Expressway snow-melting agent automatic spraying monitoring method and device, electronic equipment and storage medium

Abstract

The application discloses a highway snow-melting agent automatic spraying monitoring method and device, electronic equipment and storage medium, comprising: collecting road surface images along the snow-melting agent spraying route, and screening and preprocessing the road surface images to obtain preprocessed road surface images; performing pixel-level segmentation on the preprocessed road surface images based on a semantic segmentation model to obtain segmented road surface images; constructing a snow-melting agent identification model; identifying the segmented road surface images based on the snow-melting agent identification model to obtain the spreading density and the spreading uniformity of different snow-melting agent spreading areas; and giving a snow-melting agent spreading prompt according to the spreading labels, the spreading density and the spreading uniformity of the snow-melting agent spreading areas, so that the spreading density and the spreading uniformity of the snow-melting agent can be identified by using the semantic segmentation model and the snow-melting agent identification model in combination, the snow-melting agent can be supplemented and spread in time, the use efficiency of the snow-melting agent can be improved, resource waste can be reduced, and road safety can be ensured.

Description

Technical Field

[0001] This application relates to the field of snow melting agent spraying monitoring, and in particular to automatic snow melting agent spraying monitoring methods, devices, electronic equipment and storage media for highways. Background Technology

[0002] In winter, heavy snowfall, especially on highways, can easily lead to traffic congestion and accidents. To ensure safe passage on highways, de-icing agents are typically sprayed to accelerate snow melting and improve road capacity. Currently, relevant departments usually deploy spraying vehicles to spray de-icing agents along the road. Spraying can also be controlled by adjusting parameters such as spray density and speed, or remotely controlled using weather sensing equipment for automated spraying. However, while these methods achieve some degree of automation, they still have shortcomings. For example, there is a lack of real-time and effective monitoring of the road surface after spraying, making it difficult to understand the completion of the spraying work, detect uneven or insufficient application of de-icing agents, and provide timely replenishment reminders. This results in some sections of the road not melting sufficiently, thus affecting road safety. Summary of the Invention

[0003] The main purpose of this application is to provide an automatic monitoring method, device, electronic equipment and storage medium for snow melting agent spraying on highways, which aims to solve the technical problem that the existing technology lacks real-time and effective means of monitoring the road surface condition after spraying, making it impossible to understand the completion status of the spraying work in a timely manner and provide supplementary reminders, thus affecting road traffic safety.

[0004] To achieve the above objectives, this application proposes an automatic monitoring method for de-icing agent spraying on highways, the method comprising: Images of the road surface are collected along the de-icing agent spraying route, and the images are filtered and preprocessed to obtain preprocessed images of the road surface. The preprocessing includes at least grayscale image conversion, noise reduction, and contrast enhancement. The preprocessed road surface image is segmented at the pixel level based on the semantic segmentation model to obtain the segmented road surface image, wherein the segmented road surface image includes the spreading label, spreading particles, and position information of each de-icing agent spreading area. Construct a snow melting agent identification model; Based on the snow melting agent recognition model, the segmented road surface image is identified to obtain the spreading density and spreading uniformity of different snow melting agent spreading areas; The application of de-icing agent is indicated based on the application labels, application density, and application uniformity of each application area.

[0005] In one embodiment, the step of performing pixel-level segmentation on the preprocessed road surface image based on a semantic segmentation model to obtain a segmented road surface image includes: The preprocessed road surface image is input into a semantic segmentation model, and the semantic segmentation model is used to extract features and divide regions of the preprocessed road surface image to obtain multiple de-icing agent spreading areas. Each pixel in the preprocessed road surface image is assigned a category to obtain the category information of each pixel. Based on the category information of each pixel, generate the spreading label, spreading particles, and position information of each snow melting agent spreading area; A segmented road surface image is generated based on the de-icing agent application areas, the application labels of the de-icing agent application areas, the application particles, and the location information of the application particles.

[0006] In one embodiment, the semantic segmentation model is used to extract features and divide regions from the preprocessed road surface image to obtain multiple de-icing agent spreading areas, including: The semantic segmentation model is used to extract features from the preprocessed road surface image to obtain low-level feature maps and high-level feature maps. The pixel values ​​in the low-level feature map and the high-level feature map are averaged respectively to obtain multiple pixel average values; The average values ​​of multiple pixels are combined into a high-dimensional semantic representation; Based on the high-dimensional semantic representation, the preprocessed road surface image is divided into regions to obtain multiple candidate regions for snow melting agent application. The multiple candidate areas for de-icing agent application are screened using a non-maximum suppression algorithm to obtain multiple de-icing agent application areas. The non-maximum suppression algorithm is used to retain local maxima and suppress non-maxima to remove redundant application areas.

[0007] In one embodiment, generating the spreading label, spreading particles, and position information of each de-icing agent spreading area based on the category information of each pixel includes: Based on the category information of each pixel, the number of pixels in each category and the positional distribution of pixels in each category in the preprocessed road surface image are obtained; Based on the quantity and location distribution, determine the corresponding spreading labels for each de-icing agent spreading area; The pixel corresponding to the snow melting agent in the category information is taken as the target pixel, and the coordinates of the target pixel in the road surface image coordinate system are obtained; Target pixels with continuous coordinates are used as spreading particles, and the position information of each spreading particle is determined.

[0008] In one embodiment, constructing the de-icing agent identification model includes: Historical road surface image data is collected and labeled to obtain labeled historical road surface image data, wherein the labeled historical road surface image data includes the true values ​​of the spreading density and spreading uniformity of different de-icing agent spreading areas. The historical road surface image data is divided into a training set and a test set; The training set is input into the initial recognition model for training to obtain the trained initial recognition model; The test set is input into the initial recognition model for verification to obtain the predicted values ​​of the spreading density and spreading uniformity of different de-icing agent spreading areas; The performance indicators of the initial identification model are determined based on the predicted and actual values ​​of the spreading density and spreading uniformity in different de-icing agent spreading areas. When the performance index reaches the corresponding index threshold, the trained initial recognition model will be used as the de-icing agent recognition model. If the performance metric fails to reach the corresponding threshold, the trained initial recognition model is adjusted, and the adjusted initial recognition model is retrained and validated as the initial recognition model until the performance metric reaches the corresponding threshold.

[0009] In one embodiment, the step of identifying the segmented road surface image based on the de-icing agent identification model to obtain the spreading density and spreading uniformity of different de-icing agent spreading areas includes: The segmented road surface image is input into the de-icing agent recognition model to obtain the area of ​​each de-icing agent spreading area; Obtain the number of de-icing agent particles in each of the de-icing agent spreading areas in the segmented road surface image; The spreading density of the corresponding snow-melting agent spreading area is determined based on the area of ​​each spreading area and the number of spreading particles. The average distance between pairs of de-icing agent particles in each of the de-icing agent spreading areas in the segmented road surface image is calculated based on the position information of the particles in each of the de-icing agent spreading areas. The uniformity of the de-icing agent application area is determined based on the average distance. The formula for calculating the average distance between the pairs of spreading particles is as follows: in, For the de-icing agent application area R i The average distance between the medium-sized particles. For the de-icing agent application area Ri The total number of particles in the spread, To indicate the de-icing agent application area R i The distance between the j-th and k-th particles. The attenuation factor indicates that as the average distance between the spread particles increases, the influence between the spread particle pairs gradually decreases.

[0010] In one embodiment, the step of providing de-icing agent application guidance based on the application status, density, and uniformity of the de-icing agent includes: Determine whether the de-icing agent is currently present based on the application label of each of the aforementioned de-icing agent application areas; When the current status of the snow-melting agent is that the snow-melting agent is not present, a snow-melting agent shortage warning will be issued; When the current condition of the snow melting agent is that the snow melting agent is present, the spreading density of different snow melting agent spreading areas is compared with the preset spreading density threshold to obtain the spreading density comparison result; By comparing the uniformity of the de-icing agent application area with the preset uniformity threshold, the uniformity comparison results are obtained. Based on the comparison results of the spreading density and the spreading uniformity, a de-icing agent spreading prompt message is generated; The snow melting agent application prompt message is sent to the user terminal to prompt the user to supplement the snow melting agent application or adjust the application strategy.

[0011] Furthermore, to achieve the above objectives, this application also proposes an automatic monitoring device for de-icing agents sprayed on highways, the automatic monitoring device for de-icing agents sprayed on highways comprising: The acquisition module is used to acquire road surface images along the snow melting agent spraying route, and to filter and preprocess the road surface images to obtain preprocessed road surface images. The preprocessing includes at least grayscale image conversion, noise reduction, and contrast enhancement. The segmentation module is used to perform pixel-level segmentation on the preprocessed road surface image based on a semantic segmentation model to obtain a segmented road surface image, wherein the segmented road surface image includes the spreading labels, spreading particles, and position information of each de-icing agent spreading area. Build modules are used to construct de-icing agent identification models; The identification module is used to identify the segmented road surface image based on the snow melting agent identification model, and to obtain the spreading density and spreading uniformity of different snow melting agent spreading areas; The prompt module is used to provide prompts for the application of de-icing agent based on the application label, application density, and application uniformity of each de-icing agent application area.

[0012] In addition, to achieve the above objectives, this application also proposes an electronic device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the automatic spraying monitoring method for highway de-icing agents as described above.

[0013] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the automatic spraying monitoring method for highway de-icing agents as described above.

[0014] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the automatic spraying and monitoring method for highway de-icing agents as described above.

[0015] One or more technical solutions proposed in this application acquire road surface images along the de-icing agent spraying route, and filter and preprocess the road surface images to obtain preprocessed road surface images. The preprocessing includes at least grayscale image conversion, noise reduction, and contrast enhancement. The preprocessed road surface images are then segmented at the pixel level based on a semantic segmentation model to obtain segmented road surface images. The segmented road surface images include the spreading labels, spreading particles, and positional information of each de-icing agent spreading area. A de-icing agent recognition model is constructed. Based on the de-icing agent recognition model, the segmented road surface images are recognized to obtain the spreading density and uniformity of different de-icing agent spreading areas. De-icing agent spreading prompts are provided based on the spreading labels, spreading density, and spreading uniformity of each de-icing agent spreading area. By combining semantic segmentation and de-icing agent recognition models to perform pixel-level segmentation of road images and identify the de-icing agent spreading status, the above methods can accurately and efficiently identify the spreading density and uniformity of de-icing agents. This allows for timely understanding of the de-icing agent spraying situation and prompting for timely replenishment, thereby improving the efficiency of de-icing agent use, reducing resource waste, and ensuring safe road passage in winter. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating an embodiment of the automatic spraying monitoring method for de-icing agents on highways provided in this application; Figure 2 This is a flowchart illustrating Embodiment 2 of the method for monitoring automatic spraying of de-icing agents on highways in this application. Figure 3 This is a schematic diagram of the module structure of the automatic spraying monitoring device for de-icing agents on highways, as described in an embodiment of this application. Figure 4 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the automatic spraying monitoring method for de-icing agents on highways in the embodiments of this application.

[0019] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0021] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0022] The main solution of this application embodiment is as follows: Road surface images are acquired along the de-icing agent spraying route, and the road surface images are filtered and preprocessed to obtain preprocessed road surface images. The preprocessing includes at least grayscale image conversion, noise reduction, and contrast enhancement. The preprocessed road surface images are then segmented at the pixel level based on a semantic segmentation model to obtain segmented road surface images. The segmented road surface images include the spreading labels, spreading particles, and positional information of each de-icing agent spreading area. A de-icing agent recognition model is constructed. Based on the de-icing agent recognition model, the segmented road surface images are recognized to obtain the spreading density and uniformity of different de-icing agent spreading areas. De-icing agent spreading prompts are provided based on the spreading labels, spreading density, and spreading uniformity of each de-icing agent spreading area.

[0023] Due to the lack of existing technology to monitor the road surface conditions after spraying in real time, it is difficult to understand the completion status of the spraying work in a timely manner, and it is also difficult to detect uneven or missing de-icing agent. As a result, the snow on some road sections has not been fully melted, thus affecting road traffic safety.

[0024] This application provides a solution that combines a semantic segmentation model and a de-icing agent recognition model to perform pixel-level segmentation of road surface images and identify the de-icing agent spreading status. This solution can accurately and efficiently identify the spreading density and uniformity of the de-icing agent, provide timely information on the de-icing agent spraying status, and promptly issue reminders for respreading the de-icing agent. This improves the efficiency of de-icing agent use, reduces resource waste, and ensures safe road passage in winter.

[0025] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device capable of performing the above functions. The following description uses an electronic device as an example to illustrate this embodiment and the subsequent embodiments.

[0026] Based on this, embodiments of this application provide a method for monitoring automatic spraying of de-icing agents on highways, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the automatic spraying and monitoring method for de-icing agents on highways according to this application.

[0027] In this embodiment, the automatic spraying monitoring method for highway de-icing agents includes steps S10 to S50: Step S10: Collect road surface images along the de-icing agent spraying route, and filter and preprocess the road surface images to obtain preprocessed road surface images. The preprocessing includes at least grayscale image conversion, noise reduction, and contrast enhancement.

[0028] It should be noted that the spraying described in this embodiment is specifically for spraying de-icing agents on highways, but it can also be applied to other road scenarios; this embodiment does not impose specific limitations. De-icing agents are chemicals that can lower the melting temperature of ice and snow. They are spread on roads manually or mechanically to accelerate the melting of ice and snow and ensure smooth traffic flow. When collecting road surface images, high-definition cameras or similar devices can be used to capture images along the de-icing agent spraying route, obtaining images of the road surface including the de-icing agent's application. High-definition cameras or similar devices can be installed on the de-icing agent spraying device, for example, at the rear of the vehicle spraying the de-icing agent, or on the exterior of the spraying equipment on the vehicle, to capture images or videos of the road surface after the de-icing agent has been sprayed. This embodiment uses image acquisition as an example. If video is captured, images can also be obtained from video frames, or the video can be directly analyzed.

[0029] Specifically, the acquired road surface images are screened to remove blurry, duplicate, or unsuitable images, ensuring the accuracy and efficiency of subsequent processing. Preprocessing of the screened road surface images includes: grayscale image conversion (converting acquired color images to grayscale to reduce data volume and improve processing speed); noise reduction (removing noise points from the image to improve image quality); and contrast enhancement (adjusting the brightness and contrast of the image to make the de-icing agent application details clearer).

[0030] Step S20: Perform pixel-level segmentation on the preprocessed road surface image based on the semantic segmentation model to obtain the segmented road surface image, wherein the segmented road surface image includes the spreading labels, spreading particles, and position information of each de-icing agent spreading area.

[0031] It should be noted that a semantic segmentation model is a deep learning model used for image segmentation, which can divide an image into multiple regions with specific semantic meanings. A semantic segmentation model can be a deep learning-based model, such as a Fully Convolutional Network (FCN) or U-Net. These models, after training, can identify different semantic regions in an image. This embodiment does not impose specific limitations on this.

[0032] Understandably, semantic segmentation models are used to perform pixel-level segmentation on preprocessed road surface images to identify different objects and regions in the image, especially the areas where de-icing agents are applied. Through semantic segmentation, the model can accurately distinguish the road surface, de-icing agent application areas, and other related objects, assigning a label to each region and identifying the application particles and their location information.

[0033] It should be understood that in the segmented road surface image, the spreading label is used to identify different de-icing agent spreading areas, the spreading particles represent de-icing agent particles, and the location information records in detail the specific location of each spreading particle in the image.

[0034] Step S30: Construct a de-icing agent identification model.

[0035] It should be noted that the de-icing agent identification model is used to further analyze the segmented road surface images to identify the spreading density and uniformity of different de-icing agent spreading areas. The de-icing agent identification model can be a machine learning-based model, such as a support vector machine (SVM), random forest, or deep learning model. After training, these models can output corresponding spreading density and uniformity indices based on the de-icing agent spreading situation in the road surface images.

[0036] In one feasible implementation, step S30 may include: collecting historical road surface image data and annotating the historical road surface image data to obtain annotated historical road surface image data, wherein the annotated historical road surface image data includes the true values ​​of the spreading density and spreading uniformity of different de-icing agent spreading areas; dividing the historical road surface image data into a training set and a test set; inputting the training set into an initial recognition model for training to obtain a trained initial recognition model; inputting the test set into the initial recognition model for verification to obtain predicted values ​​of the spreading density and spreading uniformity of different de-icing agent spreading areas; determining the performance index of the initial recognition model based on the predicted and true values ​​of the spreading density and spreading uniformity of different de-icing agent spreading areas; when the performance index reaches the corresponding index threshold, using the trained initial recognition model as the de-icing agent recognition model; when the performance index does not reach the corresponding index threshold, adjusting the trained initial recognition model and retraining and verifying the adjusted initial recognition model as the initial recognition model until the performance index reaches the corresponding index threshold.

[0037] It should be noted that historical road surface image data can be downloaded from the internet or obtained from professional image databases or institutions. This data should include road surface images under different weather conditions, lighting conditions, road surface materials, and de-icing agent application scenarios to ensure the generalization ability and accuracy of the de-icing agent identification model.

[0038] It should be understood that the annotated historical road surface image data needs to include the true values ​​of spread density and spread uniformity, which can be obtained through manual annotation or measurement with professional equipment.

[0039] The purpose of dividing the training and test sets is to evaluate the model's performance and ensure that the model can perform well on different datasets. The training and test sets can be divided according to a certain ratio, such as 70% as the training set and 30% as the test set, or other ratios can be made according to actual needs.

[0040] In practice, during training, the initial recognition model continuously learns from the data in the training set, gradually adjusting its parameters to improve its ability to recognize the density and uniformity of de-icing agent application. The validation process evaluates the performance of the trained model by obtaining predicted values ​​from the test set data and comparing them with the true values ​​to determine the model's performance metrics. Performance metrics can include accuracy, recall, and F1 score, which comprehensively reflect the model's recognition performance. When a performance metric reaches its corresponding threshold, it indicates that the model has good recognition capabilities, and the trained initial recognition model can be used as the final de-icing agent recognition model. If the performance metric does not reach the corresponding threshold, the model parameters need to be adjusted, and training and validation need to be repeated until the performance metric requirements are met.

[0041] Step S40: Based on the snow melting agent recognition model, the segmented road surface image is identified to obtain the spreading density and spreading uniformity of different snow melting agent spreading areas.

[0042] It should be noted that spreading density refers to the number of de-icing agent particles per unit area, reflecting the density of the de-icing agent's distribution on the road surface. Spreading uniformity refers to whether the de-icing agent particles are evenly distributed on the road surface, without obvious clusters or sparse areas. These two indicators are important bases for evaluating the effectiveness of de-icing agent spreading and are of great significance for ensuring snow melting effect and improving road traffic safety.

[0043] It should be understood that the de-icing agent identification model counts and statistically analyzes the spread particles in each spreading area based on the segmented road surface image and their location information, thereby calculating the spreading density. Simultaneously, by performing spatial analysis on the location information of the spread particles, it assesses whether the distribution of de-icing agent particles on the road surface is uniform, thus obtaining a spreading uniformity index. These indices provide crucial data support for subsequent snow melting effect evaluation and road maintenance. In practical applications, the output of the de-icing agent identification model can be used to guide de-icing agent spreading operations, optimize spreading strategies, improve snow melting efficiency and effectiveness, and ensure road safety under adverse weather conditions.

[0044] In one feasible implementation, step S40 may include: inputting the segmented road surface image into the de-icing agent recognition model to obtain the area of ​​each de-icing agent spreading area; obtaining the number of spreading particles in each de-icing agent spreading area in the segmented road surface image; determining the spreading density of the corresponding de-icing agent spreading area based on the area and the number of spreading particles of each de-icing agent spreading area; calculating the average distance between pairs of spreading particles in the corresponding de-icing agent spreading area based on the position information of the spreading particles in each de-icing agent spreading area in the segmented road surface image; and determining the spreading uniformity of the corresponding de-icing agent spreading area based on the average distance.

[0045] It should be noted that the area of ​​the de-icing agent spreading zone can be calculated by identifying and delineating the boundaries of each spreading zone in the road image. When determining the spreading density, the number of spreading particles is divided by the area of ​​the corresponding spreading zone to obtain the number of particles per unit area, i.e., the spreading density, as shown in the following formula:

[0046] in, For the de-icing agent application area R i The spreading density, For the de-icing agent application area R i The total number of particles in the feed, For the de-icing agent application area R i The area.

[0047] The calculation of spreading uniformity is based on the positional information of the spreading particles. The uniformity of particle distribution is assessed by calculating the average distance between pairs of spreading particles. A pair of spreading particles refers to a region R... i A pair of particles is formed by any two different particles, for example, particle j and particle k constitute a pair of particles. The smaller the average distance, the more uniform the particle distribution and the higher the uniformity of spreading. Conversely, the larger the average distance, the more uneven the particle distribution and the lower the uniformity of spreading. The formula for calculating the average distance between pairs of spreading particles is:

[0048] in, For the de-icing agent application area R i The average distance between the medium-sized particles. For the de-icing agent application area R i The total number of particles in the spread, To indicate the area R where the de-icing agent is applied. i The distance between the j-th and k-th particles. The attenuation factor indicates that as the average distance between the spread particles increases, the influence between the spread particle pairs gradually decreases.

[0049] Uniformity can be represented by the reciprocal of the average distance. The higher the uniformity, the more evenly the particles are distributed. The uniformity of the de-icing agent application area is: in, For the de-icing agent application area R i The uniformity of the spread For the de-icing agent application area R i The average distance between pairs of feed particles.

[0050] Step S50: Provide de-icing agent application prompts based on the application labels, application density, and application uniformity of each de-icing agent application area.

[0051] It's important to note that the application label indicates the presence of de-icing agent, specifically whether it has been applied to the imaged area of ​​the road surface. Combining the application label, application density, and application uniformity allows for a comprehensive assessment of de-icing agent application. For example, if the application label indicates the application of de-icing agent, and both the application density and uniformity meet preset standards, the application of de-icing agent in that area can be considered appropriate. Conversely, if the application density is too low or the application is uneven, it may be necessary to reapply de-icing agent to ensure effective snow removal and road safety.

[0052] In practice, de-icing agent application prompts can be presented to users in various ways. For example, a visual interface can display the distribution of application density and uniformity, or concise text can be used to indicate whether the application of de-icing agent in a given area is appropriate. These prompts help users understand the application status of the de-icing agent in a timely manner, enabling them to take appropriate measures to improve the efficiency and effectiveness of snow melting operations.

[0053] In one feasible implementation, step S50 may include: determining whether de-melting agent is currently present based on the spreading labels of each of the de-melting agent spreading areas; issuing a de-melting agent shortage warning when the current presence of de-melting agent is that de-melting agent is absent; comparing the spreading density of different de-melting agent spreading areas with a preset spreading density threshold when the current presence of de-melting agent is present to obtain a spreading density comparison result; comparing the spreading uniformity of different de-melting agent spreading areas with a preset spreading uniformity threshold to obtain a spreading uniformity comparison result; generating de-melting agent spreading prompt information based on the spreading density comparison result and the spreading uniformity comparison result; and sending the de-melting agent spreading prompt information to the user terminal to prompt the user to supplement the spreading of de-melting agent or adjust the spreading strategy.

[0054] It should be noted that the preset spreading density threshold and preset spreading uniformity threshold are standard values ​​set based on actual road conditions and the requirements for the use of de-icing agents. The comparison results of spreading density and spreading uniformity can intuitively reflect the difference between the current de-icing agent spreading situation and the standard values. For example, when the spreading density in a certain spreading area is lower than the preset spreading density threshold, it indicates that the de-icing agent spreading in that area is insufficient, and it may be necessary to increase the amount of de-icing agent spread. Similarly, when the spreading uniformity is lower than the preset spreading uniformity threshold, it indicates that the distribution of de-icing agent in that area is uneven, and it may be necessary to adjust the spreading strategy to improve the spreading uniformity.

[0055] It should be understood that the generation and transmission of de-icing agent spreading warning messages can help users promptly identify and correct problems during de-icing agent spreading, ensuring the effective use of de-icing agents and road safety. De-icing agent spreading warning messages can be sent to user terminals via various methods such as wireless networks, SMS, and email, allowing users to receive the messages anytime, anywhere. User terminals can be devices with network communication capabilities, such as smartphones, tablets, and laptops. These devices can run dedicated applications to receive and display de-icing agent spreading warning messages, or access relevant web pages through a browser to obtain this information.

[0056] It's worth noting that through the user terminal, users can intuitively understand the application status of each de-icing agent application area, including application density, uniformity, and whether there are any issues such as insufficient de-icing agent. Based on this information, users can take timely measures, such as increasing the amount of de-icing agent applied, adjusting the application strategy, or reapplying de-icing agent, to ensure effective snow removal and safe passage.

[0057] This embodiment provides an automatic monitoring method for de-icing agent spraying on highways. Road surface images are acquired along the de-icing agent spraying route, and these images are filtered and preprocessed to obtain preprocessed road surface images. The preprocessing includes at least grayscale image conversion, noise reduction, and contrast enhancement. The preprocessed road surface images are then segmented at the pixel level using a semantic segmentation model to obtain segmented road surface images. These segmented images include the de-icing agent spraying labels, sprayed particles, and their location information for each sprayed area. A de-icing agent recognition model is constructed. Based on this model, the segmented road surface images are recognized to obtain the spraying density and uniformity of different de-icing agent spraying areas. De-icing agent spraying prompts are provided based on the spraying labels, density, and uniformity of each sprayed area. By combining semantic segmentation and de-icing agent recognition models to perform pixel-level segmentation of road images and identify the de-icing agent spreading status, the above methods can accurately and efficiently identify the spreading density and uniformity of de-icing agents. This allows for timely understanding of the de-icing agent spraying situation and prompting for timely replenishment, thereby improving the efficiency of de-icing agent use, reducing resource waste, and ensuring safe road passage in winter.

[0058] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 Step S20 includes steps S201 to S204:

[0059] Step S201: Input the preprocessed road surface image into the semantic segmentation model, and use the semantic segmentation model to extract features and divide regions of the preprocessed road surface image to obtain multiple de-icing agent spreading areas.

[0060] It should be noted that semantic segmentation models can be built based on deep learning image segmentation algorithms and pre-trained using sample images of the corresponding target scene. Feature extraction refers to the semantic segmentation model's ability to distinguish different regions in a road surface image by extracting key features from the image, such as color, texture, and shape. Based on the extracted features, it can identify which areas are areas where de-icing agents have been applied and which are not.

[0061] It should be understood that, based on feature extraction, the semantic segmentation model further divides the road image into regions, that is, subdivides the image into multiple smaller regions, each corresponding to a possible de-icing agent spreading area. Through region segmentation, the model can simplify the complex road image into a series of more easily analyzed and processed units. The accuracy of these region segments directly affects the subsequent identification of de-icing agent spreading conditions. Therefore, when designing and training a semantic segmentation model, special attention needs to be paid to its feature extraction and region segmentation capabilities to ensure that the model can accurately identify and segment the de-icing agent spreading areas in the road image.

[0062] In one feasible implementation, step S201 may include: extracting features from the preprocessed road surface image using the semantic segmentation model to obtain a low-level feature map and a high-level feature map; averaging all pixel values ​​in the low-level and high-level feature maps to obtain multiple pixel average values; combining the multiple pixel average values ​​into a high-dimensional semantic representation; dividing the preprocessed road surface image into regions based on the high-dimensional semantic representation to obtain multiple de-icing agent spreading candidate regions; and using a non-maximum suppression algorithm to filter the multiple de-icing agent spreading candidate regions to obtain multiple de-icing agent spreading regions, wherein the non-maximum suppression algorithm is used to retain local maxima and suppress non-maxima to remove redundant spreading regions.

[0063] It should be noted that the low-level feature map semantic segmentation model performs multi-level feature extraction on the preprocessed road surface image, and the resulting feature maps at different levels reflect different information in the image. Low-level feature maps mainly contain detailed information such as edges and textures, while high-level feature maps reflect more of the overall structure and semantic information of the image. By combining the pixel averages from the low-level and high-level feature maps, a high-dimensional semantic representation can be obtained, which can more comprehensively reflect the content of the image.

[0064] Understandably, based on this high-dimensional semantic representation, the preprocessed road surface image can be divided into regions, initially identifying potential de-icing agent spreading areas, i.e., candidate de-icing agent spreading regions. However, these candidate regions may contain some redundant or overlapping areas, thus requiring further filtering. The non-maximum suppression algorithm, by retaining local maxima and suppressing non-maximums, can effectively remove redundant spreading regions, thereby obtaining more accurate de-icing agent spreading regions. These regions will serve as the basis for subsequent de-icing agent spreading identification.

[0065] Step S202: Assign a category to each pixel in the preprocessed road surface image to obtain the category information of each pixel.

[0066] In practice, the semantic segmentation model assigns a category label to each pixel based on the results of feature extraction and region partitioning. These category labels represent the type of different regions or objects in the road image. For example, some pixels may be labeled as de-icing agent pixels, while others may be labeled as non-de-icing agent pixels, such as roads, vehicles, and pedestrians. In this way, the model can classify each pixel in the road image into its corresponding category, providing a foundation for the accurate identification of subsequent de-icing agent application.

[0067] Step S203: Generate the spreading label, spreading particles, and position information of each snow melting agent spreading area based on the category information of each pixel.

[0068] It should be noted that the spreading label is an identifier used to identify the presence of de-icing agent in the spreading area, the spreading particles are the set of pixels in the road image that are identified as de-icing agent, and the location information of the spreading particles refers to the specific location of these pixels in the image.

[0069] In one feasible implementation, step S203 may include: statistically analyzing the category information of each pixel to obtain the number of pixels in each category and the positional distribution of pixels in each category in the preprocessed road surface image; determining the spreading label corresponding to each de-icing agent spreading area based on the number and positional distribution; taking the pixels corresponding to the de-icing agent in the category information as target pixels and obtaining the coordinates of the target pixels in the road surface image coordinate system; taking the target pixels with continuous coordinates as spreading particles and determining the positional information of each spreading particle.

[0070] It should be noted that the category information of each pixel is statistically analyzed to obtain the number of pixels in each category and their location distribution. Based on the number and location distribution of de-icing agent pixels in each area, the presence and density of de-icing agent application in that area are determined, thereby generating a corresponding application label. The application label can be a simple binary label indicating whether de-icing agent has been applied in that area, or it can be a more complex label containing the application density level or other relevant information.

[0071] It should be understood that by taking the pixels marked as de-icing agents in the category information as target pixels, and by obtaining the coordinates of these de-icing agent pixels in the road surface image coordinate system, it is possible to determine the de-icing agent pixels with continuous coordinates, and take the de-icing agent pixels with continuous coordinates as a spreading particle, thereby obtaining multiple spreading particles.

[0072] It is worth noting that determining the position information of the spreading particles can include: taking the center point of the spreading particles as the position of the spreading particles; or determining an outer polygon or circle as the outline of the spreading particles based on the coordinates of continuous de-icing agent pixels, and taking the center of the polygon or circle as the position of the spreading particles.

[0073] Step S204: Generate a segmented road surface image based on each of the de-icing agent spreading areas, the spreading labels of each of the de-icing agent spreading areas, the spreading particles, and the location information of the spreading particles.

[0074] It should be noted that by associating the spreader labels, spreader particles, and the location information of the spreader particles with the corresponding de-icing agent spread areas, segmented road surface images are generated. In these segmented images, each de-icing agent spread area is clearly identified and accompanied by detailed spread information. This information includes spreader labels for quickly identifying whether de-icing agent has been spread in the area; spreader particles showing the specific distribution of the de-icing agent on the road surface; and the location information of the spreader particles, providing precise location of the de-icing agent spread.

[0075] This embodiment uses a semantic segmentation model to extract features and divide regions from the preprocessed road surface image, which can accurately identify the de-icing agent spreading areas. Then, each pixel in the processed road surface image is assigned, which facilitates the accurate generation of spreading labels, spreading particles, and the location information of spreading particles for each de-icing agent spreading area. This generates a segmented road surface image, enabling the de-icing agent identification model to accurately identify the spreading density and uniformity of each de-icing agent spreading area, effectively improving the monitoring efficiency and accuracy of de-icing agent spreading.

[0076] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the automatic spraying monitoring method for de-icing agents on highways in this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0077] This application also provides an automatic monitoring device for snow-melting agent spraying on highways; please refer to [reference needed]. Figure 3 The automatic monitoring device for de-icing agent spraying on highways includes: The acquisition module 10 is used to acquire road surface images along the snow melting agent spraying route, and to filter and preprocess the road surface images to obtain preprocessed road surface images. The preprocessing includes at least grayscale image conversion, noise reduction, and contrast enhancement.

[0078] The segmentation module 20 is used to perform pixel-level segmentation on the preprocessed road surface image based on a semantic segmentation model to obtain a segmented road surface image, wherein the segmented road surface image includes the spreading labels, spreading particles, and position information of each de-icing agent spreading area.

[0079] Module 30 is used to build a snow melting agent identification model.

[0080] The identification module 40 is used to identify the segmented road surface image based on the snow melting agent identification model, and to obtain the spreading density and spreading uniformity of different snow melting agent spreading areas.

[0081] The prompt module 50 is used to provide prompts for the application of de-icing agent based on the application label, application density, and application uniformity of each de-icing agent application area.

[0082] The automatic highway de-icing agent spraying monitoring device provided in this application, employing the automatic highway de-icing agent spraying monitoring method described in the above embodiments, can solve the technical problem of existing technologies lacking real-time and effective monitoring means for road surface conditions after spraying, making it impossible to promptly understand the completion status of spraying work and provide supplementary reminders, thus affecting road traffic safety. Compared with the prior art, the beneficial effects of the automatic highway de-icing agent spraying monitoring device provided in this application are the same as those of the automatic highway de-icing agent spraying monitoring method provided in the above embodiments, and other technical features in the automatic highway de-icing agent spraying monitoring device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0083] In one embodiment, the segmentation module 20 is further configured to input the preprocessed road surface image into a semantic segmentation model, and perform feature extraction and region division on the preprocessed road surface image through the semantic segmentation model to obtain multiple de-icing agent spreading areas; assign a category to each pixel in the preprocessed road surface image to obtain category information of each pixel; generate spreading labels, spreading particles, and position information of spreading particles for each de-icing agent spreading area based on the category information of each pixel; and generate a segmented road surface image based on each de-icing agent spreading area, spreading labels, spreading particles, and position information of spreading particles.

[0084] In one embodiment, the segmentation module 20 is further configured to extract features from the preprocessed road surface image using the semantic segmentation model to obtain a low-level feature map and a high-level feature map; average all pixel values ​​in the low-level and high-level feature maps to obtain multiple pixel average values; combine the multiple pixel average values ​​into a high-dimensional semantic representation; divide the preprocessed road surface image into regions based on the high-dimensional semantic representation to obtain multiple de-icing agent spreading candidate regions; and use a non-maximum suppression algorithm to filter the multiple de-icing agent spreading candidate regions to obtain multiple de-icing agent spreading regions, wherein the non-maximum suppression algorithm is used to retain local maxima and suppress non-maxima to remove redundant spreading regions.

[0085] In one embodiment, the segmentation module 20 is further configured to perform statistics based on the category information of each pixel to obtain the number of pixels of each category and the positional distribution of pixels of each category in the preprocessed road surface image; determine the spreading label corresponding to each de-icing agent spreading area based on the number and positional distribution; take the pixel corresponding to the de-icing agent in the category information as the target pixel and obtain the coordinates of the target pixel in the road surface image coordinate system; take the target pixels with continuous coordinates as spreading particles and determine the positional information of each spreading particle.

[0086] In one embodiment, the construction module 30 is further configured to collect historical road surface image data and annotate the historical road surface image data to obtain annotated historical road surface image data, wherein the annotated historical road surface image data includes the true values ​​of the spreading density and spreading uniformity of different de-icing agent spreading areas; divide the historical road surface image data into a training set and a test set; input the training set into an initial recognition model for training to obtain a trained initial recognition model; input the test set into the initial recognition model for verification to obtain predicted values ​​of the spreading density and spreading uniformity of different de-icing agent spreading areas; determine the performance index of the initial recognition model based on the predicted and true values ​​of the spreading density and spreading uniformity of different de-icing agent spreading areas; when the performance index reaches the corresponding index threshold, use the trained initial recognition model as the de-icing agent recognition model; when the performance index does not reach the corresponding index threshold, adjust the trained initial recognition model and retrain and verify the adjusted initial recognition model as the initial recognition model until the performance index reaches the corresponding index threshold.

[0087] In one embodiment, the recognition module 40 is further configured to input the segmented road surface image into the de-icing agent recognition model to obtain the area of ​​each de-icing agent spreading area; Obtain the number of de-icing agent particles in each of the de-icing agent spreading areas in the segmented road surface image; The spreading density of the corresponding snow-melting agent spreading area is determined based on the area of ​​each spreading area and the number of spreading particles. The average distance between pairs of de-icing agent particles in each of the de-icing agent spreading areas in the segmented road surface image is calculated based on the position information of the particles in each of the de-icing agent spreading areas. The uniformity of the de-icing agent application area is determined based on the average distance. The formula for calculating the average distance between the pairs of spreading particles is as follows: in, For the de-icing agent application area R iThe average distance between the medium-sized particles. For the de-icing agent application area R i The total number of particles in the spread, To indicate the de-icing agent application area R i The distance between the j-th and k-th particles. The attenuation factor indicates that as the average distance between the spread particles increases, the influence between the spread particle pairs gradually decreases.

[0088] In one embodiment, the prompting module 50 is further configured to: determine whether de-melting agent is currently present based on the spreading labels of each de-melting agent spreading area; issue a de-melting agent shortage prompt when the current presence of de-melting agent is that de-melting agent is absent; when the current presence of de-melting agent is that de-melting agent is present, compare the spreading density of different de-melting agent spreading areas with a preset spreading density threshold to obtain a spreading density comparison result; compare the spreading uniformity of different de-melting agent spreading areas with a preset spreading uniformity threshold to obtain a spreading uniformity comparison result; generate de-melting agent spreading prompt information based on the spreading density comparison result and the spreading uniformity comparison result; and send the de-melting agent spreading prompt information to the user terminal to prompt the user to supplement the spreading of de-melting agent or adjust the spreading strategy.

[0089] This application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the automatic spraying monitoring method for de-icing agents on highways as described in Embodiment 1 above.

[0090] The following is for reference. Figure 4 The diagram illustrates a structural schematic of an electronic device suitable for implementing embodiments of this application. The electronic devices in these embodiments may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0091] like Figure 4As shown, the electronic device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM (Read Only Memory) 1002 or a program loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touch screens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, LCDs (Liquid Crystal Displays), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. While electronic devices with various systems are shown in the figures, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0092] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0093] The electronic device provided in this application, employing the automatic highway de-icing agent spraying monitoring method described in the above embodiments, solves the technical problem of existing technologies lacking real-time and effective monitoring methods for road surface conditions after spraying, making it impossible to promptly understand the completion status of spraying work and provide supplementary reminders, thus affecting road traffic safety. Compared with the prior art, the beneficial effects of the electronic device provided in this application are the same as those of the automatic highway de-icing agent spraying monitoring method provided in the above embodiments, and other technical features of this electronic device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0094] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0095] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0096] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the automatic spraying monitoring method for de-icing agents on highways in the above embodiments.

[0097] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory or Flash Memory), optical fibers, CD-ROM (CD-Read Only Memory), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0098] The aforementioned computer-readable storage medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

[0099] The aforementioned computer-readable storage medium carries one or more programs that, when executed by an electronic device, cause the electronic device to: acquire road surface images along the de-icing agent spraying route, and filter and preprocess the road surface images to obtain preprocessed road surface images, wherein the preprocessing includes at least grayscale image conversion, noise reduction, and contrast enhancement; perform pixel-level segmentation of the preprocessed road surface images based on a semantic segmentation model to obtain segmented road surface images, wherein the segmented road surface images include the spreading labels, spreading particles, and position information of each de-icing agent spreading area; construct a de-icing agent recognition model; identify the segmented road surface images based on the de-icing agent recognition model to obtain the spreading density and spreading uniformity of different de-icing agent spreading areas; and provide de-icing agent spreading prompts based on the spreading labels, spreading density, and spreading uniformity of each de-icing agent spreading area.

[0100] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including LAN (Local Area Network) or WAN (Wide Area Network)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0101] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0102] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0103] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., computer programs) for executing the above-described automatic spraying monitoring method for highway de-icing agents. This solves the technical problem in the prior art where there is a lack of real-time and effective monitoring methods for road surface conditions after spraying, making it impossible to promptly understand the completion status of spraying work and provide supplementary reminders, thus affecting road traffic safety. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the automatic spraying monitoring method for highway de-icing agents provided in the above embodiments, and will not be repeated here.

[0104] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for monitoring automatic spraying of de-icing agents on highways.

[0105] The computer program product provided in this application can solve the technical problem that existing technologies lack real-time and effective monitoring methods for road surface conditions after spraying, making it impossible to promptly understand the completion status of spraying work and provide supplementary reminders, thus affecting road traffic safety. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the automatic spraying monitoring method for highway de-icing agents provided in the above embodiments, and will not be repeated here.

[0106] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for monitoring automatic spraying of de-icing agents on highways, characterized in that, The method includes: Images of the road surface are collected along the de-icing agent spraying route, and the images are filtered and preprocessed to obtain preprocessed images of the road surface. The preprocessing includes at least grayscale image conversion, noise reduction, and contrast enhancement. The preprocessed road surface image is segmented at the pixel level based on the semantic segmentation model to obtain the segmented road surface image, wherein the segmented road surface image includes the spreading label, spreading particles, and position information of each de-icing agent spreading area. Construct a snow melting agent identification model; Based on the snow melting agent recognition model, the segmented road surface image is identified to obtain the spreading density and spreading uniformity of different snow melting agent spreading areas; The application of de-icing agent is indicated based on the application labels, application density, and application uniformity of each application area.

2. The method as described in claim 1, characterized in that, The step of performing pixel-level segmentation on the preprocessed road surface image based on a semantic segmentation model to obtain a segmented road surface image includes: The preprocessed road surface image is input into a semantic segmentation model, and the semantic segmentation model is used to extract features and divide regions of the preprocessed road surface image to obtain multiple de-icing agent spreading areas. Each pixel in the preprocessed road surface image is assigned a category to obtain the category information of each pixel. Based on the category information of each pixel, generate the spreading label, spreading particles, and position information of each snow melting agent spreading area; A segmented road surface image is generated based on the de-icing agent application areas, the application labels of the de-icing agent application areas, the application particles, and the location information of the application particles.

3. The method as described in claim 2, characterized in that, The semantic segmentation model is used to extract features and divide regions from the preprocessed road surface image to obtain multiple de-icing agent spreading areas, including: The semantic segmentation model is used to extract features from the preprocessed road surface image to obtain low-level feature maps and high-level feature maps. The pixel values ​​in the low-level feature map and the high-level feature map are averaged respectively to obtain multiple pixel average values; The average values ​​of multiple pixels are combined into a high-dimensional semantic representation; Based on the high-dimensional semantic representation, the preprocessed road surface image is divided into regions to obtain multiple candidate regions for snow melting agent application. The multiple candidate areas for de-icing agent application are screened using a non-maximum suppression algorithm to obtain multiple de-icing agent application areas. The non-maximum suppression algorithm is used to retain local maxima and suppress non-maxima to remove redundant application areas.

4. The method as described in claim 2, characterized in that, The generation of the spreading label, spreading particles, and position information of each snow-melting agent spreading area based on the category information of each pixel includes: Based on the category information of each pixel, the number of pixels in each category and the positional distribution of pixels in each category in the preprocessed road surface image are obtained; Based on the quantity and location distribution, determine the corresponding spreading labels for each de-icing agent spreading area; The pixel corresponding to the snow melting agent in the category information is taken as the target pixel, and the coordinates of the target pixel in the road surface image coordinate system are obtained; Target pixels with continuous coordinates are used as spreading particles, and the position information of each spreading particle is determined.

5. The method as described in claim 1, characterized in that, The construction of the snow melting agent identification model includes: Historical road surface image data is collected and labeled to obtain labeled historical road surface image data, wherein the labeled historical road surface image data includes the true values ​​of the spreading density and spreading uniformity of different de-icing agent spreading areas. The historical road surface image data is divided into a training set and a test set; The training set is input into the initial recognition model for training to obtain the trained initial recognition model; The test set is input into the initial recognition model for verification to obtain the predicted values ​​of the spreading density and spreading uniformity of different de-icing agent spreading areas; The performance indicators of the initial identification model are determined based on the predicted and actual values ​​of the spreading density and spreading uniformity in different de-icing agent spreading areas. When the performance index reaches the corresponding index threshold, the trained initial recognition model will be used as the de-icing agent recognition model. If the performance metric fails to reach the corresponding threshold, the trained initial recognition model is adjusted, and the adjusted initial recognition model is retrained and validated as the initial recognition model until the performance metric reaches the corresponding threshold.

6. The method as described in claim 1, characterized in that, The step of identifying the segmented road surface image based on the snow-melting agent identification model to obtain the spreading density and spreading uniformity of different snow-melting agent spreading areas includes: The segmented road surface image is input into the de-icing agent recognition model to obtain the area of ​​each de-icing agent spreading area; Obtain the number of de-icing agent particles in each of the de-icing agent spreading areas in the segmented road surface image; The spreading density of the corresponding snow-melting agent spreading area is determined based on the area of ​​each spreading area and the number of spreading particles. The average distance between pairs of de-icing agent particles in each of the de-icing agent spreading areas in the segmented road surface image is calculated based on the position information of the particles in each of the de-icing agent spreading areas. The uniformity of the de-icing agent application area is determined based on the average distance. The formula for calculating the average distance between the pairs of spreading particles is as follows: in, For the de-icing agent application area R i The average distance between the medium-sized particles. For the de-icing agent application area R i The total number of particles in the feed, To indicate the area R where the de-icing agent is applied. i The distance between the j-th and k-th particles. The attenuation factor indicates that as the average distance between the spread particles increases, the influence between the spread particle pairs gradually decreases.

7. The method as described in claim 1, characterized in that, The method of providing de-icing agent application guidance based on the application status, density, and uniformity of the de-icing agent includes: Determine whether the de-icing agent is currently present based on the application label of each of the aforementioned de-icing agent application areas; When the current status of the snow-melting agent is that the snow-melting agent is not present, a snow-melting agent shortage warning will be issued; When the current condition of the snow melting agent is that the snow melting agent is present, the spreading density of different snow melting agent spreading areas is compared with the preset spreading density threshold to obtain the spreading density comparison result; By comparing the uniformity of the de-icing agent application area with the preset uniformity threshold, the uniformity comparison results are obtained. Based on the comparison results of the spreading density and the spreading uniformity, a de-icing agent spreading prompt message is generated; The snow melting agent application prompt message is sent to the user terminal to prompt the user to supplement the snow melting agent application or adjust the application strategy.

8. An automatic monitoring device for de-icing agent spraying on highways, characterized in that, The automatic snow-melting agent spraying monitoring device for highways includes: The acquisition module is used to acquire road surface images along the snow melting agent spraying route, and to filter and preprocess the road surface images to obtain preprocessed road surface images. The preprocessing includes at least grayscale image conversion, noise reduction, and contrast enhancement. The segmentation module is used to perform pixel-level segmentation on the preprocessed road surface image based on a semantic segmentation model to obtain a segmented road surface image, wherein the segmented road surface image includes the spreading labels, spreading particles, and position information of each de-icing agent spreading area. Build modules are used to construct de-icing agent identification models; The identification module is used to identify the segmented road surface image based on the snow melting agent identification model, and to obtain the spreading density and spreading uniformity of different snow melting agent spreading areas; The prompt module is used to provide prompts for the application of de-icing agent based on the application label, application density, and application uniformity of each de-icing agent application area.

9. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and an automatic highway de-icing agent spraying monitoring program stored in the memory and executable on the processor, the automatic highway de-icing agent spraying monitoring program being configured to implement the automatic highway de-icing agent spraying monitoring method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores an automatic monitoring program for de-icing agents on highways, which, when executed by a processor, implements the automatic monitoring method for de-icing agents on highways as described in any one of claims 1 to 7.

Images