STDNet: A CNN-based approach to single-/mixed-script detection View Full Text


Ontology type: schema:ScholarlyArticle     


Article Info

DATE

2021-04-27

AUTHORS

Mridul Ghosh, Himadri Mukherjee, Sk Md Obaidullah, Kaushik Roy

ABSTRACT

Script identification serves as a guide to the detection of the text of the scene through optical character recognition (OCR). But this is not a principal concern for the OCR engine. Until script identification, it is important to identify the script-type because today the text of the scene in natural images does not consist only of a single script, rather mixed-script words at character level are very often encountered. These words are also used in various ways, such as signboards, t-shirt graffiti, hoardings, and banners and often written in artistic way. In this work, a CNN-based deep learning framework, named as STDNet: Script-Type detection Network, was developed to detect single-/mixed-script images. To determine the feasibility of the system presented, tests were also undertaken with an outlier which is composed of a wide range of single scripts. Experiments were performed with over 20K images and 99.53% highest accuracy was reached. This approach was compared to a state-of-the-art deep learning techniques and handcrafted feature-based methodologies where the proposed approach obtained a better performance. More... »

PAGES

277-288

References to SciGraph publications

  • 2012. Random Forest for Image Annotation in COMPUTER VISION – ECCV 2012
  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1007/s11334-021-00395-6

    DOI

    http://dx.doi.org/10.1007/s11334-021-00395-6

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1137531013


    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": "Department of Computer Science, Shyampur Siddheswari Mahavidyalaya, Howrah, India", 
              "id": "http://www.grid.ac/institutes/None", 
              "name": [
                "Department of Computer Science, Shyampur Siddheswari Mahavidyalaya, Howrah, India"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Ghosh", 
            "givenName": "Mridul", 
            "id": "sg:person.010132450437.01", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010132450437.01"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Computer Science, West Bengal State University, Kolkata, India", 
              "id": "http://www.grid.ac/institutes/grid.419478.7", 
              "name": [
                "Department of Computer Science, West Bengal State University, Kolkata, India"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Mukherjee", 
            "givenName": "Himadri", 
            "id": "sg:person.013340464753.42", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013340464753.42"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Computer Science and Engineering, Aliah University, Kolkata, India", 
              "id": "http://www.grid.ac/institutes/grid.440546.7", 
              "name": [
                "Department of Computer Science and Engineering, Aliah University, Kolkata, India"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Obaidullah", 
            "givenName": "Sk Md", 
            "id": "sg:person.012440201362.15", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012440201362.15"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Computer Science, West Bengal State University, Kolkata, India", 
              "id": "http://www.grid.ac/institutes/grid.419478.7", 
              "name": [
                "Department of Computer Science, West Bengal State University, Kolkata, India"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Roy", 
            "givenName": "Kaushik", 
            "id": "sg:person.016332466165.00", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016332466165.00"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "sg:pub.10.1007/978-3-642-33783-3_7", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1041227192", 
              "https://doi.org/10.1007/978-3-642-33783-3_7"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2021-04-27", 
        "datePublishedReg": "2021-04-27", 
        "description": "Script identification serves as a guide to the detection of the text of the scene through optical character recognition (OCR). But this is not a principal concern for the OCR engine. Until script identification, it is important to identify the script-type because today the text of the scene in natural images does not consist only of a single script, rather mixed-script words at character level are very often encountered. These words are also used in various ways, such as signboards, t-shirt graffiti, hoardings, and banners and often written in artistic way. In this work, a CNN-based deep learning framework, named as STDNet: Script-Type detection Network, was developed to detect single-/mixed-script images. To determine the feasibility of the system presented, tests were also undertaken with an outlier which is composed of a wide range of single scripts. Experiments were performed with over 20K images and 99.53% highest accuracy was reached. This approach was compared to a state-of-the-art deep learning techniques and handcrafted feature-based methodologies where the proposed approach obtained a better performance.", 
        "genre": "article", 
        "id": "sg:pub.10.1007/s11334-021-00395-6", 
        "inLanguage": "en", 
        "isAccessibleForFree": false, 
        "isPartOf": [
          {
            "id": "sg:journal.1044675", 
            "issn": [
              "1614-5046", 
              "1614-5054"
            ], 
            "name": "Innovations in Systems and Software Engineering", 
            "publisher": "Springer Nature", 
            "type": "Periodical"
          }, 
          {
            "issueNumber": "3", 
            "type": "PublicationIssue"
          }, 
          {
            "type": "PublicationVolume", 
            "volumeNumber": "17"
          }
        ], 
        "keywords": [
          "optical character recognition", 
          "art deep learning techniques", 
          "single script", 
          "script identification", 
          "deep learning framework", 
          "deep learning techniques", 
          "feature-based methodology", 
          "learning techniques", 
          "OCR engine", 
          "learning framework", 
          "character recognition", 
          "detection network", 
          "natural images", 
          "character level", 
          "script images", 
          "CNN", 
          "better performance", 
          "high accuracy", 
          "scene", 
          "images", 
          "scripts", 
          "text", 
          "network", 
          "signboards", 
          "detection", 
          "engine", 
          "artistic way", 
          "outliers", 
          "recognition", 
          "framework", 
          "accuracy", 
          "words", 
          "way", 
          "performance", 
          "wide range", 
          "methodology", 
          "system", 
          "today", 
          "feasibility", 
          "technique", 
          "work", 
          "identification", 
          "experiments", 
          "guide", 
          "principal concern", 
          "concern", 
          "banner", 
          "state", 
          "graffiti", 
          "levels", 
          "range", 
          "test", 
          "hoarding", 
          "approach", 
          "mixed-script words", 
          "shirt graffiti", 
          "Script-Type detection Network", 
          "script detection"
        ], 
        "name": "STDNet: A CNN-based approach to single-/mixed-script detection", 
        "pagination": "277-288", 
        "productId": [
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1137531013"
            ]
          }, 
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1007/s11334-021-00395-6"
            ]
          }
        ], 
        "sameAs": [
          "https://doi.org/10.1007/s11334-021-00395-6", 
          "https://app.dimensions.ai/details/publication/pub.1137531013"
        ], 
        "sdDataset": "articles", 
        "sdDatePublished": "2022-01-01T19:04", 
        "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_913.jsonl", 
        "type": "ScholarlyArticle", 
        "url": "https://doi.org/10.1007/s11334-021-00395-6"
      }
    ]
     

    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/s11334-021-00395-6'

    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/s11334-021-00395-6'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s11334-021-00395-6'

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

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s11334-021-00395-6'


     

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

    147 TRIPLES      22 PREDICATES      84 URIs      75 LITERALS      6 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1007/s11334-021-00395-6 schema:about anzsrc-for:08
    2 anzsrc-for:0801
    3 schema:author N404c4394b3d245e986af39d6c6dda96f
    4 schema:citation sg:pub.10.1007/978-3-642-33783-3_7
    5 schema:datePublished 2021-04-27
    6 schema:datePublishedReg 2021-04-27
    7 schema:description Script identification serves as a guide to the detection of the text of the scene through optical character recognition (OCR). But this is not a principal concern for the OCR engine. Until script identification, it is important to identify the script-type because today the text of the scene in natural images does not consist only of a single script, rather mixed-script words at character level are very often encountered. These words are also used in various ways, such as signboards, t-shirt graffiti, hoardings, and banners and often written in artistic way. In this work, a CNN-based deep learning framework, named as STDNet: Script-Type detection Network, was developed to detect single-/mixed-script images. To determine the feasibility of the system presented, tests were also undertaken with an outlier which is composed of a wide range of single scripts. Experiments were performed with over 20K images and 99.53% highest accuracy was reached. This approach was compared to a state-of-the-art deep learning techniques and handcrafted feature-based methodologies where the proposed approach obtained a better performance.
    8 schema:genre article
    9 schema:inLanguage en
    10 schema:isAccessibleForFree false
    11 schema:isPartOf N9fa3157d0e6a456499d61720d6fae922
    12 Naa58d052faa343749e7552277b77806e
    13 sg:journal.1044675
    14 schema:keywords CNN
    15 OCR engine
    16 Script-Type detection Network
    17 accuracy
    18 approach
    19 art deep learning techniques
    20 artistic way
    21 banner
    22 better performance
    23 character level
    24 character recognition
    25 concern
    26 deep learning framework
    27 deep learning techniques
    28 detection
    29 detection network
    30 engine
    31 experiments
    32 feasibility
    33 feature-based methodology
    34 framework
    35 graffiti
    36 guide
    37 high accuracy
    38 hoarding
    39 identification
    40 images
    41 learning framework
    42 learning techniques
    43 levels
    44 methodology
    45 mixed-script words
    46 natural images
    47 network
    48 optical character recognition
    49 outliers
    50 performance
    51 principal concern
    52 range
    53 recognition
    54 scene
    55 script detection
    56 script identification
    57 script images
    58 scripts
    59 shirt graffiti
    60 signboards
    61 single script
    62 state
    63 system
    64 technique
    65 test
    66 text
    67 today
    68 way
    69 wide range
    70 words
    71 work
    72 schema:name STDNet: A CNN-based approach to single-/mixed-script detection
    73 schema:pagination 277-288
    74 schema:productId Ncee12c9d25874fb4aaa3ac83a373d1bf
    75 Nf9a7756218cb4b98bda4be4af3f096cf
    76 schema:sameAs https://app.dimensions.ai/details/publication/pub.1137531013
    77 https://doi.org/10.1007/s11334-021-00395-6
    78 schema:sdDatePublished 2022-01-01T19:04
    79 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    80 schema:sdPublisher N4945e836c2d8491f90decf56740a76a3
    81 schema:url https://doi.org/10.1007/s11334-021-00395-6
    82 sgo:license sg:explorer/license/
    83 sgo:sdDataset articles
    84 rdf:type schema:ScholarlyArticle
    85 N0693d2664dad46d4abe927dde9ff593e rdf:first sg:person.012440201362.15
    86 rdf:rest N42e620c0399242c7a6b75ba9f8edefe0
    87 N404c4394b3d245e986af39d6c6dda96f rdf:first sg:person.010132450437.01
    88 rdf:rest Na1e5d950da6d471d8a4533a1f3063425
    89 N42e620c0399242c7a6b75ba9f8edefe0 rdf:first sg:person.016332466165.00
    90 rdf:rest rdf:nil
    91 N4945e836c2d8491f90decf56740a76a3 schema:name Springer Nature - SN SciGraph project
    92 rdf:type schema:Organization
    93 N9fa3157d0e6a456499d61720d6fae922 schema:issueNumber 3
    94 rdf:type schema:PublicationIssue
    95 Na1e5d950da6d471d8a4533a1f3063425 rdf:first sg:person.013340464753.42
    96 rdf:rest N0693d2664dad46d4abe927dde9ff593e
    97 Naa58d052faa343749e7552277b77806e schema:volumeNumber 17
    98 rdf:type schema:PublicationVolume
    99 Ncee12c9d25874fb4aaa3ac83a373d1bf schema:name doi
    100 schema:value 10.1007/s11334-021-00395-6
    101 rdf:type schema:PropertyValue
    102 Nf9a7756218cb4b98bda4be4af3f096cf schema:name dimensions_id
    103 schema:value pub.1137531013
    104 rdf:type schema:PropertyValue
    105 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
    106 schema:name Information and Computing Sciences
    107 rdf:type schema:DefinedTerm
    108 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
    109 schema:name Artificial Intelligence and Image Processing
    110 rdf:type schema:DefinedTerm
    111 sg:journal.1044675 schema:issn 1614-5046
    112 1614-5054
    113 schema:name Innovations in Systems and Software Engineering
    114 schema:publisher Springer Nature
    115 rdf:type schema:Periodical
    116 sg:person.010132450437.01 schema:affiliation grid-institutes:None
    117 schema:familyName Ghosh
    118 schema:givenName Mridul
    119 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010132450437.01
    120 rdf:type schema:Person
    121 sg:person.012440201362.15 schema:affiliation grid-institutes:grid.440546.7
    122 schema:familyName Obaidullah
    123 schema:givenName Sk Md
    124 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012440201362.15
    125 rdf:type schema:Person
    126 sg:person.013340464753.42 schema:affiliation grid-institutes:grid.419478.7
    127 schema:familyName Mukherjee
    128 schema:givenName Himadri
    129 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013340464753.42
    130 rdf:type schema:Person
    131 sg:person.016332466165.00 schema:affiliation grid-institutes:grid.419478.7
    132 schema:familyName Roy
    133 schema:givenName Kaushik
    134 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016332466165.00
    135 rdf:type schema:Person
    136 sg:pub.10.1007/978-3-642-33783-3_7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1041227192
    137 https://doi.org/10.1007/978-3-642-33783-3_7
    138 rdf:type schema:CreativeWork
    139 grid-institutes:None schema:alternateName Department of Computer Science, Shyampur Siddheswari Mahavidyalaya, Howrah, India
    140 schema:name Department of Computer Science, Shyampur Siddheswari Mahavidyalaya, Howrah, India
    141 rdf:type schema:Organization
    142 grid-institutes:grid.419478.7 schema:alternateName Department of Computer Science, West Bengal State University, Kolkata, India
    143 schema:name Department of Computer Science, West Bengal State University, Kolkata, India
    144 rdf:type schema:Organization
    145 grid-institutes:grid.440546.7 schema:alternateName Department of Computer Science and Engineering, Aliah University, Kolkata, India
    146 schema:name Department of Computer Science and Engineering, Aliah University, Kolkata, India
    147 rdf:type schema:Organization
     




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


    ...