IntelliText OCR improves consistency and readability of a variety of substandard marks, such as inkjet marks or marks on noisy backgrounds. Significant improvements over Standard OCR can be seen in the following areas:
Improved invariance to background noise and uneven illumination;
The ability to read rotated OCR text at any angle;
The ability to handle fixed tilts and slants in characters;
Characters can be scaled up or down to make them the optimal size for reading;
The ability to create a variable match string using regular expressions;
The ability to set character width and height to resolve segmentation issues.
See the IntelliText OCR Tool Parameters section for details.
Use Default Font: Select whether to use pre-trained fonts. Uncheck to use only user-trained fonts.
Font: Select the font to use.
Confidence: Allows you to specify the minimum level of character identification confidence required before a character will be read and output. (Range: 0.0 to 1.0)
Character Polarity: Allow you so specify Dark Characters – black characters on a white background – and Light Characters – white characters on a black background.
Min./Max. Height: Allow you to configure the minimum and maximum character height limits. (Range: 10 to 200)
Min./Max. Width: Allow you to configure the minimum and maximum character width limits. (Range: 1 to 200)
Minimum Character Pixels: Allow you to configure the minimum size of a character, thereby eliminating segments that are too small to be characters by themselves. The default setting is 20 pixels.
Scaling Factor – If a symbol is very large or very small, adjust Scaling Factor to make the character height 30 pixels tall – the best height for the algorithm. For example, if a character is 60 pixels tall, set Scaling Factor to 0.5. See the Scaling Factor section for details.
Unknown Character is the character that will be returned from the font reading process when no match can be made within the selected font. The default: character is ?. For example, if the tool cannot decipher the 'H' in "MICROHAWK", the output string will be MICRO?AWK.
Multiline Separator: If the string has multiple lines, the Multiline Separator will be inserted into the string between lines.
Image Binarization: Convert grayscale images to black and white (binary) to optimize the readability of poorly segmented characters. See the Image Binarization section for details on how to adjust each parameter.
Binarization Mean Factor: Set the average coefficient for binarization. This can be set when the binarization threshold mode is the [Fixed Template] method. (Range: 0.0 to 1.0)
See the Image Binarization section for details on how to adjust each parameter.
Binarization Dynamic Range: Set the dynamic range for binarization. This can be set when the binarization threshold mode is the [Fixed Template] method. (Range: 0.0 to 1.0)
See the Image Binarization section for details on how to adjust each parameter.
Discard Boundary Characters: Select whether to ignore characters that touch the search box.
Character Slant Detection: Turn on if characters are expected to slant heavily to the right or the left. Detection is better for non-slanted characters when Character Slant Detection is turned off.
Single Line Detection: Turn on when there is only a single line of characters. This helps optimize character segmentation for difficult symbols. Often the best option for difficult symbols is to add one OCR Tool per line and then to use the String Format Tool to concatenate the strings into a single string.
Auto Teach Options: Learn Match String.
Match String: Allow you to specify the string that must be matched for the OCR Tool to pass. See the OCR Tool Match String for details. (up to 254 strings)
String Match Method: Select the method of string matching. See the OCR Tool Match String section for details.
[Disable]: Does not perform string matching.
[Basic]: Basic string matching is performed on the result string.
[Regular Expression (TRE)]: The result string is tested against a user-defined regular expression in the match string field.