arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1710.02546cs.CVcs.LGstat.ML
收藏

基于深度学习的违章停车实时检测系统

Real-Time Illegal Parking Detection System Based on Deep Learning

School of Electronic Engineering, Xidian University(西安电子科技大学电子工程学院)

Xuemei Xie, Chenye Wang, Shu Chen, Guangming Shi, Zhifu Zhao

摘要:

越来越多的违章停车已经变得越来越严重。目前检测违章停车的方法主要是基于背景分割的方法。然而,该方法鲁棒性差,对环境敏感。借鉴深度学习的思想,提出了一种新的违章停车检测系统。摄像头捕捉到的违章车辆首先通过著名的Single Shot MultiBox Detector(SSD)算法进行定位和分类。为了提高性能,我们建议通过调整默认框的宽高比来优化SSD,以更好地适应我们的数据集。在此基础上,采用运动跟踪分析的方法,对感兴趣区域(ROI)内的违章车辆进行判断。实验表明,该系统在复杂环境下能够实现99%的准确率和实时的【形】(25 FPS)检测,具有较强的鲁棒性。

英文摘要:

The increasing illegal parking has become more and more serious. Nowadays the methods of detecting illegally parked vehicles are based on background segmentation. However, this method is weakly robust and sensitive to environment. Benefitting from deep learning, this paper proposes a novel illegal vehicle parking detection system. Illegal vehicles captured by camera are firstly located and classified by the famous Single Shot MultiBox Detector (SSD) algorithm. To improve the performance, we propose to optimize SSD by adjusting the aspect ratio of default box to accommodate with our dataset better. After that, a tracking and analysis of movement is adopted to judge the illegal vehicles in the region of interest (ROI). Experiments show that the system can achieve a 99% accuracy and real-time (25FPS) detection with strong robustness in complex environments.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑