messy sheet papers (Bug #22)

Added by red sea over 13 years ago. Updated about 13 years ago.

Status:Closed Start date:03/21/2012
Priority:Normal Due date:
Assignee:- % Done:

100%
Category:-
Target version:-

Description

I'm on the order of the papers before scanned, sorted and saved in PDF file(arranged).
but become messy after i import them into AMC (data capture).

History

#1
Updated by red sea over 13 years ago

#2
Updated by red sea over 13 years ago

sorry..
i mean the sheet id 1 should be given for the first sheet paper in pdf file, and the sheet i 2 should be given for 2nd sheet paper in pdf file ....

#3
Updated by red sea over 13 years ago

and o for the correct teacher sheet.

#4
Updated by Alexis Bienvenüe over 13 years ago

In the case answer sheets were photocopied, the sheet ID created by AMC for each page (the one after the colon) cannot be reliably connected with the page number from the PDF scan. This is because of parallelized processing. Moreover, at the data capture stage, AMC does not know which is the teacher sheet, so it cannot tag it with sheet ID 0.

AMC is not designed to take care of the order the scans were given. The right way to identify the scans in AMC is to use the name field.

#5
Updated by red sea over 13 years ago

ok, Thank you for your generosity and your patience.

#6
Updated by Alexis Bienvenüe over 13 years ago

Since revision r956, when scan files are split (for example when you selected a multi-page PDF scan file), and when you don't use parallelization (set "number of processes" to 1 in Edit/Preferences/Main/System), the sheet IDs should be in the same order than the pages in the original PDF. However, this will slow down data capture on multi-processors (or multi-cores) computers...

#7
Updated by red sea over 13 years ago

thanks very much.
i don't understand the last thing?
i have i7 , what i can do to get the fast data capure ("number of processes").

best regards

#8
Updated by Alexis Bienvenüe over 13 years ago

what i can do to get the fast data capure ("number of processes").

It is enabled by default: default value for "number of processes" is 0, and that means that AMC will use all the cores in parallel during automatic data capture.

#9
Updated by Alexis Bienvenüe over 13 years ago

From revision r963, an option has been added to pre-allocate sheets IDs (available in the window where you select scan files). With this option set, when you run automatic association from some scans, sheets IDs are allocated before data capture to all pages in the scans in the right order, starting from the maximum of all sheets IDs already used plus one.

  • If all scans are from the same photocopied page and are properly recognized, all goes well.
  • If some pages are not recognized, the corresponding pre-allocated sheet ID won't be used.
  • If pages are from photocopies of different subjects, the sheet IDs will often jump when you look at the sheets IDs for one paparticular page.

This option is compatible with parallelization, so this solution is probably better than the one I suggested in note number 6 above.

  • Status changed from New to Resolved

#10
Updated by Alexis Bienvenüe about 13 years ago

  • Status changed from Resolved to Closed

Also available in: Atom PDF