Cut-in vehicle warning system exploiting multiple rotational images of SVM cameras

作者：

Highlights：

• A surround view monitoring is effective to detect cut-in events in short distances.

• The part-based object detection is adopted to detect a partially visible vehicle.

• The appearance variation of a vehicle can be reduced through virtual images.

• In the videos over 7 h, the true warning rate is over 98% with 3 false warnings.

摘要

•A surround view monitoring is effective to detect cut-in events in short distances.•The part-based object detection is adopted to detect a partially visible vehicle.•The appearance variation of a vehicle can be reduced through virtual images.•In the videos over 7 h, the true warning rate is over 98% with 3 false warnings.

论文关键词：Part-based object detection,Vehicle detection,Surround view monitoring,Cut-in warning,Rotational images

论文评审过程：Received 3 September 2018, Revised 14 January 2019, Accepted 31 January 2019, Available online 1 February 2019, Version of Record 7 February 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.01.081