API published in: S60 3rd Ed FP 1
Link against: ocrsrv.lib
Required Capabilities
None
#include <ocrsrv.h>
Inherits MOCREngineInterface.
This set of interfaces offers the OCR engine which supports Layout Analysis on full images. User shall pass the handle of the image through LayoutAnalysisL first, and then pass block ids to RecognizeL to complete the recognition.
For Example: After passing the image handle to LayoutAnalysisL, the MOCREngineObserver::LayoutComplete will be called after the process. Then through the TOCRBlockInfo array (const TOCRBlockInfo* aBlock and const TInt aBlockCount), the user knows how many blocks have been identified together with their position information. Those blocks are marked with ids from Zero to aBlockCount - 1.
Then the user must specify which blocks are going to be further processed (Note that layout analysis only gets you the position of those blocks, but does not recognize the characters inside each block). The const RArray<TInt>& aBlock in MOCREngineLayoutRecognize::RecognizeL saves all the ids that are selected to be further processed. e.g. if you have 10 blocks from LayoutComplete, you can append 1, 3, 5, 7 to the TInt array and pass it to RecognizeL
Note that this class shall be used together with TEngineType set to EEngineLayoutRecognize
Public Member Functions |
|
| virtual void | LayoutAnalysisL (const TInt aBitmapHandle, const TOCRLayoutSetting aSettings)=0 |
| This method does the layout analysis for the full image. |
|
| virtual void | RecognizeL (const TOCRRecognizeSetting aSettings, const RArray< TInt > &aBlockIdsToRecognize)=0 |
| This method does the character recognition for the image on base of the result from LayoutAnalysisL. |
|
|
||||||||||||
|
This method does the layout analysis for the full image.
|
|
||||||||||||
|
This method does the character recognition for the image on base of the result from LayoutAnalysisL.
|