|
|
-pattern recognition based on decision tree architecture, to achieve high-speed grayscale pattern matching of complex shapes or characters. Arguably the most powerful and widely used search tool.
- Two main phases, Minos Teach and Minos Search
- Classifiers (containing pos/neg info) are generated in Minos Teach and used in Minos Search
- Classifiers are created by recording sample images and marking objects to be identified in these images. Combined set of images comprises the training set, users can specify positive and negative examples and “don’t care” regions.
- Objects of all types are “learned” in a program with a user-friendly GUI
- Can match patterns simultaneously, once trained
- Special functionality for OCR/OCV classifier creation
- Includes a correlation tool to compare original and “detected pattern” correlation value for quality measurement.
<< BACK TO TOOLS |
|
|
