Handwritten character segmenting method and electronic equipment

Abstract

The invention discloses a handwritten character segmenting method and electronic equipment. The method comprises the steps that handwritten input strokes are received and displayed, external rectangles containing the handwritten input strokes are generated, and the external rectangles are defined as 'character frames'; the 'character frames' are overlapped and displayed outside the handwritten input strokes; subsequent handwritten input strokes are received, and all the strokes in the 'character frames' before the subsequent handwritten input strokes are segmented as independent characters when all the tracks of the subsequent handwritten input strokes fall outside the 'character frames'. Through the utilization of the design of the 'character frames' in the scheme, the identification of character strings is converted to the identification of single character, character segmentation does not need to be analyzed, and therefore the accuracy of the identification of handwritten character strings is improved greatly.

Description

technical field [0001] The embodiments of the present application relate to handwriting input, and in particular to a character segmentation method for handwritten character string input and related electronic equipment. Background technique [0002] With the widespread use of electronic devices such as electronic computers or mobile phones, how to effectively input information into computers or mobile phones is always a problem. Existing schemes include keyboard input and handwriting input. Keyboard input is because people's writing habits will be changed, so for people who are not familiar with keyboards, they need to be trained before they can be used. However, handwriting input has no limitation in this regard. People can input it like writing on paper, and the written handwriting is received by electronic devices and displayed or stored after being recognized. [0003] The earliest handwriting input methods are single character recognition, that is, users are required...

AI Technical Summary

Problems solved by technology

[0009] The biggest problem of Patent Document 1 is the setting of the "pause time". If it is too short, the computer may send a signal that the input is complete before a relatively complicated word is written; if it is too long, the user will wait for too long. Input is less efficient

Moreover, this method requires the user to pause for a few seconds after writing a word, which does not conform to the user habit of continuous handwriting, and is not very convenient to use

[0010] Patent Document 2 requires a special device to collect the three-dimensional motion trajectory of the input characters, so this method has great limitations

However, the algorithm in Patent Document 3 is more complex, with a large amount of calculation, and takes up more system resources.

[0011] In addition, existing handwritten string input methods (including the methods mentioned in the above three patent documents) all have a common deficiency: the prerequisite for the correct rate of handwritten string recognition in these methods is the character segmentation Correctness, once the characters are segmented incorrectly, it will cause the entire string to be recognized incorrectly (no matter how high the correct rate of individual character recognition is)

Images