초록 ▼

A system according to the invention identifies image data points defining a curve. The method comprises the steps of determining a start point and an end point for the curve, establishing a search space that includes at least the start point, the end point, and other image data elements comprising t

A system according to the invention identifies image data points defining a curve. The method comprises the steps of determining a start point and an end point for the curve, establishing a search space that includes at least the start point, the end point, and other image data elements comprising the curve, and searching the search space using a dynamic programming algorithm to locate image data elements corresponding to the curve. Another embodiment consistent with the present invention identifies image data points defining a curve. The method comprises the steps of determining a start point and an end point for the curve, generating a model of the curve, establishing a search space that includes at least the start point, the end point, and other image data elements comprising the curve, and searching the search space using a dynamic programming algorithm and the model for the curve to locate image data elements corresponding to the curve. Yet another embodiment consistent with the present invention matches a first curve to a second curve. The method comprises the steps of identifying a first curve, identifying a second curve, generating a higher order distance measure for comparing the first curve and the second curve, and matching the first curve to the second curve using the higher order distance measure.

대표청구항 ▼

A system according to the invention identifies image data points defining a curve. The method comprises the steps of determining a start point and an end point for the curve, establishing a search space that includes at least the start point, the end point, and other image data elements comprising t

A system according to the invention identifies image data points defining a curve. The method comprises the steps of determining a start point and an end point for the curve, establishing a search space that includes at least the start point, the end point, and other image data elements comprising the curve, and searching the search space using a dynamic programming algorithm to locate image data elements corresponding to the curve. Another embodiment consistent with the present invention identifies image data points defining a curve. The method comprises the steps of determining a start point and an end point for the curve, generating a model of the curve, establishing a search space that includes at least the start point, the end point, and other image data elements comprising the curve, and searching the search space using a dynamic programming algorithm and the model for the curve to locate image data elements corresponding to the curve. Yet another embodiment consistent with the present invention matches a first curve to a second curve. The method comprises the steps of identifying a first curve, identifying a second curve, generating a higher order distance measure for comparing the first curve and the second curve, and matching the first curve to the second curve using the higher order distance measure. tion, and a number of directions for a stroke from the outputted coordinate string of the handwritten strokes; a stroke candidate output unit operable to compare the obtained stroke information with each of the sets of standard stroke information and output the plurality of stroke candidates; and an evaluation value adding unit operable to add an evaluation value to each of the plurality of stroke candidates according to a degree of matching between the obtained stroke information and the standard stroke information of each of the plurality of stroke candidates, and wherein said character detecting unit comprises: a character candidate detecting unit operable to ignore stroke candidates with the evaluation value showing a poor match and detect character candidates using remaining stroke candidates; and a correct character detecting unit operable to detect a correct character out of the detected character candidates. 3. The character input apparatus of claim 2, further comprising a coordinate string detecting unit operable to detect coordinate strings, out of the coordinate strings outputted by said coordinate output unit, that are inputted at least a predetermined time after an immediately preceding coordinate string, wherein said character candidate detecting unit detects the character candidates by setting the stroke candidates corresponding to the coordinate strings detected by said coordinate string detecting unit as first strokes of characters. 4. The character input apparatus of claim 3, wherein said correct character detecting unit comprises: a word dictionary in which character orders for a plurality of words are registered, each of the character orders corresponding to a different word; and a word detecting unit operable to detect, when either one of the character candidates detected by said character candidate detecting unit or a combination of the character candidates is registered in said word dictionary, a corresponding word as the inputted handwritten characters. 5. The character input apparatus of claim 2, further comprising: a word dictionary in which character orders for a plurality of words are registered, each of the character orders corresponding to a different word; and a word detecting unit operable to detect, when either one of the character candidates detected by said character candidate detecting unit or a combination of the character candidates is registered in said word dictionary, a corresponding word as the inputted handwritten characters. 6. The character input apparatus of claim 2, wherein said correct character detecting unit comprises: a probability dictionary for showing a numeric value of a probability of each pair of adjacent characters appearing in a character string; a high probability character detecting unit operable to generate character strings by combining the character candidates detected by said character candidate detecting unit without changing an order of the character candidates, calculate a value for each of the character strings by summing up the numeric values of the probability for pairs of adjacent characters that constitute a character string and dividing the sum by a number of characters that constitute the character string, and detect a character string with a highest value as the inputted handwritten characters. 7. The character input apparatus of claim 1, further comprising a coordinate string detecting unit operable to detect coordinate strings, out of the coordinate strings outputted by said coordinate output unit, that are inputted at least a predetermined time after an immediately preceding coordinate string, wherein said character detecting unit comprises: a character candidate detecting unit operable to detect a plurality of character candidates by setting the plurality of stroke candidates corresponding to the coordinate strings detected by said coordinate string detecting unit as first strokes of characters; and a correc t character detecting unit operable to detect the character as a correct character out of the plurality of detected character candidates. 8. The character input apparatus of claim 1, wherein said character detecting unit comprises: a character candidate detecting unit operable to detect a plurality of character candidates; a word dictionary in which character orders for a plurality of words are registered, each of the character orders corresponding to a different word; and a word detecting unit operable to detect, when either one of the character candidates detected by said character candidate detecting unit or a combination of the character candidates is registered in said word dictionary, a corresponding word as the inputted handwritten characters. 9. The character input apparatus of claim 1, wherein said character input apparatus is operable to be set to a learning mode for learning handwritten characters, wherein when said character input apparatus is in learning mode, said stroke matching unit adds an evaluation value to each stroke candidate according to a degree of matching between the obtained stroke information and the standard stroke information of the stroke candidate registered in said stroke dictionary, and wherein said character input apparatus further comprises: a stroke dictionary additional register unit operable to additionally register, when the evaluation value for a stroke candidate is smaller than a predetermined evaluation value which shows a poor match, the stroke information for the stroke candidate as the standard stroke information in said stroke dictionary; and a character dictionary additional register unit operable to additionally register a stroke order that includes the stroke candidate for a character to be learned in said character dictionary. 10. The character input apparatus of claim 1, wherein said stroke dictionary stores each of the sets of standard stroke information together with a flag that indicates whether each of the plurality of strokes has a possibility of being a first stroke of a character, and said stroke matching unit compares the stroke information obtained from a first coordinate string outputted by said coordinate output unit only with the stroke information having the flag that indicates a first stroke in said stroke dictionary. 11. The character input apparatus of claim 1, further comprising a coordinate string detecting unit operable to detect coordinate strings, out of the coordinate strings outputted by said coordinate output unit, that are inputted at least a predetermined time after an immediately preceding coordinate string, wherein said stroke dictionary stores each of the sets of standard stroke information together with a flag that indicates whether each of the plurality of strokes has a possibility of being a first stroke of a character, and said stroke matching unit compares stroke information obtained from a first coordinate string outputted by said coordinate output unit only with stroke information having the flag that indicates a first stroke in said stroke dictionary. 12. The character input apparatus of claim 1, wherein said stroke matching unit comprises: an overlap detecting unit operable to detect from the coordinate string outputted by said coordinate output unit whether a stroke has a line segment that partially or completely overlaps another stroke preceding the stroke, and if said overlap detecting unit detects overlapping strokes, said character detecting unit treats the latter of the overlapping strokes as belonging to a different character than the former stroke. 13. The character input apparatus of claim 1, further comprising, an off-stroke information detecting unit operable to detect off-stroke information from an end point of one coordinate string and a start point of a following coordinate string outputted by said coordinate output unit, wherein said character dictionary further stores, when a c