Recognition

Version 2.0 of the Pen API significantly changes the way an application interacts with a recognizer. It allows an application to install multiple recognizers and use them selectively, first creating an HRC object for each recognizer to configure the recognition process. Version 2.0 provides many more recognition functions than did version 1.0, but places the full burden of recognition on the recognizer. All recognition functions are provided by the installed recognizer dynamic-link libraries (DLLs) and not by Windows.

The change in philosophy mentioned at the beginning of this chapter is particulary true with regard to recognition. Recognizers are now more clearly separate from the system and enjoy a corresponding freedom in their implementation. The Pen API defines the interaction between application and recognizer, but stops short of man-dating how a recognizer performs its tasks. The recognizer objects described in Chapter 8, "Writing a Recognizer," are called objects to emphasize that their forms are invisible to the application. The objects are "black box" entities, which the recognizer developer designs without restraint from the system.

For a description of the HRC object and other elements of recognition, see Chapter 5, "The Recognition Process."

Software for developers

Delphi Components
.Net Components
Software for Android Developers

More information resources

MegaDetailed.Net
Unix Manual Pages
Delphi Examples

Databases for Amazon shops developers

Amazon Categories Database
Browse Nodes Database