High-precision bicycle detection on single side-view image based on the geometric relationship
作者:
Highlights:
• The geometric constraint on the proposed bicycle model is discovered.
• The relationship between the bicycle wheel pair and frame is proposed.
• The entire bicycle detector is based on the ellipse and line detection.
• The resultant is applicable and accurate for single side-view images.
• This algorithm is non-iterative, easy to implement, and not training-based.
摘要
Highlights•The geometric constraint on the proposed bicycle model is discovered.•The relationship between the bicycle wheel pair and frame is proposed.•The entire bicycle detector is based on the ellipse and line detection.•The resultant is applicable and accurate for single side-view images.•This algorithm is non-iterative, easy to implement, and not training-based.
论文关键词:Bicycle,Computer vision,Ellipse detection,Triangle detection,Geometry,Algorithm
论文评审过程:Received 27 February 2016, Revised 21 August 2016, Accepted 10 October 2016, Available online 15 October 2016, Version of Record 24 October 2016.
论文官网地址:https://doi.org/10.1016/j.patcog.2016.10.012