Handwritten music sign recognition device and program
AssigneeKabushiki Kaisha Kawai Gakki Seisakusho
InventorAkihiro Fujita
Life of this patent
Estimating the term…
Abstract
A handwritten music sign recognition device (10) has: a touch information obtaining part (21) which obtains, as position information, a position of an input means on a screen every time the input means moves, from when the input means touches the screen until when the input means is moved away therefrom; a vector information creation part (22) which creates attributes of a vector indicating a trace of movement of the input means as vector information based on the position information; and a stroke characteristic amount derivation part (23) which derives a characteristic amount of a stroke based on the vector information included in the stroke information.
Claims17
A handwritten music sign recognition device, comprising:
- a touch screen which obtains, as position information, a position of an input on a screen every time the input moves, from when the input touches the screen until when the input is moved away therefrom to form a stroke;
- a processor, the processor configured to:
- divide the stroke into one or more vectors, each vector being a straight line or a point;
- create attributes of a vector indicating a trace of movement of the input as vector information based on the position information obtained by the touch screen, the position information including a start point having coordinates X1 and Y1 and an end point having coordinates X2 and Y2;
- create vector information for each vector based on an inclination of the vector calculated from the difference between Y1 and Y2 relative to the difference between X1 and X2, the vector information including a vector ID indicating one of nine types of vectors including that a direction of the vector is down to up, lower left to upper right, left to right, upper left to lower right, up to down, upper right to lower left, right to left, and lower right to upper left or that the vector is a directionless vector;
- store one or a plurality of pieces of stroke information for each object to be recognized in a memory, the stroke information including the vector information with respect to the one or more vectors based on the position information from when the input touches the screen until when the input is moved away therefrom;
- derive a characteristic amount of a stroke being a trace of movement of the input from when the input touches the screen until when the input is moved away therefrom, based on the vector information included in the stroke information;
- and
- recognize a sign indicated by the stroke based on at least one of the vector information and the characteristic amount of the stroke,
- wherein at least one of the two or more types of the characteristic amount of the stroke and the vector information with respect to the stroke is stored in a table arranged by vector ID, and
- wherein the processor derives as the characteristic amount of the stroke all of:
- a starting point-end point distance which is a distance of a straight line linking a starting point and an end point of the stroke;
- coordinates of four corners and coordinates of a center of a rectangle inscribed on the stroke;
- a connection type indicating a value corresponding to an angle formed by a vector included in the stroke and a vector inputted next to the vector;
- a rotation direction from a vector included in the stroke to a vector inputted next to the vector;
- a rotation angle from a vector included in the stroke to a vector inputted next to the vector;
- a left-right movement direction indicating a direction of left and right as a direction in which a vector included in the stroke is directed;
- an up-down movement direction indicating an up-down direction as a direction in which a vector included in the stroke is directed;
- a leftmost point index indicating a leftmost vector among vectors included in the stroke;
- a rightmost point index indicating a rightmost vector among vectors included in the stroke;
- a highest point index indicating a highest vector among vectors included in the stroke;
- and
- a lowest point index indicating a lowest vector among vectors included in the stroke, and stores the derived characteristic amount of the stroke in the stroke information.
A computer readable non-transitory recording medium recording a handwritten music sign recognition program causing a computer to execute:
- a touch information obtaining step which obtains, as position information, a position of an input step on a screen every time the input step moves, from when the input step touches the screen until when the input step is moved away therefrom to form a stroke;
- a division step which divides the stroke into one or more vectors, each vector being a straight line or a point;
- a vector information creation step which creates attributes of a vector indicating a trace of movement of the input step as vector information based on the position information obtained by the touch information obtaining step, the position information including a start point having coordinates X1 and Y1 and an end point having coordinates X2 and Y2;
- a vector information step which creates vector information for each vector based on an inclination of the vector calculated from the difference between Y1 and Y2 relative to the difference between X1 and X2, the vector information including a vector ID indicating one of nine types of vectors including that a direction of the vector is down to up, lower left to upper right, left to right, upper left to lower right, up to down, upper right to lower left, right to left, and lower right to upper left or that the vector is a directionless vector;
- a stroke information storage step which stores one or a plurality of pieces of stroke information for each object to be recognized, the stroke information including the vector information with respect to the one or more vectors created by the vector information creation step based on the position information from when the input step touches the screen until when the input step is moved away therefrom;
- a stroke characteristic amount derivation step which derives a characteristic amount of a stroke being a trace of movement of the input step from when the input step touches the screen until when the input step is moved away therefrom, based on the vector information included in the stroke information;
- and
- a sign recognition step which recognizes a sign indicated by the stroke based on at least one of the vector information and the characteristic amount of the stroke, wherein at least one of the two or more types of the characteristic amount of the stroke and the vector information with respect to the stroke is stored in a table arranged by vector ID, wherein the processor derives as the characteristic amount of the stroke all of:
- a starting point-end point distance which is a distance of a straight line linking a starting point and an end point of the stroke;
- coordinates of four corners and coordinates of a center of a rectangle inscribed on the stroke;
- a connection type indicating a value corresponding to an angle formed by a vector included in the stroke and a vector inputted next to the vector;
- a rotation direction from a vector included in the stroke to a vector inputted next to the vector;
- a rotation angle from a vector included in the stroke to a vector inputted next to the vector;
- a left-right movement direction indicating a direction of left and right as a direction in which a vector included in the stroke is directed;
- an up-down movement direction indicating an up-down direction as a direction in which a vector included in the stroke is directed;
- a leftmost point index indicating a leftmost vector among vectors included in the stroke;
- a rightmost point index indicating a rightmost vector among vectors included in the stroke;
- a highest point index indicating a highest vector among vectors included in the stroke;
- and
- a lowest point index indicating a lowest vector among vectors included in the stroke, and stores the derived characteristic amount of the stroke in the stroke information.
Dependencies are inferred from the claim text.