There are multiple sectors where it’s required to be able to read strings of text, ranging from typical serial numbers up to expiration dates in various formats.
For the operation of our artificial vision OCR systems it’s necessary to use some samples from the characters that are to be recognized, which will be used as reference for the actual readings. This also permits to use custom characters without having to restrict oneself to predefined types.
Our devices can ensure a maximum of 100,000 acquisitions per minute, without restrictions on text direction. The process is sufficiently robust to allow retrieving data even in the case of text in degraded conditions.