Automatic scale estimation for music score images
作者：
Highlights：
• Dealing with images at multiple scales in Optical Music Recognition (OMR).
• Estimation of the image scale based on Convolutional Neural Networks.
• Correction of the scale in the document analysis stage.
• Goal-directed experiments with two case studies on typical OMR steps.
• Results show the direct benefit of our approach on a standard OMR pipeline.
摘要
•Dealing with images at multiple scales in Optical Music Recognition (OMR).•Estimation of the image scale based on Convolutional Neural Networks.•Correction of the scale in the document analysis stage.•Goal-directed experiments with two case studies on typical OMR steps.•Results show the direct benefit of our approach on a standard OMR pipeline.
论文关键词：Optical music recognition,Scale estimation,Scale correction,Music score images analysis
论文评审过程：Received 11 October 2019, Revised 10 May 2020, Accepted 20 May 2020, Available online 4 June 2020, Version of Record 27 June 2020.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.113590
