Land cover classification of RADARSAT-2 SAR data using convolutional neural network View Full Text


Ontology type: schema:ScholarlyArticle     


Article Info

DATE

2016-04

AUTHORS

Wei Lin, Xiangyong Liao, Juan Deng, Yao Liu

ABSTRACT

In this paper, we propose a convolutional neural network (CNN) based on deep learning method for land cover classification of synthetic aperture radar (SAR) images. The proposed method consists of convolutional layers, pooling layers, a full connection layer and an output layer. The method acquires high-level abstractions for SAR data by using a hierarchical architecture composed of multiple non-linear transformations such as convolutions and poolings. The feature maps produced by convolutional layers are subsampled by pooling layers and then are converted into a feature vector by the full connection layer. The feature vector is then used by the output layer with softmax regression to perform land cover classification. The multi-layer method replaces hand-engineered features with backpropagation (BP) neural network algorithm for supervised feature learning, hierarchical feature extraction and land cover classification of SAR images. RADARSAT-2 ultra-fine beam high resolution HH-SAR images acquired in the rural urban fringe of the Greater Toronto Area (GTA) are selected for this study. The experiment results show that the accuracy of our classification method is about 90% which is higher than that of nearest neighbor (NN). More... »

PAGES

151-158

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/s11859-016-1152-y

DOI

http://dx.doi.org/10.1007/s11859-016-1152-y

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1091388457


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Information and Computing Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Artificial Intelligence and Image Processing", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "International School of Software, Wuhan University, Wuhan 430072, Hubei, China", 
          "id": "http://www.grid.ac/institutes/grid.49470.3e", 
          "name": [
            "International School of Software, Wuhan University, Wuhan 430072, Hubei, China"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Lin", 
        "givenName": "Wei", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "International School of Software, Wuhan University, Wuhan 430072, Hubei, China", 
          "id": "http://www.grid.ac/institutes/grid.49470.3e", 
          "name": [
            "International School of Software, Wuhan University, Wuhan 430072, Hubei, China"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Liao", 
        "givenName": "Xiangyong", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "International School of Software, Wuhan University, Wuhan 430072, Hubei, China", 
          "id": "http://www.grid.ac/institutes/grid.49470.3e", 
          "name": [
            "International School of Software, Wuhan University, Wuhan 430072, Hubei, China"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Deng", 
        "givenName": "Juan", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "International School of Software, Wuhan University, Wuhan 430072, Hubei, China", 
          "id": "http://www.grid.ac/institutes/grid.49470.3e", 
          "name": [
            "International School of Software, Wuhan University, Wuhan 430072, Hubei, China"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Liu", 
        "givenName": "Yao", 
        "type": "Person"
      }
    ], 
    "datePublished": "2016-04", 
    "datePublishedReg": "2016-04-01", 
    "description": "In this paper, we propose a convolutional neural network (CNN) based on deep learning method for land cover classification of synthetic aperture radar (SAR) images. The proposed method consists of convolutional layers, pooling layers, a full connection layer and an output layer. The method acquires high-level abstractions for SAR data by using a hierarchical architecture composed of multiple non-linear transformations such as convolutions and poolings. The feature maps produced by convolutional layers are subsampled by pooling layers and then are converted into a feature vector by the full connection layer. The feature vector is then used by the output layer with softmax regression to perform land cover classification. The multi-layer method replaces hand-engineered features with backpropagation (BP) neural network algorithm for supervised feature learning, hierarchical feature extraction and land cover classification of SAR images. RADARSAT-2 ultra-fine beam high resolution HH-SAR images acquired in the rural urban fringe of the Greater Toronto Area (GTA) are selected for this study. The experiment results show that the accuracy of our classification method is about 90% which is higher than that of nearest neighbor (NN).", 
    "genre": "article", 
    "id": "sg:pub.10.1007/s11859-016-1152-y", 
    "isAccessibleForFree": false, 
    "isPartOf": [
      {
        "id": "sg:journal.1135877", 
        "issn": [
          "1007-1202", 
          "1993-4998"
        ], 
        "name": "Wuhan University Journal of Natural Sciences", 
        "publisher": "Springer Nature", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "2", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "21"
      }
    ], 
    "keywords": [
      "convolutional neural network", 
      "full connection layer", 
      "land cover classification", 
      "convolutional layers", 
      "cover classification", 
      "connection layer", 
      "neural network", 
      "feature vectors", 
      "output layer", 
      "backpropagation neural network algorithm", 
      "high-level abstractions", 
      "hand-engineered features", 
      "supervised feature learning", 
      "multiple non-linear transformations", 
      "deep learning methods", 
      "hierarchical feature extraction", 
      "nearest neighbors", 
      "neural network algorithm", 
      "synthetic aperture radar (SAR) images", 
      "feature learning", 
      "softmax regression", 
      "feature maps", 
      "feature extraction", 
      "network algorithm", 
      "multi-layer method", 
      "learning methods", 
      "SAR data", 
      "aperture radar images", 
      "classification method", 
      "hierarchical architecture", 
      "SAR images", 
      "experiment results", 
      "RADARSAT-2 SAR data", 
      "non-linear transformation", 
      "radar images", 
      "classification", 
      "images", 
      "network", 
      "layer", 
      "architecture", 
      "algorithm", 
      "abstraction", 
      "convolution", 
      "learning", 
      "neighbors", 
      "pooling", 
      "accuracy", 
      "method", 
      "vector", 
      "Greater Toronto Area", 
      "extraction", 
      "data", 
      "maps", 
      "features", 
      "fringes", 
      "rural-urban fringe", 
      "Toronto Area", 
      "results", 
      "transformation", 
      "area", 
      "regression", 
      "study", 
      "urban fringe", 
      "paper"
    ], 
    "name": "Land cover classification of RADARSAT-2 SAR data using convolutional neural network", 
    "pagination": "151-158", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1091388457"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/s11859-016-1152-y"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1007/s11859-016-1152-y", 
      "https://app.dimensions.ai/details/publication/pub.1091388457"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2022-08-04T17:03", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20220804/entities/gbq_results/article/article_689.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://doi.org/10.1007/s11859-016-1152-y"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/s11859-016-1152-y'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/s11859-016-1152-y'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s11859-016-1152-y'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s11859-016-1152-y'


 

This table displays all metadata directly associated to this object as RDF triples.

138 TRIPLES      20 PREDICATES      89 URIs      81 LITERALS      6 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/s11859-016-1152-y schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N4c88bc0e86684972bc11a660378bf996
4 schema:datePublished 2016-04
5 schema:datePublishedReg 2016-04-01
6 schema:description In this paper, we propose a convolutional neural network (CNN) based on deep learning method for land cover classification of synthetic aperture radar (SAR) images. The proposed method consists of convolutional layers, pooling layers, a full connection layer and an output layer. The method acquires high-level abstractions for SAR data by using a hierarchical architecture composed of multiple non-linear transformations such as convolutions and poolings. The feature maps produced by convolutional layers are subsampled by pooling layers and then are converted into a feature vector by the full connection layer. The feature vector is then used by the output layer with softmax regression to perform land cover classification. The multi-layer method replaces hand-engineered features with backpropagation (BP) neural network algorithm for supervised feature learning, hierarchical feature extraction and land cover classification of SAR images. RADARSAT-2 ultra-fine beam high resolution HH-SAR images acquired in the rural urban fringe of the Greater Toronto Area (GTA) are selected for this study. The experiment results show that the accuracy of our classification method is about 90% which is higher than that of nearest neighbor (NN).
7 schema:genre article
8 schema:isAccessibleForFree false
9 schema:isPartOf N2e5d9d0ba87f44b880b900b2c032a18b
10 Ndd4467fe31d4435a8e529dfb3f577b6c
11 sg:journal.1135877
12 schema:keywords Greater Toronto Area
13 RADARSAT-2 SAR data
14 SAR data
15 SAR images
16 Toronto Area
17 abstraction
18 accuracy
19 algorithm
20 aperture radar images
21 architecture
22 area
23 backpropagation neural network algorithm
24 classification
25 classification method
26 connection layer
27 convolution
28 convolutional layers
29 convolutional neural network
30 cover classification
31 data
32 deep learning methods
33 experiment results
34 extraction
35 feature extraction
36 feature learning
37 feature maps
38 feature vectors
39 features
40 fringes
41 full connection layer
42 hand-engineered features
43 hierarchical architecture
44 hierarchical feature extraction
45 high-level abstractions
46 images
47 land cover classification
48 layer
49 learning
50 learning methods
51 maps
52 method
53 multi-layer method
54 multiple non-linear transformations
55 nearest neighbors
56 neighbors
57 network
58 network algorithm
59 neural network
60 neural network algorithm
61 non-linear transformation
62 output layer
63 paper
64 pooling
65 radar images
66 regression
67 results
68 rural-urban fringe
69 softmax regression
70 study
71 supervised feature learning
72 synthetic aperture radar (SAR) images
73 transformation
74 urban fringe
75 vector
76 schema:name Land cover classification of RADARSAT-2 SAR data using convolutional neural network
77 schema:pagination 151-158
78 schema:productId Na8d410fef9904e828eab2ddc11e9ba30
79 Ne31b85b47853493b88bc438bb7543967
80 schema:sameAs https://app.dimensions.ai/details/publication/pub.1091388457
81 https://doi.org/10.1007/s11859-016-1152-y
82 schema:sdDatePublished 2022-08-04T17:03
83 schema:sdLicense https://scigraph.springernature.com/explorer/license/
84 schema:sdPublisher N4ebdd8b04b98488abe8b419f871fedfe
85 schema:url https://doi.org/10.1007/s11859-016-1152-y
86 sgo:license sg:explorer/license/
87 sgo:sdDataset articles
88 rdf:type schema:ScholarlyArticle
89 N2cf1d4c14607412c957e56b33bf9fac5 schema:affiliation grid-institutes:grid.49470.3e
90 schema:familyName Deng
91 schema:givenName Juan
92 rdf:type schema:Person
93 N2e5d9d0ba87f44b880b900b2c032a18b schema:issueNumber 2
94 rdf:type schema:PublicationIssue
95 N4c88bc0e86684972bc11a660378bf996 rdf:first N8b49a2a639284920b9f8bddd2a1d7ba1
96 rdf:rest Na8b4442a2c1442f28d4bbefeb5b617db
97 N4ebdd8b04b98488abe8b419f871fedfe schema:name Springer Nature - SN SciGraph project
98 rdf:type schema:Organization
99 N8b49a2a639284920b9f8bddd2a1d7ba1 schema:affiliation grid-institutes:grid.49470.3e
100 schema:familyName Lin
101 schema:givenName Wei
102 rdf:type schema:Person
103 Na8b4442a2c1442f28d4bbefeb5b617db rdf:first Nd47bfa0f954047f9965797461300ff23
104 rdf:rest Ne2f8b4fea4a444bba0f6807fb9f722af
105 Na8d410fef9904e828eab2ddc11e9ba30 schema:name dimensions_id
106 schema:value pub.1091388457
107 rdf:type schema:PropertyValue
108 Nb54d69cb8062409c9f395397c10f65af rdf:first Nd76851671a3f4383a43d8b29bb71ac16
109 rdf:rest rdf:nil
110 Nd47bfa0f954047f9965797461300ff23 schema:affiliation grid-institutes:grid.49470.3e
111 schema:familyName Liao
112 schema:givenName Xiangyong
113 rdf:type schema:Person
114 Nd76851671a3f4383a43d8b29bb71ac16 schema:affiliation grid-institutes:grid.49470.3e
115 schema:familyName Liu
116 schema:givenName Yao
117 rdf:type schema:Person
118 Ndd4467fe31d4435a8e529dfb3f577b6c schema:volumeNumber 21
119 rdf:type schema:PublicationVolume
120 Ne2f8b4fea4a444bba0f6807fb9f722af rdf:first N2cf1d4c14607412c957e56b33bf9fac5
121 rdf:rest Nb54d69cb8062409c9f395397c10f65af
122 Ne31b85b47853493b88bc438bb7543967 schema:name doi
123 schema:value 10.1007/s11859-016-1152-y
124 rdf:type schema:PropertyValue
125 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
126 schema:name Information and Computing Sciences
127 rdf:type schema:DefinedTerm
128 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
129 schema:name Artificial Intelligence and Image Processing
130 rdf:type schema:DefinedTerm
131 sg:journal.1135877 schema:issn 1007-1202
132 1993-4998
133 schema:name Wuhan University Journal of Natural Sciences
134 schema:publisher Springer Nature
135 rdf:type schema:Periodical
136 grid-institutes:grid.49470.3e schema:alternateName International School of Software, Wuhan University, Wuhan 430072, Hubei, China
137 schema:name International School of Software, Wuhan University, Wuhan 430072, Hubei, China
138 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...