JP3608674B2 - Score recognition device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a score recognition apparatus, and more particularly to a score recognition apparatus capable of recognizing paragraphs and parts of a score and appropriately time-sequencing the score information.
[0002]
[Prior art]
In a conventional score recognition apparatus, for example, musical score image data read by a scanner recognizes staffs, notes, and various symbols, and generates performance data such as MIDI file data. In such a score recognition apparatus, when a score score is recognized, the number of parts is fixed in all paragraphs. In this case, if the recognized parts correspond to the staff in each paragraph in order from the top, the time series of data is completed.
[0003]
[Problems to be solved by the invention]
In the actual score, not only the number of parts in all paragraphs is the same, but also the addition of parts from a paragraph, the omission of parts, the fusion of parts (from a paragraph with two single staffs to a grand staff), Separation (the reverse), staff change (large staff ← → single staff), etc. may be performed. In such a case, there is a problem in that the data cannot be correctly time-series simply by associating the parts with the staffs in the paragraph in order from the top.
The object of the present invention is to solve the above-mentioned problems of the prior art, recognize the separation of paragraphs in the score score, recognize the correspondence of parts, and can recognize the score-recognized data in a time series for each track. To provide an apparatus.
[0004]
[Means for Solving the Problems]
The present inventionRead for each rectangular areaFrom musical score image dataThe sheet musicA musical score recognition device that recognizes various symbols.ThedetectionStaff detectionMeans,A paragraph line for detecting a paragraph line extending between adjacent staffs, a paragraph line extending upward from the uppermost staff, and a paragraph line extending downward from the lowermost staff from the musical score image data read for each rectangular areaDetecting means,In the score image data read for each rectangular area adjacent in the vertical direction, a paragraph line extending downward from the lowest staff of the score image data read in the upper rectangular area is detected, and the lower rectangle is detected. When a paragraph line extending upward from the uppermost staff of the musical score image data read in the area is detected, the lowermost staff and the uppermost staff are included in one paragraph. recognizeIt is characterized by that.
Also, paragraph detection means for detecting the presence or absence of paragraph lines between multiple staffs from the score image data read for each rectangular area, and bracket detection for detecting the presence or absence of brackets on the left side of each paragraph line And a plurality of staffs enclosed by the brackets are recognized as one part when the brackets are detected by the bracket detection means.
Furthermore, part name recognition means for recognizing the part name described in the score from the score image data, and part correspondence detection means for taking correspondence between the parts based on the part name recognized by the part name recognition means, It is provided with.
[0005]
According to the present invention, the staff can be accurately grouped for each paragraph, and each part can be matched between paragraphs. Therefore, it is possible to automatically convert notes and the like described in the score into performance information with appropriate timing and timbre.
[0006]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing the configuration of one embodiment of the score recognition apparatus of the present invention. This apparatus is obtained by adding a scanner and a MIDI interface circuit to a general computer system such as a personal computer. The CPU 1 is a central processing unit that controls the entire score recognition apparatus based on a program stored in the ROM 2 or the RAM 3. Further, a timer circuit for interrupting 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, etc. in addition to a program area. The hard disk device HDD4 and the floppy disk device 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 information input from the keyboard 8 is taken into the CPU 1 through the keyboard interface circuit 9. The printer 10 prints the print information output from the printer interface circuit 11 based on the control of the CPU 1.
[0007]
The scanner 12 optically scans (printed) musical scores and converts them into binary or gray scale image data, and uses any type such as a flatbed type, a handy type, and a feeder type. it can. Image information read by the scanner 12 is taken into the RAM 3 or the HDD 4 via the scanner interface circuit 13. The MIDI interface circuit 14 is a circuit that transmits and receives MIDI data to and from an external MIDI device such as a sound module. The bus 15 connects each circuit in the score recognition apparatus. In addition, a pointing device such as a mouse, a serial interface circuit such as RS232C, or the like may be provided.
[0008]
FIG. 3 is a flowchart showing the main processing of the CPU 1. In S <b> 1, the image of the score is taken into the RAM 3 by the scanner 12. The image is captured as a binary image. In S2, image quality smoothing processing such as figure fusion is performed in order to reduce blurring and dot noise. In S3, an image quality check process is performed. In the image quality check process, in order to obtain information on magnification and density, and in order to obtain reference data for detecting the staff in the subsequent stage, first, the width between the staffs and the width between the staffs is detected. In order to obtain the line width and the interval width, first, scanning is performed in the vertical (y) direction at several positions in the horizontal (x) direction on the image, and all the lengths of the black run (continuous black pixels) and the white run are determined. Obtain frequency distribution (histogram) data for each length.
[0009]
Since the most common symbol on the score is a staff, the line width and interval of the staff can be estimated by detecting the peaks of the created black run length histogram and white run length histogram. The magnification of the image data can be estimated from, for example, the gap width, and the density can be estimated from the ratio between the line width and the gap width. In the score recognition process, if the magnification and the density are out of the predetermined range, the recognition rate decreases. Therefore, in S3, it is checked whether or not these values are within the predetermined range. In S4, it is determined whether or not the check result in S3 is an image quality OK. If the result is not OK, the process returns to S1 and recapture is performed by changing the magnification and density.
[0010]
In S5, staff recognition described later is performed. The staff recognition process is roughly divided into staff area separation processing, staff scanning start position detection processing, and staff shift amount detection processing. In S6, a paragraph recognition process described later is performed. This process is roughly divided into a paragraph recognition process and a bracket recognition process.
[0011]
In S7, the recognition result of the paragraph is displayed and the user is checked whether or not the paragraph recognition result is correct to determine whether or not it is OK. If the result is not OK, the process proceeds to S8. The paragraph recognition result is corrected. In the score notation, in addition to the same part structure of each paragraph, there are cases where parts are omitted or added in the middle, or the single staff and grand staff change for each paragraph in the same part. Correspondence of such parts is performed by comparison with brackets. This means that two paragraphs are extracted from all the paragraphs, and the combination that minimizes the difference in the number of non-major parts between the grand staves is selected from the corresponding combinations of the grand staves of each paragraph. Although it is performed by the method, the correspondence of the parts may not be uniquely determined, so that the paragraph recognition result can be corrected in advance.
[0012]
If the staff recognition fails, the subsequent processing cannot be performed, so it is necessary to change the magnification and density and capture the image again. Therefore, in step S7, first, the recognition result of the staff is displayed to let the user determine whether or not it is correct. If it is not correct, the process returns to S1 to re-import the image. If it is recognized correctly, the paragraph recognition result may be displayed and checked.
[0013]
In S9, processing rectangle determination processing is performed. A rectangle that is fairly wide including the found staff (the staff in the staff in the case of a grand staff) is taken, and this is used as a recognition processing rectangle. The rectangle in this case is changed from the obtained staff to a rectangle with a certain width in consideration of the shift amount of the staff, and there is a certain amount of white space in the vertical direction when viewed from the staff side. If so, the rectangle is reduced.
[0014]
In S10, staff inclination correction processing is performed. In brief, based on the staff shift amount obtained previously, a rectangleInsideShift the vertical pixel column of the image up and down. In the case of a grand staff, since there are a plurality of staffs, the shift amount of the top staff may be adopted, or the average of the shift amounts may be taken. After that, if the rectangle overlaps the upper and lower staffs, delete that part, and if there is a label that touches the upper and lower ends of the rectangle and does not reach the recognized staff part The upper and lower staffThe parts ofErase as label. After performing this processing, the rectangle may be further reduced from the end of the rectangle to the portion where the black pixel exists.
[0015]
In S11 to S15, recognition processing of various symbols is performed. There are roughly the following three types of musical notation symbols in terms of shape and position. (1) A fixed type whose top and bottom positions are almost fixed (clef, time signature, etc.). (2) It is a standard type and has a degree of freedom in the vertical position (temporary symbols, rests, etc.). (3) An indefinite type and an indefinite position (note, slur, tie, etc.). These are recognized in the order of clef, time signature recognition, note recognition, fixed symbol recognition, character string recognition, slur, and tie recognition in a method suitable for each.
[0016]
The clef and time signature recognition is performed first in order to reduce the processing cost of later recognition by performing recognition with low processing cost first and then deleting this symbol. This is to reduce misrecognition in later recognition. In addition, standard symbol recognition after note recognition is a recognition method that is not easily affected by label contact, and by deleting this note it is possible to recognize temporary symbols that touch the note It is to make it. The reason for the slur and tie recognition last is to reduce as many labels as possible for slur recognition with high processing costs. Further, by making only the labels around the previously detected notes a target for slur and tie recognition, the processing cost of slur and tie recognition can be further reduced, and the misrecognition of slur and tie can also be reduced.
[0017]
In S11, a clef and a time signature are recognized as symbols at a fixed position with respect to the staff. In this process, first, a histogram of black pixels is taken vertically in the rectangular area including the found staff, and a band where the black pixel amount is equal to or greater than a certain threshold is a place where a symbol may exist. As a target of matching. Matching is performed by peripheral features in the horizontal direction at several points between the staffs. Peripheral features refer to the distance from the left and right ends of a rectangular area that contains only the symbol to be matched to several black pixel areas between the staffs inward and the distance to the black pixel area. It is obtained up to the first) or several times (after the second). If matching fails, the adjacent bands are merged and recognized again. The recognized symbol is deleted from the image data.
[0018]
In S12, note recognition is performed. First, a rectangular area is scanned horizontally to detect and separate black runs of a predetermined length or less. Since the separated image data is a thin portion, a vertical line that is a candidate for a stem or a bar line is detected from here. Next, a black run having a predetermined length or less is detected and separated vertically. Since the separated image data is a portion constituting a thin horizontal line, a horizontal line that is a candidate for an additional line or a crescendo is detected from here. Finally, by removing the vertical and horizontal thin runs detected from the original image data, a thick part (hereinafter referred to as a thick label) in the image can be extracted. In the case of a musical score, note heads (hereinafter referred to as black ball heads) and continuations (bands connecting a plurality of notes) shorter than four-part notes can be separated.
[0019]
The black ball head obtains coordinate chain data for the border line of the thick label, calculates an elliptic formula from the coordinate data by a known method, and recognizes the matching degree with the shape and the thick label. For the two-part notes and all note heads (hereinafter, white note heads), an elliptic formula is calculated from the coordinate chain of the hole in the image.
[0020]
Finally, a note is detected by combining with the previously determined note candidate. Renren detects a thick label existing around the tail of a note that does not consider the flag that has been determined so far, and calculates the number of consecutive labels from this shape. Also, other notes connected to the chain are also detected. If there is no other note to connect, it is considered as a note with a single flag. The note information is changed according to the number of reams. After this, the separated horizontal lines are used to recognize the pitch of the sound (additional lines), crescendo, repeated horizontal and vertical lines, etc. Recognize bar lines from the remaining vertical lines. Then, the recognized symbol is deleted from the image.
[0021]
In S13, standard symbol recognition is performed. In this process, first, the well-known contour load direction index is taken, the label size and the load direction index matching degree are calculated for each symbol data in the dictionary, the matching degree is normalized, and the integrated result is the most. Output high symbols. In addition to the size and load direction index, other features such as peripherals may be used. In addition, as a countermeasure against a broken label due to staff erasure, a label that has been erased by staff erasure is registered in the dictionary, and if it is recognized as this symbol, the surrounding labels are combined and re-recognized. . The recognized symbol is deleted from the image.
[0022]
In S14, character string recognition is performed. To recognize a character string such as a speed symbol, the alphabet or other symbol recognized by the standard symbol recognition is used, and a character string in which the rectangles surrounding the symbol are arranged is extracted. Matching makes it possible to recognize a character string-like symbol even if each constituent character is slightly wrong.
[0023]
In S15, slur tie recognition is performed. In this process, among the remaining labels, the labels around the detected notes are thinned and approximated by a multi-arc. Then, since the line may be cut off due to a previously erased symbol, the obtained multiple arcs are connected. Finally, slurs and ties are recognized based on the arc shape obtained, the thickness of the original image, and the relationship with the notes. Since this is the last routine in recognition, the recognized symbols do not have to be deleted from the image, but if the recognized slurs and ties are deleted and then the standard symbol recognition is performed again, the slurs and ties The touched symbol can be recognized.
[0024]
In S16, for example, based on the recognition result, the score image data is synthesized and displayed, and it is determined whether or not it is OK by letting the user check whether it is correct. If the result is not OK, the process goes to S17. Then, the recognition result is manually corrected using a mouse, a keyboard, or the like. In S18, performance data creation processing is performed. In this processing, based on the recognized various symbols and note information, for example, MIDI file data in a known performance data format is generated.
[0025]
FIG. 4 is a flowchart showing details of the staff recognition process in S5 of FIG. In the staff recognition process, first, a paragraph rectangle is detected in order to facilitate recognition of a coder-like staff arranged in the horizontal direction. Since this rectangle detection process detects an accurate paragraph line later, it may be a rough one, and the paragraph rectangle may be smaller than the correct one, for example, separated for every staff.
In S20, the number of black pixels for each horizontal axis of the image is counted and a histogram is created. FIG. 2 is an explanatory diagram of a histogram. The histogram of (b) is obtained by counting the number of black pixels at each y coordinate of the musical score image of FIG. 2 (a) (notes are present but not shown). This histogram has a large value at the location where the staff is present, but since the increase and decrease of the value is severe, it is difficult to make a determination if it remains as it is. Therefore, in S21, in order to smooth the waveform, an addition histogram is created by summing the histogram values in the y-coordinate range having a predetermined width for each y-coordinate. The width may be, for example, about the width between two adjacent lines of the staff, and the addition range may be, for example, a predetermined range lower than an arbitrary y coordinate or centered on the y coordinate It may be in the upper and lower range. FIG. 2C is an example of an addition histogram.
[0026]
In S22, the addition histogram is searched from the top (the one with the smaller y coordinate), and the addition histogram value is normalized by the line width of the staves previously obtained and the width between the five shown by the dotted line in FIG. The y coordinate (y1) that is equal to or greater than the predetermined threshold value is detected. If the lower end of the image is reached in S22, the process proceeds to S27. In S23, the addition histogram is searched downward from y1, and the y coordinate (y2) at which the addition histogram value is equal to or smaller than a predetermined threshold value is detected. Further, the addition histogram is searched from y2 below, and the y coordinate (y3) at which the addition histogram value is again equal to or greater than a predetermined threshold is detected. However, if (y3−y2) is equal to or smaller than a predetermined value, that is, if the width of the region with few black pixels is narrow, the y2 and y3 are discarded, and the search for y2 is performed again below y3. If the lower end is reached during the y3 search, it is assumed that y3 has been detected sufficiently downward, and the process proceeds to S25.
In S25, if (y2-y1) is a predetermined value, for example, the width of the entire staff or more, the coordinate information of the rectangle is stored as a rectangle estimated to include a paragraph or staff. In S26, it is determined whether or not the lower end of the image has been reached. If the result is negative, the process returns to S22 to search for the remaining area.
[0027]
In S27 to S30, a staff break is searched in the horizontal axis direction. First, in S27, a histogram of the number of black pixels for each vertical axis is taken for each rectangle detected in S25. In S28, the histogram is searched from the left, and the x coordinate (x1) at which the histogram value is equal to or greater than a predetermined value is detected. For example, if the threshold is set to the number of black pixels for four staffs, the threshold value is always exceeded if there is a staff. In S29, the histogram is searched from x1 to the right, and the x coordinate (x2) at which the histogram value is equal to or smaller than a predetermined value is detected. In S30, if (x2-x1) is greater than or equal to a predetermined value, the rectangle is divided. Since a symbol or the like may exist on the right side of the staff, the shortest rightmost rectangle is included in the left rectangle.
[0028]
In S31, staff recognition is performed for each paragraph rectangle. First, the paragraph rectangle is set as a staff detection rectangle, and the width of the staff and the width between the five are recalculated in the rectangle. Then, the start position of the staff shift amount scanning is obtained, and further the staff shift amount is obtained. From the first staff obtained in this way to the upper and lower ends of the staff detection rectangle, the next staff detection rectangle is used. Similarly, the line width of the staff and the width between the five are recalculated to detect the staff. The reason for recalculating the line width and the width between the five lines in each rectangle is that the widths of all the lines in the paragraph rectangle are not always constant. This process is performed until there is no rectangle in which the staff can be detected in each paragraph rectangle.
[0029]
The outline of the staff scanning start position detection process is as follows. At a certain position in the x-axis direction, the widths of black pixels and white pixels are obtained in order, and the positions where the obtained line widths and inter-widths are arranged in the form of a staff Detecting in consideration of some error. Then, in order to remove the influence of the additional line (the horizontal line added to describe the note that protrudes from the staff), a condition that there is a white pixel width larger than the width between both sides of the staff line is added. The midpoint of each black run at the x position where the arrangement of black and white pixels meeting this condition is set as the staff scanning start position.
[0030]
In S32, the staff is traced and a shift amount for correction is detected. The outline of the staff shift amount detection process will be described. The staff is traced by changing the position to the left and right one dot at a time from the staff scanning start position (5 black pixel positions) of the obtained x position. When the number of black pixels becomes less than a certain number (for example, 3 or 4) out of 5 points, the 5 pixels are shifted up and down to check the number of black pixels, and the y coordinate increases in the direction in which the ratio of black pixels increases. To shift. The shift amount from the start position is taken as the staff shift amount. The staff is detected by tracing from the staff scanning start position to the left and right until the position where the number of black pixels becomes zero.
[0031]
FIG. 5 is a flowchart showing details of the paragraph recognition process in S6 of FIG. In S40, the detected staff information is arranged in order from the top. In S41, the absolute value d of the difference between the x-coordinates at the left end of the staff is obtained for adjacent staffs. Note that the staff in the paragraph is not necessarily adjacent in the y-coordinate, such as when the paragraph is cut off in the middle, so check all the combinations of staffs, or for a certain staff All staffs within a predetermined y-axis distance are checked.
[0032]
In S42, it is determined whether or not the distance d is less than the predetermined threshold value THR. If the result is negative, the process proceeds to S46, but if the result is affirmative, the process proceeds to S43. In S43, it is checked whether or not there is a continuous paragraph line in the region between the left ends of the two staffs. The presence of a paragraph line can be determined by checking whether the black pixels are continuous in 8 connections from top to bottom, but in order to be able to recognize the presence of a paragraph line even if the image is faint, Even if a certain number of black pixels are interrupted, if it continues thereafter, it is determined that a paragraph line is present.
[0033]
FIG. 6 is an explanatory diagram showing a paragraph line detection area. The paragraph line is detected in a region A between the left ends of the upper and lower staffs. Since the coordinates of the left end of the two staffs are known, the area A can be obtained from the coordinate data. When a musical score image is read by a narrow scanner such as a handy scanner, image data may be divided in the middle of a paragraph. In such a case, it is checked whether or not there is a paragraph line extending to the end of the image data in the upper area B of the uppermost staff or the lower area C of the lowermost staff. Detect continuity with data. For example, if a paragraph line exists in the area C and a paragraph line also exists in the area B of the next image data, it is determined that the two line lines of the image data are connected.
[0034]
In S44, it is determined whether or not the check result in S43 is a paragraph line. If the result is negative, the process proceeds to S46, but if the result is affirmative, the process proceeds to S45. In S45, the upper and lower staffs are set to the same paragraph number. In S46, it is determined whether or not there is any remaining staff combination. If there is still a combination of the staffs, the process returns to S41 and the paragraph line check is repeated. If the obtained paragraph rectangles do not overlap in the horizontal axis direction but overlap in the vertical axis direction and the number of staffs is equal, it is assumed that the paragraph is a coder-like paragraph. In this case, two paragraphs are combined into one paragraph, or as another paragraph, symbols and notes are recognized and time series processing is performed later.
[0035]
In S47, brackets are recognized in order to distinguish whether the staff is a grand staff or a single staff in each paragraph where a paragraph line is detected. Symbol recognition is performed while performing label extraction in the area to the left of the left end of the staff of the obtained paragraph. At this time, since the brackets may be in contact with the paragraph line, the label is cut in advance at the part of the paragraph line. The symbol recognition at this time may be a recognition rate that can distinguish between a large arc and a large arc line. Also, since the brackets are often separated by a thin portion at the center, separate dictionary data is also prepared, and the recognition results are combined when the upper and lower separation brackets are recognized. Then, among the staffs in the paragraph, the group of staffs recognized to be enclosed in brackets are assigned the same part number.
[0036]
Although the embodiments have been disclosed above, the following modifications are also conceivable. The image data has been disclosed as a method of recognizing the image data read by the scanner one by one in order. However, when the image data is read by the hand scanner, for example, there is a high possibility that one song is captured as a plurality of image files. In such a case, a plurality of image data files may be combined into a single file and recognized. Further, the recognition processing of the present invention may be performed after the image data is divided into a plurality of pieces in advance.
[0037]
In the paragraph recognition process, a method for checking the presence or absence of a paragraph line has been disclosed. However, a black pixel histogram is taken in the x-axis and y-axis directions in advance, and the existence of a paragraph is estimated by detecting a complete blank portion thereof. May be.
[0038]
During paragraph recognition processing, character string recognition for recognizing part names may be performed simultaneously with recognition of brackets and arcs. In this case, alphanumeric characters (Japanese) are added to the dictionary with a wider range than the recognition of the brackets being recognized. When the character is recognized, the recognition result of the character string target is extracted from the recognition rectangle sequence, and the part name is recognized by the maximum value selection threshold value determination of the character string matching point. A character string dictionary of part names for this is also prepared. In many cases, the part name is abbreviated by paragraph, so that the correspondence between the abbreviation and the other part name can be taken.
[0039]
Recognition processLabels that are separated from the top and bottom edges of the rectangle and from the staff in a rectangleThat is, the components of the staffUp and downRecognition processMay be in both rectangles. In this case, the previousRecognition processThe symbol information recognized by the rectangleRecognition processCheck when the rectangle is recognized, thenRecognition processIf the same figure in the rectangle is erased in advance, it can contribute to speeding up the recognition. Further, for example, downward slur a rectangular bottom, some symbols, because there is what can determine belongs in advance either part, in this case, putting the mark information only to the corresponding part.
In the embodiment, the shift amount is calculated for each staff and the shift correction is performed in the rectangular image. However, one shift amount may be calculated for the entire captured image, and the entire image may be shifted. .
[0040]
【The invention's effect】
As described above, according to the present invention, staffs can be grouped for each paragraph, and each part can be matched between paragraphs. Therefore, there is an effect that it is possible to automatically convert notes and the like described in the score into performance information with appropriate timing and tone color. In addition, there is an effect that it is possible to achieve a high recognition rate even if the score magnification and the symbol shape are different.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an embodiment of a score recognition apparatus of the present invention.
FIG. 2 is an explanatory diagram of a histogram.
FIG. 3 is a flowchart showing main processing of the CPU 1;
FIG. 4 is a flowchart showing details of the staff recognition process in S5 of FIG. 3;
FIG. 5 is a flowchart showing details of a paragraph recognition process in S6 of FIG. 3;
FIG. 6 is an explanatory diagram showing a paragraph line detection area;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... CPU, 2 ... ROM, 3 ... RAM, 4 ... Hard disk device, 5 ... Floppy disk device, 6 ... CRT display device, 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

Claims (3)

In a score recognition device for recognizing various symbols of a score from score image data read for each rectangular area ,
Musical score image data read for each of the rectangular area, and the staff detecting means for detecting the staff,
Musical score image data read for each of the rectangular area, to detect a paragraph line extending paragraph line extending between adjacent staves, from staff paragraph line and lowermost portion which extends from the top of the staff upward downward Paragraph line detection means ,
In the score image data read for each rectangular area adjacent in the vertical direction, a paragraph line extending downward from the lowest staff of the score image data read in the upper rectangular area is detected, and the lower rectangle is detected. When a paragraph line extending upward from the uppermost staff of the musical score image data read in the area is detected, the lowermost staff and the uppermost staff are included in one paragraph. music recognition and wherein the recognizing. In a score recognition device for recognizing various symbols of a score from score image data read for each rectangular area ,
Musical score image data read for each of the rectangular area, and the staff detecting means for detecting the staff,
Musical score image data read for each of the rectangular area, and the paragraph detecting means for detect the presence or absence of paragraph lines between a plurality of the staff,
On the left side of each paragraph line , there is a bracket detection means for detecting the presence or absence of brackets,
Above when it is detected that there is bracket by brackets detecting means, the musical score recognition apparatus characterized by recognizing a plurality of staffs and one part which is enclosed by the brackets. Musical score image data, comprising a part name recognition means for recognizing the part names listed in the score, based on the part name recognized by the part name recognition means, and a part corresponding detecting means to take corresponding parts between paragraphs The musical score recognition apparatus according to claim 2, wherein:

Images