Data fusion of multi-beam sonar and side-scan sonar base on feature contour registration

Abstract

This paper proposed a same-name feature point registration method based on feature contour. By analyzing the map of multi-beam water depth false-gray image and side-scan sonar image, and adopting CANNY operator to extract contour line. Then select the turning points which cater the constraints as the feature points, and deal the feature points with quadratic polynomial algorithm to obtain the deformation model to realize image registration between the two. Finally, process the fusion result by wavelet decomposition and evaluate it by cross entropy algorithm. Data simulation shows that the result is reasonable and the method proposed in this paper is practicable.

Design and Experiments of a Portable Seabed Integrated Detection Sonar

“…When transmitting a multibeam sound source, only the two linear arrays (No. 18,No.19) in the middle of the parametric array need to be enabled. Based on the theory of multibeam sound sources, two channels simultaneously send sine signal pulses with a 65 degrees phase difference when 0 ≤ t ≤ T the sonar behaves as a phase-controlled sound source, while sonar is a single-channel sound source when T ≤ t ≤ 2T Then the two-stage directivities are spliced in the time domain to form a multibeam sound source with a wide directivity as described in Equation ( 3), the synthetic directivity is shown in Figure 6a.…”

“…First, due to the different installation forms and working principles of the transducers of this equipment (for example, SSS is usually towed behind the surveying ship, while MBSS is fixed to the ship), the attitude information of these devices is different, so it is impossible to observe the same measurement point at the same time. Second, the uncertain factors introduced by time-varying and spatial-varying characteristics of the marine environment make it difficult or impossible to accurately integrate the data measured by multiple instruments [ 17 , 18 , 19 ]. On the other hand, different sonar equipment and their auxiliary equipment are complicated to install and must be repeatedly calibrated and matched.…”

Combined imaging matching method of side scan sonar images with prior position knowledge

“…The approach improves the reliability of the matching, but the segmenting is carried out manually. Hao and Han [18] extracted the feature contours from images of an SSS and matched them with those from an MBS. This approach can get the position of the current location relative to the reference point and deduce the location of their platforms, but it is complicated and feature point extraction needs to be improved when there are not enough significant features in the images.…”

Curvelet Approach for Deep-sea Sonar Image Denoising, Contrast Enhancement and Fusion

“…In [9], Y. Han et al proposed a wavelet based multi-beam sonar and side-scan sonar fusion method. It adapted dyadic wavelet transform for fusion.…”