SUPPORT VECTOR MACHINE AND DECISION TREE BASED CLASSIFICATION OF SIDE-SCAN SONAR MOSAICS USING TEXTURAL FEATURES

Abstract. The diversity and heterogeneity of coastal, estuarine and stream habitats has led to them becoming a prevalent topic for study. Woody ruins are areas of potential riverbed habitat, particularly for fish. Therefore, the mapping of those areas is of interest. However, due to the limited visibility in some river systems, satellites, airborne or other camera-based systems (passive systems) cannot be used. By contrast, sidescan sonar is a popular underwater acoustic imaging system that is capable of providing high- resolution monochromatic images of the seafloor and riverbeds. Although the study of sidescan sonar imaging using supervised classification has become a prominent research subject, the use of composite texture features in machine learning classification is still limited. This study describes an investigation of the use of texture analysis and feature extraction on side-scan sonar imagery in two supervised machine learning classifications: Support Vector Machine (SVM) and Decision Tree (DT). A combination of first- order texture and second-order texture is investigated to obtain the most appropriate texture features for the image classification. SVM, using linear and Gaussian kernels along with Decision Tree classifiers, was examined using selected texture features. The results of overall accuracy and kappa coefficient revealed that SVM using a linear kernel leads to a more promising result, with 77% overall accuracy and 0.62 kappa, than SVM using either a Gaussian kernel or Decision Tree (60% and 73% overall accuracy, and 0.39 and 0.59 kappa, respectively). However, this study has demonstrated that SVM using linear and Gaussian kernels as well as a Decision Tree makes it capable of being used in side-scan sonar image classification and riverbed habitat mapping.

[1]  Natanael Nunes de Moura,et al.  Novelty detection in passive SONAR systems using support vector machines , 2015 .

[2]  Wenjia Wang,et al.  A Texture Analysis Approach to Identifying Sabellaria Spinulosa Colonies in Sidescan Sonar Imagery , 2011, 2011 Irish Machine Vision and Image Processing Conference.

[3]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[4]  Rhinelander Jason,et al.  Feature extraction and target classification of side-scan sonar images , 2016 .

[5]  L. Parson,et al.  TexAn: textural analysis of sidescan sonar imagery and generic seafloor characterisation , 1998, IEEE Oceanic Engineering Society. OCEANS'98. Conference Proceedings (Cat. No.98CH36259).

[6]  Ute Beyer,et al.  Remote Sensing And Image Interpretation , 2016 .

[8]  Daniel Ierodiaconou,et al.  Evaluation of Four Supervised Learning Methods for Benthic Habitat Mapping Using Backscatter from Multi-Beam Sonar , 2012, Remote. Sens..

[9]  Daniel Buscombe,et al.  Shallow water benthic imaging and substrate characterization using recreational-grade sidescan-sonar , 2017, Environ. Model. Softw..

[10]  Barry Haack,et al.  Comparison and Integration of Radar and Optical Data for Land Use/Cover Mapping , 2006 .

[11]  Yun Zhang,et al.  Texture-Integrated Classification of Urban Treed Areas in High-Resolution Color-Infrared Imagery , 2001 .

[12]  Vladimir A. Krylov,et al.  Textural lacunarity for semi-supervised detection in sonar imagery , 2014 .

[13]  Timothy A. Warner,et al.  Kernel-Based Texture in Remote Sensing Image Classification , 2011 .

[14]  Adam J. Kaeser,et al.  USING LOW‐COST SIDE‐SCAN SONAR FOR BENTHIC MAPPING THROUGHOUT THE LOWER FLINT RIVER, GEORGIA, USA , 2013 .

[15]  Changjing Shang,et al.  Feature-based texture classification of side-scan sonar images using a neural network approach , 1992 .

[16]  Yue Pan,et al.  Underwater acoustic target recognition using SVM ensemble via weighted sample and feature selection , 2016, 2016 13th International Bhurban Conference on Applied Sciences and Technology (IBCAST).

[17]  Jacques Wainer,et al.  Empirical comparison of cross-validation and internal metrics for tuning SVM hyperparameters , 2017, Pattern Recognit. Lett..

[18]  G. Fader,et al.  Benthic habitat mapping on the Scotian Shelf based on multibeam bathymetry, surficial geology and sea floor photographs , 2001 .

[19]  Philippe Blondel,et al.  The Handbook of Sidescan Sonar , 2009 .

[20]  Barry Haack,et al.  A Comparison of Land Use/Cover Mapping with Varied Radar Incident Angles and Seasons , 2007 .

[21]  M. Lianantonakis,et al.  Sidescan Sonar Segmentation Using Texture Descriptors and Active Contours , 2007, IEEE Journal of Oceanic Engineering.

[22]  Mustafa Neamah Jebur,et al.  Spatial prediction of flood susceptible areas using rule based decision tree (DT) and a novel ensemble bivariate and multivariate statistical models in GIS , 2013 .

[23]  G. T. Uber,et al.  Side Scan Sonar Object Classification Algorithms , 1989, Proceedings of the 6th International Symposium on Unmanned Untethered Submersible Technology,.

[24]  Alexandre C. G. Schimel,et al.  Backscatter measurements by seafloor‐mapping sonars: guidelines and recommendations , 2015 .

[25]  E. Saleh,et al.  Marine habitat mapping at Labuan Marine Park, Federal Territory of Labuan, Malaysia , 2015, Ocean Science Journal.

[26]  Chris Eliasmith,et al.  Hyperopt: a Python library for model selection and hyperparameter optimization , 2015 .

[27]  J. Wheaton,et al.  Alluvial substrate mapping by automated texture segmentation of recreational-grade side scan sonar imagery , 2018, PloS one.

[28]  .A Vikas.  Detection and Removal of Shadows for Side Scan Sonar Images By Effective Image Processing Algorithms , 2017 .

[29]  D. Ierodiaconou,et al.  Marine benthic habitat mapping using Multibeam data, georeferencedvideo and image classification techniques in Victoria, Australia , 2007 .

[30]  Ning Fang,et al.  Radar target classification using support vector machine and subspace methods , 2015 .

[31]  Annette R. Grilli,et al.  Toward wind farm monitoring optimization: assessment of ecological zones from marine landscapes using machine learning algorithms , 2014, Hydrobiologia.

[32]  Giles M. Foody,et al.  The use of small training sets containing mixed pixels for accurate hard image classification: Training on mixed spectral responses for classification by a SVM , 2006 .

[33]  Iain Parnum,et al.  Benthic habitat mapping using multibeam sonar systems , 2007 .

[34]  Verónica Bolón-Canedo,et al.  Ensembles for feature selection: A review and future trends , 2019, Inf. Fusion.

[35]  I. Kröncke,et al.  Relationships between spatial patterns of macrofauna communities, sediments and hydroacoustic backscatter data in a highly heterogeneous and anthropogenic altered environment , 2017 .

[36]  Eirik Malnes,et al.  Automatic detection of snow avalanche debris in central Svalbard using C-band SAR data , 2017 .

[37]  Esther Dura.  Image Processing Techniques For the Detection and Classification of Man Made Objects in Side-Scan Sonar Images , 2011 .

[38]  Kurt Prospere Ab,et al.  Substrate mapping of three rivers in a Ramsar wetland in Jamaica: a comparison of data collection (hydroacoustic v. grab samples), classification and kriging methods , 2015 .

[39]  Chih-Jen Lin,et al.  A Practical Guide to Support Vector Classication , 2008 .

[40]  Dave Reid,et al.  Acoustic seabed classification: current practice and future directions , 2008 .

[41]  W. Loh,et al.  SPLIT SELECTION METHODS FOR CLASSIFICATION TREES , 1997 .

[42]  Daniel Buscombe,et al.  Automated Riverbed Sediment Classification Using Low-Cost Sidescan Sonar , 2016 .