Abstract
With the development of science and technology, the application of vehicle object detection in intelligent video monitoring and vehicle assisted driving has become more extensive. Traditional vehicle object detection algorithms have some inefficiencies in generalization ability and recognition rate. To solve this problem, this paper proposes an improved vehicle detection method based on YOLO(You Only Look Once) V3. Our model improves the YOLOV3 algorithm, integrating the label smoothing, and adopts K-means++ algorithm to analyze the data aggregation and avoid the disappearance of the gradient, make our train model has better generalization ability. Finally, we use the UA-DETRAC data set for training, and the experimental results showed that our method has a better detection speed and a higher recognition rate for vehicle detection, which is 6% mAP higher than the original YOLOV3 algorithm.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
