Difference between revisions of "Publications/chen.21.dgmm"

From LRDE

(Created page with "{{Publication | published = true | date = 2021-02-16 | authors = Yizi Chen, Edwin Carlinet, Joseph Chazalon, Clément Mallet, Bertrand Duménieu, Julien Perret | title = Combi...")
 
Line 8: Line 8:
 
| address = Uppsala, Sweden
 
| address = Uppsala, Sweden
 
| publisher = Springer
 
| publisher = Springer
  +
| pages = 79 to 92
 
| abstract = The digitization of historical maps enables the study of ancient, fragile, unique, and hardly accessible information sources. Main map features can be retrieved and tracked through the time for subsequent thematic analysis. The goal of this work is the vectorization stepi.e., the extraction of vector shapes of the objects of interest from raster images of maps. We are particularly interested in closed shape detection such as buildingsbuilding blocks, gardens, rivers, etc. in order to monitor their temporal evolution. Historical map images present significant pattern recognition challenges. The extraction of closed shapes by using traditional Mathematical Morphology (MM) is highly challenging due to the overlapping of multiple map features and texts. Moreoverstate-of-the-art Convolutional Neural Networks (CNN) are perfectly designed for content image filtering but provide no guarantee about closed shape detection. Also, the lack of textural and color information of historical maps makes it hard for CNN to detect shapes that are represented by only their boundaries. Our contribution is a pipeline that combines the strengths of CNN (efficient edge detection and filtering) and MM (guaranteed extraction of closed shapes) in order to achieve such a task. The evaluation of our approach on a public dataset shows its effectiveness for extracting the closed boundaries of objects in historical maps.
 
| abstract = The digitization of historical maps enables the study of ancient, fragile, unique, and hardly accessible information sources. Main map features can be retrieved and tracked through the time for subsequent thematic analysis. The goal of this work is the vectorization stepi.e., the extraction of vector shapes of the objects of interest from raster images of maps. We are particularly interested in closed shape detection such as buildingsbuilding blocks, gardens, rivers, etc. in order to monitor their temporal evolution. Historical map images present significant pattern recognition challenges. The extraction of closed shapes by using traditional Mathematical Morphology (MM) is highly challenging due to the overlapping of multiple map features and texts. Moreoverstate-of-the-art Convolutional Neural Networks (CNN) are perfectly designed for content image filtering but provide no guarantee about closed shape detection. Also, the lack of textural and color information of historical maps makes it hard for CNN to detect shapes that are represented by only their boundaries. Our contribution is a pipeline that combines the strengths of CNN (efficient edge detection and filtering) and MM (guaranteed extraction of closed shapes) in order to achieve such a task. The evaluation of our approach on a public dataset shows its effectiveness for extracting the closed boundaries of objects in historical maps.
 
| lrdepaper = http://www.lrde.epita.fr/dload/papers/chen.2021.dgmm.pdf
 
| lrdepaper = http://www.lrde.epita.fr/dload/papers/chen.2021.dgmm.pdf
Line 15: Line 16:
 
| type = inproceedings
 
| type = inproceedings
 
| id = chen.21.dgmm
 
| id = chen.21.dgmm
  +
| identifier = doi:10.1007/978-3-030-76657-3_5
 
| bibtex =
 
| bibtex =
 
@InProceedings<nowiki>{</nowiki> chen.21.dgmm,
 
@InProceedings<nowiki>{</nowiki> chen.21.dgmm,
Line 28: Line 30:
 
address = <nowiki>{</nowiki>Uppsala, Sweden<nowiki>}</nowiki>,
 
address = <nowiki>{</nowiki>Uppsala, Sweden<nowiki>}</nowiki>,
 
publisher = <nowiki>{</nowiki>Springer<nowiki>}</nowiki>,
 
publisher = <nowiki>{</nowiki>Springer<nowiki>}</nowiki>,
  +
pages = <nowiki>{</nowiki>79--92<nowiki>}</nowiki>,
 
abstract = <nowiki>{</nowiki>The digitization of historical maps enables the study of
 
abstract = <nowiki>{</nowiki>The digitization of historical maps enables the study of
 
ancient, fragile, unique, and hardly accessible information
 
ancient, fragile, unique, and hardly accessible information
Line 54: Line 57:
 
extracting the closed boundaries of objects in historical
 
extracting the closed boundaries of objects in historical
 
maps.<nowiki>}</nowiki>,
 
maps.<nowiki>}</nowiki>,
note = <nowiki>{</nowiki>Accepted<nowiki>}</nowiki>
+
note = <nowiki>{</nowiki>Accepted<nowiki>}</nowiki>,
  +
doi = <nowiki>{</nowiki>10.1007/978-3-030-76657-3_5<nowiki>}</nowiki>
 
<nowiki>}</nowiki>
 
<nowiki>}</nowiki>
   

Revision as of 19:00, 21 May 2021

Abstract

The digitization of historical maps enables the study of ancient, fragile, unique, and hardly accessible information sources. Main map features can be retrieved and tracked through the time for subsequent thematic analysis. The goal of this work is the vectorization stepi.e., the extraction of vector shapes of the objects of interest from raster images of maps. We are particularly interested in closed shape detection such as buildingsbuilding blocks, gardens, rivers, etc. in order to monitor their temporal evolution. Historical map images present significant pattern recognition challenges. The extraction of closed shapes by using traditional Mathematical Morphology (MM) is highly challenging due to the overlapping of multiple map features and texts. Moreoverstate-of-the-art Convolutional Neural Networks (CNN) are perfectly designed for content image filtering but provide no guarantee about closed shape detection. Also, the lack of textural and color information of historical maps makes it hard for CNN to detect shapes that are represented by only their boundaries. Our contribution is a pipeline that combines the strengths of CNN (efficient edge detection and filtering) and MM (guaranteed extraction of closed shapes) in order to achieve such a task. The evaluation of our approach on a public dataset shows its effectiveness for extracting the closed boundaries of objects in historical maps.

Documents

Bibtex (lrde.bib)

@InProceedings{	  chen.21.dgmm,
  author	= {Yizi Chen and Edwin Carlinet and Joseph Chazalon and
		  Cl\'ement Mallet and Bertrand Dum\'enieu and Julien Perret},
  title		= {Combining Deep Learning and Mathematical Morphology for
		  Historical Map Segmentation},
  booktitle	= {IAPR International Conference on Discrete Geometry and
		  Mathematical Morphology (DGMM)},
  year		= {2021},
  series	= {Lecture Notes in Computer Science},
  month		= may,
  address	= {Uppsala, Sweden},
  publisher	= {Springer},
  pages		= {79--92},
  abstract	= {The digitization of historical maps enables the study of
		  ancient, fragile, unique, and hardly accessible information
		  sources. Main map features can be retrieved and tracked
		  through the time for subsequent thematic analysis. The goal
		  of this work is the vectorization step, i.e., the
		  extraction of vector shapes of the objects of interest from
		  raster images of maps. We are particularly interested in
		  closed shape detection such as buildings, building blocks,
		  gardens, rivers, etc. in order to monitor their temporal
		  evolution. Historical map images present significant
		  pattern recognition challenges. The extraction of closed
		  shapes by using traditional Mathematical Morphology (MM) is
		  highly challenging due to the overlapping of multiple map
		  features and texts. Moreover, state-of-the-art
		  Convolutional Neural Networks (CNN) are perfectly designed
		  for content image filtering but provide no guarantee about
		  closed shape detection. Also, the lack of textural and
		  color information of historical maps makes it hard for CNN
		  to detect shapes that are represented by only their
		  boundaries. Our contribution is a pipeline that combines
		  the strengths of CNN (efficient edge detection and
		  filtering) and MM (guaranteed extraction of closed shapes)
		  in order to achieve such a task. The evaluation of our
		  approach on a public dataset shows its effectiveness for
		  extracting the closed boundaries of objects in historical
		  maps.},
  note		= {Accepted},
  doi		= {10.1007/978-3-030-76657-3_5}
}