Musical Score Repeat Mark Recognition via Bar Line Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing musical score reproduction systems face challenges in accurately associating musical score data with performance data, particularly with repeat symbols and marks, leading to difficulties in smooth performance and correct recognition of repeat marks within the score.
Innovation Solution
A system that includes a storing unit for music and image data, a measure specifying unit, a music-data generating unit, a repeat-mark position specifying unit, and a musical-score element data generating unit, which extracts and associates musical score elements like staffs, bar lines, and repeat marks to generate unit music data and element data, ensuring proper alignment and removal of overlapping data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If repeat marks are included in the musical score, then the musical score can represent repeated measures, but the complex figure of repeat marks reduces the possibility of correct recognition in image data
Solution Approach 1:
The patent segments the musical score image into multiple candidate regions using bar lines as delimiters. Each candidate region corresponds to a potential measure containing a repeat mark. This segmentation allows the system to focus recognition efforts on specific localized areas, improving the reliability of repeat mark detection despite their complex figures.
Solution Approach 2:
The patent introduces bar lines as intermediary elements that structure the musical score into discrete measures. These bar lines serve as reliable reference points that facilitate the identification and isolation of repeat marks within specific measure boundaries, making the complex repeat mark figures more manageable for recognition.
2Productivity
If musical notes on the musical score are used to correspond to events in music data, then performance can be generated, but without repeat information in music data, smooth performance cannot be achieved
Solution Approach 1:
The patent performs preliminary analysis of the musical score image to identify repeat marks and their positions before music performance generation. This advance identification allows the system to pre-process the music data by inserting appropriate repeat instructions at the correct positions, ensuring smooth performance reproduction without requiring real-time complex processing.
Solution Approach 2:
The patent creates a feedback loop where the recognized repeat mark positions from image data are used to modify and enhance the music data. This feedback mechanism ensures that the final performance data accurately reflects the repeat structures indicated in the original musical score, improving overall performance reliability.
3Loss of information
If image data of musical score is used, then visual representation is provided, but correct recognition of repeat marks becomes difficult due to their complex figures
Solution Approach 1:
The patent divides the musical score image into discrete measure segments using bar lines as boundaries. This segmentation transforms the continuous complex image into manageable discrete units, making it easier to detect and recognize repeat marks within each segment without being overwhelmed by the overall complexity of the full score.
Solution Approach 2:
The patent employs dynamic processing where candidate regions for repeat marks are identified and extracted based on the detected bar line positions. This dynamic region extraction adapts to the specific structure of each musical score, allowing the recognition system to focus computational resources on relevant areas containing potential repeat marks.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A musical score extracting unit specifies areas of each of measures and the measure number of each measure on a musical score based on positions of musical-score composing elements such as part lines, staffs and bar lines. A music-data dividing unit divides music-data file based on time information in the music-data file to generate plural unit music-data files containing pitch information and time information for one measure. The music-data dividing unit specifies measures where repeat marks are placed based on sorts and positions of the repeat marks and positions of the part lines, staffs and bar lines on the musical score, thereby removing unit music-data file to repeat as instructed by the repeat marks from the plural music-data files to obtain a final unit music-data files associated with the respective measure numbers. --