An adaptive image-stitching algorithm for an underwater observation system

Published: 08th May 2020
Views: N/A

Currently the majority of nations are devoted to focusing on observation that is underwater. On the other hand, the acquisition of submerged pictures is performed in noisy surroundings with low visibility because natural light is not available and even if artificial light is applied, the observable range is restricted [1]. Acoustic cameras can provide extremely high resolutions (for sonar) and rapid refresh rates [2]. Therefore, sonar systems are broadly used to obtain images of the seabed or another objects that were submerged. It's not conducive to tracking items or environments submerged through sonar gear alone since the observation angle of high-resolution sonar gear is so small that only a part of the submerged scene can be discovered, i.e., the horizontal view angle of DIDSON (double-frequency identification sonar) is 28.8[degrees]. Therefore, most sonar equipment is frequently installed to have scene info that was distinct; as such, sonar images are stitched to enlarge the monitoring horizon of the submerged environment [3]. Hence, image mosaicing technology plays a significant part in monitoring systems that are underwater. There has been much research into optical image-stitching; yet, little research has been dedicated to stitching for sonar pictures. Image stitching on an optical picture may be separated into algorithms on the basis of the pixel-level, the frequency power spectrum and features [4]. The computation of the pixel-level algorithm is fast and simple, but the result that is mosaicing is just not perfect if there aren't enough attributes or when the pictures are particularly noisy. The computation of the frequency power spectrum algorithm is also fast and noise correlation interference can be beat, but it requires a satisfactory overlap width. Meanwhile, feature points can easily be taken out by the characteristic algorithm, which is appropriate for stitching the images with enough features but the expense of a substantial computation load. It isn't proper to adopt a pixel-level algorithm for stitching sonar images with plenty of noise. Several modified stage correlation algorithms on the basis of the Fourier transform have been proposed in the literature [57]. The algorithms enable the measurement of rotation, translation and scaling variables between two pictures. The stitching resulting from this algorithm is not bad on paper. However, the precision of the stitching result for sonar pictures with characteristics that are sufficient is lower than for the feature algorithm. The SIFT algorithm has been examined for extracting invariant attributes that were distinguishing from images which may be used to perform dependable fitting between different views of an object or scene [8]. Because the SIFT feature points are described as 128 -dimensional vectors, the speed of the algorithm is slow. Mair et al. [9] have described a new corner detection approach, called the FAST (attributes from gifted section evaluation) algorithm. The FAST algorithm just uses the surrounding pixels information to get the feature points, which is relatively rapid and easy.

Report this article Ask About This Article

More to Explore