I think that the Tesserct uses the text outline as the OCR classification features might have problem when the font size is relative small, E.G. less than 40 pixel height and 15 pixel width.
Please comment? On Apr 13, 3:31 pm, mw18888 <[email protected]> wrote: > The tesseract extracts the micro feature from image blob. Eventually, > the function will find out the outlines of the character. Each outline > is defined as two points, the start and the end. > > My question is accuracy of defining these list of two-points if the > character is a small size font, E.G x<10 and y<20 ? > > Thank. > > Code segment in mfx.cpp. > > / > *--------------------------------------------------------------------------- > */ > MICROFEATURE ExtractMicroFeature(MFOUTLINE Start, MFOUTLINE End) { > /* > ** Parameters: > ** Start starting point of micro-feature > ** End ending point of micro-feature > ** Globals: none > ** Operation: > ** This routine computes the feature parameters which > describe > ** the micro-feature that starts and Start and ends at > End. > ** A new micro-feature is allocated, filled with the > feature > ** parameters, and returned. The routine assumes that > ** Start and End are not the same point. If they are > the > ** same point, NULL is returned, a warning message is > ** printed, and the current outline is dumped to stdout. > ** Return: New micro-feature or NULL if the feature was > rejected. > ** Exceptions: none > ** History: 7/26/89, DSJ, Created. > ** 11/17/89, DSJ, Added handling for Start and End same > point. > */ > MICROFEATURE NewFeature; > MFEDGEPT *P1, *P2; > > P1 = PointAt (Start); > P2 = PointAt (End); > > NewFeature = NewMicroFeature (); > NewFeature[XPOSITION] = AverageOf (P1->Point.x, P2->Point.x); > NewFeature[YPOSITION] = AverageOf (P1->Point.y, P2->Point.y); > NewFeature[MFLENGTH] = DistanceBetween (P1->Point, P2->Point); > NewFeature[ORIENTATION] = > NormalizedAngleFrom (&((P1)->Point), &((P2)->Point), 1.0); > ComputeBulges(Start, End, NewFeature); > return (NewFeature); > > > > > > > > } /* ExtractMicroFeature */ -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/tesseract-ocr?hl=en
