System Overview

GiDoc is implemented as a set of Gimp plug-ins, and thus it has to be installed first into Gimp . As Gimp , GiDoc is licensed under the GNU General Public License, and it can be downloaded from iDoc . To run GiDoc , we must first run Gimp and open a document image. Gimp will come up with its high-end user interface, which is often configured to only show the main toolbox (with docked dialogs) and an image window. GiDoc can be accessed from the menubar of the image window (see Figure 1).

Figure 1: Image window and GiDoc menu.

As shown in Figure 1, the GiDoc menu includes six entries:

• Advanced options: second-level menu where experimental features of GiDoc are grouped.
• 0: Preferences: opens a dialog to configure global options, as well as more specific options for preprocessing, training and recognition.
• 1: Block Detection: performs block detection on the current image.
• 2: Line Detection: detects (straight) text baselines in the current block.
• 3: Training: reads all text line images that have been already transcribed and trains from them statistical models for prediction of transcriptions.
• 4: Transcription: opens the interactive-predictive transcription dialog of GiDoc .