Optical Character Verification, or OCV, is a machine vision software tool used to inspect the print or marked quality of an Optical Character Recognition (OCR) string and confirm its legibility. In addition to checking that the content of the presented text string is correct, it will also inspect for quality, contrast and sharpness, and will flag or reject poor quality samples.
A key application using OCV is date/lot code verification (either printed on a label or directly marked on the package or product) for pharmaceutical, medical device, cosmetics and other consumer goods packaging.
OCR software is used to read a text string and will attempt to read even poor quality and damaged text, making a best guess at the data. On the other hand, OCV software inspects the quality and confirms the legibility of the text. This inspection is most often used to confirm that the printed date/lot codes are high quality and will be legible by a user in case of a product recall—a major concern in most packaging applications.
A font-based OCV tool, such as the one available in Visionscape I-PAK inspection system, uses stored templates of acceptable levels of character quality, which allows text to be easily validated. Capabilities such tolerance of rotated text, image brightness variations, breaks or dropouts in characters, and differentiating similar characters (B vs. 8, O vs. 0, or O vs. D, etc.) are highly desirable in any OCV vision software.
|Rotated text tolerance||Examples of print quality defects|