Semantic Mapping on Underwater Environment Using Sonar Data

Machado, Mauro Luís da Silva; Drews, Paul; Núñez, Pedro; Botelho, Silvia Silva da Costa

doi:10.1109/lars-sbr.2016.48

Cited by 6 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This paper extends the contributions proposed in [12], with modifications on the segmentation methodology. Now, the local parameter adjustment averages a window of bins in order to find peaks of intensities.…”

Section: Introductionsupporting

confidence: 53%

Object Classification in Semi Structured Enviroment Using Forward-Looking Sonar

Santos

Ribeiro

Núñez

et al. 2017

Sensors

View full text Add to dashboard Cite

The submarine exploration using robots has been increasing in recent years. The automation of tasks such as monitoring, inspection, and underwater maintenance requires the understanding of the robot’s environment. The object recognition in the scene is becoming a critical issue for these systems. On this work, an underwater object classification pipeline applied in acoustic images acquired by Forward-Looking Sonar (FLS) are studied. The object segmentation combines thresholding, connected pixels searching and peak of intensity analyzing techniques. The object descriptor extract intensity and geometric features of the detected objects. A comparison between the Support Vector Machine, K-Nearest Neighbors, and Random Trees classifiers are presented. An open-source tool was developed to annotate and classify the objects and evaluate their classification performance. The proposed method efficiently segments and classifies the structures in the scene using a real dataset acquired by an underwater vehicle in a harbor area. Experimental results demonstrate the robustness and accuracy of the method described in this paper.

show abstract

Section: Introductionsupporting

confidence: 53%

Object Classification in Semi Structured Enviroment Using Forward-Looking Sonar

Santos

Ribeiro

Núñez

et al. 2017

Sensors

View full text Add to dashboard Cite

show abstract

“…It is assumed that a component operates reliably when the wear does not exceed the acceptable (limit. This process includes corrections of the signal amplification (such as Time Varying Gain (TVG)) and geometric distortions of the sonar image (Slant Range Corrections (SRC)) [23,24]. After SRC, the sonar images are geometrically corrected across-track; the along-track corrections account for the variations in platform speed.…”

Section: Factors Affecting the Quality Of Sonar Datamentioning

confidence: 99%

A Method for the Interpretation of Sonar Data Recorded during Autonomous Underwater Vehicle Missions

Zieja

Wawrzyński

Sigiel³

2022

Polish Maritime Research

View full text Add to dashboard Cite

Image acquisition from autonomous underwater vehicles (AUVs) is useful for mapping objects on the seabed. However, there are few studies on the interpretation of data collected with side-scan sonar during autonomous underwater vehicle missions. By recording the seabed with 3D multibeam sonar, a large number of survey points can be obtained. The collected data are processed using applications based on remote sensing image processing. The data collected during AUV missions (or other sonar carriers) needs to be pre-processed to reach the proper effectiveness level. This process includes corrections of signal amplification (Time Varying Gain, or TVG) and geometric distortions of sonar images (Slant Range Corrections). It should be mentioned that, when carrying out the interpretation process for structures on the sea floor, sonar users need to understand the process of visualising seabed projections and depressions, as well as the resolution limitations of the sonar sensors.

show abstract

“…Forward-looking sonar devices provide real-time sonar images for underwater target detection, navigation, surveillance, and inspection [ 34 , 35 , 36 ]. In combination with semantic segmentation algorithms, forward-looking sonar can present the underwater scene clearly, providing an important basis for target localization and identification [ 37 ]. At present, semantic segmentation of forward-looking sonar images has the following challenges [ 38 ]: (1) serious noise interference, which makes it difficult to segment target areas accurately, especially when they are small; and (2) many images are required to obtain sufficient data to achieve high segmentation accuracy, and improvements are required to achieve high accuracy from a limited number of images.…”

Section: Introductionmentioning

confidence: 99%

Feature Pyramid U-Net with Attention for Semantic Segmentation of Forward-Looking Sonar Images

Zhao

Chen

et al. 2022

Sensors

View full text Add to dashboard Cite

Forward-looking sonar is a technique widely used for underwater detection. However, most sonar images have underwater noise and low resolution due to their acoustic properties. In recent years, the semantic segmentation model U-Net has shown excellent segmentation performance, and it has great potential in forward-looking sonar image segmentation. However, forward-looking sonar images are affected by noise, which prevents the existing U-Net model from segmenting small objects effectively. Therefore, this study presents a forward-looking sonar semantic segmentation model called Feature Pyramid U-Net with Attention (FPUA). This model uses residual blocks to improve the training depth of the network. To improve the segmentation accuracy of the network for small objects, a feature pyramid module combined with an attention structure is introduced. This improves the model’s ability to learn deep semantic and shallow detail information. First, the proposed model is compared against other deep learning models and on two datasets, of which one was collected in a tank environment and the other was collected in a real marine environment. To further test the validity of the model, a real forward-looking sonar system was devised and employed in the lake trials. The results show that the proposed model performs better than the other models for small-object and few-sample classes and that it is competitive in semantic segmentation of forward-looking sonar images.

show abstract

Semantic Mapping on Underwater Environment Using Sonar Data

Cited by 6 publications

References 9 publications

Object Classification in Semi Structured Enviroment Using Forward-Looking Sonar

Object Classification in Semi Structured Enviroment Using Forward-Looking Sonar

A Method for the Interpretation of Sonar Data Recorded during Autonomous Underwater Vehicle Missions

Feature Pyramid U-Net with Attention for Semantic Segmentation of Forward-Looking Sonar Images

Contact Info

Product

Resources

About