Compensating over- and underexposure in optical target pose determination
作者：
Highlights：
• A wrong (over or under) exposure leads to a bias in optical target pose determination.
• Concentric circles targets allow detecting and compensating for any bias caused by exposure.
• A rigorous sub-pixel edge detector is formulated based on a pixel-level image formation model.
• A new algorithm uses geometric constraints of concentric circles to determine the exposure bias.
• Experimental evaluation in a controlled environment confirms unbiased pose determination.
摘要
•A wrong (over or under) exposure leads to a bias in optical target pose determination.•Concentric circles targets allow detecting and compensating for any bias caused by exposure.•A rigorous sub-pixel edge detector is formulated based on a pixel-level image formation model.•A new algorithm uses geometric constraints of concentric circles to determine the exposure bias.•Experimental evaluation in a controlled environment confirms unbiased pose determination.
论文关键词：Optical target,Target orientation,Image processing algorithm,Geometry,Ellipse fitting,Computer vision,Overexposure,Exposure compensation,Resection
论文评审过程：Received 9 April 2019, Revised 20 October 2020, Accepted 28 February 2021, Available online 5 March 2021, Version of Record 12 April 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.107930
