JPH1138965A - Sheet of music recognition method and recording medium recording sheet of music recognition program and readable by computer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a sheet of music recognition method enabling such a sheet of music recognition that an accurate timing of utterance can be obtained and a computer-readable record medium in which the sheet of music recognition program is stored. SOLUTION: In the sheet of music recognition method of reading sheet of music images, recognizing the sheet of music signs, and converting them into a performable and/or expressible sheet of music data form, notes and rests relating to utterance during performance of measures, where tuplet number exists, are taken out from the recognition result in which five staffs, bar lines, notes, rests, tuplet signs, and other signs coexist in the sheet of music, and are grouped, etc., relating to the delimitation of beats, based on information on linked flags information and stem information, and are combined with other groups for recognizing a connected note with hooks as an object most suitable for position, time value of composed sound, and phrasing for notation, so as to satisfy a tone length condition as tuplet number object so that, in groups combined by centering a reference group close to the tuplet number, sum of tone length of composing notes becomes a multiple of integer of the tuplet number to a shortest tone length in the group.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for recognizing a musical score including a tuplet and a computer-readable recording medium on which a program for recognizing a musical score is recorded.

[0002]

2. Description of the Related Art When a musical score is mechanically recognized and converted into a playable musical score data format, it is necessary to determine a sounding timing that accurately corresponds to a chord or a tuplet. To get it, it is necessary to determine the note to be tupled. Similarly, when converting to musical score data that can be displayed, if the note to be a tuplet is not correctly recognized, an incorrect display will be made. Conventionally, in order to determine a tuplet target, as shown in FIG. 12, note heads held in recognition results of notes and rests in a bar where a tuplet numeral (3 in the drawing) exist. Utilizing feature points such as the center 20 and the stem end point 21 and the like, the one having a close positional relationship with the feature point of the tuplet numeral (display position of the tuplet numeral) is selected, and the selected note together with the same stem chord. It was targeted for tuplets. In addition, a method of avoiding inconsistency in pitches by selecting notes / rests of the same pitch and selecting the number of tuplets when selecting the target is adopted.

[0003]

In an actual musical score, FIG.
As shown in FIG. 5, there is a continuous hook 22 that represents a phrasal phrase (beat division), and those connected with the same number of digits within the continuous hook also constitute a phrase. On the other hand, a tuplet combines several notes and replaces them with a sharp beat, so that the tuplet object does not cut this phrase or straddle another phrase. However, in the above-mentioned method, since such a regularity is ignored, as shown in FIG. 12, the judgment is made only by the coordinate values, so that the note which ignores the phrase is selected, and the wrong target is selected. Often. Also, in an actual musical score, the object of the tuplet number and the position of the tuplet number may be displayed such that only one of the objects is close and the other objects are far away. In the above method, as shown in FIG. 13, even if the closest object is selected, it is unlikely that the remaining objects will be selected. In the example shown in the figure, a note close to the tuplet number 3 is selected as a target in the drawing, but notes on both sides of the tuplet number are separated from the tuplet number, and are not targeted for tuplet. Further, in the case of a triplet, as shown in FIG. 14, notes and rests having a note length of 1: 2 are often targeted for tuplets (in the drawing, quarter rests and eighth notes). Is a triplet), but in this case, the tuplet numbers and the constituent notes
Since the position of the rest is often separated (the position of the tuplet numeral 3 and the quarter rest and the eighth note are separated on the drawing), they are often not recognized as targets. In addition, tuplet objects are not always limited to the same note length, and the conventional method does not cope with such a case.

The present invention has been made in view of the above-mentioned problems of the prior art, and automatically detects an accurate tuplet object which is not inconsistent with a phrase or a pitch in a notation, thereby performing and / or playing. Alternatively, it is necessary to provide a music score recognition method and a computer readable recording medium storing a music score recognition program that enables recognition of a music score that can obtain accurate pronunciation timing when converted into a music data format that can be displayed. Is what you do.

[0005]

Therefore, the configuration of the invention according to claim 1 of the present application (the same applies to the configuration of a computer-readable recording medium according to claim 7) reads a score image and recognizes the score symbol. The present invention relates to a music score recognition method for converting musical data into a musical score data format that can be played and / or displayed, wherein recognition of mixed staffs, bar lines, notes, rests, tuplets, and other symbols in the music. From the results,
The notes and rests related to the pronunciation during performance in the bar where the tuplet number exists are extracted, and the grouping related to the beat delimiter is performed based on the hook information and the stem information. In the case of 3, a pair of adjacent notes whose pitch is 1: 2
At a middle position of the rest, a virtual group having these notes and rests as constituent elements is generated, and this is added to the group as a dummy group, and from a tuplet numeral to a note head center or a stem end point. If the tuplet number is in parentheses, the group is further divided into a reference group that can be a reference for a tuplet target search and an additional group that cannot be a reference according to the distance of the tuplet. A note of the following formula 1 as a tuplet object in which the total note length of the constituent notes is an integral multiple of the tuplet number in the group to be combined, centering on the reference group close to the note number. The basic feature is to perform group combination with other groups so as to satisfy the length condition, and to recognize the tuplet object most suitable for position, duration of constituent sounds, and phrasing in notation by the group combination. .

[0006]

[Equation 1] (sum of note length in group) = (shortest note length in group) × (tuplet number) × (arbitrary integer)

According to a seventh aspect of the present invention, there is provided a recording medium according to the first aspect, wherein the above-described procedure is a grouping function, a dummy group adding function, a reference group determining function, and a group combining function, and these functions are executed. The present invention is provided as a computer-readable recording medium storing a musical score recognition program.

[0008] Unlike the note lengths which each of them has as a single note / rest alone, a tuplet object composed of various note lengths combining several of these notes and replacing them with a sharp beat number Is not exactly recognizable when judged based on the position of the tuplet numeral and the position of the note / rest. However, in this configuration, by using the concept of a group divided based on a phrasal phrase (beat delimiter), it is possible to eliminate errors in recognition of a tuplet target based on the above determination. become able to. In other words, by using groups based on phrases as search units for tuplets,
Phrase inconsistency of the target note is eliminated, and the tuplet number becomes more accurate (by selecting this group as the minimum unit of the tuplet target, there is no selection of the target that divides the phrase). Furthermore, taking advantage of the property that the target of the tuplet will always satisfy the above-mentioned pitch condition,
Since the above-described group combination is performed, a tuplet object that is far from a tuplet numeral is also evaluated in connection with the reference group, so that various tuplet shapes can be handled. As a result, it is possible to recognize the tuplet object most suitable for the position, the sound length of the constituent sound, and the phrasing in notation. In addition, when performing musical score recognition corresponding to consecutive hook notes, it is necessary to include consecutive hook information as shown in FIG. 3 described later in the recognition result, and the recognition result is used as it is for the above grouping. it can. In the case of triplets, notes and rests with a preceding and succeeding note length of 1: 2 are often subject to tuplets, and as described above, a combination of them is added as a dummy group. By doing so, even if these are displayed at distant positions, they can be targeted for tuplets.

In the above-described musical score recognition method (recording medium), when performing the grouping related to beat division (by the grouping function), as shown in FIG. And This is because beat divisions are indicated by the same hook. Notes and rests that are not consecutive hooks are in a single group, and chords that have the same stem and are pronounced simultaneously are in the same group. Of course, they cannot be in separate groups, nor are they of a nature that can be separately grouped into other groups. As described above, the same chord chords are put into the same group, but chords that do not share the stems are put into another group, and the groups having the same gate sound (chord) are ignored when the groups are combined. Therefore, it is possible to avoid selecting an object over a plurality of voice parts. When grouping the same series of hooks, as shown in FIG. 2, those connected with the same number of digits are grouped. This group is hierarchical according to the number of digits.

[0010] At the time of grouping of the same hook, each piece of recognized note information holds a unique ID for each hook, and furthermore, the note within the same hook holds the number of digits to be joined to the preceding and following notes. By using this, those connected with the same number of digits in the hook are grouped. As shown in FIG. 3, each note data holds three consecutive hook information of a consecutive hook ID, a consecutive hook connection digit number (left), and a consecutive hook connection digit number (right). Among these, the hook ID is a unique ID for each note group sharing one or more hooks (the same hook ID).
= "Same hook"). The consecutive hook ID of data relating to a single note / rest that is not a consecutive hook is set to 0 to distinguish it from a note that is a consecutive hook. The number of connected digits indicates the number of left and right consecutive hook digits at the end of the note. Further, in the case of the same note chord, the consecutive hook information of each note data is equal.

As described above, the group is set to be the minimum unit of the tuplet object, so that the determination of the tuplet object is naturally made in the group unit. Here, since the feature points in the group are set based on the note / rest that constitutes the group and that is closest to the tuplet number, it is accurate regardless of the position of other notes / rests in the group. It is possible to evaluate various positional relationships. For example, as shown in FIG. 4, only the characteristic points (positions indicated by arrows) of each group are reflected in the distance between the group and the tuplet numeral, and the distance between the other notes / rests and the tuplet numeral is Has no meaning. On the other hand, in the dummy group consisting of a pair of notes and rests having a note length of 1: 2, note points are set not at note positions or rest positions but at intermediate positions between constituent notes or constituent rests.・ Even if the rest font is far from the tuplet digits, an accurate positional relationship with the tuplet digits can be obtained. Therefore, even in the case as shown in FIG. 14, a triplet object can be accurately recognized.

As a prerequisite for the group combination, based on the tuplet numbers and the positional relationship between the groups, a group within a certain range is set as a reference group based on the tuplet numbers, and a group outside the range is set as an additional group. I do. Then, the combination of the groups is always performed at the beginning of the reference group, so that the tuplet object always includes one reference group. At least one of the groups subject to tuplets
This is because one (reference group) is located near a tuplet numeral. An additional group can be subject to tuplets only by combining with the reference group. As a result, an object that is far from the tuplet numeral is evaluated in connection with the reference group regardless of the position from the tuplet numeral. When combining a reference group with another group, be sure to use the group within the chain hook as a reference and combine it with the group at the next higher level in the same chain when combining with another group. . As a result, selection over a phrase is not performed.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the accompanying drawings. FIG. 5 shows an embodiment of a music score recognition apparatus which operates by reading a computer-readable recording medium storing a music score recognition program according to the present invention into an external storage device (or a CD-ROM drive or the like) such as a flexible disk drive FDD5 described later. It is a block diagram showing an example composition. This device is obtained by adding a scanner and a MIDI interface circuit to the configuration of an electronic computer such as a personal computer. The CPU 1 is in the ROM 2 or RA
It is a central processing unit that controls the entire score recognition apparatus based on a program stored in M3. Further, a timer circuit that interrupts the CPU 1 at a predetermined cycle set in advance is incorporated. The RAM 3 is used as an image data buffer, a work area, and the like in addition to the program area. The hard disk drive HDD4 and the flexible disk drive FDD5 store programs, image data, performance data, and the like. The CRT 6 displays video information output from the CRT interface circuit 7 based on the control of the CPU 1, and the information input from the keyboard 8 is sent to the CPU via the keyboard interface circuit 9.
It is taken into 1. The printer 10 prints print information output from the printer interface circuit 11 under the control of the CPU 1.

The scanner 12 optically scans a (printed) score, for example, and converts it into binary or gray scale image data. The scanner 12 may be of any type such as a flatbed type, a handy type, and a feeder type. Scanners can be used. The image information read by the scanner 12 is sent to the RA via the scanner interface circuit 13.
M3 or HDD4. The MIDI interface circuit 14 includes an external MID such as a sound source module.
This is a circuit for transmitting and receiving MIDI data to and from an I device. The bus 15 connects each circuit in the musical score recognition device,
Various data, programs, addresses, etc. are exchanged. In addition, a pointing device such as a mouse or a serial interface circuit such as RS232C may be provided.

FIG. 6 is a flowchart showing the main processing of the CPU 1. In step S1, the scanner 1
2, the image of the musical score is loaded into the RAM 3. The image is captured as binary image data. In step S2, a staff is detected from the musical score image binarized and captured in this manner. In step S3, inclination correction is performed for each staff based on the amount of shift with respect to the horizontal line.

In step S4, labels are extracted, and in step S5, symbols such as notes and rests are recognized. Thereafter, in step S6, the recognition of the tuplet target is performed according to the program read from the recording medium of the present invention. Finally, in step S7, conversion to the musical score data format is performed, and the process ends.

The recognition of the tuplet object in step S6 is performed as shown in FIG. That is, step S
In step 8, grouping is performed on a tuplet basis based on a hook. Those connected by the same series of hooks are regarded as one group, and notes / rests which are not consecutive hooks are regarded as one group alone. At the same time, based on the stem information, the same stem chords are grouped so as to belong to the same group.

In step S9, of the groups obtained in step S8, the group connected by the same hook is further divided into hierarchical groups according to the number of digits of the hook. As shown in FIG. 2 described above, this means that, based on the target consecutive hooked notes, the same number of digits are sequentially shifted from the lowest layer to the upper layer (from the first layer to the third layer in the drawing). Take things out and group them.

At step S10, it is determined whether or not the tuplet numeral is three. If the numeral is three, step S11 is executed.
, A pair of adjacent notes / rests having a note length of 1: 2 is detected, and a dummy group which is a virtual group having these notes / rests as constituent elements is generated at an intermediate position. And add it to the group.

In step S12, among the notes and rests constituting each group, the center position of the note head or the position of the end point of the stem that is closest to the tuplet numeral is set as the characteristic point of the group. The dummy group has the feature point at the center of the rectangle formed by the two notes / rests (in the case of FIG. 14, the center of the quarter rest and the eighth note is the feature point).

In step S13, if the distance between the feature point of the group and the tuplet numeral is within a certain range, this is set as a reference group, and if it is outside the range, it is set as an additional group. In step S14, each is sorted by the distance from the tuplet numeral. At this time, the dummy group that does not become the reference group is discarded. This is because if the dummy group is to be a tuplet, it can be established only by the dummy group alone. The tuplet numbers are shown in FIGS.
If there is a parenthesis as shown in (1), only the reference group is narrowed down according to the direction of the parenthesis.

In step S15, a process for recognizing a tuplet object in group units is performed. In step S16, it is checked whether or not a tuplet object has been found. If the tuplet object is not found, the tuplet number is erroneously recognized in step S17. Conversely, if a tuplet object is found, the tuplet object is determined in step S18.

The tuplet object recognition processing in group units described above is performed as shown in FIG. That is, it is checked in step S19 whether or not the i-th reference group exists in the order of sorting from i = 0. If it does not exist, it is determined that the processing has been completed to the end of the sort order, and error processing is performed in step S20. If the ith reference group exists, step S2
In step 1, the i-th reference group is set as the current group (the groups are extracted in the order of the nearest tuplet numeral and set as the current group). In step S22, a step check based on the following equation (2) is performed (where n is an integer).

[0024]

[Equation 2] (sum of note length in group) / (shortest note length in group) = n × (tuplet number)

In this step check, when the total sound length in the group is small (LESS in the figure), the processing proceeds to step S25 and thereafter to perform group combination. However, in step S23, it is checked whether the current group is a dummy group. If the current group is a dummy group, the current group is discarded, i is incremented, and the process returns to step S19. If the total length of notes in the group exceeds the value in the step check (OVER in the figure), the current group is discarded, i is similarly incremented, and the process returns to step S19. If the step check is OK (FIT in the figure), the process proceeds to step S24, where the notes / rests in the current group are set as the target notes / rests of the tuplet numeral, and the tuplet target search processing ends.

Before performing group combination, step S25
It is checked whether or not the current group is a continuous hook. If the current group is a continuous hook, the process proceeds to step S26. If not, the process proceeds to the group combining process of step S28.

In step S26, a group joining process within the same hook is performed. That is, as shown in FIG. 9, in step S30, it is checked whether or not there is a group Grp 'in the next higher hierarchy. If the group Grp 'exists, in step S31, the group Grp' of the next higher hierarchy is joined to the current group Grp. Next, in step S32, the same step check as described above is performed, and when the total sound length in the group is small (LESS in the figure), the process returns to step S30 to perform group combination with a further higher hierarchy, and If the sum of the note lengths in the box exceeds (OVER in the figure), step S
At 33, the current group is discarded, and the process returns to step S19. If the step check is OK (FI in the figure)
T), the process proceeds to step S34, where the notes / rests in the current group are set as the target notes / rests of the tuplet numeral, and the tuplet target search process ends. Even if the group combination up to the top layer in the same tuplet is still small, when the total length in the group is small,
Proceeding to step S35, the process proceeds to step S28 in order to perform connection with a group that is not a continuous hook. The example of FIG. 10 shows an example of the above-described group connection in the hook. A step check is performed on the reference group of the hierarchy 1, and the group duration is 1/4 beat and the shortest duration is 1/8 beat, so that group duration / minimum duration = 2 and tuplet Since the value is smaller than the integer n times the value of the number 12, it is combined with the group of the next higher hierarchy 2. Here also, a step check is performed, and since group length / shortest length = 4, the value is smaller than the integer n times the number of the tuplet number 12, so that the connection with the group of the next higher hierarchy 3 is performed. Here, since the group length / shortest length = 12, it matches the integer n times the value of the tuplet numeral 12, and the above-described group combining process in the chain hook is completed.

At the time of the continuous hook check in step S25,
When it is determined that the current group is not a chain hook, and when the total length of notes in the group is small even after the group combination up to the top layer in the same tuplet (steps S27 and S3)
In step S28, in step S28, a group to be combined is selected from the reference group and the additional group in the order closer to the tuplet number, and combined with the current group.

That is, as shown in FIG.
At 40, it is checked whether or not there is a reference group next to the current group j (j = i + 1). If there is no reference group, the process proceeds to step S41 to determine whether an additional group exists at the k-th (first k = 0). It is checked whether or not. If there is no additional group, it is determined that the processing has failed. In step S40, the existence of the next reference group (Grp ') is confirmed, or in step S41, the additional group (Grp') is determined.
If the presence of p ′) is confirmed, the process proceeds to step S42,
A check is made to see if the following join condition is satisfied.

The group to be joined must satisfy the following conditions. It is not a dummy group (since a dummy group with a duration of 1: 2 is not linked to another group and becomes a tuplet target). In the case of a grand staff, if the current group is a treble staff, the joining group also belongs to the treble staff, and if the current group is the bass staff, the joining group also belongs to the bass staff. If the current group includes hooks, the joining group is not a hook. That is, when the current group includes a chain hook, the group to be connected next must not have a chain hook. Does not include notes / rests in the current group and notes / rests that are chords. That is, if there is a chord that does not have a common stem, it is a chord of a different voice part, so that it is not included.

If it is determined that the above condition is not satisfied, j and k are incremented, and the process returns to step S40. On the other hand, if it is determined that the combination condition is satisfied, in step S43, the above-described next reference group to additional group Grp ′ are combined with the current group Grp.

In step S44, a step check is also performed on the current group after group combination. In this step check, when the total note length in the group is small (LE in the figure)
SS) proceeds to step S45 to check whether or not the group (Grp ') lastly joined to the current group is a continuous hook in order to join the groups further. When the total length of notes in the group exceeds the value in the step check (OV in the figure)
ER) returns to step S40 in step S48 except for the last group joined from the current group. If the step check is OK (FIT in the figure), the note / rest in the current group is set as the target note / rest of the tuplet numeral, and the above tuplet target search processing is terminated.

If it is determined in step S45 that the group (Grp ') most recently joined to the current group is a continuous hook, then in step S46, group joining processing within the same continuous hook from step S30 is performed. In the next step S47, it is checked whether or not the above processing is successful. If the processing is successful, the notes / rests in the current group are set as the target notes / rests of the tuplet numeral, and the tuplet target search processing ends. I do. If the process fails, in step S48, except for the last group joined from the current group, step S40
Return to. Further, in the same hook connection process, if the total length of the current group is still small and the total length of the current group is not completed even in the group connection up to the top layer, the process returns to step S40.

(Other Embodiments) In each group connected by the same number of digits in the same hook, the tone length in the group is 1 / n of a beat (n is an integer), except for tuplets. So
From the same grouping, it is possible to anticipate the presence of a tuplet, erroneous recognition of a note, and erroneous recognition of the number of digits. If priority is given to the recognition of a note or a hook digit, a place where the pitches in the group contradict each other can be recognized as a tuplet without recognizing the tuplet numeral.
When a tuplet numeral is pasted on a musical score input device, a process of selecting a tuplet target is required. If the tuplet target is automatically recognized as in the present invention, the target note selection process becomes unnecessary.

[0035]

According to the configuration of the present invention described in detail above, an accurate tuplet object which is not inconsistent with a phrase or a pitch in notation is automatically detected, and thereby a musical score which can be played and / or displayed. It enables the recognition of music scores that can provide accurate pronunciation timings when converting to a data format.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram showing a state of grouping based on stem information and hook information;

FIG. 2 is an explanatory diagram showing a transition of grouping using a hook girder within the same hook.

FIG. 3 is an explanatory diagram of hook information held by note data;

FIG. 4 is an explanatory diagram showing how to set feature points in a group.

FIG. 5 is a block diagram showing the configuration of an embodiment of a musical score recognizing apparatus that operates by reading a computer-readable recording medium storing a musical score recognizing program according to the present invention into an external storage device.

FIG. 6 is a flowchart illustrating main processing of a CPU.

FIG. 7 is a flowchart showing a tuplet object recognition process.

FIG. 8 is a flowchart showing a tuplet object recognition process in group units.

FIG. 9 is a flowchart showing group joining processing in the same hook.

FIG. 10 is an explanatory diagram showing a state of group connection in a hook.

FIG. 11 is a flowchart illustrating group combination processing.

FIG. 12 is an explanatory diagram showing a failure example when a tuplet target is determined only by coordinate values.

FIG. 13 is an explanatory diagram showing a failure example when a tuplet target is determined based on only coordinate values.

FIG. 14 is an explanatory diagram showing an example of failure in determining a tuplet object in the case of triplets.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 CPU 2 ROM 3 RAM 4 Hard disk drive 5 Flexible disk drive 6 CRT 7 CRT interface circuit 8 Keyboard 9 Keyboard interface circuit 10 Printer 11 Printer interface circuit 12 Scanner 13 Scanner interface circuit 14 MIDI interface circuit 15 Bus 20 Center of stems 21 Suffix End point 22 consecutive hook

Claims (14)

[Claims] 1. A music score recognition method for reading a music score image, recognizing the music score symbol, and converting it into a music data format that can be played and / or displayed, comprising: a staff, a bar line, a note, a rest, From the recognition result in which tuplet symbols and other symbols are mixed, take out notes and rests related to pronunciation during performance in measures where tuplet numbers exist.
Based on the hook information and stalk information, grouping related to beat delimitation is performed. If the tuplet number is 3, the note length is 1: 2, the middle of a pair of adjacent notes and rests. At the position, a virtual group having these notes / rests as constituent elements is generated, and this is added to the group as a dummy group. Depending on the distance from the tuplet numeral to the note head center or the stem end point, Furthermore, when the tuplet number is in parentheses, the group is divided into a reference group that can be a reference for a tuplet target search and an additional group that cannot be a reference by referring to the direction of the parenthesis, and is close to a tuplet number. With the reference group at the center, the total length of the constituent notes in the group to be combined must be an integral multiple of the tuplet number with respect to the shortest duration in the group. Group join with By a group binding, location, tone length component sounds, music recognition method and recognizes the most suitable tuplet subject phrasing of notational. 2. The music score recognition method according to claim 1, wherein
2. The musical score recognition method according to claim 1, wherein, when performing the grouping related to a beat division, the same hooks are grouped into the same group. 3. The music score recognition method according to claim 2,
3. The musical score recognition method according to claim 2, wherein, when grouping the same series of hooks, those connected with the same number of digits are grouped and hierarchical grouping is performed according to the number of digits. 4. The music score recognition method according to claim 3,
At the time of grouping of the same series of hooks, the recognized individual note information holds a unique ID for each series of hooks, and furthermore, the note within the same series of hooks holds the number of digits connected to the preceding and following notes, 4. The score recognizing method according to claim 3, wherein the groups connected by the same number of digits in the hook are used as a group. 5. The musical score recognizing method according to claim 1, wherein, when performing said grouping relating to a beat division, notes and rests which are not consecutive hooks are a single group. Music score recognition method. 6. The score recognizing method according to claim 1, wherein, when performing said grouping relating to beat delimiters, chords having the same suffix which are pronounced simultaneously are in the same group. The music score recognition method according to any one of claims 1 to 5, wherein 7. The score recognizing method according to claim 1, wherein when the group of the reference group and another group are combined, the same group is combined with another group based on the group in the hook. 7. The score recognizing method according to claim 1, wherein the musical score is combined with the next higher hierarchical group in the chain hook. 8. A computer-readable recording medium storing a score recognition program for reading a score image, recognizing the score symbol, and converting the score into a playable and / or displayable score data format. From the recognition results where bar lines, notes, rests, tuplets, and other symbols are mixed, the notes and rests related to the pronunciation during performance in the bar where the tuplet digits are present are extracted.
A grouping function that performs grouping related to beat delimiters based on tuplet information and stem information, and a pair of adjacent notes whose pitch is 1: 2 when the tuplet number is 3. A dummy group adding function that generates a virtual group having these notes and rests as constituent elements at the middle position of the rest and adds this to the group as a dummy group, and a note head center from a tuplet numeral Or, depending on the distance to the stem end point, if the tuplet numeral is in parentheses, also refer to the direction of the parentheses, and refer to the reference group that can be the reference for the tuplet target search and the additional group that cannot be the reference. And the reference group judgment function that divides the reference note into a tuplet number, and the sum of the note lengths of the constituent notes in the combined group is an integral multiple of the tuplet number with respect to the shortest note length in the group Tuplet pair To perform the group combination with other groups so as to satisfy the duration condition as, and to execute the group combination function that recognizes the tuplet target that is most suitable for the position, the duration of the constituent sounds, and the phrasing on the notation A computer-readable recording medium storing a musical score recognition program. 9. A computer-readable recording medium recording the musical score recognition program according to claim 8,
9. The computer-readable recording medium storing a musical score recognition program according to claim 8, wherein, in the grouping function for performing the grouping relating to beat division, the same hooks are grouped into the same group. 10. A computer-readable recording medium on which the music score recognition program according to claim 9 is recorded, wherein a grouping function for grouping same hooks is performed by grouping those connected by the same number of digits. ,
10. A computer-readable recording medium storing a musical score recognition program according to claim 9, wherein hierarchical grouping is performed according to the number of digits. 11. A computer-readable recording medium on which a musical score recognition program according to claim 10 is recorded, wherein individual note information recognized by a grouping function for grouping the same continuous hooks is stored in units of continuous hooks. , A unique ID shall be retained, and the same number of digits connected to the preceding and following notes shall be retained for notes within the same hook, and those connected by the same number of digits within the hook shall be grouped using this. A computer-readable recording medium storing the musical score recognition program according to claim 10. 12. A computer-readable recording medium on which a musical score recognition program according to claim 8 is recorded, wherein said grouping function for performing said grouping relating to beat delimiters, wherein notes and rests which are not consecutively hooked are used. A computer-readable recording medium recording the musical score recognition program according to any one of claims 8 to 11, which is a single group. 13. A computer-readable recording medium on which a musical score recognition program according to claim 8 is recorded. 9. The chords having the same group.
13. A computer-readable recording medium on which the musical score recognition program according to any one of claims 12 to 12 is recorded. 14. A computer-readable recording medium on which a musical score recognizing program according to claim 8 is recorded, wherein when a group of a reference group is combined with another group, a group in a hook is used as a reference. 14. The computer-readable recording medium storing a musical score recognition program according to claim 8, wherein when combining with the group of the music score, the combination is made with the next higher hierarchical group in the same chain hook.