Abstract
In order to meet the needs of complex seafloor scenes target detection, an image registration method based on sensitive region is proposed which considering the fuzzy boundary, poor contrast and low overall gray value of forward-looking sonar image. In this method, gradient enhancement algorithm is used to enhance the boundary of the target region in forward-looking sonar(FLS) image, and morphological operation method is used to remove a large number of noise points in the image background. And then, the coordinates and the sensitive region which contains a part of obvious grayscale changes is located in the floating image. The combined method of particle swarm optimization(PSO) algorithm and mutual information(MI) algorithm are used for initial registration, and the re-registration method of mutual information is used to optimize the registration error and improve the registration accuracy. The experimental results show that this method has the advantages of accurate location of the sensitive region, low computation time and high registration accuracy.
Highlights
Image registration is a simultaneous interpreting process, which aligns two images collected at different times, different viewpoints, and different sensors or instruments in space [1], [2]
Sonar image registration is the basis of sonar image fusion [4], which can be used for motion estimation, terrain detection and mapping in the process of AUV underwater operation
EXTRACTION VERIFICATION OF SENSITIVE REGION 1) NECESSITY VERIFICATION OF SENSITIVE REGION a: VERTICAL COMPARISON The vertical comparison is carried out in the MI registration based on sensitive region and global MI registration.The adjacent frames of forward-looking sonar (FLS) data are selected registration experiment
Summary
Image registration is a simultaneous interpreting process, which aligns two images collected at different times, different viewpoints, and different sensors or instruments in space [1], [2]. Through affine transformation and rigid transformation between image coordinate systems, the displacement and angle between images are calculated [10] This kind of algorithm is suitable for image registration with abundant speckle features and corner features. The grayscalebased image registration method calculates the similarity between grayscale images by the whole gray information, and searches for the transformation parameters of the maximum similarity measure by searching strategy [11]. It is suitable for image registration where the grayscale changes are less obvious
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
