Procedure for using the StringReader step

A StringReader context groups the string models and fonts definitions and settings in a StringReader file.

When a StringReader step is first inserted into your project, you will be able to choose to Import or Create a new String Reader context using an interactive utility from the Configuration pane. For more information on how to create one, see the Creating and testing a context using the MIL String Reader Interactive Utility section later in this chapter.

Once you have created or imported a context, you can adjust overall settings which will affect all the results of the StringReader step.

1. Add, edit, rename, or delete a string model, if necessary.

2. Set the Total number of strings to be read for the context.

3. You can also set the number of strings to read for each individual string model. See the Adjust the string model specific settings subsection of this section.

4. Draw the search region, if needed. To do so interactively, use the Define a region () toolbar button in the Project toolbar. See the Search regions section in Chapter 2: Building a project for more information.

5. Change the Search speed depending on the quality of the source image. A higher speed can be used with clean, high-contrast images.

   Increasing the speed can decrease the robustness of the operation if your image is not clean.

6. Set a Timeout (in milliseconds). If the read operation times out, all the strings read up to that point will be returned.

7. If your characters are very faint, or there is a lot of background noise, you might need to adjust the Minimum contrast. It sets the minimum intensity difference between character and background. Increasing the minimum contrast will typically speed up the read operation, as long as the level is less than the contrast of the actual characters.

8. You can set an alternate Space character or Separator character to insert into the result string, for example an underscore or a linefeed. The character must be specified with its ASCII code (95 for an underscore for example).

A string model is a container, within the StringReader context, that holds all the settings and constraints that control how to read one or many specific strings in the target image.

1. Set the Number of strings to read for this string model.

   Note that if the context's number of string to read is reached, the read operation stops, regardless of any outstanding string model setting.

2. Optimize by stopping the search as soon as one candidate with a score above the Certainty level is found.

3. Set the Foreground property to Black or White depending on the color of the text to read. If it is not always the same, set it to BlackOrWhite.

4. Set Minimum and Maximum size to the minimum and maximum number of characters the string can contain.

5. Specify the Maximum consecutive spaces to control whether several groups of characters on the same line should be considered part of the same string. If they are separated by less than this number of spaces, they will be included in the same string, and no space characters are reported in the result.

6. Modify the default string and character Acceptance, Homogeneity, Similarity score values if needed. The default values are generally appropriate for most applications. For more information on these properties, see the StringReader step scores section later in this chapter.

7. Set Constraints rules for specific character locations in the string. The rules can include which font, which type of characters (for example digits or uppercase), or even specific characters that must appear in certain positions.

For more information on some of the advanced settings available for string models, see the Creating and testing a context using the MIL String Reader Interactive Utility section later in this chapter.