You are viewing limited content. For full access, please sign in.

Question

Question

fix a corrupted image

asked on May 9, 2016 Show version history

I have a document that I am trying to break apart in Quick Fields, its not importing the whole document using the capture engine and giving "unknown error". I have tracked down the page that comes right after the last page that shows in quick fields.

It was missing OCR, so I OCR'd it and the image displays in LF and the OCR worked but LF client gave me the error "2016 Council Minutes (2), page 62: Corrupt image may have adversely affected OCR results"

So this is probably what quick fields is freaking about. I have already copied the full document and it has the same issue, is there a way to fix this "corrupted image" without just removing it and re-scanning it? 

 

This document is from this year, so it wouldn't have gone through many upgrades, however another document seems to have the same issue and its a few years old.

Edit: Also the tiff opens fine from windows explorer.

0 0

Answer

SELECTED ANSWER
replied on May 12, 2016

I was able to fix the issue by opening it in the back end. Editing it in a picture editor, just anything, so that I could save over it. That fixed it, so yes there was something in the file that that was wrong, however as I said it did open in the editor fine, so there was something specifically that the LF tiff viewer didn't like.

0 0

Replies

replied on May 11, 2016

Hi Joel,

Please open a support case and attach the document with the case. We'll follow up this issue with the support case.

Although the image may have looked fine, there may be hidden issues such as:

  • Image file being truncated at the end, causing the loss of some image pages or part of the last page.
  • Problems or inconsistency with the image header.
  • Some image pages encoded in a file format or pixel format that Laserfiche cannot decode.

 

This warning is a safeguard against silent image corruption or data loss. When this warning is issued, it is recommended to check the processed result carefully and determine if any actual data loss has occurred. If it is determined that there is no actual data loss, the warning can be ignored. This should be decided on a case-by-case basis because the same warning can be triggered by different reasons.

In order to intelligently suppress the same warning for all images having the same symptom, we will need to analyze a sample document with the problem. After we determine that the symptom is harmless, we will then implement a software change for future versions of Laserfiche. Opening a support case and providing a sample document is the first step.

1 0
You are not allowed to follow up in this post.

Sign in to reply to this post.