Barcode OCR

On-the-fly processing is actually a preferable way of reading barcodes since it does not noticeably decrease scan speed. The recognition will be the same whether the barcode is processed on the fly or as a post-process.

A single job step in Capture can be configured to break a document, index one or more values, and then delete the page used for breaking/indexing. This could be configured in a Capture Scan step, Barcode 1D or 2D step, or a Capture OCR (zonal) step. The steps below describe the setting changes that need to be made in either a Barcode or OCR step.

1. In the Capture job definition, add a Barcode step after the job’s start step.
2. Set the Auto Document Break Mode to ‘Barcode’.
3. Click the ellipses button next to Barcode Zones to open the Edit Zones dialog. Import or scan a break sheet and draw your zone.
4. The default barcode type is Code 39; change this if necessary. An optional setting is “Required for Delete,” which makes your zone required to be read successfully for the break page to be deleted from the document. Save and close the Edit Zones dialog when finished.
5. Set the Save Page property to ‘False’.
6. Now, under the Indexes section, click the ellipses to add a new index field.
7. Name the field, then click the ellipses next to Barcode Zones to draw the zone.
8. Save the index configuration.
9. Repeat Steps 6 through 8 for any additional index fields that need to be populated by barcodes on this page.
10. When finished, add and configure any additional job steps, then save the job, activate it, and check it in for testing.

The job is now configured to break on the barcode (or OCR value) configured in Step 3. Then the index(s) configured to be read off this page will be read, and once completed, the page will be deleted due to the ‘False’ setting configured in Step 5.

You can have as many barcodes as you can fit on a sheet of paper, as long as the barcode is big enough so that the software can read it. If is too small on the paper to begin with, when it gets scanned, the pixels are so small that they get compressed together even though to the eye it still looks like a barcode. When the scanner scans the data, it can’t read it because to the software, the barcode (depending on size) looks like a black line instead of multiple black lines. It then considers it an invalid barcode.