A Method Based on Data Fusion of Multiple Sensors to Evaluate Road Cleanliness

For the supervision and evaluation of the quality of garbage detection and sanitation cleaning of urban roads, it has been carried out on a road-by road inspection by manual means for a long time, and then the mobile phone is used to locate the garbage and quantify the score. However, there are many problems in manual supervision. Not only is the work efficiency very low, but also a lot of manpower and material resources are wasted. This has not been able to meet the current smart city governance needs. Therefore, we propose a method of integrating multi-sensor data to replace the manual detection and evaluation of road cleanliness automatically and intelligently, and design a complete system, which can be loaded on the car to work on the road efficiently and directly output the score. Experiments show that our method can replace manual well to realize the identification and classification of garbage on the road, the calculation of garbage area, and the calculation of the latitude and longitude of the garbage, then display the above information of garbage in real time on the web map side.

[1]  Yixin Chen,et al.  Vehicle Tracking and Distance Estimation Based on Multiple Image Features , 2007, Fourth Canadian Conference on Computer and Robot Vision (CRV '07).

[2]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

[4]  Gaurav Mittal,et al.  SpotGarbage: smartphone app to detect garbage using deep learning , 2016, UbiComp.

[5]  Zhanyi Hu,et al.  PnP Problem Revisited , 2005, Journal of Mathematical Imaging and Vision.

[6]  Zhengyou Zhang,et al.  A Flexible New Technique for Camera Calibration , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Ali Farhadi,et al.  YOLOv3: An Incremental Improvement , 2018, ArXiv.