Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Urban road problem well lid patrol method and system
The invention relates to the field of road engineering, and particularly discloses an urban road problem well lid patrol method and system, and the method comprises the steps: collecting a pavement image according to a preset image collection frequency, carrying out the neural network image recognition of the pavement image, and judging whether a well lid is recognized or not; when the manhole cover is identified, recording position data of a manhole cover frame and position data of a disease frame; and copying corresponding contents in the road surface image based on the manhole cover frame, inputting the copied contents into a trained deep learning classification neural network model, and judging whether the manhole cover and the road surface have defects or not. The invention provides a rapid detection means, the well lids are rapidly screened at high frequency, the well lids with problems can be found and handled in time, the number of the well lids with the problems can be obviously reduced, the maintenance cost is reduced, and the driving comfort and safety on a road are improved.
本发明涉及道路工程领域，具体公开了一种城市道路问题井盖巡查方法和系统，所述方法包括根据预设的图像采集频率采集路面图像，对所述路面图像进行神经网络图像识别，判断是否识别到井盖；当识别到井盖时，记录井盖边框的位置数据和病害边框的位置数据；基于所述井盖边框在所述路面图像中拷贝相应内容，并将拷贝的内容输入训练好的深度学习分类神经网络模型中，判断井盖和路面是否存在缺陷。本发明提供了一种快速的检测手段，高频次地对井盖进行快速筛查，及时发现、处置问题井盖，可以实现问题井盖数量明显减少，降低养护费用，提高道路的行车舒适性、安全性。
Urban road problem well lid patrol method and system
The invention relates to the field of road engineering, and particularly discloses an urban road problem well lid patrol method and system, and the method comprises the steps: collecting a pavement image according to a preset image collection frequency, carrying out the neural network image recognition of the pavement image, and judging whether a well lid is recognized or not; when the manhole cover is identified, recording position data of a manhole cover frame and position data of a disease frame; and copying corresponding contents in the road surface image based on the manhole cover frame, inputting the copied contents into a trained deep learning classification neural network model, and judging whether the manhole cover and the road surface have defects or not. The invention provides a rapid detection means, the well lids are rapidly screened at high frequency, the well lids with problems can be found and handled in time, the number of the well lids with the problems can be obviously reduced, the maintenance cost is reduced, and the driving comfort and safety on a road are improved.
本发明涉及道路工程领域，具体公开了一种城市道路问题井盖巡查方法和系统，所述方法包括根据预设的图像采集频率采集路面图像，对所述路面图像进行神经网络图像识别，判断是否识别到井盖；当识别到井盖时，记录井盖边框的位置数据和病害边框的位置数据；基于所述井盖边框在所述路面图像中拷贝相应内容，并将拷贝的内容输入训练好的深度学习分类神经网络模型中，判断井盖和路面是否存在缺陷。本发明提供了一种快速的检测手段，高频次地对井盖进行快速筛查，及时发现、处置问题井盖，可以实现问题井盖数量明显减少，降低养护费用，提高道路的行车舒适性、安全性。
Urban road problem well lid patrol method and system
一种城市道路问题井盖巡查方法和系统
ZHANG XIAOMING (Autor:in) / HONG GANG (Autor:in) / ZHONG SHENG (Autor:in) / YANG KANG (Autor:in) / DU JIAQI (Autor:in)
26.04.2022
Patent
Elektronische Ressource
Chinesisch
PATROL SUPPORT SYSTEM, PATROL SUPPORT METHOD, AND PATROL SUPPORT PROGRAM
| Europäisches Patentamt | 2022