration for underwater images with higher speed and better performance than other classical methods.
Underwater Image Registration Based on Polar Complex Exponential Transform
Lect. Notes Electrical Eng.
International Conference on Autonomous Unmanned Systems ; 2021 ; Changsha, China September 24, 2021 - September 26, 2021
Proceedings of 2021 International Conference on Autonomous Unmanned Systems (ICAUS 2021) ; Chapter : 120 ; 1203-1213
2022-03-18
11 pages
Article/Chapter (Book)
Electronic Resource
English
Image Registration using Polar Wavelets
IEEE | 2008
|British Library Online Contents | 2013
|Log-polar height maps for multiple range image registration
British Library Online Contents | 2009
|Multiple Range Image Registration by Matching Local Log-Polar Range Images
British Library Conference Proceedings | 2006
|