Automated Stone Detection on Side-Scan Sonar Mosaics Using Haar-Like Features

Abstract

Stony grounds form important habitats in the marine environment, especially for sessile benthic organisms. For the purpose of habitat demarcation and monitoring, knowledge of the position and abundance of individual stones is necessary. This is especially the case in areas with a scattered occurrence of stones in an environment which is otherwise characterized by relatively mobile sandy sediments. Exposed stones can be detected using side-scan sonar (SSS) data. However, apart from laborious manual identification, there is as yet no automated or semi-automated method available for a fast and spatially resolved detection of stones. In this study, a Haar-like feature detector was trained to identify individual stones on an SSS mosaic (~12 km2) showing heterogeneous sediment distribution. The results of this method were compared with those of manually derived stones. Our study shows that the Haar-like feature detector was able to detect up to 62% of the overall occurrence of stones within the study area. Even though the sheer number of correctly identified stones was influenced by, e.g., the type of sediments and the number of grey values of the mosaic, Haar-like feature detectors provide a relatively easy and fast method to identify stones on SSS mosaics when compared to the manual investigation.

Super Resolution by Deep Learning Improves Boulder Detection in Side Scan Sonar Backscatter Mosaics

“…The performance of the 0.25 m @ × 2 mosaic is affected most in the area with many small boulders, indicating that the super resolution fails to restore very small, clustered objects, a trend that is commonly observed in other areas of application [17]. On the other hand, and important for practical application, few false positives are introduced and the precision exceeds 0.85 for all natural facies, and a drop in performance depending on the background sediment, as observed in previous studies using Haar-like features for boulder detection [3], was not observed. However, in contrast to the good performance on natural seafloor, many false positives around plough marks created by anthropogenic impact exist, comprising a noticeable percentage of the total number of detections by the model between 2% and 4%, depending on the mosaic.…”

“…The need for improving the resolution of images is widespread, including tasks in medical applications, object detection, or remote sensing [1]. In marine habitat mapping by acoustic remote sensing specifically, a recent topic of interest is the detection of individual boulders for purposes of hard ground delineation for marine spatial planning purposes as well as ecosystem research [2][3][4][5]. The detection of boulders is typically based on the interpretation of backscatter intensity mosaics derived from acoustic remote sensing by side scan sonars or multibeam echo sounders [6].…”

Hydroacoustic Mapping of Geogenic Hard Substrates: Challenges and Review of German Approaches

“…Sawas et al [104] and Barngrover et al [105] have trained Haar-Like features (equivalent to convolutional neural networks with pre-selected kernel values) to automatically detect objects, specifically mines, using real and synthetic SSS images to increase the sample size. Applying Haar-Like features to the identification of stones based oñ 22,000 positive images and~340,000 negative samples, Michaelis et al [26] showed that training data in terms of different acoustic backscatter signals of the underlying seafloor have a strong influence on the detectability of stones in heterogeneous environments. Detection of mines using convolutional neural networks was attempted by Dzieciuch et al [106].…”

“…In times of highly sophisticated computer technology it is still common that experts manually interpret stone signatures on hydroacoustic backscatter data which is, however, very time-consuming and not practical for large and heterogeneous areas. Recently, research has focused on the automated identification of stone signatures in SSS backscatter data by means of machine learning techniques for selected study sites [21,26]. The future perspective is to improve publicly available training data sets to make the models more accurate and applicable for a variation of geological sites and large scale mapping campaigns.…”

Limitations of Boulder Detection in Shallow Water Habitats Using High-Resolution Sidescan Sonar Images

“…Those survey setups generate SSS image resolutions ranging from 0.25 m/pixel to 1 m/pixel, thus are insufficient to identify individual stones or boulders, especially in areas with coarse, mixed sediments [16]. Despite recent advances in methods for the automated detection of individual stones and boulders [27,28] methodological constraints of survey settings (acoustic frequencies, horizontal resolution and range, survey speeds, etc.) remain undiscussed.…”