views:

968

answers:

3

Hello, I know this has been discussed a few times on stackoverflow but I wasn't able to find a sufficient answer. For a project, I am seeking the best possible piece of software/api/library that can recognise barcode numbers from the input image. It will be used on mobile phones(iphone, windows mobile, android etc.) so it should be good at catching barcodes from bad angles, blurry images and such bad conditions. Being open source or commercial doesn't matter as long as it does the job well (indeed open source would be better :) )

If you have any experience with this topic, please share with us:)

Thanks

+2  A: 

Google has an Open Source project for this zxing

I'm familiar enough with it to know if it will do everything you need it to do, but I imagine it would be a really great place for you to start.

jlech
zxing does not work with blurry images
endolith
Granted, but as with any image recognition software, there are thresholds as to how good of a sample image is needed in order for it to work. This is exactly why CAPTCHA images will be intentionally blurred, malformed, and have added noise to the image. If it can be helped, of course try to supply the sharpest image possible, but I think it's a bit much to expect any IR software to be able to perfectly recognize images regardless of poor sample quality.
jlech
+1  A: 

see http://sourceforge.net/projects/zbar/

aron
Do you know how well zbar performs compared to zxing?
Rafael Vega
+1  A: 

There is a number of projects, some open source, listed at http://code.google.com/p/zxing/wiki/InterestingLinks

To read blurry barcodes such as the ones captured by a fixed focus camera (notably the iPhone 2G/3G), I think you need a commercial library, like the one used by the pic2shop iPhone app: http://www.visionsmarts.com/products/products.html

Disclosure: I am the lead developer of that library.

Benoit