Machine vision module toolkit added

Feature-based OCR module has been added to a software development toolkit for machine vision, for image capture, processing, analysis, display and archiving.

Matrox Imaging has announced the addition of a new module to its award-winning software development toolkit for machine vision - Matrox Imaging Library (MIL).

This high-level programming library offers an extensive set of optimised functions for image capture, processing, analysis, display and archiving.

Feature-based optical character recognition (OCR) is a powerful tool for applications that require character reading or verification.

"Feature-based OCR is extremely robust tool, reading character strings at any scale factor in both x and y, including different aspect ratios, allowing it to tolerate some perspective distortion " says Stephane Baldo, MIL Processing Software Leader.

"Feature-based OCR is powerful enough to read strings in poor or non-uniform lighting conditions, and automatically locates the string, even if it is rotated".

Feature-based OCR is ideal for applications where the location of the string and font are unpredictable, such as license plate reading.

The module includes a universal font which can be used in simple applications, bypassing font training altogether, and since MIL's feature-based OCR module supports reading multiple fonts simultaneously, additional fonts and/or characters can be used in conjunction with the universal font.

Back to top