Overlapped handwriting input method

摘要

An overlapped handwriting input method, which includes: moving a touch unit on a touch screen, recording and displaying the track of a stroke on a handwriting area of the screen; determining whether the stroke currently being written and all previously input strokes having a color A belong to the same character; if yes, keeping the current stroke color unchanged; otherwise, determining whether an inactive visual layer character is present on the screen; if yes, clearing an image of the inactive character and related stroke information; otherwise, defining as the inactive character all strokes having the color A, excluding the current stroke; submitting data of the current inactive character to a recognition engine, and outputting a recognition result. Thus, the interface friendliness for users to continuously input handwriting according to their natural writing practice is ensured, and multiple handwritten characters can be written continuously without pause, thereby improving handwriting input efficiency.

主权项

1. An overlapped handwriting input method, comprising:
Step 100: performing initialization process: clearing a screen, and clearing an inactive visual layer character; Step 110: enabling a handwriting pen or finger to touch a touch screen, and starting to input a stroke; Step 120: moving the handwriting pen or finger on the touch screen, recording and displaying the track of the stroke on a handwriting area of the touch screen, and drawing the track of the current stroke by using a first color value; Step 130: enabling the handwriting pen or finger to leave the touch screen, and ending the input of the current stroke; Step 140: determining whether the stroke being currently written and all previously input strokes having the first color value belong to the same character, and if yes, going to Step 240; otherwise, going to Step 150, wherein whether the stroke being currently written and the previously input strokes belong to the same character is determined according to a relationship between geometric position information of the stroke being currently written and geometric position information of a character constituted by all the previously input strokes; assistant determination is performed in combination with a recognition result, which specifically comprises: performing recognition on the character constituted by all the previously written strokes having the first color value except for the current stroke, and if the recognition reliability is high, determining that the current stroke and the previous strokes do not belong to the same character; the method of determining the recognition reliability of a recognition engine recognizing a certain handwritten character is: calculating a difference between a probability of the character to be recognized belonging to a first candidate character and a probability of the character to be recognized belonging to a second candidate character, and if the difference is greater than a certain set threshold, determining that the reliability of the recognition result provided by the recognition engine is high; the method of calculating the probability of the character to be recognized belonging to a certain candidate recognition result character is provided by a recognition classifier; a quadratic discriminant function classifier is used as the classifier, and the probability is approximate to a negative exponent of a recognition distance provided by the quadratic discriminant function classifier; and Step 140 comprises the following steps: Step 141: determining whether the current stroke is a first stroke input by a user; if yes, going to Step 146; otherwise, going to Step 142; Step 142: determining whether the current stroke is a stroke of a character being newly written at the right of the previously stroke; if yes, going to Step 145; otherwise, going to Step 143; Step 143: determining whether the current stroke overlaps another stroke having been written; if yes, going to Step 144; otherwise, going to Step 145; Step 144: determining whether the overlapping degree of the current stroke and the stroke having been written is greater than a certain given threshold; if yes, going to Step 146; otherwise, going to Step 145; Step 145: returning a determination result that the currently input stroke and the previously input stroke may belong to the same character; and Step 146: returning a determination result that the currently input stroke and the previously input stroke do not belong to a same character; Step 150: determining whether an inactive visual layer character is present on the screen, wherein the inactive visual layer character refers to the last completely written character that has undergone recognition and color transformation, and the character is defined in Step S170; if yes, going to Step 160; otherwise, going to Step 170; Step 160: clearing an image of the inactive visual layer character and related stroke information; Step 170: defining a character constituted by all strokes having a first color value, except for the current stroke, as the inactive visual layer character; Step 180: submitting data of the current inactive visual layer character to the recognition engine for recognition, and output the recognition result; Step 190; determining whether the recognition result provided by the recognition engine is reliable enough; if yes, going to Step S200; otherwise, going to Step 220, wherein the method of determining the recognition reliability of the recognition engine recognizing a certain handwritten character comprises: calculating a difference between a probability of the character to be recognized belonging to a first candidate character and a probability of the character to be recognized belonging to a second candidate character, and if the difference is greater than a certain set threshold, determining that the reliability of the recognition result provided by the recognition engine is high; the method of calculating the probability of the character to be recognized belonging to a certain candidate recognition result character is provided by a recognition classifier; a quadratic discriminant function classifier is used as the classifier, and the probability is approximate to a negative exponent of a recognition distance provided by the quadratic discriminant function classifier; Step 200: fading the color of the strokes of the current inactive visual layer character until disappearance of the color; Step 210: clearing the image of the inactive visual layer character and the related stroke information; going to Step 230; Step 220: transforming the color of the strokes of the current inactive visual layer character to a second color for display; Step 230: determining whether a new stroke is input, and if yes, waiting for the user to input the new stroke, and going to Step 120; otherwise, going to Step 260; Step 240: keeping the color of the current stroke unchanged; Step 250: determining whether a new stroke is input, and if yes, waiting for the user to input the new stroke, and going to Step 120; otherwise, going to Step 260; Step 260: submitting the data of the character being currently written to the recognition engine for recognition, and outputting a recognition result; and Step 270: ending.