How to solve color and B&W mixed batch problem?
Hi,
my client wants to store color images in FileNet, but only on subset of documents, all others should go in as black-and-white.
So each batch (scanner, not VScan) will have some documents that should be scanned and stored as color images. All others should be stored as B&W.
I, obviously, don't want to store images that should be B&W as color.
I was thinking that I could scan all as color and then convert some to B&W. But how can I find out which one is which when I cannot do PageID on color images to find out?
How would you do this?
EDIT:
I'll try like this and report back the findings:
- scan all as color images
- convert all to b&w (keeping the copies)
- do PageID
- b&w: store the converted b&w
- color: store the color originals
The one thing that's not clear to me is how to change which images get uploaded to FileNet when I call FNP8_Upload()?
I think what you'd have to do is set the DCO status of the ones you do not want to export and then have another rule where you delete all child objects with that DCO status you set. This would remove the images you don't want to export from the batch.
Hi Markobonaci
According to me if you want your original image to be exported to Filenet whether its coloured or B&W, you can you this method FNP8_UploadDir()
which is inside the FileNetP8 action library.This method would upload all the images which are in the path assigned in the method.
Hope it solves your problem.
Regards
Deekshith
I'm talking about the actual physical scanner you're using to scan images into Datacap - the settings for that. I'm assuming you might be able to increase the dpi on it and get a better image.
If you're using ISIS drivers for your scanner, you should be able to set automatic color detection as the color mode - that way, you don't have to do any development.
The problem is that the document appearance (whether it's in color or not) doesn't mean nothing.
Not all color documents need to be stored in color and some b&w documents, on the other hand, need to be stored in color (e.g. if they contain stamp or signature, or maybe that type of document always comes with a notary seal on the back).
So the b&w/color decision is not a dynamic thing, it's made in advance, based on document type, which is why I need PageID.
Do you have any idea how to manipulate which image is used for FileNet export?
Thank you both.
FYI, when I tried to scan all images as color and then convert to B&W I noticed that the conversion process (compared to scanning in b&w) degrades the image quality by aprox. 10 - 30 % (my rough estimate).
Original color image is scanned from DotScan as LZW compressed tif, with 24 bpp and 200 dpi.
Original b&w image was scanned from DotScan as G4 compressed tif with 1bpp and 200 dpi.
Image shared on my Google Drive:
Converted image is on the left.
Does it have to be like that?
So due to this "conversion with quality degradation problem", since there will be way more b&w then color doc types, I think we're going to (Dot) scan all b&w docs and then use DotScan actions to insert additional color docs in the same batch.
If we have the luxury that there's only a single document in a logical group or that all document types in a group need to be scanned the same way (all color or all b&w), then we're going to place color and b&w documents in separate Datacap applications.
What do you think?
Mark, can you get at the scanner settings? I'm guessing there's something that could fix your quality issue.
What do you mean by scanner settings?
I scan image in color and then convert to B&W and it's visibly lower quality than the same image scanned in B&W at the scanner.
I'd rather say that Datacap Color to B&W conversion (I'm using OCRA_ConvertImage2BW and C2BW_Convert gave me the same result) might maybe be better, cuz' both images I get from scanner seem fine to me.
Can you explain, what do you think might be wrong with scanner settings?