Current Chinese font library generation systems bear a major drawback that the user is required to write all characters contained in the font library, which is rather boring and time consuming. This paper proposes a system to automatically generate Chinese font libraries of high quality based on component assembling. An input set of a few characters for users is selected to write according to the instructive information of Chinese characters. Then components of each written characters are extracted. Several selected components are combined to construct each unwritten character. Finally the complete Chinese personal font library is obtained, which contains 6763 Chinese characters according to the GB2312 standard. Experimental results show that the proposed system can generate personal Chinese font libraries with dramatically shorter time and still keep excellent quality.

Abstract This paper presents a benchmark, which includes a manually-constructed database and evaluation tools.
Specifically, the database contains a number of images of Chinese characters represented in four commonly-used
font styles and corresponding stroke images manually segmented from character images. Performance of a given
stroke extraction method can be evaluated by calculating dissimilarities of the automatic segmentation results and
the ground truth using two specially-designed metrics. Moreover, the authors also propose a new method based on
Delaunay triangulation to effectively extract strokes from Chinese characters. Experimental results obtained by
comparing three algorithms demonstrate that the benchmark works well for the evaluation of stroke extraction
approaches and the proposed method performs considerably well in the application of stroke extraction for Chinese
characters.