Grass band detection in soccer images for improved image registration

作者：

Highlights：

• Detection of boundaries between the different grass bands that can be seen in images of soccer matches.

• Local analysis using partially overlapping blocks.

• Grouping of lines based on their common vanishing points.

• Key-points from grass bands can be exploited to build more robust camera calibration or pitch registration algorithms.

摘要

•Detection of boundaries between the different grass bands that can be seen in images of soccer matches.•Local analysis using partially overlapping blocks.•Grouping of lines based on their common vanishing points.•Key-points from grass bands can be exploited to build more robust camera calibration or pitch registration algorithms.

论文关键词：Soccer,Football,Line mark,Grass band,Turf,Green,Pattern,Key point,Segmentation,Line detection,Registration

论文评审过程：Received 23 November 2021, Revised 21 June 2022, Accepted 1 August 2022, Available online 6 August 2022, Version of Record 29 August 2022.

论文官网地址：https://doi.org/10.1016/j.image.2022.116837