Abstract

In desktop publishing, fonts are essential components in each document design. With the development of font design software and tools, there are thousands of digital fonts. Increasing the number of available fonts makes selecting an appropriate font, which best serves the objective of a design, not an intuitive issue. Designers can search for a font like any other file types by using general information such as name and file format. But for document design purposes, the design features or visual characteristics of fonts are more meaningful for designers than font file information. Therefore, representing fonts’ design features by searchable and comparable data would facilitate searching and selecting a desirable font. One solution is to represent a font’s design features by a code composed of several digits. This solution has been implemented as a computerized system called PANOSE-1 for Latin script fonts. PANOSE-1 is a system for classifying and matching typefaces based on design features. It is composed of 10 digits, where each digit represents a specific design feature. It is used within several font management tools as an option for ordering and searching fonts based on their design features. It is also used in font replacement processes when an application or an operating system detects a missing font in an immigrant document or website. Currently, PANOSE-1 is only defined for fonts that have Latin characters. Therefore there is a need for providing a model that describes and classifies fonts with Arabic characters.
In this research, a new model PANOSE-A is defined to extend PANOSE-1 coverage to support Arabic characters. The model defines eight digits in addition to the first digit of PANOSE-1which indicates the font script and family type. Each digit describes a visual or a design feature and takes value between 0-15. The meaning of 0 and 1 values is similar to what is defined for PANOSE-1. Each of the remaining values indicates a specific variation of its represented feature. Weight and contrast are two essential features in any font design. Two digits of the models describe the common variations of the weight and contrast features for text fonts. Another four digits describe the shape of some strokes that usually vary in their design between fonts. One digit describes the end shape of terminal strokes using three letters with different terminal strokes. Another digit describes the shape of the bowl stroke while the third digit describes the shape of curved stroke. The thresholds that used to define each shape class are taken from Naskh calligraphy guidelines. The fourth digit describes the shape of rounded strokes with enclosed counter. The shape classes of this digit are based on how the counter shape is similar to one of five geometric shapes. These basic geometric shapes are triangle, square, rectangular, oval and circle. The last two digits describe the characteristics of two important vertical references of the Arabic font design which are tooth and loop heights. The reliability of the model was evaluated by conducting two clustering processes on 30 fonts of Naskh style. The proposed PANOSE-A model was used to construct a similarity matrix for one the clustering processes while the other clustering process used a similarity matrix that was produced by using a font matching tool. The result clusters of the two clustering processes have been evaluated by silhouette coefficient. Silhouette is a method to measure data consistency validation within clusters. It indicates how objects are similar in their own cluster compared to other clusters. Clustering result based on the similarity matrix produced by font matching tool got 97.04 while using clustering result based on the similarity measured by PANOSE-A model got 98.24.The similarity between the results of the two clustering processes has been estimated, indicating that the model succeeded in classifying 85% of the fonts as a font matching tool.