Squeezed fire binary segmentation model using convolutional neural network for outdoor images on embedded device View Full Text


Ontology type: schema:ScholarlyArticle     


Article Info

DATE

2021-09-27

AUTHORS

Kyungmin Song, Han-Soo Choi, Myungjoo Kang

ABSTRACT

Even though image-based prediction of fire events is widely used, the current predictive methods are difficult to implement due to low performance and high specifications. In this work designed to overcome such problems, we propose binary semantic segmentation for fire images by employing deep learning that can be applied to embedded devices such as Jetson TX2. To reduce the parameters and consequently the model size while maintaining the performance, we replaced regular convolution with depthwise separable convolution and 1×1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$1 \times 1 $$\end{document} convolution. Moreover, the addition operation in the long skip connection was replaced with the concatenation operation to properly convey the information in the encoding phase. Besides, we propose the confusion block that can execute the model to proceed training more actively. From these approaches, we achieved a significantly small-sized network for fire segmentation with the highest performance. We compared the performance of the proposed method with various deep learning-based binary segmentation networks and image processing algorithm. Extensive experimental results on the FiSmo Dataset and Corsican Fire Database demonstrated that the proposed network outperforms other models with fewer parameters and is suitable for application in embedded devices. More... »

PAGES

120

References to SciGraph publications

  • 2018-06-02. Fire Detection Algorithm Combined with Image Processing and Flame Emission Spectroscopy in FIRE TECHNOLOGY
  • 2016-09-27. The Importance of Skip Connections in Biomedical Image Segmentation in DEEP LEARNING AND DATA LABELING FOR MEDICAL APPLICATIONS
  • 2019-02-27. Smoke Detection on Video Sequences Using 3D Convolutional Neural Networks in FIRE TECHNOLOGY
  • 2014. Visualizing and Understanding Convolutional Networks in COMPUTER VISION – ECCV 2014
  • 2014. Microsoft COCO: Common Objects in Context in COMPUTER VISION – ECCV 2014
  • 2015-11-18. U-Net: Convolutional Networks for Biomedical Image Segmentation in MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION – MICCAI 2015
  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1007/s00138-021-01242-1

    DOI

    http://dx.doi.org/10.1007/s00138-021-01242-1

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1141419905


    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/2008/17", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Psychology and Cognitive 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"
          }, 
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/1701", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Psychology", 
            "type": "DefinedTerm"
          }
        ], 
        "author": [
          {
            "affiliation": {
              "alternateName": "Optimization & Analytics, SK Innovation, Seoul, Republic of Korea", 
              "id": "http://www.grid.ac/institutes/grid.507563.2", 
              "name": [
                "Optimization & Analytics, SK Innovation, Seoul, Republic of Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Song", 
            "givenName": "Kyungmin", 
            "id": "sg:person.0602362310.21", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0602362310.21"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Research Institute of Mathematics, Seoul National University, Seoul, Republic of Korea", 
              "id": "http://www.grid.ac/institutes/grid.31501.36", 
              "name": [
                "Research Institute of Mathematics, Seoul National University, Seoul, Republic of Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Choi", 
            "givenName": "Han-Soo", 
            "id": "sg:person.013326206753.08", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013326206753.08"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Mathematical Science, Seoul National University, GwanakRo 1, 151-747, Gwanak-Gu, Seoul, Korea", 
              "id": "http://www.grid.ac/institutes/grid.31501.36", 
              "name": [
                "Mathematical Science, Seoul National University, GwanakRo 1, 151-747, Gwanak-Gu, Seoul, Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Kang", 
            "givenName": "Myungjoo", 
            "id": "sg:person.014467157631.45", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014467157631.45"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "sg:pub.10.1007/978-3-319-10590-1_53", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1032233097", 
              "https://doi.org/10.1007/978-3-319-10590-1_53"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s10694-018-0727-x", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1104370356", 
              "https://doi.org/10.1007/s10694-018-0727-x"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-319-46976-8_19", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1084920305", 
              "https://doi.org/10.1007/978-3-319-46976-8_19"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-319-24574-4_28", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1017774818", 
              "https://doi.org/10.1007/978-3-319-24574-4_28"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s10694-019-00832-w", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1112435274", 
              "https://doi.org/10.1007/s10694-019-00832-w"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-319-10602-1_48", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1045321436", 
              "https://doi.org/10.1007/978-3-319-10602-1_48"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2021-09-27", 
        "datePublishedReg": "2021-09-27", 
        "description": "Even though image-based prediction of fire events is widely used, the current predictive methods are difficult to implement due to low performance and high specifications. In this work designed to overcome such problems, we propose binary semantic segmentation for fire images by employing deep learning that can be applied to embedded devices such as Jetson TX2. To reduce the parameters and consequently the model size while maintaining the performance, we replaced regular convolution with depthwise separable convolution and 1\u00d71\\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym}\n\t\t\t\t\\usepackage{amsfonts}\n\t\t\t\t\\usepackage{amssymb}\n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$1 \\times 1 $$\\end{document} convolution. Moreover, the addition operation in the long skip connection was replaced with the concatenation operation to properly convey the information in the encoding phase. Besides, we propose the confusion block that can execute the model to proceed training more actively. From these approaches, we achieved a significantly small-sized network for fire segmentation with the highest performance. We compared the performance of the proposed method with various deep learning-based binary segmentation networks and image processing algorithm. Extensive experimental results on the FiSmo Dataset and Corsican Fire Database demonstrated that the proposed network outperforms other models with fewer parameters and is suitable for application in embedded devices.", 
        "genre": "article", 
        "id": "sg:pub.10.1007/s00138-021-01242-1", 
        "inLanguage": "en", 
        "isAccessibleForFree": false, 
        "isPartOf": [
          {
            "id": "sg:journal.1045266", 
            "issn": [
              "0932-8092", 
              "1432-1769"
            ], 
            "name": "Machine Vision and Applications", 
            "publisher": "Springer Nature", 
            "type": "Periodical"
          }, 
          {
            "issueNumber": "6", 
            "type": "PublicationIssue"
          }, 
          {
            "type": "PublicationVolume", 
            "volumeNumber": "32"
          }
        ], 
        "keywords": [
          "binary semantic segmentation", 
          "depthwise separable convolution", 
          "convolutional neural network", 
          "long skip connections", 
          "image processing algorithms", 
          "Extensive experimental results", 
          "image-based prediction", 
          "small-sized networks", 
          "Jetson TX2", 
          "fire segmentation", 
          "separable convolution", 
          "semantic segmentation", 
          "segmentation network", 
          "deep learning", 
          "skip connections", 
          "fire images", 
          "outdoor images", 
          "segmentation model", 
          "neural network", 
          "regular convolution", 
          "processing algorithms", 
          "model size", 
          "concatenation operation", 
          "addition operation", 
          "current predictive methods", 
          "network", 
          "high performance", 
          "segmentation", 
          "such problems", 
          "experimental results", 
          "convolution", 
          "high specification", 
          "images", 
          "lower performance", 
          "TX2", 
          "performance", 
          "devices", 
          "algorithm", 
          "dataset", 
          "fire database", 
          "predictive methods", 
          "specification", 
          "learning", 
          "operation", 
          "database", 
          "model", 
          "information", 
          "applications", 
          "method", 
          "training", 
          "work", 
          "block", 
          "prediction", 
          "fire events", 
          "connection", 
          "parameters", 
          "results", 
          "size", 
          "events", 
          "phase", 
          "problem", 
          "approach", 
          "confusion block", 
          "deep learning-based binary segmentation networks", 
          "learning-based binary segmentation networks", 
          "binary segmentation networks", 
          "FiSmo Dataset", 
          "Corsican Fire Database", 
          "Squeezed fire binary segmentation model", 
          "fire binary segmentation model", 
          "binary segmentation model"
        ], 
        "name": "Squeezed fire binary segmentation model using convolutional neural network for outdoor images on embedded device", 
        "pagination": "120", 
        "productId": [
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1141419905"
            ]
          }, 
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1007/s00138-021-01242-1"
            ]
          }
        ], 
        "sameAs": [
          "https://doi.org/10.1007/s00138-021-01242-1", 
          "https://app.dimensions.ai/details/publication/pub.1141419905"
        ], 
        "sdDataset": "articles", 
        "sdDatePublished": "2022-01-01T19:03", 
        "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
        "sdPublisher": {
          "name": "Springer Nature - SN SciGraph project", 
          "type": "Organization"
        }, 
        "sdSource": "s3://com-springernature-scigraph/baseset/20220101/entities/gbq_results/article/article_916.jsonl", 
        "type": "ScholarlyArticle", 
        "url": "https://doi.org/10.1007/s00138-021-01242-1"
      }
    ]
     

    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/s00138-021-01242-1'

    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/s00138-021-01242-1'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s00138-021-01242-1'

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

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s00138-021-01242-1'


     

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

    180 TRIPLES      22 PREDICATES      104 URIs      88 LITERALS      6 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1007/s00138-021-01242-1 schema:about anzsrc-for:08
    2 anzsrc-for:0801
    3 anzsrc-for:17
    4 anzsrc-for:1701
    5 schema:author N762b77a3ba994607a3f4656088bd6107
    6 schema:citation sg:pub.10.1007/978-3-319-10590-1_53
    7 sg:pub.10.1007/978-3-319-10602-1_48
    8 sg:pub.10.1007/978-3-319-24574-4_28
    9 sg:pub.10.1007/978-3-319-46976-8_19
    10 sg:pub.10.1007/s10694-018-0727-x
    11 sg:pub.10.1007/s10694-019-00832-w
    12 schema:datePublished 2021-09-27
    13 schema:datePublishedReg 2021-09-27
    14 schema:description Even though image-based prediction of fire events is widely used, the current predictive methods are difficult to implement due to low performance and high specifications. In this work designed to overcome such problems, we propose binary semantic segmentation for fire images by employing deep learning that can be applied to embedded devices such as Jetson TX2. To reduce the parameters and consequently the model size while maintaining the performance, we replaced regular convolution with depthwise separable convolution and 1×1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$1 \times 1 $$\end{document} convolution. Moreover, the addition operation in the long skip connection was replaced with the concatenation operation to properly convey the information in the encoding phase. Besides, we propose the confusion block that can execute the model to proceed training more actively. From these approaches, we achieved a significantly small-sized network for fire segmentation with the highest performance. We compared the performance of the proposed method with various deep learning-based binary segmentation networks and image processing algorithm. Extensive experimental results on the FiSmo Dataset and Corsican Fire Database demonstrated that the proposed network outperforms other models with fewer parameters and is suitable for application in embedded devices.
    15 schema:genre article
    16 schema:inLanguage en
    17 schema:isAccessibleForFree false
    18 schema:isPartOf N75797ff8030d41efa200156cfd7f6e5e
    19 N89c4384f0e184821be60ff420fceca57
    20 sg:journal.1045266
    21 schema:keywords Corsican Fire Database
    22 Extensive experimental results
    23 FiSmo Dataset
    24 Jetson TX2
    25 Squeezed fire binary segmentation model
    26 TX2
    27 addition operation
    28 algorithm
    29 applications
    30 approach
    31 binary segmentation model
    32 binary segmentation networks
    33 binary semantic segmentation
    34 block
    35 concatenation operation
    36 confusion block
    37 connection
    38 convolution
    39 convolutional neural network
    40 current predictive methods
    41 database
    42 dataset
    43 deep learning
    44 deep learning-based binary segmentation networks
    45 depthwise separable convolution
    46 devices
    47 events
    48 experimental results
    49 fire binary segmentation model
    50 fire database
    51 fire events
    52 fire images
    53 fire segmentation
    54 high performance
    55 high specification
    56 image processing algorithms
    57 image-based prediction
    58 images
    59 information
    60 learning
    61 learning-based binary segmentation networks
    62 long skip connections
    63 lower performance
    64 method
    65 model
    66 model size
    67 network
    68 neural network
    69 operation
    70 outdoor images
    71 parameters
    72 performance
    73 phase
    74 prediction
    75 predictive methods
    76 problem
    77 processing algorithms
    78 regular convolution
    79 results
    80 segmentation
    81 segmentation model
    82 segmentation network
    83 semantic segmentation
    84 separable convolution
    85 size
    86 skip connections
    87 small-sized networks
    88 specification
    89 such problems
    90 training
    91 work
    92 schema:name Squeezed fire binary segmentation model using convolutional neural network for outdoor images on embedded device
    93 schema:pagination 120
    94 schema:productId Nf0d1c5f66cbc41079a8560bd7775974d
    95 Nf49e9719d7624ca0a70cbde49c9fe259
    96 schema:sameAs https://app.dimensions.ai/details/publication/pub.1141419905
    97 https://doi.org/10.1007/s00138-021-01242-1
    98 schema:sdDatePublished 2022-01-01T19:03
    99 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    100 schema:sdPublisher Ndeea4c558a3c4a67b5e19337879be2c7
    101 schema:url https://doi.org/10.1007/s00138-021-01242-1
    102 sgo:license sg:explorer/license/
    103 sgo:sdDataset articles
    104 rdf:type schema:ScholarlyArticle
    105 N3b4fa71539214f9d89f22dbcc08eb494 rdf:first sg:person.013326206753.08
    106 rdf:rest Nbf6ad694d0344dfa8a1c947bb251d63c
    107 N75797ff8030d41efa200156cfd7f6e5e schema:volumeNumber 32
    108 rdf:type schema:PublicationVolume
    109 N762b77a3ba994607a3f4656088bd6107 rdf:first sg:person.0602362310.21
    110 rdf:rest N3b4fa71539214f9d89f22dbcc08eb494
    111 N89c4384f0e184821be60ff420fceca57 schema:issueNumber 6
    112 rdf:type schema:PublicationIssue
    113 Nbf6ad694d0344dfa8a1c947bb251d63c rdf:first sg:person.014467157631.45
    114 rdf:rest rdf:nil
    115 Ndeea4c558a3c4a67b5e19337879be2c7 schema:name Springer Nature - SN SciGraph project
    116 rdf:type schema:Organization
    117 Nf0d1c5f66cbc41079a8560bd7775974d schema:name dimensions_id
    118 schema:value pub.1141419905
    119 rdf:type schema:PropertyValue
    120 Nf49e9719d7624ca0a70cbde49c9fe259 schema:name doi
    121 schema:value 10.1007/s00138-021-01242-1
    122 rdf:type schema:PropertyValue
    123 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
    124 schema:name Information and Computing Sciences
    125 rdf:type schema:DefinedTerm
    126 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
    127 schema:name Artificial Intelligence and Image Processing
    128 rdf:type schema:DefinedTerm
    129 anzsrc-for:17 schema:inDefinedTermSet anzsrc-for:
    130 schema:name Psychology and Cognitive Sciences
    131 rdf:type schema:DefinedTerm
    132 anzsrc-for:1701 schema:inDefinedTermSet anzsrc-for:
    133 schema:name Psychology
    134 rdf:type schema:DefinedTerm
    135 sg:journal.1045266 schema:issn 0932-8092
    136 1432-1769
    137 schema:name Machine Vision and Applications
    138 schema:publisher Springer Nature
    139 rdf:type schema:Periodical
    140 sg:person.013326206753.08 schema:affiliation grid-institutes:grid.31501.36
    141 schema:familyName Choi
    142 schema:givenName Han-Soo
    143 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013326206753.08
    144 rdf:type schema:Person
    145 sg:person.014467157631.45 schema:affiliation grid-institutes:grid.31501.36
    146 schema:familyName Kang
    147 schema:givenName Myungjoo
    148 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014467157631.45
    149 rdf:type schema:Person
    150 sg:person.0602362310.21 schema:affiliation grid-institutes:grid.507563.2
    151 schema:familyName Song
    152 schema:givenName Kyungmin
    153 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0602362310.21
    154 rdf:type schema:Person
    155 sg:pub.10.1007/978-3-319-10590-1_53 schema:sameAs https://app.dimensions.ai/details/publication/pub.1032233097
    156 https://doi.org/10.1007/978-3-319-10590-1_53
    157 rdf:type schema:CreativeWork
    158 sg:pub.10.1007/978-3-319-10602-1_48 schema:sameAs https://app.dimensions.ai/details/publication/pub.1045321436
    159 https://doi.org/10.1007/978-3-319-10602-1_48
    160 rdf:type schema:CreativeWork
    161 sg:pub.10.1007/978-3-319-24574-4_28 schema:sameAs https://app.dimensions.ai/details/publication/pub.1017774818
    162 https://doi.org/10.1007/978-3-319-24574-4_28
    163 rdf:type schema:CreativeWork
    164 sg:pub.10.1007/978-3-319-46976-8_19 schema:sameAs https://app.dimensions.ai/details/publication/pub.1084920305
    165 https://doi.org/10.1007/978-3-319-46976-8_19
    166 rdf:type schema:CreativeWork
    167 sg:pub.10.1007/s10694-018-0727-x schema:sameAs https://app.dimensions.ai/details/publication/pub.1104370356
    168 https://doi.org/10.1007/s10694-018-0727-x
    169 rdf:type schema:CreativeWork
    170 sg:pub.10.1007/s10694-019-00832-w schema:sameAs https://app.dimensions.ai/details/publication/pub.1112435274
    171 https://doi.org/10.1007/s10694-019-00832-w
    172 rdf:type schema:CreativeWork
    173 grid-institutes:grid.31501.36 schema:alternateName Mathematical Science, Seoul National University, GwanakRo 1, 151-747, Gwanak-Gu, Seoul, Korea
    174 Research Institute of Mathematics, Seoul National University, Seoul, Republic of Korea
    175 schema:name Mathematical Science, Seoul National University, GwanakRo 1, 151-747, Gwanak-Gu, Seoul, Korea
    176 Research Institute of Mathematics, Seoul National University, Seoul, Republic of Korea
    177 rdf:type schema:Organization
    178 grid-institutes:grid.507563.2 schema:alternateName Optimization & Analytics, SK Innovation, Seoul, Republic of Korea
    179 schema:name Optimization & Analytics, SK Innovation, Seoul, Republic of Korea
    180 rdf:type schema:Organization
     




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


    ...