An Evaluation Framework of Optical Music Recognition in Numbered Music Notation

2016 ◽  
Author(s):  
Fu-Hai Frank Wu
Keyword(s):  
Evaluation Framework ◽  
Music Notation ◽  
Optical Music Recognition ◽  
Music Recognition

Optical Music Recognition with Wavelets

2005 ◽  
pp. 2225-2229
Author(s):  
Susan E. George
Keyword(s):  
Wavelet Transform ◽  
General Purpose ◽  
Music Notation ◽  
Optical Music Recognition ◽  
Music Recognition ◽  
Logical Representation ◽  
High Level

The aim of optical music recognition (OMR) is to “recognise” images of music notation and capture the “meaning” of the music. When OMR is successful it will be able to automatically extract a logical representation of printed or handwritten music captured in an image. There are a variety of reasons why OMR is required. Chiefly, it is convenient for swift input of music notation and might be subsequently edited, performed, used as a search or other. There are many stages before that final high-level interpretation can be made and recognition of the primitive symbols contained in the notation is primary. One of the biggest challenges in OMR is the super-imposition of music notation symbols – notes and other – upon stave lines in the music image. This article examines a general-purpose knowledge-free method in the wavelet transform, to deal with super-imposition in images of typeset music.

scholarly journals Selected Research in Computer Music

2017 ◽  
Author(s):  
Michael Droettboom
Keyword(s):  
Signal Processing ◽  
Digital Signal Processing ◽  
Programming Language ◽  
Video Processing ◽  
Motion Tracking ◽  
Digital Signal ◽  
Music Notation ◽  
Optical Music Recognition ◽  
Music Interpretation ◽  
Music Recognition

This thesis describes three interrelated projects that cut across the author’s interests in musical information representation and retrieval, programming language theory, machine learning and human/computerinteraction.I. Optical music recognition. This first part introduces an optical music interpretation (OMI) system that derives musical information from the symbols on sheet music.The first chapter is an introduction to OMI’s parent field of optical music recognition (OMR), and to the present implementation as created for the Levy project. It is important that OMI has a representation standard in which to create its output. Therefore, the second chapter is a somewhat tangential but necessary study of computer-based musical representation languages, with particular emphasis on GUIDO and Mudela. The third and core chapter describes the processes involved in the present optical music interpretation system. While there are some details related to its implementation in the Python programming language, most of the material involves issues surrounding music notation rather thancomputer programming. The fourth chapter demonstrates how the logical musical data generated by the OMI system can be used as part of a musical search engine.II. Tempo extraction. The second part presents a system to automatically obtain the tempo and rubato curves from recorded performances, by aligning them to strict-tempo MIDI renderings of the same piece of music. The usefulness of such a system in the context of current musicological research is explored.III. Realtime digital signal processing programming environment. Lastly, a portable and flexible system for realtime digital signal processing (DSP) is presented. This system is both easy-to-use and powerful, in large part because it takes advantage of existing mature technologies. This framework provides the foundation for easier experimentation in new directions in audio and video processing, including physical modeling and motion tracking.

Optical Music Recognition

2011 ◽  
pp. 80-110 ◽  
Author(s):  
Pierfrancesco Bellini ◽  
Ivan Bruno ◽  
Paolo Nesi
Keyword(s):  
Image Processing ◽  
Pattern Recognition ◽  
Research And Development ◽  
European Commission ◽  
Image Understanding ◽  
Development Project ◽  
Music Notation ◽  
Optical Music Recognition ◽  
Digital World ◽  
Music Recognition

Optical music recognition is a key problem for coding western music sheets in the digital world. This problem has been addressed in several manners obtaining suitable results only when simple music constructs are processed. To this end, several different strategies have been followed, to pass from the simple music sheet image to a complete and consistent representation of music notation symbols (symbolic music notation or representation). Typically, image processing, pattern recognition and symbolic reconstruction are the technologies that have to be considered and applied in several manners the architecture of the so called OMR (Optical Music Recognition) systems. In this chapter, the O3MR (Object Oriented Optical Music Recognition) system is presented. It allows producing from the image of a music sheet the symbolic representation and save it in XML format (WEDELMUSIC XML and MUSICXML). The algorithms used in this process are those of the image processing, image segmentation, neural network pattern recognition, and symbolic reconstruction and reasoning. Most of the solutions can be applied in other field of image understanding. The development of the O3MR solution with all its algorithms has been partially supported by the European Commission, in the IMUTUS Research and Development project, while the related music notation editor has been partially funded by the research and development WEDELMUSIC project of the European Commission. The paper also includes a methodology for the assessment of other OMR systems. The set of metrics proposed has been used to assess the quality of results produce by the O3MR with respect the best OMR on market.

Applying Machine Learning in Optical Music Recognition of Numbered Music Notation

2017 ◽  
Vol 8 (3) ◽  
pp. 21-41 ◽  
Author(s):  
Fu-Hai Frank Wu
Keyword(s):  
Machine Learning ◽  
Image Processing ◽  
Objective Function ◽  
Common Ground ◽  
Recognition Accuracy ◽  
Daily Life ◽  
Sight Reading ◽  
Music Notation ◽  
Optical Music Recognition ◽  
Music Recognition

Although research of optical music recognition (OMR) has existed for few decades, most of efforts were put in step of image processing to approach upmost accuracy and evaluations were not in common ground. And major music notations explored were the conventional western music notations with staff. On contrary, the authors explore the challenges of numbered music notation, which is popular in Asia and used in daily life for sight reading. The authors use different way to improve recognition accuracy by applying elementary image processing with rough tuning and supplementing with methods of machine learning. The major contributions of this work are the architecture of machine learning specified for this task, the dataset, and the evaluation metrics, which indicate the performance of OMR system, provide objective function for machine learning and highlight the challenges of the scores of music with the specified notation.

An Off-Line Optical Music Sheet Recognition

2011 ◽  
pp. 40-77 ◽  
Author(s):  
Pierfrancesco Bellini ◽  
Ivan Bruno ◽  
Paolo Nesi
Keyword(s):  
Recognition System ◽  
Music Notation ◽  
Original Music ◽  
Optical Processing ◽  
Optical Music Recognition ◽  
Digital World ◽  
Music Recognition ◽  
Critical Phase ◽  
Graphic Elements ◽  
Music Symbol

The optical music recognition is a key problem for coding music sheets of western music in the digital world. The most critical phase of the optical music recognition process is the first analysis of the image sheet. In optical processing of music or documents, the first analysis consists of segmenting the acquired sheet into smaller parts in order to recognize the basic symbols that allow reconstructing the original music symbol. In this chapter, an overview of the main issues and a survey of the main related works are discussed. The O3MR system (Object Oriented Optical Music Recognition) system is also described. The used approach in such system is based on the adoption of projections for the extraction of basic symbols that constitute graphic elements of the music notation. Algorithms and a set of examples are also included to better focus concepts and adopted solutions.

scholarly journals Exploiting the Two-Dimensional Nature of Agnostic Music Notation for Neural Optical Music Recognition

2021 ◽  
Vol 11 (8) ◽  
pp. 3621
Author(s):  
María Alfaro-Contreras ◽  
Jose J. Valero-Mas
Keyword(s):  
Holistic Approach ◽  
Vertical Position ◽  
Music Notation ◽  
Early Music ◽  
Optical Music Recognition ◽  
Music Recognition ◽  
End To End ◽  
Line Level ◽  
Double Nature ◽  
Music Representation

State-of-the-art Optical Music Recognition (OMR) techniques follow an end-to-end or holistic approach, i.e., a sole stage for completely processing a single-staff section image and for retrieving the symbols that appear therein. Such recognition systems are characterized by not requiring an exact alignment between each staff and their corresponding labels, hence facilitating the creation and retrieval of labeled corpora. Most commonly, these approaches consider an agnostic music representation, which characterizes music symbols by their shape and height (vertical position in the staff). However, this double nature is ignored since, in the learning process, these two features are treated as a single symbol. This work aims to exploit this trademark that differentiates music notation from other similar domains, such as text, by introducing a novel end-to-end approach to solve the OMR task at a staff-line level. We consider two Convolutional Recurrent Neural Network (CRNN) schemes trained to simultaneously extract the shape and height information and to propose different policies for eventually merging them at the actual neural level. The results obtained for two corpora of monophonic early music manuscripts prove that our proposal significantly decreases the recognition error in figures ranging between 14.4% and 25.6% in the best-case scenarios when compared to the baseline considered.

Staff Line Removal Algorithm and Research Based on Run-Length Graph Slice and Topological Structure of Music

2013 ◽  
Vol 760-762 ◽  
pp. 1429-1433
Author(s):  
Yin Xian Yang ◽  
Li Zhao ◽  
Cai Rong Zou ◽  
Yin Xian Yang
Keyword(s):  
Topological Structure ◽  
Recognition Rate ◽  
Careful Analysis ◽  
Music Notation ◽  
Optical Music Recognition ◽  
Run Length ◽  
Music Recognition ◽  
Low Efficiency ◽  
And Function ◽  
Staff Line

Staff line removal is a key step before segmentation and recognition of music image and plays an important role in OMR (Optical Music Recognition) research, the result of staff line removal directly influences the performance and function of the whole OMR system. However, over-removal and under-removal often occurs in the processing and leads to the low efficiency of music recognition rate. So, in order to solve the arduous problem, an approach based on run-length graph slice and topological structure of music is put forward by careful analysis of staff line and music notation structure. Experience results show the validity and practicality of the presented algorithm fast and effectively.

Optical Music Recognition for Numbered Music Notation with Multimodal Reconstruction

2013 ◽  
Vol 479-480 ◽  
pp. 943-947
Author(s):  
Fu Hai Frank Wu ◽  
Jyh Shing Roger Jang
Keyword(s):  
Information Retrieval ◽  
Small Group ◽  
Music Information Retrieval ◽  
Music Notation ◽  
Modern Music ◽  
Notation System ◽  
Optical Music Recognition ◽  
Neighborhood Relationship ◽  
Music Recognition ◽  
Music Information

Optical music recognition (OMR) is attracted a lot attention on different music notation system which could be so focused on Back’s C-Clefs; in contrast, it could handle complete modern music symbols. One of notation system, numbered music notation, which is literally call “simplified notation”, is popular in many Asia countries. There is a traditional Chinese hymnbook, which usually used in small group of worship, in which one page has several hymns. We propose algorithms for the recognition of those notations in camera images of the hymn, which could effectively identify score zone and lyric zone, segment notation image, classify music notation, and reconstruct scores from classified notation by their coordinates and neighborhood relationship. Those algorithms comprise the preliminary demo system by which we provide a solution for music information retrieval and reconstruction.

Wavelets for Dealing with Super-Imposed Objects in Recognition of Music Notation

2011 ◽  
pp. 78-107
Author(s):  
Susan E. George
Keyword(s):  
Image Processing ◽  
Theoretical Perspective ◽  
Image Filtering ◽  
General Purpose ◽  
Music Notation ◽  
Two Dimensional ◽  
Optical Music Recognition ◽  
Music Recognition ◽  
Wavelet Image

This chapter is concerned with a problem that arises in Optical Music Recognition (OMR) when notes and other music notation symbols are super-imposed upon stavelines in the music image. We investigate a general-purpose, knowledge-free method of image filtering using two-dimensional wavelets to separate the super-imposed objects. Some background is given to the area of wavelets and a demonstration of how stavelines can be located in a wavelet-filtered image. We also explore the separation of foreground objects (notes) from the background (stavelines) over a variety of image resolutions, in binary and greyscale images using a pixel-based truth representation of the image to evaluate the accuracy with which symbols are identified. We find that the Coifmann family of wavelets appear most suitable for vertical image components, and the Daubechies for the horizontal. The motivation for this chapter stems from the desire to (i) make an original wavelet application in image processing, (ii) provide a fresh (theoretical) perspective on the problem of super-imposed objects in music notation, recognizing the duality of the segregation task that exists with staveline removal/symbol extraction and (iii) evaluate how beneficial wavelet image filtering might be to the OMR process.

Sign in / Sign up

Export Citation Format

Share Document