Difference between revisions of "Marine Lives guide to creating a Transkribus Ground Truth"

From MarineLives
Jump to: navigation, search
(Using Transkribus Lite version 2.0)
 
(32 intermediate revisions by the same user not shown)
Line 1: Line 1:
 
'''This page is the main page of a guide that Marine Lives is creating to cover practical aspects of creating a Transkribus Ground Truth'''
 
'''This page is the main page of a guide that Marine Lives is creating to cover practical aspects of creating a Transkribus Ground Truth'''
 
+
----
 
__TOC__
 
__TOC__
 
+
----
 
==Objective==
 
==Objective==
  
Our objective is to create a C17th English secretarial hand HTR model, which we will use on our collection of 80,000 images of English High Court of Admiralty depositions.
+
[[File:Transkribus Lite Search Bahia 03032022.png|750px|thumb|left|Transkribus web capability has simple, but useful search functionality]]
  
We are aiming to create two models. The first based on a Ground Truth of 500,000 words (roughly 1,000 manuscript images). The second based on one million words (roughly 2,000 manuscript images).
+
[[File:Transkribus Lite Search Bahia Full TexT03032022.png|750px|thumb|left|We want to have 80,000 images covering HCA 13/20to HCA 13/79 publicly available and searchable by end 2022l]]
  
For our first model, we are using existing semi-diplomatic transcriptions of the HCA 13/72 volume [late 1650s], made between 2013 and 2015 by Marine Lives volunteers
+
'''Our overall objective is to make 80,000 images of English High Court of Admiralty depositions covering 1570 to 1690 publicly available and searchable.'''
----
+
==Tools==
+
----
+
===Using Transkribus Expert Client===
+
  
We are using Transkribus Expert Client as our main tool to perform automated layout recognition, manual correction of these layouts, and to enter and modify existing semi-diplomatic transcriptions of material in HCA 13/72.
+
'''To do this our immediate objective is to create a C17th English secretarial hand HTR model, which we will use on our collection of 80,000 images of English High Court of Admiralty depositions.'''
  
We are then using Transkribus Lite version 2.0 to view completed Ground Truth pages, and to keep an overview of our work.
+
We are aiming to create two models. The first based on a Ground Truth of 500,000 words (roughly 1,000 manuscript images). The second based on one million words (roughly 2,000 manuscript images).
  
 +
For our first model, we are using existing semi-diplomatic transcriptions of the HCA 13/72 volume [late 1650s], made between 2013 and 2015 by Marine Lives volunteers. We have to convert these semi-diplomatic transcriptions back to full diplomatic, and mark up contractions, as part of the process of creating a highly reliable Ground Truth. We also have to treat interlineation differently, with each interline shown separately and the insertion point or points marked inm the line on which the interlineation depends.
 
----
 
----
===Using Transkribus Lite version 2.0===
+
==Tools==
  
[[File:Transkribus Lite HCA1372 Done Thumbnails 03032022.png|750px|thumb|left|Transkribus Lite Version 2.0 Thumbnail display showing images with status "Done" in our work processl]]
+
We are working with several related Transkribus Tools and with our own semantic media wiki
  
 +
[http://www.marinelives.org/wiki/Tools_to_create_our_Ground_Truth#Using_Transkribus_Expert_Client 1. Transkribus Expert Client]
  
Transkribus has recently introduced an improved version of its web browser interface ([https://transkribus.eu/lite/ Transkribus Liter Version 2.0]).
+
[http://www.marinelives.org/wiki/Tools_to_create_our_Ground_Truth#Using_Transkribus_Lite_version_2.0 2. Transkribus Lite version 2.0]
  
Transkribus has a [https://readcoop.eu/transkribus/howto/getting-started-with-transkribus-lite/ useful online guide to using Transkribus Lite Version 2.0].
+
[http://www.marinelives.org/wiki/Tools_to_create_our_Ground_Truth#Marine_Lives_wiki 3. Marine Lives semantic media wiki]
  
We are finding this improved browser interface to be pretty responsive in terms of short lag times as we browse images.
+
----
 +
==Work Process==
  
The browser interface also has useful functionality not available in Transkribus Expert Client.
+
[http://www.marinelives.org/wiki/Ground_Truth_Work_Process#Automatic_layout_recognition_of_all_1518_images_in_HCA_13.2F72 1 Automatic layout recognition of all 1518 images in HCA 13/72]
  
Most useful to date are:
+
[http://www.marinelives.org/wiki/Ground_Truth_Work_Process#Input_of_existing_semi-diplomatic_transcriptions_of_HCA_13.2F72_manuscript_pages_into_Transkribus_Expert_Client 2 Input of existing semi-diplomatic transcriptions of HCA 13/72 manuscript pages into Transkribus Expert Client]
  
# Large thumbnails
+
==We are experimenting==
  
# Ability to display thumbnails by staus of manuscript pages within our work process
+
We are experimenting with a range of Transkribus tools related to [http://www.marinelives.org/wiki/Customized_Structural_Analysis layout analysis and HTR].
----
+
 
+
===Marine Lives wiki===
+
 
+
The [http://www.marinelives.org/wiki/MarineLives Marine Lives wiki] is a Semantic Media Wiki. It is organised into volumes and pages.
+
 
+
We are working from volume [http://www.marinelives.org/wiki/HCA_13/72 HCA 13/72] and are inputing existing semi-diplomatic transcriptions from this volume by hand into Transkribus xpert Cleient.
+
  
 
----
 
----
==Work process==
+
==Questions==
  
We have set up a simple work process
+
'''We are developing a [http://www.marinelives.org/wiki/Running_List_of_Questions running list of questions.]'''
  
1. Automatic layout recognition of all 1518 images in HCA 13/72
+
'''Some of these questions we will probably be able to answer ourselves, as we get more experience of building our Ground Truth.'''
 
+
- Used the CITlab Advanced Tool
+
 
+
[[File:CITlab Advanced Tool ML 03032022.png|500px|thumb|left|Layout Analysis controls in Tools section of Transkribus Expert Client controls panel]]
+
 
+
- Modified the layout page by page after manual inspection of automatically generated layouts
+
 
+
* We are only just beginning to think through what makes sense in terms of use of Text Regions when creating our Ground Truth
+
* We are finding that the automatic tool is typically producing between one and three Text Regions per manucript image
+
* Typically the tool is NOT identifying text blocks on the left hand side of an image as separate from structurally separate text in the main body of text
+
* Ideally, we would train the automatic layout recognition tool to be sensitive to the typical structures of HCA legal depositions, and we are looking into this
+
* In the short term, we are manually adding Text Regions, and changing the shape and size of Text Regions
+
* However, lines of text have already been recognised and allocated to specific text regions. So we, are having to reallocate lines of text to our new Text Regions after we have entered the text. This is cumbersome, so we are looking into alternatives
+
 
+
[[File:Transkribus Expert Client Layout HCA 1372 f.14v.png|750px|thumb|left|Layout out HCA 13/72 f.14v once we have manually adjusted the Text Regions, creating six Text Regions and reallocating lines to those regions]]
+
 
+
----
+
2. Input of existing semi-diplomatic transcriptions of HCA 13/72 manuscript pages into Transkribus Expert Client
+
  
[ADD TEXT]
+
'''But, in the meantime, we would appreciate sugegstions from fellow Transkribus users.'''

Latest revision as of 22:25, March 3, 2022

This page is the main page of a guide that Marine Lives is creating to cover practical aspects of creating a Transkribus Ground Truth






Objective


Transkribus web capability has simple, but useful search functionality
We want to have 80,000 images covering HCA 13/20to HCA 13/79 publicly available and searchable by end 2022l

Our overall objective is to make 80,000 images of English High Court of Admiralty depositions covering 1570 to 1690 publicly available and searchable.

To do this our immediate objective is to create a C17th English secretarial hand HTR model, which we will use on our collection of 80,000 images of English High Court of Admiralty depositions.

We are aiming to create two models. The first based on a Ground Truth of 500,000 words (roughly 1,000 manuscript images). The second based on one million words (roughly 2,000 manuscript images).

For our first model, we are using existing semi-diplomatic transcriptions of the HCA 13/72 volume [late 1650s], made between 2013 and 2015 by Marine Lives volunteers. We have to convert these semi-diplomatic transcriptions back to full diplomatic, and mark up contractions, as part of the process of creating a highly reliable Ground Truth. We also have to treat interlineation differently, with each interline shown separately and the insertion point or points marked inm the line on which the interlineation depends.



Tools


We are working with several related Transkribus Tools and with our own semantic media wiki

1. Transkribus Expert Client

2. Transkribus Lite version 2.0

3. Marine Lives semantic media wiki



Work Process


1 Automatic layout recognition of all 1518 images in HCA 13/72

2 Input of existing semi-diplomatic transcriptions of HCA 13/72 manuscript pages into Transkribus Expert Client

We are experimenting


We are experimenting with a range of Transkribus tools related to layout analysis and HTR.



Questions


We are developing a running list of questions.

Some of these questions we will probably be able to answer ourselves, as we get more experience of building our Ground Truth.

But, in the meantime, we would appreciate sugegstions from fellow Transkribus users.