Side-scan sonar image registration for AUV navigation

Vandrish, Peter; Vardy, Andrew; Walker, Dan; Dobre, Octavia A.

doi:10.1109/ut.2011.5774096

Cited by 34 publications

(23 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…based on OpenCV. Features are typically matched using the strategy of nearest-neighbor distance ratio (NNDR), which was developed by Lowe [6] and the strategy of random sample consensus (RANSAC), which was developed by Fischler and Bolles [22]. Figure 7 shows two side-scan images at different scales obtained by side-scan sonar from the opposite direction, which are corrupted by noise.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

A feature-matching method for side-scan sonar images based on nonlinear scale space

Zhang

et al. 2015

J Mar Sci Technol

View full text Add to dashboard Cite

We report a novel feature-matching method for side-scan sonar images. The method uses nonlinear diffusion filtering to build a nonlinear scale space. The noisereduction performance is enhanced via nonlinear diffusion filtering, and the improved Perona-Malik diffusion equation results in a more distinct edge and line texture in the side-scan sonar image. The modified feature descriptor reduces the dimensionality of the feature vector so that the computational expense is reduced. Experimental results show that the method provides improved noise-reduction performance and better accuracy than SIFT, SURF, and other state-of-the-art feature-matching algorithms.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Vanish used the scale-invariant feature transform (SIFT) algorithm with side-scan sonar image registration and found that it performed better than previous methods [6]. The SIFT algorithm, which was first developed by Lowe in [7], is perhaps the most important achievement in image matching.…”

Section: Introductionmentioning

confidence: 99%

A feature-matching method for side-scan sonar images based on nonlinear scale space

Zhang

et al. 2015

J Mar Sci Technol

View full text Add to dashboard Cite

show abstract

“…Matching sonar images with high accuracy has been a unsolved problem for a long time [7] [1] [8]. This is due to specifics issues in sonar imaging, such as viewpoint dependence, non-uniform insonification, low signal-to-noise ratio, low resolution, and low feature repeatability [9].…”

Section: Related Workmentioning

confidence: 99%

Improving sonar image patch matching via deep learning

Valdenegro-Toro

2017

2017 European Conference on Mobile Robots (ECMR)

View full text Add to dashboard Cite

Matching sonar images with high accuracy has been a problem for a long time, as sonar images are inherently hard to model due to reflections, noise and viewpoint dependence. Autonomous Underwater Vehicles require good sonar image matching capabilities for tasks such as tracking, simultaneous localization and mapping (SLAM) and some cases of object detection/recognition. We propose the use of Convolutional Neural Networks (CNN) to learn a matching function that can be trained from labeled sonar data, after pre-processing to generate matching and non-matching pairs. In a dataset of 39K training pairs, we obtain 0.91 Area under the ROC Curve (AUC) for a CNN that outputs a binary classification matching decision, and 0.89 AUC for another CNN that outputs a matching score. In comparison, classical keypoint matching methods like SIFT, SURF, ORB and AKAZE obtain AUC 0.61 to 0.68. Alternative learning methods obtain similar results, with a Random Forest Classifier obtaining AUC 0.79, and a Support Vector Machine resulting in AUC 0.66.

show abstract

“…This mechanism is very important for the detection and classification of bottom mines. Because of their very low target strengths, echoes from mines are very difficult to detect against a background of bottom reverberation; the shadowgraph can, however, indicate the presence of a mine even when its echo is hidden by the reverberation, see [9]- [10].…”

Section: Side Scan Sonar Operationmentioning

confidence: 99%

DESIGN OF 100 kHz SONAR FOR MARINE MAPPING APPLICATIONS

Sathishkumar¹,

Gupta²,

Babu³

et al. 2013

Int. J. of Pure and Appl. Math.

View full text Add to dashboard Cite

Side scan sonar (SSS) is an important mapping tool in underwear imaging system. This paper provides sufficient information on SSS design. SSS is known for very high quality, high resolution, ocean bottom imaging. It is able to rapidly survey large ocean areas for bottom and suspended sea-mines or other kinds of threats. Another operational aspect of these systems is that they allow autonomous underwater vehicles (AUVs) to conduct operations, mostly in shallow water and near land. It produces a representation in image form of the seabed; however, no closed form analytic expression or direct modeling technique exists to represent this complex process. In this paper the basic principle is presented for a model for the simulation of SSS, which produces as its output a qualitative image of the form generated by the actual SSS. The model considers the underlying physical processes including the effects of the refractive water medium, the scattering from the seabed and the transducer motion and directivity characteristics. Further, this work concerns the design and analysis of SSS via MATLAB simulation in order to satisfy the operational requirements. Moreover, this research has tried to indicate the various problems that may arise when a SSS operates in its near-field region and suggests certain solutions. The active sonar equation and its factors were explained and were evaluated.

show abstract

Side-scan sonar image registration for AUV navigation

Abstract: peteR king/mUn mUn explorer AUV platform with the marine institute's holyrood marine base in the background.

Cited by 34 publications

References 15 publications

A feature-matching method for side-scan sonar images based on nonlinear scale space

A feature-matching method for side-scan sonar images based on nonlinear scale space

Improving sonar image patch matching via deep learning

DESIGN OF 100 kHz SONAR FOR MARINE MAPPING APPLICATIONS

Contact Info

Product

Resources

About