Detection of deep water benthic macroalgae using image-based classification techniques on multibeam backscatter at Cashes Ledge, Gulf of Maine, USA

McGonigle, Chris; Grabowski, Jonathan H.; Brown, Craig J.; Weber, Thomas C.; Quinn, Rory

doi:10.1016/j.ecss.2010.10.016

Cited by 48 publications

(41 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The comparison between two categorical maps is commonly applied in terrestrial applications, and is becoming increasingly important for applications in benthic habitat mapping [38][39][40]. In this study, map comparison did not identify the most accurate classifier; rather it measured the relative agreement between two thematic maps.…”

Section: Discussionmentioning

confidence: 99%

Evaluation of Four Supervised Learning Methods for Benthic Habitat Mapping Using Backscatter from Multi-Beam Sonar

2012

View full text Add to dashboard Cite

An understanding of the distribution and extent of marine habitats is essential for the implementation of ecosystem-based management strategies. Historically this had been difficult in marine environments until the advancement of acoustic sensors. This study demonstrates the applicability of supervised learning techniques for benthic habitat characterization using angular backscatter response data. With the advancement of multibeam echo-sounder (MBES) technology, full coverage datasets of physical structure over vast regions of the seafloor are now achievable. Supervised learning methods typically applied to terrestrial remote sensing provide a cost-effective approach for habitat characterization in marine systems. However the comparison of the relative performance of different classifiers using acoustic data is limited. Characterization of acoustic backscatter data from MBES using four different supervised learning methods to generate benthic habitat maps is presented. Maximum Likelihood Classifier (MLC), Quick, Unbiased, Efficient Statistical Tree (QUEST), Random Forest (RF) and Support Vector Machine (SVM) were evaluated to classify angular backscatter response into habitat classes using training data acquired from underwater video observations. Results for biota classifications indicated that SVM and RF produced the highest accuracies, followed by QUEST and MLC, respectively. The most important backscatter data were from the moderate incidence angles between 30 and 50. This study presents initial results for understanding how acoustic backscatter from MBES can be optimized for the characterization of marine benthic biological habitats.

show abstract

Section: Discussionmentioning

confidence: 99%

Evaluation of Four Supervised Learning Methods for Benthic Habitat Mapping Using Backscatter from Multi-Beam Sonar

2012

View full text Add to dashboard Cite

show abstract

“…Substantial efforts have been made in trying to obtain seafloor type or physical characteristics from the measurement of the backscatter level (e.g., Brown and Blondel 2009;De Moustier and Matsumoto 1993;Kloser et al 2010;Lamarche et al 2011;McGonigle and Collier 2014;Preston et al 2001), in many cases according to its angle dependence (Haniotis et al 2015). Some original applications include recognition and mapping algae (McGonigle et al 2011), oil spills (Medialdea et al 2008), or hydrothermal vents and seeps (Durand et al 2006;Klaucke et al 2008).…”

Section: Backscatter Interpretationmentioning

confidence: 99%

Recommendations for improved and coherent acquisition and processing of backscatter data from seafloor-mapping sonars

Lamarche

Lurton

2017

Mar Geophys Res

View full text Add to dashboard Cite

“…In order to validate and evaluate the two classification approaches, groundtruth information collected by diving prospection and independent echo-sounder data derived from the Furuno System were examined. This study links different mapping approaches which act on different scales such as optical remote sensing in intertidal areas (Hennig et al, 2007;Oppelt et al, 2012) and MBES techniques in deeper waters (Komatsu et al, 2003;Rattray et al, 2009;Ierodiaconou et al, 2011;McGonigle et al, 2011;;van Rein et al, 2011).…”

Section: Discussionmentioning

confidence: 99%

“…Recently, many hydroacoustic studies applied and compared singlebeam echosounders (SBES), or multibeam echo sounders (MBES), and sidescan sonars in order to detect and classify seafloor vegetation in different regions of the world. Parnum et al, 2009;Rattray et al (2009), Ierodiaconou et al (2011), McGonigle et al (2011), andvan Rein et al (2011), for example, used MBES and partially also sidescan sonar to identify benthic habitats and discriminate between different seaweed species in water depth of w40 m. While it was simple to distinguish between physical habitats, discrimination between biological habitats was still afflicted with uncertainties (e.g. Méléder et al, 2010;Minami et al, 2010).…”

Section: Introductionmentioning

confidence: 99%

Predicting spatial kelp abundance in shallow coastal waters using the acoustic ground discrimination system RoxAnn

Mielck

Bartsch

Hass

et al. 2014

Estuarine, Coastal and Shelf Science

View full text Add to dashboard Cite

a b s t r a c tKelp forests represent a major habitat type in coastal waters worldwide and their structure and distribution is predicted to change due to global warming. Despite their ecological and economical importance, there is still a lack of reliable spatial information on their abundance and distribution. In recent years, various hydroacoustic mapping techniques for sublittoral environments evolved. However, in turbid coastal waters, such as off the island of Helgoland (Germany, North Sea), the kelp vegetation is present in shallow water depths normally excluded from hydroacoustic surveys. In this study, single beam survey data consisting of the two seafloor parameters roughness and hardness were obtained with RoxAnn from water depth between 2 and 18 m. Our primary aim was to reliably detect the kelp forest habitat with different densities and distinguish it from other vegetated zones. Five habitat classes were identified using underwater-video and were applied for classification of acoustic signatures. Subsequently, spatial prediction maps were produced via two classification approaches: Linear discriminant analysis (LDA) and manual classification routine (MC). LDA was able to distinguish dense kelp forest from other habitats (i.e. mixed seaweed vegetation, sand, and barren bedrock), but no variances in kelp density. In contrast, MC also provided information on medium dense kelp distribution which is characterized by intermediate roughness and hardness values evoked by reduced kelp abundances. The prediction maps reach accordance levels of 62% (LDA) and 68% (MC). The presence of vegetation (kelp and mixed seaweed vegetation) was determined with higher prediction abilities of 75% (LDA) and 76% (MC). Since the different habitat classes reveal acoustic signatures that strongly overlap, the manual classification method was more appropriate for separating different kelp forest densities and low-lying vegetation. It became evident that the occurrence of kelp in this area is not simply linked to water depth. Moreover, this study shows that the two seafloor parameters collected with RoxAnn are suitable indicators for the discrimination of different densely vegetated seafloor habitats in shallow environments.

show abstract

Detection of deep water benthic macroalgae using image-based classification techniques on multibeam backscatter at Cashes Ledge, Gulf of Maine, USA

Cited by 48 publications

References 47 publications

Evaluation of Four Supervised Learning Methods for Benthic Habitat Mapping Using Backscatter from Multi-Beam Sonar

Evaluation of Four Supervised Learning Methods for Benthic Habitat Mapping Using Backscatter from Multi-Beam Sonar

Recommendations for improved and coherent acquisition and processing of backscatter data from seafloor-mapping sonars

Predicting spatial kelp abundance in shallow coastal waters using the acoustic ground discrimination system RoxAnn

Contact Info

Product

Resources

About