Hi, I am working on a feasibility study for an automatic label inspection. I made a good expirience with vision, but less with OCR. I faced several problems and I cannot understand whether is a matter of settings or a matter of image quality.The first problem is the splitting of the chars, the second is the stability and the repeatabilty of the reading. I fixed the splitting of the chars implementing a new algorithm. I was not able to find a configuration of settings that was able to work properly. I think that the "Bounding Rect Width" does not work as it should.The stability of the reading is still an open point. If I run continuously the test acquiring always the same images on the same pieces the results can be 80% pass and 20% fail. It happens that one char that for several previous run had a high score (about 910) suddenly fails (score 0) Could you take a look to the images and tell me if they are good enough?In Runtime I am using the same setting than I used for training: FixedThreshold, Read Strategy, Read Resolution, Acceptance Level.Do you have any suggestion about the stability? Thanks a lot. Marry Christmas.

Machine Vision

OCR for label inspection

To upsample, use the resample function and set the new dimensions to values proportionally larger than the old dimensions (x2, x3, etc).

Classification is a more generic version of OCR. It can handle rotations and identifies objects in any position. OCR requires sequential characters on a single line. I'm not sure how similar the actual identification algorithms are.

Bruce

Bruce Ammons
Ammons Engineering

Message 11 of 11

(961 Views)

Machine Vision

OCR for label inspection

Re: OCR for label inspection