Two-dimensional forward-looking sonar image registration by maximization of peripheral mutual information

Abstract

Monitoring the field of operation of an underwater vehicle is crucial during missions near the sea floor. The forwardlooking sonar is often the only available sensor for the observation of the ambient turbid water environment. Sonar image registration is not only a first step towards a panoramic mosaic but it also provides an initial motion parameter estimation for the vehicle self-localization and navigation. In this article, a peripheral mutual information (PMI) maximization method is proposed for the sonar image registration. Peripheral mutual information is inspired by regional mutual information (RMI) which makes use of the closed-form solution for the Shannon entropy by the assumption that the data vectors made of neighbouring pixels are normally distributed, an assumption that ignores correlations between the pixels in sonar images. To accommodate the fact that the neighbouring pixels show dependencies due to acoustic reverberation and dispersion, only the peripheral information in the neighbourhood of a pixel is used in peripheral mutual information for the calculation of the mutual information. Experiments show that the peripheral mutual information registration function is much smoother than that of regional mutual information. Further experiments on the two-dimensional forward-looking sonar image registration demonstrate the efficiency of peripheral mutual information.

Mutual Information Re-Registration of Sensitive Region in Forward-Looking Sonar Images Combined With Particle Swarm Optimization Algorithm

“…The criterion of image registration based on the result value is to find the maximum MI between the images as the result of registration [21]. The MI of two images reflects the mutual inclusion degree of information between them by their entropy and joint entropy [2] [22].…”

Scan registration for underwater mechanical scanning imaging sonar using symmetrical Kullback–Leibler divergence

“…However, the high-frequency information, such as edges and contours, becomes blurred if the consecutive scans are registered to form a panoramic view, due to the rapid accumulation of local errors. 30 To assess the abilities of different methods in reducing the accumulation error, we registered five consecutive sonar scans to form a panoramic map. A short sonar scan sequence, ranging from 21st to 25th, was selected for registration because the error from INS of this sequence was more severe than other segments.…”

A Survey of Underwater Acoustic SLAM System

“…Hence, the algorithm based on optical flow between consecutive sonar frames was proposed to register the sonar image in [12]. With the fact that neighboring pixels of sonar image show dependencies due to acoustic reverberation and dispersion, [41] only used the peripheral information in the neighborhood of a pixel to calculate the mutual information. The transformation parameters were pursued by maximizing mutual information.…”