Image Processing and Data Storage

The preprocessing of a codex page normally entails the following steps

1. Preprocessing of the digitized images of medieval manuscripts: this contains basic image processing steps like

a) Color Calibration: this involves calibrating the colors of the digitized images of manuscripts scanned from different scanners to a standard color space.

b) Spatial Calibration: calibrating the images of different dpi i.e. to common dpi

c) Filtering: During analog to digital process conversion. Noise is often introduced to the images which are random or little disturbances of the images. Filtering works to remove this random disturbances.

d) Scaling and Duplication: In order to decrease the processing time, the digitized images are scaled to different resolutions like 75 dpi, 150 dpi to get preprocessed images of different resolutions.

2. Segmentation: Segmentation helps in separating the image objects (e.g. writing and illustrations) from the background. The complexity of this process can vary and it is one of the most elaborated operations in digital image processing.

3. Feature Extraction: Features describing the whole page and segmented objects like page size, written space, pictorial space can be measured using algorithm library.

4. Data storage: After the extraction, the features will be stored in the Metadata of the codex in XML Schema according to TEI.