We handle MICR checks in our office. Right now it's done by hand. The paper is stored forever in some dungeon somewhere.

I'm thinking scanning in and burning the hard copy.

Being a hard stompin open source minion- I looked for micr on cpan first. Looks like nothing.

I looked at sourceforge. There are some java tools there for doing things with already scanned checks- some drivers for scanners.

My two options are using proprietary software/hardware (any suggestions?) - or writing a system- it would work with any scanner, it would automatically chop the scan of maybe multiple checks per page to individual pdfs etc etc. And I'd have to piece together a very specific set of character matching tools.. just for this doc type. Any suggestions for where to look - Anyone working for finance that would perhaps also be intererested in doing this?

update

I am using gocr and I put together some scripts and a Finance::MICR::LineParser module, and a scratchy Finance::MICR::GOCR module.

I am using a canon machine, that's already available in most offices. These have adapters on top that let you feed papers in stacks and put it all to one file.

I route the check to my desktop with another tool we have, and then a script using imagemagick formats the scans to nice little files right way up.

Then another script isolates the MICR lines, and feeds them to gocr with a trained MICR library, and passes that to my Finance::MICR::LineParser module, which tells me if it can extract the MICR code, and gives me a minor rating of cleanliness etc. Then the master file image is renamed by check number , bank account, etc.. or xxx_xxx_xx etc to let the user know it cannot read the MICR line properly.

I was hitting 92.5% accuracy last night with about two hundred. Most of the ones not read properly were either turned by scanner mistake- or would not be read by a commercial MICR reading machine (i've read up on it).

This morning I ran a more real world test, was a little messy feeding them in on purpose, and ran about 989 checks. I am hitting 85%. None of the hits reported are wrong- that is.. no mistakes have been made (i read every single one) (ie, system reports that check is number 123 when it is in fact 124, that does not happen although it did at first ).

Not so bad for 27 or so hrs work. But MICR standards report the accepted fail ratio as 1/1000 - so I'm far from it.

In reply to (OT) MICR check scanning by leocharre

Title:
Use:  <p> text here (a paragraph) </p>
and:  <code> code here </code>
to format your post, it's "PerlMonks-approved HTML":


  • Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!
  • Titles consisting of a single word are discouraged, and in most cases are disallowed outright.
  • Read Where should I post X? if you're not absolutely sure you're posting in the right place.
  • Please read these before you post! —
  • Posts may use any of the Perl Monks Approved HTML tags:
    a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr
  • You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)
            For:     Use:
    & &amp;
    < &lt;
    > &gt;
    [ &#91;
    ] &#93;
  • Link using PerlMonks shortcuts! What shortcuts can I use for linking?
  • See Writeup Formatting Tips and other pages linked from there for more info.