Abstract:
The deep learning object detection algorithms have become one of the powerful tools for road
vehicle detection in autonomous driving where vehicles are capable of sensing its environment and
operating without human involvement where they can go anywhere a traditional car goes and do
everything that an experienced human driver does. However, the limitation of the number of high-
quality labeled training samples makes the single-object detection algorithms unable to achieve
satisfactory accuracy in road vehicle detection. In this paper, by comparing the pros and cons of
various object detection algorithms, an ensemble is attenuated with multiple models and they are
selected for a method where the first step groups the overlapping regions. Subsequently, a voting
strategy is applied to discard some of those groups, this can further reduce the vehicle misdetection
of the target detection algorithm, helps in obtaining a better detection result with a Non-maximum
Suppression algorithm for the final prediction.
