A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Pavement disease duplicate removal method based on high-precision positioning and deep neural network
The invention relates to a pavement disease duplicate removal method based on high-precision positioning and a deep neural network, and the method comprises the following steps: 1, enabling vehicle-mounted intelligent patrol equipment to carry a positioning module, a vehicle-mounted camera, and a central industrial control computer for collection processing and uploading, and carrying out the overlook transformation external parameter calibration of the vehicle-mounted camera before the vehicle-mounted intelligent patrol equipment is started; 2, starting equipment to collect images and GPS data, and identifying road pavement diseases by using a neural network algorithm; 3, calculating the GPS coordinates of the road diseases by combining the calibration parameters of the vehicle-mounted camera and the GPS of the vehicle-mounted camera with the pixel coordinates of the road diseases in image recognition; and step 4, performing duplicate checking by using the GPS coordinates and the disease types of the diseases and database data, if the diseases are duplicated, determining the diseases as repeated diseases, and if the diseases are not duplicated, determining the diseases as newly-added diseases, and writing the diseases into a database. According to the method, high-precision positioning and information labeling of road diseases can be realized quickly and accurately, and the method is used for judging the same diseases in daily inspection.
本发明涉及一种基于高精度定位及深度神经网络的路面病害去重方法，包括以下步骤：步骤1、车载智能巡查设备搭载定位模块、车载相机、用于采集处理和上传的中央工控机，车载智能巡查设备启动前对车载相机进行俯视变换外参标定；步骤2、启动设备采集图像和GPS数据,并利用神经网络算法识别道路路面病害；步骤3、结合车载相机标定参数和车载相机的GPS结合图像识别中的道路病害像素坐标,计算出道路病害的GPS坐标的计算；步骤4、利用病害的GPS坐标和病害类型与数据库数据进行查重,若重复则判定为重复病害,若不重复则判定为新增病害,并写入数据库。其能够实现快速、准确的做到道路病害的高精度定位和标签信息，并用于日常巡查的相同病害判断。
Pavement disease duplicate removal method based on high-precision positioning and deep neural network
The invention relates to a pavement disease duplicate removal method based on high-precision positioning and a deep neural network, and the method comprises the following steps: 1, enabling vehicle-mounted intelligent patrol equipment to carry a positioning module, a vehicle-mounted camera, and a central industrial control computer for collection processing and uploading, and carrying out the overlook transformation external parameter calibration of the vehicle-mounted camera before the vehicle-mounted intelligent patrol equipment is started; 2, starting equipment to collect images and GPS data, and identifying road pavement diseases by using a neural network algorithm; 3, calculating the GPS coordinates of the road diseases by combining the calibration parameters of the vehicle-mounted camera and the GPS of the vehicle-mounted camera with the pixel coordinates of the road diseases in image recognition; and step 4, performing duplicate checking by using the GPS coordinates and the disease types of the diseases and database data, if the diseases are duplicated, determining the diseases as repeated diseases, and if the diseases are not duplicated, determining the diseases as newly-added diseases, and writing the diseases into a database. According to the method, high-precision positioning and information labeling of road diseases can be realized quickly and accurately, and the method is used for judging the same diseases in daily inspection.
本发明涉及一种基于高精度定位及深度神经网络的路面病害去重方法，包括以下步骤：步骤1、车载智能巡查设备搭载定位模块、车载相机、用于采集处理和上传的中央工控机，车载智能巡查设备启动前对车载相机进行俯视变换外参标定；步骤2、启动设备采集图像和GPS数据,并利用神经网络算法识别道路路面病害；步骤3、结合车载相机标定参数和车载相机的GPS结合图像识别中的道路病害像素坐标,计算出道路病害的GPS坐标的计算；步骤4、利用病害的GPS坐标和病害类型与数据库数据进行查重,若重复则判定为重复病害,若不重复则判定为新增病害,并写入数据库。其能够实现快速、准确的做到道路病害的高精度定位和标签信息，并用于日常巡查的相同病害判断。
Pavement disease duplicate removal method based on high-precision positioning and deep neural network
一种基于高精度定位及深度神经网络的路面病害去重方法
ZHANG XIAOMING (author) / YANG KANG (author) / ZHONG SHENG (author) / LI MENGKE (author) / LI XIANGYONG (author)
2022-11-25
Patent
Electronic Resource
Chinese
IPC:
G01N
Untersuchen oder Analysieren von Stoffen durch Bestimmen ihrer chemischen oder physikalischen Eigenschaften
,
INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
/
E01C
Bau von Straßen, Sportplätzen oder dgl., Decken dafür
,
CONSTRUCTION OF, OR SURFACES FOR, ROADS, SPORTS GROUNDS, OR THE LIKE
/
G01S
RADIO DIRECTION-FINDING
,
Funkpeilung
/
G06N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
,
Rechnersysteme, basierend auf spezifischen Rechenmodellen
/
G06T
Bilddatenverarbeitung oder Bilddatenerzeugung allgemein
,
IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
Estimating Precision Using Duplicate Measurements
| British Library Conference Proceedings | 2009
Estimating Precision Using Duplicate Measurements
| British Library Conference Proceedings | 2009
Estimating Precision Using Duplicate Measurements
| Taylor & Francis Verlag | 2009
Vision-Based Crack Detection of Asphalt Pavement Using Deep Convolutional Neural Network
| Springer Verlag | 2021
Automatic classification of pavement crack using deep convolutional neural network
| Taylor & Francis Verlag | 2020