E-Mail Alert

Add your e-mail address to receive forthcoming issues of this journal:

Journal Browser

Journal Browser

Special Issue "Advances in Object-Based Image Analysis—Linking with Computer Vision and Machine Learning"

A special issue of Remote Sensing (ISSN 2072-4292).

Deadline for manuscript submissions: closed (1 January 2017)

Special Issue Editors

Guest Editor
Prof. Dr. Norman Kerle

Department of Earth Systems Analysis (ESA), Faculty of Geo-Information Science and Earth Observation (ITC), University of Twente, P.O. Box 6, Hengelosestraat 99, 7500 AA Enschede, Netherlands
Website1 | Website2 | E-Mail
Interests: damage assessment; vulnerability; disaster risk management; UAV; resilience; recovery; OBIA; object-oriented analysis; VGI
Guest Editor
Prof. Dr. Markus Gerke

Institute of Geodesy and Photogrammetry, Technical University of Braunschweig, Braunschweig, Germany
Website | E-Mail
Phone: +31 53 4874522
Interests: image geometry; scene understanding; UAVs; oblique airborne photogrammetry
Guest Editor
Prof. Dr. Sébastien Lefèvre

IRISA - Université Bretagne Sud, Campus de Tohannic, BP 573, 56017 Vannes Cedex France
Website | E-Mail
Phone: +33 2 97 01 72 66
Interests: image analysis and processing, multiscale representations, computer vision, pattern recognition, machine learning, data mining, remote sensing

Special Issue Information

Dear Colleagues,

During the last 10 years the GEOBIA community has grown from a niche discipline to a recognized and vibrant branch of geoinformation science, and methods developed by the growing community have helped to tackle problems in virtually all domains where geographic data are used. The growing importance of image processing, be it of traditional airborne or satellite data, or complex hyperspectral data stacks, videos, or image data used by other communities, has resulted in a multitude of methodological approaches. Object-based approaches have turned out to be an excellent way to incorporate process and feature knowledge, in addition to providing an effective way of dealing with multi-scale data. As a consequence, hundreds of scientific publications have greatly enriched the geoinformation science domain over the past decade. Following a range of very successful biennial conferences, the GEOBIA 2016 conference took place from 14-16 September at the University of Twente in Enschede, the Netherlands. It focused on two broad issues where more progress is needed: (i) Solutions: while GEOBIA has undoubtedly advanced our understanding of a wide range of environmental processes or anthropogenic activities, many studies are marked by great technical complexity, a reliance on proprietary software, and have resulted in methods and procedures that are difficult to replicate and use for non-expert stakeholders. Especially in countries with limited resources and great environmental challenges, GEOBIA has to date only played a minor role in actual problem solving, which was addressed by the conference. GEOBIA 2016 also included a benchmarking effort, connected to an ongoing ISPRS initiative, aimed at stimulating the development of optimized, generic and transferable methods for standard GEOBIA problems. (ii) Synergies: Segmentation-based analysis is not an exclusive GEOBIA domain. On the contrary, such methodological approaches play a vital role in the biomedical and pharmaceutical communities already mentioned, but also feature very strongly in the computer vision domain. Object-based analysis is as critical to the analysis of lidar or image-derived point clouds as to the processing of satellite images. As such, those domains have a lot to share with the GEOBIA field, and vice versa. A core effort of GEOBIA 2016 was to find synergies between the disciplines that share an interest in object-based data processing.

This Special Issue will primarily feature selected papers from the GEOBIA 2016 conference. Authors wishing to have their work considered for this issue, including those not able to present at the conference, should contact the guest editors.

Authors are required to check and follow the specific Instructions to Authors, http://www.mdpi.com/journal/remotesensing/instructions.

Dr. Norman Kerle
Dr. Markus Gerke
Prof. Dr. Sébastien Lefèvre
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Published Papers (12 papers)

View options order results:
result details:
Displaying articles 1-12
Export citation of selected articles as:

Research

Open AccessArticle Good Practices for Object-Based Accuracy Assessment
Remote Sens. 2017, 9(7), 646; doi:10.3390/rs9070646
Received: 3 January 2017 / Revised: 26 May 2017 / Accepted: 19 June 2017 / Published: 22 June 2017
PDF Full-text (3477 KB) | HTML Full-text | XML Full-text
Abstract
Thematic accuracy assessment of a map is a necessary condition for the comparison of research results and the appropriate use of geographic data analysis. Good practices of accuracy assessment already exist, but Geographic Object-Based Image Analysis (GEOBIA) is based on a partition of
[...] Read more.
Thematic accuracy assessment of a map is a necessary condition for the comparison of research results and the appropriate use of geographic data analysis. Good practices of accuracy assessment already exist, but Geographic Object-Based Image Analysis (GEOBIA) is based on a partition of the spatial area of interest into polygons, which leads to specific issues. In this study, additional guidelines for the validation of object-based maps are provided. These guidelines include recommendations about sampling design, response design and analysis, as well as the evaluation of structural and positional quality. Different types of GEOBIA applications are considered with their specific issues. In particular, accuracy assessment could either focus on the count of spatial entities or on the area of the map that is correctly classified. Two practical examples are given at the end of the manuscript. Full article
Figures

Open AccessArticle Segment-before-Detect: Vehicle Detection and Classification through Semantic Segmentation of Aerial Images
Remote Sens. 2017, 9(4), 368; doi:10.3390/rs9040368
Received: 28 December 2016 / Revised: 5 April 2017 / Accepted: 7 April 2017 / Published: 13 April 2017
Cited by 1 | PDF Full-text (5318 KB) | HTML Full-text | XML Full-text
Abstract
Like computer vision before, remote sensing has been radically changed by the introduction of deep learning and, more notably, Convolution Neural Networks. Land cover classification, object detection and scene understanding in aerial images rely more and more on deep networks to achieve new
[...] Read more.
Like computer vision before, remote sensing has been radically changed by the introduction of deep learning and, more notably, Convolution Neural Networks. Land cover classification, object detection and scene understanding in aerial images rely more and more on deep networks to achieve new state-of-the-art results. Recent architectures such as Fully Convolutional Networks can even produce pixel level annotations for semantic mapping. In this work, we present a deep-learning based segment-before-detect method for segmentation and subsequent detection and classification of several varieties of wheeled vehicles in high resolution remote sensing images. This allows us to investigate object detection and classification on a complex dataset made up of visually similar classes, and to demonstrate the relevance of such a subclass modeling approach. Especially, we want to show that deep learning is also suitable for object-oriented analysis of Earth Observation data as effective object detection can be obtained as a byproduct of accurate semantic segmentation. First, we train a deep fully convolutional network on the ISPRS Potsdam and the NZAM/ONERA Christchurch datasets and show how the learnt semantic maps can be used to extract precise segmentation of vehicles. Then, we show that those maps are accurate enough to perform vehicle detection by simple connected component extraction. This allows us to study the repartition of vehicles in the city. Finally, we train a Convolutional Neural Network to perform vehicle classification on the VEDAI dataset, and transfer its knowledge to classify the individual vehicle instances that we detected. Full article
Figures

Open AccessArticle An Open-Source Semi-Automated Processing Chain for Urban Object-Based Classification
Remote Sens. 2017, 9(4), 358; doi:10.3390/rs9040358
Received: 19 December 2016 / Revised: 5 April 2017 / Accepted: 6 April 2017 / Published: 11 April 2017
PDF Full-text (9785 KB) | HTML Full-text | XML Full-text
Abstract
This study presents the development of a semi-automated processing chain for urban object-based land-cover and land-use classification. The processing chain is implemented in Python and relies on existing open-source software GRASS GIS and R. The complete tool chain is available in open access
[...] Read more.
This study presents the development of a semi-automated processing chain for urban object-based land-cover and land-use classification. The processing chain is implemented in Python and relies on existing open-source software GRASS GIS and R. The complete tool chain is available in open access and is adaptable to specific user needs. For automation purposes, we developed two GRASS GIS add-ons enabling users (1) to optimize segmentation parameters in an unsupervised manner and (2) to classify remote sensing data using several individual machine learning classifiers or their prediction combinations through voting-schemes. We tested the performance of the processing chain using sub-metric multispectral and height data on two very different urban environments: Ouagadougou, Burkina Faso in sub-Saharan Africa and Liège, Belgium in Western Europe. Using a hierarchical classification scheme, the overall accuracy reached 93% at the first level (5 classes) and about 80% at the second level (11 and 9 classes, respectively). Full article
Figures

Open AccessArticle Object-Based Detection of Lakes Prone to Seasonal Ice Cover on the Tibetan Plateau
Remote Sens. 2017, 9(4), 339; doi:10.3390/rs9040339
Received: 30 November 2016 / Revised: 20 March 2017 / Accepted: 30 March 2017 / Published: 2 April 2017
PDF Full-text (7452 KB) | HTML Full-text | XML Full-text
Abstract
The Tibetan Plateau, the world’s largest orogenic plateau, hosts thousands of lakes that play prominent roles as water resources, environmental archives, and sources of natural hazards such as glacier lake outburst floods. Previous studies have reported that the size of lakes on the
[...] Read more.
The Tibetan Plateau, the world’s largest orogenic plateau, hosts thousands of lakes that play prominent roles as water resources, environmental archives, and sources of natural hazards such as glacier lake outburst floods. Previous studies have reported that the size of lakes on the Tibetan Plateau has changed rapidly in recent years, possibly because of atmospheric warming. Tracking these changes systematically with remote sensing data is challenging given the different spectral signatures of water, the potential for confusing lakes with glaciers, and difficulties in classifying frozen or partly frozen lakes. Object-based image analysis (OBIA) offers new opportunities for automated classification in this context, and we have explored this method for mapping lakes from LANDSAT images and Shuttle Radar Topography Mission (SRTM) elevation data. We tested our algorithm for most of the Tibetan Plateau, where lakes in tectonic depressions or blocked by glaciers and sediments have different surface colours and seasonal ice cover in images obtained in 1995 and 2015. We combined a modified normalised difference water index (MNDWI) with OBIA and local topographic slope data in order to classify lakes with an area >10 km2. Our method derived 323 water bodies, with a total area of 31,258 km2, or 2.6% of the study area (in 2015). The same number of lakes had covered only 24,892 km2 in 1995; lake area has increased by ~26% in the past two decades. The classification had estimated producer’s and user’s accuracies of 0.98, with a Cohen’s kappa and F-score of 0.98, and may thus be a useful approximation for quantifying regional hydrological budgets. We have shown that our method is flexible and transferable to detecting lakes in diverse physical settings on several continents with similar success rates. Full article
Figures

Open AccessArticle Stratified Template Matching to Support Refugee Camp Analysis in OBIA Workflows
Remote Sens. 2017, 9(4), 326; doi:10.3390/rs9040326
Received: 30 December 2016 / Revised: 22 March 2017 / Accepted: 27 March 2017 / Published: 30 March 2017
PDF Full-text (9642 KB) | HTML Full-text | XML Full-text
Abstract
Accurate and reliable information about the situation in refugee or internally displaced person camps is very important for planning any kind of help like health care, infrastructure, or vaccination campaigns. The number and spatial distribution of single dwellings extracted semi-automatically from very high-resolution
[...] Read more.
Accurate and reliable information about the situation in refugee or internally displaced person camps is very important for planning any kind of help like health care, infrastructure, or vaccination campaigns. The number and spatial distribution of single dwellings extracted semi-automatically from very high-resolution (VHR) satellite imagery as an indicator for population estimations can provide such important information. The accuracy of the extracted dwellings can vary quite a lot depending on various factors. To enhance established single dwelling extraction approaches, we have tested the integration of stratified template matching methods in object-based image analysis (OBIA) workflows. A template library for various dwelling types (template samples are taken from ten different sites using 16 satellite images), incorporating the shadow effect of dwellings, was established. Altogether, 18 template classes were created covering typically occurring dwellings and their cast shadows. The created template library aims to be generally applicable in similar conditions. Compared to pre-existing OBIA classifications, the approach could increase the producer’s accuracy by 11.7 percentage points on average and slightly increase the user’s accuracy. These results show that the stratified integration of template matching approaches in OBIA workflows is a possibility to further improve the results of semi-automated dwelling extraction, especially in complex situations. Full article
Figures

Open AccessArticle An Object-Based Semantic Classification Method for High Resolution Remote Sensing Imagery Using Ontology
Remote Sens. 2017, 9(4), 329; doi:10.3390/rs9040329
Received: 30 December 2016 / Revised: 17 March 2017 / Accepted: 24 March 2017 / Published: 30 March 2017
PDF Full-text (10139 KB) | HTML Full-text | XML Full-text
Abstract
Geographic Object-Based Image Analysis (GEOBIA) techniques have become increasingly popular in remote sensing. GEOBIA has been claimed to represent a paradigm shift in remote sensing interpretation. Still, GEOBIA—similar to other emerging paradigms—lacks formal expressions and objective modelling structures and in particular semantic classification
[...] Read more.
Geographic Object-Based Image Analysis (GEOBIA) techniques have become increasingly popular in remote sensing. GEOBIA has been claimed to represent a paradigm shift in remote sensing interpretation. Still, GEOBIA—similar to other emerging paradigms—lacks formal expressions and objective modelling structures and in particular semantic classification methods using ontologies. This study has put forward an object-based semantic classification method for high resolution satellite imagery using an ontology that aims to fully exploit the advantages of ontology to GEOBIA. A three-step workflow has been introduced: ontology modelling, initial classification based on a data-driven machine learning method, and semantic classification based on knowledge-driven semantic rules. The classification part is based on data-driven machine learning, segmentation, feature selection, sample collection and an initial classification. Then, image objects are re-classified based on the ontological model whereby the semantic relations are expressed in the formal languages OWL and SWRL. The results show that the method with ontology—as compared to the decision tree classification without using the ontology—yielded minor statistical improvements in terms of accuracy for this particular image. However, this framework enhances existing GEOBIA methodologies: ontologies express and organize the whole structure of GEOBIA and allow establishing relations, particularly spatially explicit relations between objects as well as multi-scale/hierarchical relations. Full article
Figures

Open AccessArticle Reproducibility and Practical Adoption of GEOBIA with Open-Source Software in Docker Containers
Remote Sens. 2017, 9(3), 290; doi:10.3390/rs9030290
Received: 30 December 2016 / Revised: 22 February 2017 / Accepted: 6 March 2017 / Published: 18 March 2017
PDF Full-text (15511 KB) | HTML Full-text | XML Full-text
Abstract
Geographic Object-Based Image Analysis (GEOBIA) mostly uses proprietary software,
but the interest in Free and Open-Source Software (FOSS) for GEOBIA is growing. This interest stems not only from cost savings, but also from benefits concerning reproducibility and collaboration. Technical challenges hamper practical reproducibility,
[...] Read more.
Geographic Object-Based Image Analysis (GEOBIA) mostly uses proprietary software,
but the interest in Free and Open-Source Software (FOSS) for GEOBIA is growing. This interest stems not only from cost savings, but also from benefits concerning reproducibility and collaboration. Technical challenges hamper practical reproducibility, especially when multiple software packages are required to conduct an analysis. In this study, we use containerization to package a GEOBIA workflow in a well-defined FOSS environment. We explore the approach using two software stacks to perform an exemplary analysis detecting destruction of buildings in bi-temporal images of a conflict area. The analysis combines feature extraction techniques with segmentation and object-based analysis to detect changes using automatically-defined local reference values and to distinguish disappeared buildings from non-target structures. The resulting workflow is published as FOSS comprising both the model and data in a ready to use Docker image and a user interface for interaction with the containerized workflow. The presented solution advances GEOBIA in the following aspects: higher transparency of methodology; easier reuse and adaption of workflows; better transferability between operating systems; complete description of the software environment; and easy application of workflows by image analysis experts and non-experts. As a result, it promotes not only the reproducibility of GEOBIA, but also its practical adoption. Full article
Figures

Open AccessArticle Geometric Refinement of ALS-Data Derived Building Models Using Monoscopic Aerial Images
Remote Sens. 2017, 9(3), 282; doi:10.3390/rs9030282
Received: 28 December 2016 / Revised: 9 March 2017 / Accepted: 12 March 2017 / Published: 16 March 2017
PDF Full-text (5806 KB) | HTML Full-text | XML Full-text
Abstract
Airborne laser scanning (ALS) has proven to be a strong basis for 3D building reconstruction. While ALS data allows for a highly automated processing workflow, a major drawback is often in the point spacing. As a consequence, the precision of roof plane and
[...] Read more.
Airborne laser scanning (ALS) has proven to be a strong basis for 3D building reconstruction. While ALS data allows for a highly automated processing workflow, a major drawback is often in the point spacing. As a consequence, the precision of roof plane and ridge line parameters is usually significantly better than the precision of gutter lines. To cope with this problem, the paper presents an approach for geometric refinement of building models reconstructed from ALS data using monoscopic aerial images. The core idea of the proposed modeling method is to obtain refined roof edges by intersecting roof planes accurately and reliably extracted from 3D point clouds with viewing planes assigned with building edges detected in a high resolution aerial image. In order to minimize ambiguities that may arise during the integration of modeling cues, the ALS data is used as the master providing initial information about building shape and topology. We evaluate the performance of our algorithm by comparing the results of 3D reconstruction executed using only laser scanning data and reconstruction enhanced by image information. The assessment performed within a framework of the International Society for Photogrammetry and Remote Sensing (ISPRS) benchmark shows an increase in the final quality indicator up to 8.7%. Full article
Figures

Open AccessArticle A Classification-Segmentation Framework for the Detection of Individual Trees in Dense MMS Point Cloud Data Acquired in Urban Areas
Remote Sens. 2017, 9(3), 277; doi:10.3390/rs9030277
Received: 29 December 2016 / Revised: 6 March 2017 / Accepted: 9 March 2017 / Published: 16 March 2017
Cited by 1 | PDF Full-text (4871 KB) | HTML Full-text | XML Full-text
Abstract
In this paper, we present a novel framework for detecting individual trees in densely sampled 3D point cloud data acquired in urban areas. Given a 3D point cloud, the objective is to assign point-wise labels that are both class-aware and instance-aware, a task
[...] Read more.
In this paper, we present a novel framework for detecting individual trees in densely sampled 3D point cloud data acquired in urban areas. Given a 3D point cloud, the objective is to assign point-wise labels that are both class-aware and instance-aware, a task that is known as instance-level segmentation. To achieve this, our framework addresses two successive steps. The first step of our framework is given by the use of geometric features for a binary point-wise semantic classification with the objective of assigning semantic class labels to irregularly distributed 3D points, whereby the labels are defined as “tree points” and “other points”. The second step of our framework is given by a semantic segmentation with the objective of separating individual trees within the “tree points”. This is achieved by applying an efficient adaptation of the mean shift algorithm and a subsequent segment-based shape analysis relying on semantic rules to only retain plausible tree segments. We demonstrate the performance of our framework on a publicly available benchmark dataset, which has been acquired with a mobile mapping system in the city of Delft in the Netherlands. This dataset contains 10.13 M labeled 3D points among which 17.6 % are labeled as “tree points”. The derived results clearly reveal a semantic classification of high accuracy (up to 90.77 %) and an instance-level segmentation of high plausibility, while the simplicity, applicability and efficiency of the involved methods even allow applying the complete framework on a standard laptop computer with a reasonable processing time (less than 2.5 h). Full article
Figures

Open AccessArticle Fast Segmentation and Classification of Very High Resolution Remote Sensing Data Using SLIC Superpixels
Remote Sens. 2017, 9(3), 243; doi:10.3390/rs9030243
Received: 31 December 2016 / Revised: 21 February 2017 / Accepted: 2 March 2017 / Published: 5 March 2017
Cited by 1 | PDF Full-text (32422 KB) | HTML Full-text | XML Full-text
Abstract
Speed and accuracy are important factors when dealing with time-constraint events for disaster, risk, and crisis-management support. Object-based image analysis can be a time consuming task in extracting information from large images because most of the segmentation algorithms use the pixel-grid for the
[...] Read more.
Speed and accuracy are important factors when dealing with time-constraint events for disaster, risk, and crisis-management support. Object-based image analysis can be a time consuming task in extracting information from large images because most of the segmentation algorithms use the pixel-grid for the initial object representation. It would be more natural and efficient to work with perceptually meaningful entities that are derived from pixels using a low-level grouping process (superpixels). Firstly, we tested a new workflow for image segmentation of remote sensing data, starting the multiresolution segmentation (MRS, using ESP2 tool) from the superpixel level and aiming at reducing the amount of time needed to automatically partition relatively large datasets of very high resolution remote sensing data. Secondly, we examined whether a Random Forest classification based on an oversegmentation produced by a Simple Linear Iterative Clustering (SLIC) superpixel algorithm performs similarly with reference to a traditional object-based classification regarding accuracy. Tests were applied on QuickBird and WorldView-2 data with different extents, scene content complexities, and number of bands to assess how the computational time and classification accuracy are affected by these factors. The proposed segmentation approach is compared with the traditional one, starting the MRS from the pixel level, regarding geometric accuracy of the objects and the computational time. The computational time was reduced in all cases, the biggest improvement being from 5 h 35 min to 13 min, for a WorldView-2 scene with eight bands and an extent of 12.2 million pixels, while the geometric accuracy is kept similar or slightly better. SLIC superpixel-based classification had similar or better overall accuracy values when compared to MRS-based classification, but the results were obtained in a fast manner and avoiding the parameterization of the MRS. These two approaches have the potential to enhance the automation of big remote sensing data analysis and processing, especially when time is an important constraint. Full article
Figures

Open AccessArticle Generating Topographic Map Data from Classification Results
Remote Sens. 2017, 9(3), 224; doi:10.3390/rs9030224
Received: 30 December 2016 / Accepted: 25 February 2017 / Published: 2 March 2017
PDF Full-text (5481 KB) | HTML Full-text | XML Full-text
Abstract
The use of classification results as topographic map data requires cartographic enhancement and checking of the geometric accuracy. Urban areas are of special interest. The conversion of the classification result into topographic map data of high thematic and geometric quality is subject of
[...] Read more.
The use of classification results as topographic map data requires cartographic enhancement and checking of the geometric accuracy. Urban areas are of special interest. The conversion of the classification result into topographic map data of high thematic and geometric quality is subject of this contribution. After reviewing the existing literature on this topic, a methodology is presented. The extraction of point clouds belonging to line segments is solved by the Hough transform. The mathematics for deriving polygons of orthogonal, parallel and general line segments by least squares adjustment is presented. A unique solution for polylines, where the Hough parameters are optimized, is also given. By means of two data sets land cover maps of six classes were produced and then enhanced by the proposed method. The classification used the decision tree method applying a variety of attributes including object heights derived from imagery. The cartographic enhancement is carried out with two different levels of quality. The user’s accuracies for the classes “impervious surface” and “building” were above 85% in the “Level 1” map of Example 1. The geometric accuracy of building corners at the “Level 2” maps is assessed by means of reference data derived from ortho-images. The obtained root mean square errors (RMSE) of the generated coordinates (x, y) were RMSEx = 1.2 m and RMSEy = 0.7 m (Example 1) and RMSEx = 0.8 m and RMSEy = 1.0 m (Example 2) using 31 and 62 check points, respectively. All processing for Level 1 (raster data) could be carried out with a high degree of automation. Level 2 maps (vector data) were compiled for the classes “building” and “road and parking lot”. For urban areas with numerous classes and of large size, universal algorithms are necessary to produce vector data fully automatically. The recent progress in sensors and machine learning methods will support the generation of topographic map data of high thematic and geometric accuracy. Full article
Figures

Open AccessArticle Scalable Bag of Subpaths Kernel for Learning on Hierarchical Image Representations and Multi-Source Remote Sensing Data Classification
Remote Sens. 2017, 9(3), 196; doi:10.3390/rs9030196
Received: 31 December 2016 / Revised: 9 February 2017 / Accepted: 15 February 2017 / Published: 24 February 2017
PDF Full-text (9572 KB) | HTML Full-text | XML Full-text
Abstract
The geographic object-based image analysis (GEOBIA) framework has gained increasing interest for the last decade. One of its key advantages is the hierarchical representation of an image, where object topological features can be extracted and modeled in the form of structured data. We
[...] Read more.
The geographic object-based image analysis (GEOBIA) framework has gained increasing interest for the last decade. One of its key advantages is the hierarchical representation of an image, where object topological features can be extracted and modeled in the form of structured data. We thus propose to use a structured kernel relying on the concept of bag of subpaths to directly cope with such features. The kernel can be approximated using random Fourier features, allowing it to be applied on a large structure size (the number of objects in the structured data) and large volumes of data (the number of pixels or regions for training). With the so-called scalable bag of subpaths kernel (SBoSK), we also introduce a novel multi-source classification approach performing machine learning directly on a hierarchical image representation built from two images at different resolutions under the GEOBIA framework. Experiments run on an urban classification task show that the proposed approach run on a single image improves the classification overall accuracy in comparison with conventional approaches from 2% to 5% depending on the training set size and that fusing two images allows a supplementary 4% accuracy gain. Additional evaluations on public available large-scale datasets illustrate further the potential of SBoSK, with overall accuracy rates improvement ranging from 1% to 11% depending on the considered setup. Full article
Figures

Journal Contact

MDPI AG
Remote Sensing Editorial Office
St. Alban-Anlage 66, 4052 Basel, Switzerland
E-Mail: 
Tel. +41 61 683 77 34
Fax: +41 61 302 89 18
Editorial Board
Contact Details Submit to Remote Sensing Edit a special issue Review for Remote Sensing
loading...
Back to Top