Recognition and autocompletion of partially drawn symbols by using polar histograms as spatial relation descriptors

• We present an approach for recognizing multi-stroke partially drawn symbols.
• It is invariant with respect to scale, number and order of strokes.
• It exploits a novel spatial descriptor to represent relations between two primitives.
• It is effective and efficient enough to enable the design of a system for symbol autocompletion.
• It outperforms the existing approaches in the recognition of partially drawn symbols.

We present an approach for recognizing multi-stroke hand-drawn symbols. The main feature of the approach is its capacity of recognizing partially drawn symbols. Furthermore, it is invariant with respect to scale, and supports symbol recognition independently from the number and order of strokes. The recognition technique is based on subgraph isomorphism and exploits a novel spatial descriptor, based on polar histograms, to represent relations between two stroke primitives. Using different symbol sets, both hand-drawn and artificially deformed, we evaluated the effectiveness of the approach in recognizing the symbols as a function of the number of primitives already drawn by the users. The results show that the approach gives a satisfactory recognition rate with partially drawn symbols, also with a very low level of drawing completion, and outperforms the existing approaches proposed in the literature. We also report the results of a user study aimed at evaluating whether the users can efficiently exploit symbol autocompletion.

Figure optionsDownload high-quality image (202 K)Download as PowerPoint slide