A real-time early warning system for road icing based on big data

Abstract

This invention discloses a real-time road icing early warning system based on big data, comprising: a data acquisition module, a processing module, an image recognition and comparison module, and a prediction module. The data acquisition module includes a real-time monitoring module and a historical data acquisition module; the data acquisition module is the foundation of the processing module and is used to acquire real-time monitoring data and historical data; the image recognition and comparison module is used to identify specific moving targets in captured road images. This invention belongs to the field of real-time road icing early warning technology, and its purpose is to solve the problem that existing technologies can only roughly predict whether the entire road will be icy, and cannot accurately predict the road surface area within a specified range. The achieved technical effect is: it can calculate the road icing rate of the road surface area within a specified range, and reduce the deviation between the predicted result and the actual result within the range of the historical road icing threshold F.

Description

Technical Field

[0001] This invention relates to the field of real-time road icing early warning technology, and specifically to a real-time road icing early warning system based on big data. Background Technology

[0002] In meteorology, road icing refers to the accumulation of snow or ice caused by rain, snow, freezing rain, or fog droplets falling onto the ground where the temperature is below 0°C. Humidity and wind speed are also major factors influencing road icing. Under high humidity conditions, water vapor in the air is more likely to condense on the road surface and form an ice layer. When the relative humidity approaches or reaches 100%, the likelihood of road icing increases significantly. Simultaneously, increased wind speed accelerates heat loss from the road surface, promoting the icing process. Road icing significantly reduces the coefficient of friction between vehicle tires and the road surface, affecting normal vehicle operation and increasing the risk of skidding and traffic accidents. When the road surface is dry, it has the lowest brightness and a grayish hue; when icy, it becomes grayish-white or transparent. Statistics show that traffic accidents caused by road icing account for 35% of all traffic accidents in winter. Therefore, a real-time road icing early warning system is needed to predict road icing conditions and reduce the occurrence of traffic accidents.

[0003] Most current real-time road icing warning systems rely on a single meteorological sensor to collect road surface condition data and meteorological information to roughly predict whether the entire road will be icy. They cannot accurately predict the road surface area within a specified range, thus compromising the accuracy of the final prediction. Summary of the Invention

[0004] To address these issues, the present invention provides a real-time road icing early warning system based on big data, thereby solving the aforementioned problems in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] According to a first aspect of the present invention, a real-time road icing early warning system based on big data includes:

[0007] The system includes a data acquisition module, a processing module, an image recognition and comparison module, and a prediction module. The data acquisition module includes a real-time monitoring module and a historical data acquisition module.

[0008] The data acquisition module is the foundation of the processing module, used to acquire real-time monitoring data and historical data;

[0009] The image recognition and comparison module is used to identify specific moving targets in captured road images and can further extract depth information. It can compare images taken at different times, from different angles, or under different conditions to find the same or similar images.

[0010] The specific implementation steps of the real-time road icing warning system include the following steps:

[0011] Step 1: The real-time monitoring module collects the temperature and humidity condition information, road surface wind speed information, road surface pictures, and air pressure data of the predicted road surface at each time period, obtains the temperature and humidity condition information and road surface wind speed information data at each time period and makes them into curves, and based on the big data historical data, the historical data acquisition module obtains the historical data and gets the historical data curve;

[0012] Step 2: Both the real-time monitoring information and the historical data are transmitted to the processing module. The processing module analyzes and compares the historical data curve and the curve made in Step 1, determines the road icing threshold F in the historical data and the corresponding countermeasures for the road icing threshold F, predicts the relationship between the road condition and road icing, and constructs a road icing calculation model;

[0013] The calculation formula of the road icing calculation model is:

[0014]

[0015] Among them, A is the road icing rate, C is the precipitation within the time period, V is the density of moist air, is a fixed coefficient value (J / kg), R is the number of road base layers (R≥0), r is the number of road surface layers (r≥1), and D is the monitored road surface area;

[0016] The prediction module compares the road icing rate A calculated by the road icing calculation model with the road icing threshold F. If F + 1 > A > F, it is determined that the road icing rate A is a moderate icing warning. If A≥F + 1, it is determined that the road icing rate A is a severe icing warning. If A≤F, it is determined that the road icing rate A is a severe icing warning.

[0017] Furthermore, the processing module screens the road surface picture data taken by the camera at each time period, uses the Gaussian mixture model to extract and delete the moving targets, retains the road surface pictures containing the complete road surface and transmits them to the image recognition and comparison module. The image recognition and comparison module selects the road surface pictures with relatively clear image saturation and hue at each time period, and through processing operations such as parameter calculation, parameter image reconstruction, difference, and filtering, and performs binary classification to determine the threshold F1 for image comparison.

[0018] Furthermore, when the real-time image brightness obtained in the processing module < F1, it is predicted that the road surface in this time period is a dry road surface and the picture is deleted. When the real-time image brightness ≥ F1, it is predicted that the road surface in this time period is an icing road surface, and the road icing calculation model is used to calculate the road surface icing situation, which is convenient to improve the accuracy of the prediction result and provide timely and accurate road condition information.

[0019] Furthermore, the prediction module disseminates early warning information through broadcasting, the internet, and mobile apps to remind drivers and relevant departments, ensuring the safety of travel during icy weather to the greatest extent possible.

[0020] Furthermore, the formula for calculating the density V of moist air is:

[0021]

[0022] Where P is the total pressure during the time period, T is the temperature in Celsius +273.15 during the time period, and K is the relative humidity during the time period.

[0023] Furthermore, the processing module is used to extract, transform, calculate and analyze valuable data information from a large amount of raw data, and can also compare historical data curves with the curves created in step one.

[0024] The present invention has the following advantages:

[0025] 1. Through the real-time monitoring module and image recognition comparison module, the temperature and humidity module, wind speed sensor module, camera and atmospheric pressure sensor module can collect information on the temperature and humidity of the road surface at different time periods, wind speed information of the road surface, road surface images and air pressure data, respectively, to realize real-time monitoring and early warning of road icing, and provide accurate road condition information.

[0026] 2. The road icing calculation model can calculate the road icing rate of the road surface area within a specified range, improve the overall prediction accuracy, and reduce the deviation between the prediction results and the actual results within the range of the road icing threshold F in historical data. Attached Figure Description

[0027] Figure 1 The flowchart illustrates a real-time early warning system for road icing based on big data, provided for some embodiments of the present invention. Detailed Implementation

[0028] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] Example 1

[0030] like Figure 1As shown, a real-time road icing early warning system based on big data according to a first aspect embodiment of the present invention includes: a data acquisition module, a processing module, an image recognition and comparison module, and a prediction module. The data acquisition module includes a real-time monitoring module and a historical data acquisition module, and the modules are connected to each other by electrical signals.

[0031] The data acquisition module is the foundation of the processing module and is used to acquire real-time monitoring data and historical data.

[0032] The processing module is used to extract, transform, calculate and analyze valuable data information from a large amount of raw data, and can also compare historical data curves with the curves created in step one.

[0033] The image recognition and comparison module is used to identify specific moving targets in captured road images and can further extract depth information. It can compare images taken at different times, from different angles, or under different conditions to find the same or similar images.

[0034] The specific implementation steps of this real-time road icing early warning system include the following:

[0035] Step 1: Collect information on the temperature and humidity of the road surface, wind speed, road surface images, and air pressure at different time periods using the temperature and humidity module, wind speed sensor module, camera, and atmospheric pressure sensor module respectively. Obtain the temperature and humidity information and wind speed information of the road surface at different time periods and generate curves. Based on big data historical data, obtain historical data through the historical data acquisition module to obtain historical data curves.

[0036] Step 2: Real-time monitoring information and historical data are transmitted to the processing module. The processing module analyzes and compares the historical data curves and the curves created in Step 1 to determine the road icing threshold F in the historical data, as well as the corresponding countermeasures for the road icing threshold F, predict the relationship between road conditions and road icing, and construct a road icing calculation model.

[0037] S1. The calculation formula for the road icing calculation model is:

[0038]

[0039] Where A is the road icing rate, C is the precipitation during the time period, and V is the moist air density, 3.34*10 5 Here, R is a fixed coefficient value (J / kg), R is the number of road base layers (R≥0), r is the number of road pavement layers (r≥1), and D is the monitored pavement area.

[0040] S2. The formula for calculating the density V of moist air is:

[0041]

[0042] Among them, P is the total pressure within the time period, T is the Celsius temperature + 273.15 within the time period, and K is the relative humidity within the time period;

[0043] Step 3: The processing module filters the road surface image data captured by the camera in each time period, uses the Gaussian mixture model to extract and delete moving targets, retains the road surface images containing the complete road surface and transmits them to the image recognition and comparison module. The image recognition and comparison module selects the road surface images with relatively clear image saturation and hue in each time period, and performs processing operations such as parameter calculation, parameter image reconstruction, differential, and filtering, and conducts binary classification to determine the threshold F1 for image comparison. When the brightness of the acquired real-time image < F1, it is predicted that the road surface in this time period is a dry road surface, and this image is deleted. When the real-time image brightness ≥ F1, it is predicted that the road surface in this time period is an ice-covered road surface, and the ice-covered condition of the road surface is calculated through the road ice calculation model to improve the accuracy of the prediction results and provide timely and accurate road condition information;

[0044] Step 4: The prediction module compares the road ice formation rate A calculated by the road ice calculation model with the road ice formation threshold F. If F + 1 > A > F, it is determined that the road ice formation rate A is a moderate ice formation warning. If A ≥ F + 1, it is determined that the road ice formation rate A is a severe ice formation warning. If A ≤ F, it is determined that the road ice formation rate A is a severe ice formation warning;

[0045] Step 5: Warning information is released through channels such as radio, Internet, and mobile phone APPs to remind drivers and relevant departments, ensuring the safety of travel on roads prone to icing to the greatest extent. By integrating modern sensor technology, data processing technology, and intelligent image recognition technology, real-time monitoring and warning of road icing conditions can be achieved, providing accurate road condition information, thereby reducing traffic accidents caused by road icing.

Claims

1. A real-time early warning system for road icing based on big data, characterized in that, Including: A data acquisition module, a processing module, an image recognition and comparison module, and a prediction module. The data acquisition module includes a real-time monitoring module and a historical data acquisition module; The data acquisition module is the basis of the processing module and is used to acquire real-time monitoring data and historical data; The image recognition and comparison module is used to identify specific moving targets in the captured road pictures, and can further extract depth information, and can compare images taken at different times, from different perspectives or under different conditions to find the same or similar images; The specific implementation steps of this real-time road icing warning system include the following steps: Step 1: The real-time monitoring module collects the temperature and humidity condition information, road surface wind speed information, road surface pictures, and air pressure data of the predicted road surface at each time period, obtains the temperature and humidity condition information and road surface wind speed information data at each time period and makes them into curves, and based on big data historical data, the historical data acquisition module obtains historical data to get a historical data curve; Step 2: Both the real-time monitoring information and the historical data are transmitted to the processing module. The processing module analyzes and compares the historical data curve and the curve made in Step 1 to determine the road icing threshold F in the historical data and the corresponding countermeasures for the road icing threshold F, predicts the relationship between the road condition and road icing, and constructs a road icing calculation model; The calculation formula of the road icing calculation model is: Where, A is the road icing rate, C is the precipitation within the time period, V is the density of moist air, is a fixed coefficient value (J / kg), R is the number of layers of the road base (R≥0), r is the number of layers of the road surface (r≥1), and D is the monitored road surface area; The prediction module compares the road icing rate A calculated by the road icing calculation model with the road icing threshold F. If F + 1 > A > F, it is judged that the road icing rate A is a moderate icing warning. If A≥F + 1, it is judged that the road icing rate A is a severe icing warning. If A≤F, it is judged that the road icing rate A is a severe icing warning.

2. The real-time road icing early warning system based on big data according to claim 1, characterized in that, The processing module screens the road surface picture data taken by the camera at each time period, uses the Gaussian mixture model to extract and delete moving targets, retains the road surface pictures containing the complete road surface and transmits them to the image recognition and comparison module. The image recognition and comparison module selects the road surface pictures with relatively clear image saturation and hue at each time period, and after processing operations such as parameter calculation, parametric image reconstruction, difference, and filtering, and performs binary classification to determine the threshold F1 for image comparison.

3. The real-time road icing early warning system based on big data according to claim 2, characterized in that, When the brightness of the real-time image obtained by the processing module < F1, it is predicted that the road surface in this time period is a dry road surface and the picture is deleted. When the brightness of the real-time image ≥ F1, it is predicted that the road surface in this time period is an icing road surface, and the road icing calculation model is used to calculate the road surface icing situation to improve the accuracy of the prediction result and provide timely and accurate road condition information.

4. The real-time road icing early warning system based on big data according to claim 1, characterized in that, The prediction module publishes warning information through channels such as broadcasts, the Internet, and mobile phone APPs to remind drivers and relevant departments to ensure the safety of traveling on roads prone to icing to the greatest extent.

5. A real-time road icing early warning system based on big data according to claim 1, characterized in that, The calculation formula of the density V of the moist air is: Where P is the total pressure during the time period, T is the temperature in Celsius +273.15 during the time period, and K is the relative humidity during the time period.

6. The real-time road icing early warning system based on big data according to claim 1, characterized in that, The processing module is used to extract, transform, calculate and analyze valuable data information from a large amount of raw data, and can also compare historical data curves with the curves created in step one.

Images