Matrox Imaging Library (MIL) Tools

Character recognition

MIL X provides three tools for character recognition: SureDotOCR, String Reader, and OCR. These tools combine to read text that is engraved, etched, marked, printed, punched, or stamped on surfaces.
The SureDotOCR tool is uniquely designed for the specific challenge of reading dot-matrix text produced by inkjet printers and dot peen markers. Its use is straightforward—users simply need to specify the dot size­­­­­­ and the dimension, but not the location, of the text region. The tool reads text of variable length3, at any angle, with varying contrast, and/or on an uneven background. It interprets distorted and touching characters as well as characters of varying scale. It also accepts constituent dots that are of varying size and touching each other3. The tool recognizes punctuation marks and blank spaces. It supports the creation and editing of character fonts while including pre-defined fonts. The tool automatically handles multiple lines of text where each line can utilize a different font. The ability to set user-defined constraints, overall and at specific character positions, further enhances recognition rates. The SureDotOCR tool provides greater robustness and flexibility than case-specific techniques that convert dot-matrix characters into solid ones for reading with traditional character recognition tools.
The String Reader tool is based on a sophisticated technique that uses geometric features to quickly locate and read text made up of solid characters in images where these characters are well separated from the background and from one another. The tool handles text strings with a known or unknown number of evenly or proportionally spaced characters. It accommodates changes in character angle with respect to the string, aspect ratio, scale, and skew, as well as contrast reversal. Strings can be located across multiple lines and at a slight angle. The tool reads from multiple pre-defined (TrueType™ and Postscript™) or user-defined Latin-based fonts. Also included are ready-made Latin-based unified contexts for automatic number plate recognition (ANPR) and machine print. In addition, strings can be subject to user-defined constraints, overall and at specific character positions, to further increase recognition rates. The tool is designed for ease-of-use and includes String Expert, a utility to help fine-tune settings and troubleshoot poor results.
The OCR tool utilizes a template-matching method to quickly read text with a known number of evenly spaced characters. Once calibrated, the tool reliably reads text strings with a consistent character size even if the strings themselves are at an angle. Characters can come from one of the provided OCR-A, OCR-B, MICR CMC-7, MICR E-13B, SEMI M12-92, and SEMI M13-88 fonts or a user-defined font. Strings can be subject to user-defined constraints, overall and at specific character positions, to further increase recognition rates.