CN112329582B - Method and system for road water depth monitoring based on big data analysis and mechanism model - Google Patents

Abstract

Translated fromChinese

本发明公开了一种大数据分析与机理模型协同的道路积水深度监测方法，包括步骤：S1.采集道路积水信息；S2.对所述道路积水区域图像进行识别，得到道路积水区域实际面积，并对所述降雨量、所述道路单点积水深度以及所述道路积水区域实际面积进行处理，得到道路积水深度；S3.输出所述道路积水深度。一种大数据分析与机理模型协同的道路积水深度监测系统，包括参数采集模块、数据处理模块以及输出模块。本发明能够准确地计算出道路积水区域的最大深度以及预测最大深度，有效地防止出现道路积水深度误报、错报的情况，弥补了其他道路积水监测方式的不足，为道路排水建设提供了有力的参考。

The invention discloses a method for monitoring the depth of road water accumulation in cooperation with big data analysis and mechanism model, comprising the steps of: S1. collecting road water accumulation information; S2. The actual area is obtained, and the rainfall, the water depth at a single point of the road, and the actual area of the road water area are processed to obtain the road water depth; S3. Output the road water depth. A road water depth monitoring system coordinated by big data analysis and mechanism model includes a parameter acquisition module, a data processing module and an output module. The invention can accurately calculate the maximum depth of the road water accumulation area and predict the maximum depth, effectively prevent the occurrence of misreporting and misreporting of the road water accumulation depth, make up for the insufficiency of other road water accumulation monitoring methods, and provide a good solution for road drainage construction. Provides a strong reference.

Description

Translated fromChinese

大数据分析与机理模型协同的道路积水深度监测方法及系统Method and system for road water depth monitoring based on big data analysis and mechanism model

技术领域technical field

本发明涉及道路监测领域，具体涉及一种大数据分析与机理模型协同的道路积水深度监测方法及系统。The invention relates to the field of road monitoring, in particular to a method and system for monitoring the depth of road water accumulation in which big data analysis and mechanism models are coordinated.

背景技术Background technique

随着我国城市化建设的不断推进，城市气候和地表下垫面条件均发生明显的变化。在城市排水管网系统的建设滞后和排水系统规划不够合理的影响下，一旦遭遇台风、暴雨等恶劣天气，就会导致城市道路形成积水，严重影响了城市居民出行。With the continuous advancement of urbanization in my country, the urban climate and the underlying surface conditions have undergone significant changes. Under the influence of the lag in the construction of the urban drainage network system and the unreasonable planning of the drainage system, once encountering severe weather such as typhoons and rainstorms, it will lead to the formation of water on urban roads, which seriously affects the travel of urban residents.

现有的城市道路积水的监测方式，主要分为设备监测和数据挖掘分析两类。其中，设备监测通过部署水位传感器对道路积水进行监测；而数据挖掘分析通过对城市道路形成的原因以及过程进行处理，并结合城市道路地表高程数据、道路排水管网等信息来判断道路积水产生的可能性以及道路积水深度等。The existing monitoring methods of urban road water accumulation are mainly divided into two categories: equipment monitoring and data mining analysis. Among them, equipment monitoring monitors road water accumulation by deploying water level sensors; data mining analysis processes the causes and processes of urban road formation, and combines urban road surface elevation data, road drainage pipe network and other information to determine road water accumulation The possibility of occurrence and the depth of road water accumulation, etc.

但设备监测的方式只能对传感器测量处的积水深度进行测量，难于准确地测量城市道路积水区域的最大深度或有效深度，容易产生误报、错报。而数据挖掘分析需要城市道路的全部高精度数字高程数据、城市排水管网等信息，存在数据不好获取、数据不全、数据复杂、数据不好更新以及识别精度不够的问题。However, the method of equipment monitoring can only measure the depth of water at the location measured by the sensor, and it is difficult to accurately measure the maximum depth or effective depth of the water accumulation area on urban roads, which is prone to false alarms and false alarms. However, data mining analysis requires all high-precision digital elevation data of urban roads, urban drainage pipe network and other information. There are problems such as difficult data acquisition, incomplete data, complex data, difficult data updating and insufficient identification accuracy.

发明内容SUMMARY OF THE INVENTION

有鉴于此，本发明的目的是克服现有技术中的缺陷，提供大数据分析与机理模型协同的道路积水深度监测方法及系统，能够准确地计算出城市道路积水区域的最大深度以及预测最大深度，有效地防止出现道路积水深度误报、错报的情况，弥补了其他道路积水监测方式的不足，为道路排水建设提供了有力的参考。In view of this, the purpose of the present invention is to overcome the defects in the prior art, to provide a method and system for monitoring the depth of road water accumulation in coordination with big data analysis and a mechanism model, which can accurately calculate the maximum depth and prediction of urban road water accumulation areas. The maximum depth can effectively prevent the misreporting and misreporting of road water depth, make up for the deficiencies of other road water monitoring methods, and provide a powerful reference for road drainage construction.

本发明的大数据分析与机理模型协同的道路积水深度监测方法，包括如下步骤：The method for monitoring the depth of road water accumulation in collaboration with big data analysis and mechanism model of the present invention includes the following steps:

S1.采集道路积水信息；所述道路积水信息包括降雨量、道路单点积水深度以及道路积水区域图像；S1. Collect road water accumulation information; the road water accumulation information includes rainfall, water accumulation depth at a single point on the road, and images of the road accumulation area;

S2.对所述道路积水区域图像进行识别，得到道路积水区域实际面积，并对所述降雨量、所述道路单点积水深度以及所述道路积水区域实际面积进行处理，得到道路积水深度；所述道路积水深度包括道路积水区域的最大深度以及最大预测深度；S2. Identify the image of the road water accumulation area to obtain the actual area of the road water accumulation area, and process the rainfall, the single point water accumulation depth of the road and the actual area of the road water accumulation area to obtain the road water accumulation area. Water accumulation depth; the road water accumulation depth includes the maximum depth of the road water accumulation area and the maximum predicted depth;

S3.输出所述道路积水深度。S3. Output the water depth of the road.

进一步，步骤S2中，对所述道路积水区域图像进行识别，得到道路积水区域实际面积，具体包括：Further, in step S2, the image of the road water accumulation area is identified to obtain the actual area of the road water accumulation area, which specifically includes:

获取道路积水区域样本，并对所述道路积水区域样本进行特征标记处理，得到处理后的道路积水区域样本；Acquiring a road water accumulation area sample, and performing feature marking processing on the road water accumulation area sample to obtain a processed road water accumulation area sample;

将所述处理后的道路积水区域样本输入至向量机进行训练学习，得到道路积水识别模型；Inputting the processed road water area samples into a vector machine for training and learning to obtain a road water identification model;

使用道路积水识别模型对道路积水区域图像进行识别处理，得到道路积水区域实际面积。Using the road water identification model to identify the road water area image, the actual area of the road water area is obtained.

进一步，根据如下公式确定道路积水区域实际面积S：Further, the actual area S of the road water accumulation area is determined according to the following formula:

S＝k·S_像素；S=k·S_pixels ;

其中，k为道路积水区域实际面积的换算系数；S_像素为道路积水区域在图像中的像素面积；所述道路积水区域实际面积的换算系数

S_参考物为1m×1m的参考物在图像中的像素面积。Among them, k is the conversion factor of the actual area of the road water accumulation area; S_pixel is the pixel area of the road water accumulation area in the image; the conversion coefficient of the actual area of the road water accumulation area

The S_{reference is} the pixel area of the 1m × 1m reference in the image.

进一步，步骤S2中，根据如下公式确定所述道路积水区域的最大深度：Further, in step S2, the maximum depth of the road water accumulation area is determined according to the following formula:

其中，H(m)为第m分钟的道路积水区域的最大深度；h(m)为第m分钟的道路单点积水深度；

为辅助参数；Among them, H(m) is the maximum depth of the road water accumulation area in the mth minute; h(m) is the single point water accumulation depth of the road in the mth minute;

is an auxiliary parameter;

根据如下公式确定所述道路积水区域的最大预测深度：The maximum predicted depth of the road water area is determined according to the following formula:

其中，H(m+M)为第m分钟时M分钟后的道路积水区域的最大预测深度；

以及

均为辅助参数；R(j)为第j分钟的降雨量；S(m)为第m分钟的道路积水区域实际面积；Among them, H(m+M) is the maximum predicted depth of the road water area after M minutes at the mth minute;

as well as

are auxiliary parameters; R(j) is the rainfall in the jth minute; S(m) is the actual area of the road water accumulation area in the mth minute;

根据如下式子确定所述辅助参数

以及

The auxiliary parameter is determined according to the following formula

as well as

其中，x_1i为第i次在降雨开始后累计测量得到的降雨量；x_2i为第i次计算得到的降雨开始时间与当前时间的差值；x_3i为第i次计算得到的道路积水区域实际面积的负值；y_i为第i次计算得到的道路积水区域实际面积与道路单点积水深度的乘积；n为测量或计算的总次数；i＝1,2,…,n。Among them, x_1i is the accumulated rainfall measured after the i-th rain; x_2i is the difference between the rain start time and the current time calculated in the i-th time; x_3i is the road water accumulated in the i-th calculation. Negative value of the actual area of the area; y_i is the product of the actual area of the road water accumulation area obtained by the i-th calculation and the water accumulation depth at a single point of the road; n is the total number of measurements or calculations; i=1,2,…,n .

一种大数据分析与机理模型协同的道路积水深度监测系统，包括参数采集模块、数据处理模块以及输出模块；A road water depth monitoring system in which big data analysis and mechanism model are coordinated, comprising a parameter acquisition module, a data processing module and an output module;

所述参数采集模块，用于采集道路积水信息；所述道路积水信息包括降雨量、道路单点积水深度以及道路积水区域图像；The parameter collection module is used to collect road water accumulation information; the road water accumulation information includes rainfall, water accumulation depth at a single point on the road and images of the road accumulation area;

所述数据处理模块，用于处理所述道路积水信息，得到道路积水深度；所述道路积水深度包括道路积水区域的最大深度以及最大预测深度；The data processing module is configured to process the road water accumulation information to obtain the road water accumulation depth; the road accumulation depth includes the maximum depth and the maximum predicted depth of the road water accumulation area;

所述输出模块，用于输出所述道路积水深度。The output module is used for outputting the road water depth.

进一步，所述参数采集模块包括翻斗式雨量传感器、摄像头以及压力液位传感器；Further, the parameter acquisition module includes a tipping bucket rain sensor, a camera and a pressure level sensor;

所述翻斗式雨量传感器，用于测量降雨量；The tipping bucket rain sensor is used for measuring rainfall;

所述摄像头，用于拍摄道路积水区域图像；The camera is used for capturing images of road water accumulation areas;

所述压力液位传感器，用于测量道路单点积水深度。The pressure liquid level sensor is used to measure the water depth of a single point on the road.

进一步，所述数据处理模块包括嵌入式处理模块以及与所述嵌入式处理模块通信连接的远程数据传输模块；Further, the data processing module includes an embedded processing module and a remote data transmission module communicatively connected to the embedded processing module;

所述嵌入式处理模块，用于对所述道路积水区域图像进行识别，得到道路积水区域实际面积，并对所述降雨量、所述道路单点积水深度以及所述道路积水区域实际面积进行处理，得到道路积水深度；The embedded processing module is used to identify the image of the road water accumulation area, obtain the actual area of the road water accumulation area, and analyze the rainfall, the road water accumulation depth at a single point and the road accumulation area. The actual area is processed to obtain the road water depth;

所述远程数据传输模块，用于接收所述道路积水深度，并发送降雨预报信息至所述嵌入式处理模块。The remote data transmission module is used for receiving the road water depth and sending rainfall forecast information to the embedded processing module.

进一步，所述嵌入式处理模块通过NB-IoT通信模组与所述远程数据传输模块进行通信。Further, the embedded processing module communicates with the remote data transmission module through an NB-IoT communication module.

进一步，所述嵌入式处理模块通过如下方法确定所述道路积水区域的最大深度：Further, the embedded processing module determines the maximum depth of the road water accumulation area by the following method:

其中，H(m)为第m分钟的道路积水区域的最大深度；h(m)为第m分钟的道路单点积水深度；

为辅助参数；Among them, H(m) is the maximum depth of the road water accumulation area in the mth minute; h(m) is the single point water accumulation depth of the road in the mth minute;

is an auxiliary parameter;

所述嵌入式处理模块通过如下方法确定所述道路积水区域的最大预测深度：The embedded processing module determines the maximum predicted depth of the road water accumulation area by the following method:

其中，H(m+M)为第m分钟时M分钟后的道路积水区域的最大预测深度；

以及

均为辅助参数；R(j)为第j分钟的降雨量；S(m)为第m分钟的道路积水区域实际面积；Among them, H(m+M) is the maximum predicted depth of the road water area after M minutes at the mth minute;

as well as

are auxiliary parameters; R(j) is the rainfall in the jth minute; S(m) is the actual area of the road water accumulation area in the mth minute;

所述嵌入式处理模块通过如下方法确定所述辅助参数

以及

The embedded processing module determines the auxiliary parameter by the following method

as well as

其中，x_1i为第i次在降雨开始后累计测量得到的降雨量；x_2i为第i次计算得到的降雨开始时间与当前时间的差值；x_3i为第i次计算得到的道路积水区域实际面积的负值；y_i为第i次计算得到的道路积水区域实际面积与道路单点积水深度的乘积；n为测量或计算的总次数；i＝1,2,…,n。Among them, x_1i is the accumulated rainfall measured after the i-th rain; x_2i is the difference between the rain start time and the current time calculated in the i-th time; x_3i is the road water accumulated in the i-th calculation. Negative value of the actual area of the area; y_i is the product of the actual area of the road water accumulation area obtained by the i-th calculation and the water accumulation depth at a single point of the road; n is the total number of measurements or calculations; i=1,2,…,n .

进一步，根据如下公式确定第m分钟的道路积水区域实际面积S(m)：Further, the actual area S(m) of the road water accumulation area in the mth minute is determined according to the following formula:

S(m)＝k·S_像素(m)；S(m)=k·S_pixel (m);

其中，k为道路积水区域实际面积的换算系数；S_像素(m)为第m分钟的道路积水区域在图像中的像素面积；所述道路积水区域实际面积的换算系数

S_参考物为1m×1m的参考物在图像中的像素面积。Among them, k is the conversion factor of the actual area of the road water accumulation area; S_pixel (m) is the pixel area of the road water accumulation area in the image in the mth minute; the conversion factor of the actual area of the road water accumulation area

The S_{reference is} the pixel area of the 1m × 1m reference in the image.

本发明的有益效果是：本发明公开的一种大数据分析与机理模型协同的道路积水深度监测方法及系统，通过机器学习对采集的道路积水区域图像进行大数据分析，保证了获取的道路积水区域实际面积的可靠性，通过严谨科学的机理模型对道路积水区域实际面积、降雨量、道路单点积水深度进行处理，进而准确地计算出道路积水区域的积水最大深度以及预测最大深度，有效地防止出现道路积水深度误报、错报的情况，弥补了其他道路积水监测方式的不足，为道路排水建设提供了有力的参考。The beneficial effects of the present invention are as follows: the method and system for monitoring the depth of road water accumulation in collaboration with big data analysis and mechanism models disclosed in the present invention perform big data analysis on the collected road water accumulation area images through machine learning, ensuring that the acquired The reliability of the actual area of the road water accumulation area is processed through a rigorous and scientific mechanism model to process the actual area of the road water accumulation area, rainfall, and the depth of the single point water accumulation on the road, and then accurately calculate the maximum water accumulation depth of the road water accumulation area. As well as predicting the maximum depth, it can effectively prevent the misreporting and misreporting of road water depth, make up for the deficiencies of other road water monitoring methods, and provide a powerful reference for road drainage construction.

附图说明Description of drawings

下面结合附图和实施例对本发明作进一步描述：Below in conjunction with accompanying drawing and embodiment, the present invention is further described:

图1为本发明的方法流程示意图；Fig. 1 is the method flow schematic diagram of the present invention;

图2为本发明的系统结构示意图；Fig. 2 is the system structure schematic diagram of the present invention;

图3为本发明的道路积水监测终端结构示意图；3 is a schematic structural diagram of the road water monitoring terminal of the present invention;

其中，1-支撑灯杆；2-翻斗式雨量传感器；3-道路积水监测终端；4-摄像头；5-积水信息显示屏；6-压力液位传感器。Among them, 1- support light pole; 2- tipping bucket rain sensor; 3- road water monitoring terminal; 4- camera; 5- water information display screen; 6- pressure level sensor.

具体实施方式Detailed ways

以下结合说明书附图对本发明做出进一步的说明，如图所示：The present invention is further described below in conjunction with the accompanying drawings of the description, as shown in the figure:

本发明的大数据分析与机理模型协同的道路积水深度监测方法，如图1所示，包括如下步骤：The method for monitoring the depth of road water accumulation in which big data analysis and mechanism model are coordinated, as shown in FIG. 1 , includes the following steps:

S1.采集道路积水信息；所述道路积水信息包括降雨量、道路单点积水深度以及道路积水区域图像；S1. Collect road water accumulation information; the road water accumulation information includes rainfall, water accumulation depth at a single point on the road, and images of the road accumulation area;

S2.对所述道路积水区域图像进行识别，得到道路积水区域实际面积，并对所述降雨量、所述道路单点积水深度以及所述道路积水区域实际面积进行处理，得到道路积水深度；所述道路积水深度包括道路积水区域的最大深度以及最大预测深度；S2. Identify the image of the road water accumulation area to obtain the actual area of the road water accumulation area, and process the rainfall, the single point water accumulation depth of the road and the actual area of the road water accumulation area to obtain the road water accumulation area. Water accumulation depth; the road water accumulation depth includes the maximum depth of the road water accumulation area and the maximum predicted depth;

S3.输出所述道路积水深度。S3. Output the water depth of the road.

本发明通过对采集的道路积水信息数据进行大数据分析处理，得到分析处理后的数据；并使用机理模型对分析处理后的数据进行计算，进而得到道路积水区域的最大深度以及最大预测深度。其中，所述大数据包括结构化数据以及非机构化数据，所述结构化数据包括降雨量以及道路单点积水深度；所述非机构化数据包括道路积水区域图像；所述机理模型包括所述道路积水区域的最大深度的计算公式以及所述道路积水区域的最大预测深度的计算公式。The present invention obtains the analyzed and processed data by performing big data analysis and processing on the collected road water accumulation information data; and uses the mechanism model to calculate the analyzed and processed data, thereby obtaining the maximum depth and the maximum predicted depth of the road water accumulation area . Wherein, the big data includes structured data and non-institutionalized data, the structured data includes rainfall and the depth of water accumulation at a single point on the road; the non-institutionalized data includes images of road water accumulation areas; the mechanism model includes The calculation formula of the maximum depth of the road water accumulation area and the calculation formula of the maximum predicted depth of the road water accumulation area.

本实施例中，步骤S2中，对所述道路积水区域图像进行识别，得到道路积水区域实际面积，具体包括：In this embodiment, in step S2, the image of the road water accumulation area is identified to obtain the actual area of the road water accumulation area, which specifically includes:

获取道路积水区域的图像，进而得到大量的正样本(道路积水区域)、负样本(道路非积水区域)，其中，正样本与负样本在颜色特征、纹理特征上有明显的不同，对所述道路积水区域的正、负样本分别进行颜色特征与纹理特征的标记处理，得到处理后的道路积水正、负区域样本；Obtain the image of the road water area, and then obtain a large number of positive samples (road water area) and negative samples (road non-water area). Among them, the positive samples and negative samples are obviously different in color features and texture features. Marking processing of color features and texture features is performed on the positive and negative samples of the road water accumulation area, respectively, to obtain processed road water accumulation positive and negative area samples;

采用HSV颜色空间识别和SVM纹理特征识别相结合的方式对所述处理后的道路积水区域正、负样本进行识别，得到道路积水识别模型；所述HSV颜色空间识别以及所述SVM纹理特征识别采用现有技术，在此不再赘述；The positive and negative samples of the processed road water area are identified by a combination of HSV color space recognition and SVM texture feature recognition to obtain a road water recognition model; the HSV color space recognition and the SVM texture feature The identification adopts the prior art, which will not be repeated here;

使用道路积水识别模型对道路积水区域图像进行识别处理，得到道路积水区域实际面积，进而实现了基于道路积水区域图像的大数据分析。The road water accumulation recognition model is used to identify and process the road water accumulation area image, and the actual area of the road accumulation area is obtained, and then the big data analysis based on the road accumulation area image is realized.

本实施例中，根据如下公式确定道路积水区域实际面积S：In this embodiment, the actual area S of the road water accumulation area is determined according to the following formula:

S＝k·S_像素；S=k·S_pixels ;

其中，k为道路积水区域实际面积的换算系数；S_像素为道路积水区域在图像中的像素面积；所述道路积水区域实际面积的换算系数

S_参考物为1m×1m大小的正方形参考物在图像中的像素面积。Among them, k is the conversion factor of the actual area of the road water accumulation area; S_pixel is the pixel area of the road water accumulation area in the image; the conversion coefficient of the actual area of the road water accumulation area

The S_{reference is} the pixel area of the 1m × 1m square reference in the image.

本实施例中，步骤S2中，根据如下公式确定所述道路积水区域的最大深度：In this embodiment, in step S2, the maximum depth of the road water accumulation area is determined according to the following formula:

其中，H(m)为第m分钟的道路积水区域的最大深度；h(m)为第m分钟的道路单点积水深度；

为辅助参数；Among them, H(m) is the maximum depth of the road water accumulation area in the mth minute; h(m) is the single point water accumulation depth of the road in the mth minute;

is an auxiliary parameter;

根据如下公式确定所述道路积水区域的最大预测深度：The maximum predicted depth of the road water area is determined according to the following formula:

其中，H(m+M)为第m分钟时M分钟后的道路积水区域的最大预测深度；

以及

均为辅助参数；R(j)为第j分钟的降雨量，其中，第m分钟到第m+M分钟之间某个时刻的降雨量为降雨预报提供的预测降雨量；S(m)为第m分钟的道路积水区域实际面积；所述第m分钟的道路积水区域实际面积S(m)为在第m分钟时测量得到的道路积水区域实际面积S。Among them, H(m+M) is the maximum predicted depth of the road water area after M minutes at the mth minute;

as well as

are auxiliary parameters; R(j) is the rainfall in the jth minute, where the rainfall at a certain moment between the mth minute and the m+Mth minute is the predicted rainfall provided by the rainfall forecast; S(m) is the The actual area of the road water accumulation area in the mth minute; the actual area S(m) of the road water accumulation area in the mth minute is the actual area S of the road water accumulation area measured at the mth minute.

根据如下式子确定所述辅助参数

以及

The auxiliary parameter is determined according to the following formula

as well as

其中，x_1i为第i次在降雨开始后累计测量得到的降雨量，所述降雨开始后累计测量得到的降雨量为

t₀为降雨开始时间，t_c为当前时间；x_2i为第i次计算得到的降雨开始时间与当前时间的差值，所述降雨开始时间与当前时间的差值为t₀-t_c；x_3i为第i次计算得到的当前时间的道路积水区域实际面积的负值，所述当前时间的道路积水区域实际面积为S(t_c)，其对应的负值为-S(t_c)，所述S(t_c)为在t_c时刻测量得到的道路积水区域实际面积S；y_i为第i次计算得到的当前时间的道路积水区域实际面积与道路单点积水深度的乘积，所述当前时间的道路积水区域实际面积与道路单点积水深度的乘积为S(t_c)·h(t_c)，h(t_c)为当前时间的道路单点积水深度；n为测量或计算的总次数；i＝1,2,…,n；为了保证测量或计算的有效性及准确性，n的取值不小于4。Among them, x_1i is the accumulated rainfall measured after the i-th rain starts, and the accumulated rainfall after the rainfall starts is:

t₀ is the rain start time, t_c is the current time; x_2i is the difference between the rain start time calculated for the i-th time and the current time, and the difference between the rain start time and the current time is t₀ -t_c ; x_3i is the negative value of the actual area of the road stagnant area at the current time calculated for the i-th time, the actual area of the road stagnant area at the current time is S(t_c ), and its corresponding negative value is -S(t_c ), the S(t_c ) is the actual area S of the road water accumulation area measured at time t_c ; y_i is the actual area of the road water accumulation area at the current time calculated for the i-th time and the road water accumulation at a single point The product of the depth, the product of the actual area of the road water accumulation area at the current time and the road single point water accumulation depth is S(t_c ) h(t_c ), and h(t_c ) is the road single point product at the current time. Water depth; n is the total number of measurements or calculations; i=1,2,...,n; in order to ensure the validity and accuracy of the measurement or calculation, the value of n is not less than 4.

一种大数据分析与机理模型协同的道路积水深度监测系统，包括参数采集模块、数据处理模块以及输出模块；A road water depth monitoring system in which big data analysis and mechanism model are coordinated, comprising a parameter acquisition module, a data processing module and an output module;

所述参数采集模块，用于采集道路积水信息；所述道路积水信息包括降雨量、道路单点积水深度以及道路积水区域图像；The parameter collection module is used to collect road water accumulation information; the road water accumulation information includes rainfall, water accumulation depth at a single point on the road and images of the road accumulation area;

所述数据处理模块，用于处理所述道路积水信息，得到道路积水深度；所述道路积水深度包括道路积水区域的最大深度以及最大预测深度；The data processing module is configured to process the road water accumulation information to obtain the road water accumulation depth; the road accumulation depth includes the maximum depth and the maximum predicted depth of the road water accumulation area;

所述输出模块，用于输出所述道路积水深度；本实施例中，如图2所示，所述输出模块为积水信息显示屏5，所述积水信息显示屏5为一种LED显示屏，其显示的信息有5行，内容由上至下依次为：积水提示信息、道路单点积水深度、道路积水区域的最大深度、道路积水区域的预测最大深度以及本地实时时间。The output module is used to output the depth of the road water accumulation; in this embodiment, as shown in FIG. 2 , the output module is the water accumulationinformation display screen 5, and the water accumulationinformation display screen 5 is a kind of LED The display screen has 5 lines of information, and the contents from top to bottom are: water reminder information, single-point water depth on the road, maximum depth of the road water area, predicted maximum depth of the road water area, and local real-time time.

本实施例中，如图2所示，所述参数采集模块包括翻斗式雨量传感器2、摄像头4以及压力液位传感器6；其中，所述翻斗式雨量传感器2以及所述摄像头4分别安装在支撑灯杆1的第一悬臂以及第二悬臂上，所述第一悬臂以及第二悬臂的长度均为1米；所述压力液位传感器6布置在道路路面上；所述支撑灯杆1为道路照明常用设备，所述支撑灯杆1的高度在8～12米范围内，其安装在道路一侧。In this embodiment, as shown in FIG. 2 , the parameter collection module includes a tippingbucket rain sensor 2 , a camera 4 and a pressureliquid level sensor 6 ; wherein the tippingbucket rain sensor 2 and the camera 4 are respectively installed on a support On the first cantilever and the second cantilever of the light pole 1, the lengths of the first cantilever and the second cantilever are both 1 meter; the pressureliquid level sensor 6 is arranged on the road surface; the supporting light pole 1 is a road Common lighting equipment, the height of the support light pole 1 is in the range of 8-12 meters, and it is installed on one side of the road.

所述翻斗式雨量传感器2为脉冲型雨量传感器，与所述数据处理模块连接，用于测量道路积水路段的降雨量，并将降雨量输入所述数据处理模块；The tipping buckettype rain sensor 2 is a pulse-type rain sensor, which is connected to the data processing module and is used to measure the rainfall in the road section with accumulated water, and input the rainfall into the data processing module;

所述摄像头4，与所述数据处理模块连接，用于对道路积水区域进行监控并周期性地录制道路积水区域视频或拍摄道路积水区域图像，并将道路积水区域图像输入所述数据处理模块；所述摄像头4的像素为500万，可扩展至800万及以上；其中，所述录制的道路积水区域视频可以分割成多张静态的图像，进而实现对图像的处理；The camera 4, connected to the data processing module, is used to monitor the road water area and periodically record the road water area video or shoot the road water area image, and input the road water area image into the a data processing module; the pixels of the camera 4 are 5 million, which can be expanded to 8 million and above; wherein, the recorded video of the road stagnant area can be divided into multiple static images, thereby realizing image processing;

所述压力液位传感器6，与所述数据处理模块连接，用于测量道路单点积水深度，并将道路单点积水深度输入所述数据处理模块。The pressureliquid level sensor 6 is connected to the data processing module, and is used to measure the water accumulation depth at a single point of the road, and input the water accumulation depth at a single point of the road into the data processing module.

本实施例中，所述数据处理模块包括嵌入式处理模块以及与所述嵌入式处理模块通信连接的远程数据传输模块；In this embodiment, the data processing module includes an embedded processing module and a remote data transmission module communicatively connected to the embedded processing module;

如图2、3所示，所述嵌入式处理模块为道路积水监测终端3，所述道路积水监测终端3包括嵌入式微控制器以及设置于嵌入式微控制器上的若干个通信接口；所述若干个通信接口包括显示屏接口、供电电源接口、翻斗式雨量传感器接口、压力液位传感器接口、摄像头接口以及USART串行接口；所述嵌入式微控制器具有图像处理功能，所述嵌入式微控制器可为树莓派3b、3b+、4b或4b+。As shown in Figures 2 and 3, the embedded processing module is a roadwater monitoring terminal 3, and the roadwater monitoring terminal 3 includes an embedded microcontroller and several communication interfaces provided on the embedded microcontroller; The several communication interfaces include a display interface, a power supply interface, a tipping bucket rain sensor interface, a pressure liquid level sensor interface, a camera interface and a USART serial interface; the embedded microcontroller has an image processing function, and the embedded microcontroller has an image processing function. The device can be a Raspberry Pi 3b, 3b+, 4b or 4b+.

所述道路积水监测终端3通过翻斗式雨量传感器接口接收翻斗式雨量传感器2输入的降雨量；所述道路积水监测终端3通过摄像头接口接收摄像头4输入的道路积水区域图像；所述道路积水监测终端3通过压力液位传感器接口接收压力液位传感器6输入的道路单点积水深度。The roadwater monitoring terminal 3 receives the rainfall input by the tippingbucket rain sensor 2 through the tipping bucket rain sensor interface; the roadwater monitoring terminal 3 receives the road water area image input by the camera 4 through the camera interface; the road The wateraccumulation monitoring terminal 3 receives the water accumulation depth at a single point of the road input by the pressureliquid level sensor 6 through the pressure liquid level sensor interface.

所述道路积水监测终端3通过所述供电电源接口接收外部电源的供电，所述外部电源为市电220V交流电源；所述道路积水监测终端3通过所述显示屏接口将道路积水深度输入积水信息显示屏5。The roadwater monitoring terminal 3 receives the power supply from an external power source through the power supply interface, and the external power source is a 220V AC power supply of mains; the roadwater monitoring terminal 3 monitors the road water depth through the display interface. Enter the stagnantwater information display 5.

所述道路积水监测终端3用于对所述道路积水区域图像进行识别，得到道路积水区域实际面积，并对所述降雨量、所述道路单点积水深度以及所述道路积水区域实际面积进行处理，得到道路积水深度；The roadwater monitoring terminal 3 is used to identify the image of the road water area, obtain the actual area of the road water area, and monitor the rainfall, the water depth at a single point on the road and the road water. The actual area of the area is processed to obtain the water depth of the road;

所述远程数据传输模块为云服务平台，所述云服务平台用于接收所述道路积水监测终端3输入的道路积水深度，并发送降雨预报信息至所述道路积水监测终端3；所述降雨预报信息包括道路积水区域所在地的降雨量；所述云服务平台采用现有技术，在此不再赘述。The remote data transmission module is a cloud service platform, and the cloud service platform is used to receive the road water depth input by the roadwater monitoring terminal 3, and send rainfall forecast information to the roadwater monitoring terminal 3; The rainfall forecast information includes the rainfall at the location of the road stagnant area; the cloud service platform adopts the existing technology, which will not be repeated here.

所述道路积水监测终端3以及所述积水信息显示屏5均安装在支撑灯杆1上。The road wateraccumulation monitoring terminal 3 and the water accumulationinformation display screen 5 are both installed on the supporting light pole 1 .

本实施例中，所述道路积水监测终端3通过NB-IoT通信模组与所述云服务平台进行通信。所述NB-IoT通信模组通过USART串行接口接入到所述道路积水监测终端3；所述NB-IoT通信模组为窄带物联网(Narrow Band Internet of Things,NB-IoT)，其支持接入的设备更多，功能损耗更低，信号覆盖能力更强，从而保证了道路积水监测终端3与云服务平台之间信息传递的稳定与可靠。所述NB-IoT通信模组采用现有技术，在此不再赘述。In this embodiment, the roadwater monitoring terminal 3 communicates with the cloud service platform through an NB-IoT communication module. The NB-IoT communication module is connected to the roadwater monitoring terminal 3 through the USART serial interface; the NB-IoT communication module is Narrow Band Internet of Things (NB-IoT), which There are more devices that support access, lower functional loss, and stronger signal coverage, thereby ensuring the stability and reliability of information transmission between the roadwater monitoring terminal 3 and the cloud service platform. The NB-IoT communication module adopts the prior art, and details are not repeated here.

本实施例中，所述道路积水监测终端3通过如下方法确定所述道路积水区域的最大深度：In this embodiment, the roadwater monitoring terminal 3 determines the maximum depth of the road water area by the following method:

其中，H(m)为第m分钟的道路积水区域的最大深度；h(m)为第m分钟的道路单点积水深度；

为辅助参数；Among them, H(m) is the maximum depth of the road water accumulation area in the mth minute; h(m) is the single point water accumulation depth of the road in the mth minute;

is an auxiliary parameter;

所述道路积水监测终端3通过如下方法确定所述道路积水区域的最大预测深度：The roadwater monitoring terminal 3 determines the maximum predicted depth of the road water area by the following method:

其中，H(m+M)为第m分钟时M分钟后的道路积水区域的最大预测深度；

以及

均为辅助参数；R(j)为第j分钟的降雨量，其中，第m分钟到第m+M分钟之间某个时刻的降雨量为降雨预报提供的预测降雨量；S(m)为第m分钟的道路积水区域实际面积；Among them, H(m+M) is the maximum predicted depth of the road water area after M minutes at the mth minute;

as well as

are auxiliary parameters; R(j) is the rainfall in the jth minute, where the rainfall at a certain moment between the mth minute and the m+Mth minute is the predicted rainfall provided by the rainfall forecast; S(m) is the The actual area of the road water accumulation area in the mth minute;

所述道路积水监测终端3通过如下方法确定所述辅助参数

以及

The roadwater monitoring terminal 3 determines the auxiliary parameters through the following methods

as well as

其中，x_1i为第i次在降雨开始后累计测量得到的降雨量，所述降雨开始后累计测量得到的降雨量为

t₀为降雨开始时间，t_c为当前时间，所述降雨量由翻斗式雨量传感器2测量得到；x_2i为第i次计算得到的降雨开始时间与当前时间的差值，所述降雨开始时间与当前时间的差值为t₀-t_c；x_3i为第i次计算得到的当前时间的道路积水区域实际面积的负值，所述当前时间的道路积水区域实际面积为S(t_c)，其对应的负值为-S(t_c)，所述S(t_c)为在t_c时刻测量得到的道路积水区域实际面积S；y_i为第i次计算得到的当前时间的道路积水区域实际面积与道路单点积水深度的乘积，所述当前时间的道路积水区域实际面积与道路单点积水深度的乘积为S(t_c)·h(t_c)，h(t_c)为当前时间的道路单点积水深度，所述h(t_c)为在t_c时刻由压力液位传感器6测量得到；n为测量或计算的总次数；i＝1,2,…,n；为了保证测量或计算的有效性及准确性，n的取值不小于4。Among them, x_1i is the accumulated rainfall measured after the i-th rain starts, and the accumulated rainfall after the rainfall starts is:

t₀ is the rain start time, t_c is the current time, and the rainfall is measured by the tippingbucket rain sensor 2; x_2i is the difference between the rain start time calculated for the i-th time and the current time, the rain start time The difference with the current time is t₀ -t_c ; x_3i is the negative value of the actual area of the road stagnant area at the current time calculated for the i-th time, and the actual area of the road stagnant area at the current time is S(t_c ), the corresponding negative value is -S(t_c ), the S(t_c ) is the actual area S of the road water accumulation area measured at the time t_c ; y_i is the current time calculated for the i-th time The product of the actual area of the road water accumulation area and the water accumulation depth of a single point of the road, the product of the actual area of the road accumulation area at the current time and the water accumulation depth of a single point of the road is S(t_c )·h(t_c ), h(t_c ) is the water depth at a single point of the road at the current time, and the h(t_c ) is measured by the pressureliquid level sensor 6 at the time t_c ; n is the total number of measurements or calculations; i=1, 2,...,n; in order to ensure the validity and accuracy of measurement or calculation, the value of n is not less than 4.

本实施例中，根据如下公式确定第m分钟的道路积水区域实际面积S(m)：In this embodiment, the actual area S(m) of the road water accumulation area in the mth minute is determined according to the following formula:

S(m)＝k·S_像素(m)；S(m)=k·S_pixel (m);

其中，k为道路积水区域实际面积的换算系数；S_像素(m)为第m分钟的道路积水区域在图像中的像素面积；所述道路积水区域实际面积的换算系数

S_参考物为1m×1m大小的正方形参考物在图像中的像素面积。Among them, k is the conversion factor of the actual area of the road water accumulation area; S_pixel (m) is the pixel area of the road water accumulation area in the image in the mth minute; the conversion factor of the actual area of the road water accumulation area

The S_{reference is} the pixel area of the 1m × 1m square reference in the image.

最后说明的是，以上实施例仅用以说明本发明的技术方案而非限制，尽管参照较佳实施例对本发明进行了详细说明，本领域的普通技术人员应当理解，可以对本发明的技术方案进行修改或者等同替换，而不脱离本发明技术方案的宗旨和范围，其均应涵盖在本发明的权利要求范围当中。Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be Modifications or equivalent substitutions without departing from the spirit and scope of the technical solutions of the present invention should be included in the scope of the claims of the present invention.

Claims (4)

1. A road ponding depth monitoring method based on big data analysis and mechanism model cooperation is characterized in that: the method comprises the following steps:

s1, collecting road ponding information; the road ponding information comprises rainfall, road single-point ponding depth and road ponding area images;

s2, identifying the road ponding area image to obtain the actual area of the road ponding area, and processing the rainfall, the road single-point ponding depth and the actual area of the road ponding area to obtain the road ponding depth; the road ponding depth comprises the maximum depth and the maximum prediction depth of a road ponding area; in step S2, identifying the road ponding region image to obtain an actual area of the road ponding region, specifically including:

acquiring a road ponding area sample, and performing characteristic marking processing on the road ponding area sample to obtain a processed road ponding area sample;

inputting the processed road ponding area sample to a vector machine for training and learning to obtain a road ponding recognition model;

identifying the road ponding area image by using a road ponding identification model to obtain the actual area of the road ponding area;

determining the actual area S of the road ponding area according to the following formula:

S＝k·S_pixel ；

Wherein k is a conversion coefficient of the actual area of the road waterlogging area; s_Pixel The pixel area of the road water accumulation area in the image is shown; conversion coefficient of actual area of road ponding region

S_{Reference object} Pixel area of a reference object in an image of 1m × 1 m;

in step S2, the maximum depth of the road water accumulation region is determined according to the following formula:

wherein H (m) is the maximum depth of the road water accumulation area at the mth minute; h (m) is the road single-point water accumulation depth at the mth minute;

is an auxiliary parameter;

determining the maximum prediction depth of the road water accumulation area according to the following formula:

wherein H (M + M) is the maximum predicted depth of the road water accumulation area at the mth minute after M minutes;

and

are all auxiliary parameters; r (j) is the rainfall at the j minute; (m) is the actual area of the road waterlogging area in the mth minute;

determining the auxiliary parameter according to the following equation

And

wherein x is_1i The measured rainfall is accumulated after the beginning of rainfall for the ith time; x is the number of_2i Calculating the difference value between the rainfall starting time and the current time obtained by the ith calculation; x is the number of_3i Calculating the actual area of the road waterlogged area for the ith time; y is_i Calculating the product of the actual area of the road ponding area and the road single-point ponding depth for the ith time; n is the total number of measurements or calculations; 1,2, …, n;

and S3, outputting the road accumulated water depth.

2. The utility model provides a road ponding degree of depth monitoring system that big data analysis and mechanism model are collaborative which characterized in that: the device comprises a parameter acquisition module, a data processing module and an output module;

the parameter acquisition module is used for acquiring road ponding information; the road ponding information comprises rainfall, road single-point ponding depth and road ponding area images;

the data processing module is used for processing the road ponding information to obtain the road ponding depth; the road ponding depth comprises the maximum depth and the maximum prediction depth of a road ponding area; the data processing module comprises an embedded processing module and a remote data transmission module in communication connection with the embedded processing module;

the embedded processing module is used for identifying the road ponding area image to obtain the actual area of the road ponding area, and processing the rainfall, the road single-point ponding depth and the actual area of the road ponding area to obtain the road ponding depth;

the remote data transmission module is used for receiving the road ponding depth and sending rainfall forecast information to the embedded processing module;

the output module is used for outputting the road ponding depth;

the embedded processing module determines the maximum depth of the road water accumulation area by the following method:

wherein H (m) is the maximum depth of the road water accumulation area at the mth minute; h (m) is the road single-point water accumulation depth at the mth minute;

is an auxiliary parameter;

the embedded processing module determines the maximum prediction depth of the road ponding area by the following method:

wherein H (M + M) is the maximum predicted depth of the road water accumulation area at the mth minute after M minutes;

and

are all auxiliary parameters; r (j) is the rainfall at the j minute; (m) is the actual area of the road waterlogging area in the mth minute;

the embedded processing module determines the auxiliary parameter by the following method

And

wherein x is_1i The measured rainfall is accumulated after the beginning of rainfall for the ith time; x is the number of_2i Rainfall onset time calculated for the ith timeA difference from a current time; x is the number of_3i Calculating the actual area of the road waterlogged area for the ith time; y is_i Calculating the product of the actual area of the road ponding area and the road single-point ponding depth for the ith time; n is the total number of measurements or calculations; 1,2, …, n;

determining the actual area S (m) of the road waterlogging area at the m minute according to the following formula:

S(m)＝k·S_pixel (m)；

Wherein k is a conversion coefficient of the actual area of the road waterlogging area; s. the_Pixel (m) is the pixel area of the road ponding region in the image at the mth minute; conversion coefficient of actual area of road ponding region

S_{Reference object} The area of a pixel in the image of a reference 1m × 1 m.

3. The big data analysis and mechanism model coordinated roadway water depth monitoring system according to claim 2, wherein: the parameter acquisition module comprises a tipping bucket type rainfall sensor, a camera and a pressure liquid level sensor;

the tipping bucket type rainfall sensor is used for measuring rainfall;

the camera is used for shooting road ponding area images;

and the pressure liquid level sensor is used for measuring the single-point accumulated water depth of the road.

4. The big data analysis and mechanism model coordinated roadway water depth monitoring system according to claim 2, wherein: the embedded processing module communicates with the remote data transmission module through an NB-IoT communication module.

Images