16+
DOI: 10.18413/2518-1092-2017-2-4-50-58

VEHICLE DETECTION ON HIGHWAY IMAGES BASED ON SINGLE SHOT MULTIBOX DETECTOR

In this article we consider the application of the modern object detection method – Single Shot Multibox Detector. We have trained the convolutional neural network for vehicle detection on a sample of 3000 images with marked areas where are the vehicles are placed. A network quality check was performed on 7000 test images. The test and training samples contain images made by a monocular camera mounted in a vehicle moving along suburban highways during daylight hours. Recall and precision of object detection on the test sample is correspondingly more than 88% and 78%. Recognition of one frame takes 28.5 milliseconds. Experiment was performed on a graphics processor using NVidia CUDA technology. The obtained results can be applied in driver assistance systems and monitoring of the traffic situations.

Number of views: 4155 (view statistics)
Количество скачиваний: 8142
Full text (PDF)To articles list
  • User comments
  • Reference lists

While nobody left any comments to this publication.
You can be first.

Leave comment: