CN117011692A

CN117011692A - Road identification method and related device

Info

Publication number: CN117011692A
Application number: CN202211273778.0A
Authority: CN
Inventors: 李亚宁; 高树峰
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2022-10-18
Filing date: 2022-10-18
Publication date: 2023-11-07

Abstract

The embodiment of the application provides a road identification method and a related device, which can be applied to various scenes such as artificial intelligence, intelligent traffic, auxiliary driving, internet of vehicles and the like. According to the method, the traffic flow characteristics and the traffic speed characteristics can be taken into consideration on the basis of the remote sensing images, the difficulty of road type identification can be reduced, and the identification precision and recall rate are improved. The method at least relates to the technology of machine learning in artificial intelligence and the like. The method comprises the following steps: acquiring traffic flow characteristics and traffic speed characteristics of a road area to be identified; acquiring a remote sensing image of a road area to be identified; performing feature extraction processing on the remote sensing image to obtain road network features of the road area to be identified; and carrying out road identification processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics based on the target road identification model to obtain a target classification result, wherein the target classification result is used for indicating the road type of the road in the road area to be identified.

Description

Road identification method and related device

Technical Field

The embodiment of the application relates to the technical field of artificial intelligence, in particular to a road identification method and a related device.

Background

Road identification is an essential step in road network generation and road loss mining. Along with the development of deep learning, the perception capability of road identification is also greatly improved.

In the related scheme, road identification is mainly performed on the remote sensing image of the road through a deep learning model such as a segtormer model, so that the type of the road is determined. However, the road structure is complex and is easy to be blocked by objects such as factories, houses or trees, and the remote sensing image is identified by adopting the existing mode, only roads which are not blocked by the objects can be identified, and the road types in the area blocked by the objects have larger identification difficulty, so that the identification precision is lower and the identification recall rate is low.

Disclosure of Invention

The embodiment of the application provides a road identification method and a related device, which can reduce the difficulty of road type identification and improve the identification precision and recall rate.

In a first aspect, an embodiment of the present application provides a method for identifying a road. The method comprises the following steps: acquiring traffic flow characteristics and traffic speed characteristics of a road area to be identified, wherein the traffic flow characteristics are used for indicating traffic flow of vehicles passing through the road area to be identified, and the traffic speed characteristics are used for indicating running speed of the vehicles passing through the road area to be identified; acquiring a remote sensing image of the road area to be identified; performing feature extraction processing on the remote sensing image to obtain road network features of the road area to be identified; and carrying out road identification processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics based on a target road identification model to obtain a target classification result, wherein the target classification result is used for indicating the road type of the road in the road area to be identified.

In a second aspect, an embodiment of the present application provides a road identifying apparatus. The road identification device comprises an acquisition unit and a processing unit. The system comprises an acquisition unit, a traffic flow feature and a traffic speed feature, wherein the acquisition unit is used for acquiring traffic flow features of a road area to be identified, the traffic flow features are used for indicating traffic flow of vehicles passing through the road area to be identified, and the traffic speed features are used for indicating driving speeds of vehicles passing through the road area to be identified. And the acquisition unit is used for acquiring the remote sensing image of the road area to be identified. And the processing unit is used for carrying out feature extraction processing on the remote sensing image to obtain the road network features of the road area to be identified. The processing unit is used for carrying out road identification processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics based on a target road identification model to obtain a target classification result, wherein the target classification result is used for indicating the road type of the road in the road area to be identified.

In some alternative embodiments, the processing unit is configured to: traversing each track point on the road area to be identified, wherein each track point is used for indicating the positioning position of a vehicle passing through the road area to be identified; and carrying out accumulated summation processing on each track point to acquire the traffic flow characteristics of the road area to be identified.

In other alternative embodiments, the processing unit is configured to: traversing each track point on the road area to be identified, wherein each track point is used for indicating the positioning position of a vehicle passing through the road area to be identified; and calculating the passing speed characteristic based on the projection speeds of each track point in at least two directions of the road area to be identified.

In other alternative embodiments, the processing unit is configured to: calculating the projection speed of each track point in a first direction of the road area to be identified and the projection speed of each track point in a second direction of the road area to be identified, wherein the first direction is perpendicular to the second direction, and the first direction and the second direction are any two of the at least two directions; respectively carrying out average cumulative calculation on the projection speed of each track point in the first direction of the road area to be identified and the projection speed of each track point in the second direction of the road area to be identified to obtain a first projection speed and a second projection speed, wherein the first projection speed is used for indicating the passing speed component of the vehicle in the first direction, and the second projection speed is used for indicating the passing speed component of the vehicle in the second direction; and normalizing the first projection speed and the second projection speed to obtain the passing speed characteristics.

In other alternative embodiments, the acquisition unit is further configured to: and acquiring a traffic flow characteristic representation, a traffic speed characteristic representation and a road network characteristic representation of a training sample, wherein the training sample is a sample marked with a preset road type on a fusion characteristic obtained by fusing the traffic flow characteristic representation, the traffic speed characteristic representation and the road network characteristic representation, and the road network characteristic representation is obtained from a remote sensing image of the training sample. The processing unit is further used for taking the traffic flow characteristic representation, the traffic speed characteristic representation and the road network characteristic representation as the input of an initial road identification model to obtain a target road type; calculating the difference between the target road type and the preset road type to obtain a target loss value; and updating the model parameters of the initial road recognition model based on the target loss value to obtain the target road recognition model.

In other alternative embodiments, the processing unit is configured to: carrying out fusion processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics to obtain target fusion characteristics; and taking the target fusion characteristic as the input of the target road recognition model to obtain the target classification result.

In other alternative embodiments, the target road identification model includes a D-linkNet network. The processing unit is used for: inputting the target fusion characteristic into the D-linkNet network, and performing coding processing on the target fusion characteristic through an encoder sub-network in the D-linkNet network to obtain a first output characteristic; inputting the first output characteristics into a characteristic extraction sub-network in the D-linkNet network to perform characteristic extraction processing to obtain second output characteristics, wherein the characteristic extraction sub-network is composed of a cavity convolution and convolution block attention module; and inputting the first output characteristic and the second output characteristic into a decoder sub-network in the D-linkNet network so as to perform decoding processing through the decoder sub-network in the D-linkNet network, thereby obtaining the target classification result.

In other alternative embodiments, the processing unit is configured to: taking the traffic flow characteristic as a first channel layer, the traffic speed characteristic as a second channel layer and the road network characteristic as a third channel layer, wherein the first channel layer, the second channel layer and the third channel layer form an RGB channel; modeling the first channel layer, the second channel layer and the third channel layer to obtain target fusion characteristics.

In other alternative embodiments, the processing unit is configured to: carrying out road identification processing on the remote sensing image based on a preset identification model to obtain a road segmentation map of the road area to be identified; and taking the road segmentation map of the road area to be identified as the road network characteristics of the road area to be identified.

In other alternative embodiments, the remote sensing image comprises a multispectral remote sensing image, a hyperspectral remote sensing image, or a hyperspectral remote sensing image.

A third aspect of an embodiment of the present application provides a road identifying apparatus, including: memory, input/output (I/O) interfaces, and memory. The memory is used for storing program instructions. The processor is configured to execute the program instructions in the memory to perform the method for road identification according to the embodiment of the first aspect.

A fourth aspect of the embodiments of the present application provides a computer-readable storage medium having instructions stored therein, which when run on a computer, cause the computer to perform to execute the method corresponding to the embodiment of the first aspect described above.

A fifth aspect of the embodiments of the present application provides a computer program product comprising instructions which, when run on a computer or processor, cause the computer or processor to perform the method described above to perform the embodiment of the first aspect described above.

From the above technical solutions, the embodiment of the present application has the following advantages:

in the embodiment of the application, a remote sensing image of a road area to be identified is obtained, and the remote sensing image is subjected to characteristic extraction processing to obtain road network characteristics of the road area to be identified; the traffic flow characteristics can indicate the traffic flow of vehicles passing through the road area to be identified, and the traffic speed characteristics can indicate the running speed of vehicles passing through the road area to be identified, so that the traffic flow characteristics and the traffic speed characteristics of the road area to be identified can be obtained. And then, carrying out road identification processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics based on the target road identification model, and further obtaining a target classification result, so that the road type of the road area to be identified is indicated through the target classification result. By the method, the track data of the vehicle can be drawn due to the traffic flow characteristics and the traffic speed characteristics, so that the influence of other objects on the road when the road is identified based on the remote sensing image can be eliminated. Therefore, on the basis of the remote sensing image, the traffic flow characteristics and the traffic speed characteristics are taken into consideration, so that the road type of the road area to be identified, which is finally identified by the target road identification model, is closer to reality, the method can be applied to the area scene which is shielded by the object, and can also be applied to the area scene which is not shielded by the object, the defect caused by the fact that the road type is only characterized by the remote sensing image is overcome to a large extent, the difficulty of road type identification is reduced, and the identification precision and recall rate are improved.

Drawings

In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flowchart of a method for road identification according to an embodiment of the present application;

FIG. 2 shows a trace static flow diagram provided by an embodiment of the present application;

FIG. 3 illustrates a modeled density schematic provided by an embodiment of the present application;

FIG. 4 shows a model training flowchart of a target road recognition model provided by an embodiment of the present application;

FIG. 5 is a schematic flow chart of generating training samples according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of a target road recognition model according to an embodiment of the present application;

FIG. 7 shows a schematic diagram of the structure of a feature extraction sub-network in a D-linkNet network;

FIG. 8 is a first comparative schematic diagram of the recognition result of the present application and the recognition result of the prior art scheme;

FIG. 9 is a second comparative schematic diagram of the recognition result of the present application with the recognition result of the prior art scheme;

fig. 10 is a schematic structural diagram of a road recognition device according to an embodiment of the present application;

fig. 11 shows a schematic hardware structure of a road recognition device according to an embodiment of the present application.

Detailed Description

It will be appreciated that in the specific embodiments of the present application, related data such as user information, personal data of a user, etc. are involved, and when the above embodiments of the present application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data is required to comply with relevant laws and regulations and standards of relevant countries and regions.

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

The road identification method provided by the embodiment of the application is realized based on artificial intelligence (artificial intelligence, AI). Artificial intelligence is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and expand human intelligence, sense the environment, acquire knowledge and use the knowledge to obtain optimal results. In other words, artificial intelligence is an integrated technology of computer science that attempts to understand the essence of intelligence and to produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence, i.e. research on design principles and implementation methods of various intelligent machines, enables the machines to have functions of sensing, reasoning and decision.

The artificial intelligence technology is a comprehensive subject, and relates to the technology with wide fields, namely the technology with a hardware level and the technology with a software level. Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

In the embodiments of the present application, the artificial intelligence techniques mainly include the above-mentioned directions of machine learning and the like. For example, deep learning (ML) in machine learning (machine learning) may be involved, including convolutional neural networks and the like.

The road identification method provided by the application can be applied to a road identification device with data processing capability, such as a terminal device, a server and the like. The terminal device may include, but is not limited to, a smart phone, a desktop computer, a notebook computer, a tablet computer, a smart speaker, a vehicle-mounted device, a smart watch, and the like. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server or the like for providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (content delivery network, CDN), basic cloud computing services such as big data and artificial intelligent platforms, and the application is not limited in particular. In addition, the terminal device and the server may be directly connected or indirectly connected by wired communication or wireless communication, and the present application is not particularly limited.

The road identification device mentioned above may be provided with the capability to implement the computer vision technique mentioned above. The mentioned computer vision technology is a science of researching how to make the machine "look at", and further, a camera and a computer are used to replace human eyes to perform machine vision such as recognition, trace tracing and measurement on the target, and further perform graphic processing, so that the computer processing becomes an image more suitable for human eyes to observe or transmit instrument detection. As a scientific discipline, computer vision research-related theory and technology has attempted to build artificial intelligence systems that can acquire information from images or multidimensional data. Computer vision techniques typically include image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, video content/behavior recognition, three-dimensional object reconstruction, 3D techniques, virtual reality, augmented reality, synchronous positioning, and map construction, among others, as well as common biometric recognition techniques such as face recognition, fingerprint recognition, and others. In the embodiment of the application, the road recognition device can process remote sensing images and the like through the computer vision technology.

The road recognition device mentioned above may be provided with machine learning capabilities. Machine Learning (ML) is a multi-domain interdisciplinary, involving multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. It is specially studied how a computer simulates or implements learning behavior of a human to acquire new knowledge or skills, and reorganizes existing knowledge structures to continuously improve own performance. Machine learning is the core of artificial intelligence, a fundamental approach to letting computers have intelligence, which is applied throughout various areas of artificial intelligence. Machine learning and deep learning typically involve neural networks and the like.

The artificial intelligent model is adopted in the road identification method provided by the embodiment of the application, which mainly relates to application of a neural network, and the road type of a road area to be identified is identified through the neural network.

In the process of mining basic road data of an electronic map, accurate, rapid and stable recognition of road types is performed, and then road networks in corresponding ranges are generated as a popular research direction. The road in the track is accurately identified, the road and the background are automatically segmented, and the method has critical significance for the road shape and the road shortage excavation production line. In the real world, the track is complex and various, and different grades of roads have different width and shape characteristics, such as national roads, provincial roads, villages, mountain roads, suburban areas and the like. Meanwhile, the road area is influenced by a plurality of factors such as shielding of buildings, trees, road central green belts and the like, so that accurate extraction of road information is still a research front and technical difficulty in the track information extraction field. At present, in the related scheme, road identification is performed on a remote sensing image of a road mainly through a deep learning model such as a segtormer model, so as to determine the type of the road. However, the road structure is complex and is easy to be blocked by objects such as factories, houses or trees, and the remote sensing image is identified by adopting the existing mode, only roads which are not blocked by the objects can be identified, and the road types in the area blocked by the objects have larger identification difficulty, so that the identification precision is lower and the identification recall rate is low.

Based on the above, the embodiment of the application provides a road identification method. The road identification method can take the traffic flow characteristics and the traffic speed characteristics into consideration on the basis of the remote sensing image, so that the road type of the road area to be identified, which is finally identified by the target road identification model, is closer to reality, can be suitable for an area scene shielded by an object, can also be suitable for an area scene not shielded by the object, overcomes the defect brought by the fact that the road type is only depicted by the remote sensing image to a large extent, reduces the difficulty of road type identification, and improves the identification precision and recall rate. The road recognition method provided by the embodiment of the application can be applied to various scenes such as maps, intelligent traffic, automatic driving, artificial intelligence, auxiliary driving, internet of vehicles and the like.

It should be noted that the intelligent transportation system (Intelligent Traffic System, ITS) in the described intelligent transportation field is also called as intelligent transportation system (Intelligent Transportation System), and is a comprehensive transportation system that uses advanced scientific technologies (information technology, computer technology, data communication technology, sensor technology, electronic control technology, automatic control theory, operation study, artificial intelligence, etc.) effectively and comprehensively for transportation, service control and vehicle manufacturing, and enhances the connection among vehicles, roads and users, so as to form a comprehensive transportation system for guaranteeing safety, improving efficiency, improving environment and saving energy. Or alternatively;

The intelligent vehicle-road cooperative system (Intelligent Vehicle Infrastructure Cooperative Systems, IVICS), which is simply called a vehicle-road cooperative system, is one development direction of an Intelligent Transportation System (ITS). The vehicle-road cooperative system adopts advanced wireless communication, new generation internet and other technologies, carries out vehicle-vehicle and vehicle-road dynamic real-time information interaction in all directions, develops vehicle active safety control and road cooperative management on the basis of full-time idle dynamic traffic information acquisition and fusion, fully realizes effective cooperation of people and vehicles and roads, ensures traffic safety, improves traffic efficiency, and forms a safe, efficient and environment-friendly road traffic system.

The following describes a road recognition method according to an embodiment of the present application with reference to the accompanying drawings. Fig. 1 shows a flowchart of a method for road identification according to an embodiment of the present application. As shown in fig. 1, the method for identifying a road may include the steps of:

101. and acquiring traffic flow characteristics and traffic speed characteristics of the road area to be identified, wherein the traffic flow characteristics are used for indicating the traffic flow of vehicles passing through the road area to be identified, and the traffic speed characteristics are used for indicating the running speed of vehicles passing through the road area to be identified.

In this example, since the vehicle trajectory includes both spatially distributed features and implicit dynamic features, a static flow map of the trajectory may be constructed based on the vehicle trajectory over a period of time. The track static flow diagram comprises the traffic flow and speed of the track in the road area to be identified. The described spatial distribution features may include, but are not limited to, traffic flow features. The kinetic characteristics may include, but are not limited to, traffic speed characteristics, and the like. The traffic flow characteristics described can be used to indicate the traffic flow of vehicles passing over the road area to be identified, and the traffic speed characteristics can indicate the speed of travel of vehicles passing over the road area to be identified.

Illustratively, the described trajectory static flow map may be understood with reference to the schematic diagram shown in fig. 2. As shown in fig. 2, in the constructed trace static flow diagram, 3 channel layers, namely, R, G and B1 channel layers, are included.

The R-channel map layer can be understood as the actual flow value of the road area to be identified in the geospatial area. For example, the road recognition device can traverse each track point on the road area to be recognized, and perform cumulative summation processing on each track point, so as to obtain the traffic flow characteristics of the road area to be recognized. It should be noted that each of the described track points, sometimes also referred to as GPS points, can be used to indicate the location of vehicles traveling over the road area to be identified, such as p1 to p7 shown in fig. 2, etc. Therefore, after the traffic flow characteristics of the road area to be identified are obtained, the traffic flow characteristics can be regarded as an R-channel layer. In addition, the larger the value of the R channel layer (e.g., 0 to 255), the larger the traffic flow of the road area to be identified is, and the larger the red component is in visual effect.

The B1 channel map is understood to be the projected speed in the first direction of the actual travel speed of a vehicle travelling over the road area to be identified in the geospatial area. The G1 channel map is understood to be the projected speed in the second direction of the actual travel speed of the vehicle traveling over the road area to be identified in the geospatial area. The first direction is perpendicular to the second direction, for example, if the first direction is the north direction, the second direction may be the east direction, which is not particularly limited in the embodiment of the present application. In addition, the larger the value of the B1 channel layer (e.g., 0 to 255), the larger the component of the passing speed of the vehicle in the current position along the first direction is, and the blue component is represented as a larger proportion in visual effect. Similarly, the larger the value of the B1 channel layer (e.g., 0 to 255), the larger the component of the passing speed of the vehicle in the first direction at the current position is, and the blue component appears to be a larger proportion in visual effect. In addition, since the B1 channel layer and the G1 channel layer are used to represent the speed of the vehicle, the values of the B1 channel layer and the G1 channel layer may be normalized to be the final B channel layer.

In some examples, the road recognition device may traverse each track point on the road area to be recognized and calculate the traffic speed feature according to the projection speed of each track point in at least two directions of the road area to be recognized. For example, the projection speed of each track point in the first direction of the road area to be identified and the projection of each track point in the second direction of the road area to be identified are calculated. Then, the road recognition device respectively carries out average cumulative calculation on the projection speed of each track point in the first direction of the road area to be recognized and the projection speed of each track point in the second direction of the road area to be recognized so as to obtain a first projection speed and a second projection speed, wherein the first projection speed is used for indicating the passing speed component of the vehicle in the first direction, and the second projection speed is used for indicating the passing speed component of the vehicle in the second direction. And finally, the road recognition device normalizes the first projection speed and the second projection speed to acquire the traffic speed characteristics. It should be noted that after the traffic speed feature of the road area to be identified is obtained, the traffic speed feature may be regarded as a B-channel layer in the subsequent modeling process.

102. And acquiring a remote sensing image of the road area to be identified.

In this example, the remote sensing image may also be referred to as a remote sensing image, which is not limited in the embodiment of the present application. The remote sensing image of the road area to be identified may include a remote sensing image of each road in the road area to be identified.

The described remote sensing image may include, but is not limited to, multispectral remote sensing image, hyperspectral remote sensing image, low-spectrum remote sensing image, etc., and the embodiment of the application is not limited thereto.

103. And carrying out feature extraction processing on the remote sensing image to obtain road network features of the road area to be identified.

In this example, if the original color remote sensing image is directly selected as the road network feature of the road area to be identified, the remote sensing image itself has many other objects such as buildings, roads, farms, forests, rivers, etc., so that the road information in the remote sensing image is weakened, and larger noise data is generated for the labeling process. Therefore, after the remote sensing image of the road area to be identified is obtained, it is necessary to perform binarization processing on the remote sensing image, so as to extract the corresponding features. For example, after the remote sensing image of the road area to be identified is obtained, feature extraction processing is required to be performed on the remote sensing image, so as to extract road network features of the road area to be identified.

In some examples, the road recognition device performs feature extraction processing on the remote sensing image to obtain road network features of the road area to be recognized, which can be implemented by adopting a deep learning technology. The road recognition device may perform road recognition processing on the remote sensing image through a preset recognition model after the remote sensing image is acquired, so as to obtain a road segmentation map of the road area to be recognized. Then, the road recognition device can take the road segmentation map of the road area to be recognized as the road network feature of the road area to be recognized. The described road segmentation map can be understood as a segmentation map formed after road segmentation of all roads in the road area to be identified. In addition, the described preset recognition model may include, but is not limited to, a segvormer model, and the like, and the embodiment of the present application is not limited thereto.

The road recognition device may, for example, also consider the road network feature of the road area to be recognized as a G-channel layer after extracting the road network feature.

104. And carrying out road identification processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics based on the target road identification model to obtain a target classification result, wherein the target classification result is used for indicating the road type of the road in the road area to be identified.

In this example, after the traffic flow characteristic and the traffic speed characteristic of the road area to be identified are obtained and the corresponding road network characteristic is extracted, the road identification device may use the traffic flow characteristic, the traffic speed characteristic and the road network characteristic as the input of the target road model, and further perform the road identification processing on the traffic flow characteristic, the traffic speed characteristic and the road network characteristic through the target road model, so as to obtain a target classification result, where the target classification result can indicate the road type of the road in the road area to be identified.

The road recognition processing is performed on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics based on the target road recognition model, so as to obtain a target classification result, which can be achieved by the following modes: carrying out fusion processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics to obtain target fusion characteristics; and taking the target fusion characteristic as the input of a target road identification model to obtain a target classification result. For example, the traffic flow characteristic may be used as a first channel layer, the traffic speed characteristic may be used as a second channel layer, the road network characteristic may be used as a third channel layer, and then the first channel layer, the second channel layer and the third channel layer may be modeled to obtain the target fusion characteristic. It should be noted that the first channel layer may be understood as the R channel layer described above, the second channel layer may be understood as the B channel layer described above, and the third channel layer may be understood as the G channel layer. By three-channel modeling of the R channel layer (i.e. traffic flow characteristics), the B channel layer (traffic speed characteristics) and the G channel layer (i.e. road network characteristics), the traffic flow characteristics, the traffic speed characteristics and the road network characteristics can be better fused. For example, the density map obtained after modeling can be understood with reference to the schematic diagram depicted in fig. 3. As can be seen from fig. 3, the density map not only includes space-time information and speed information of the road, but also includes rich remote sensing space information.

In the application, the modeling description is performed by taking the first channel layer as the R channel layer, the second channel layer as the B channel layer and the third channel layer as the G channel layer as an example. In practical application, the first channel layer may also be a G channel layer or a B channel layer, etc., the second channel layer may also be an R channel layer or a G channel layer, etc., and the third channel layer may also be an R channel layer or a B channel layer, etc., in this embodiment, the first channel layer, the second channel layer, and the third channel layer may form an RGB channel layer, and in particular, what type of channel layer is not specifically limited in this application.

The above-mentioned target road recognition model, the training process of which can be understood with reference to the content of the training flow shown in fig. 4. As shown in fig. 4, the process of model training of the target road recognition model at least includes the following steps:

401. and obtaining a traffic flow characteristic representation, a traffic speed characteristic representation and a road network characteristic representation of a training sample, wherein the training sample is a sample marked with a preset road type on the fusion characteristic obtained after the traffic flow characteristic representation, the traffic speed characteristic representation and the road network characteristic representation are fused, and the road network characteristic representation is obtained from a remote sensing image of the training sample.

In this example, the flow of generating training samples may be understood with reference to the schematic depicted in FIG. 5. As shown in fig. 5, the road recognition apparatus may acquire a training sample, and acquire trajectory data and a remote sensing image included in the training sample. Then, the road recognition device can extract the traffic flow characteristic representation and the traffic speed characteristic representation by analyzing the track data. And the road recognition device takes the remote sensing image as the input of a model such as a segrormer model and the like, and further extracts and obtains the related road network characteristic representation. It should be understood that, in order to achieve a better training effect, the embodiment of the application selects road samples of special areas such as suburbs with sparse tracks and urban areas with dense tracks, partial intersections and the like as training samples. The road recognition device also needs to perform fusion processing on the traffic flow characteristic representation, the traffic speed characteristic representation and the road network characteristic representation, further performs manual marking on the fusion characteristics obtained by fusion, and adds a sample with a preset road type manually marked into a training set so as to obtain a complete training sample.

402. And taking the traffic flow characteristic representation, the traffic speed characteristic representation and the road network characteristic representation of the training sample as the input of the initial road identification model to obtain the target road type.

In this example, after the traffic flow characteristic representation, the traffic speed characteristic representation, and the road network characteristic representation are obtained, the traffic flow characteristic representation, the traffic speed characteristic representation, and the road network characteristic representation may be subjected to fusion processing to obtain fusion characteristics. And then, taking the fusion characteristic as the input of the initial road identification model, and identifying the corresponding target road type through the initial road identification model.

403. And calculating the difference between the target road type and the preset road type to obtain a target loss value.

In this example, since it is desirable that the output of the deep neural network (such as the initial road recognition model described above) is as close as possible to the truly desired value, the weight vector of each layer of the neural network can be updated by comparing the predicted value of the current network with the truly desired target value, and then based on the difference between the two (of course, there is typically an initialization process before the first update, i.e. pre-configuring parameters for each layer in the deep neural network), for example, if the predicted value of the network is higher, the weight vector is adjusted to make it predict lower, and the adjustment is continued until the neural network can predict the truly desired target value. Thus, it is necessary to define in advance "how to compare the difference between the predicted value and the target value", which is a loss function (loss function) or an objective function (objective function), which are important equations for measuring the difference between the predicted value and the target value. Taking the loss function as an example, the higher the output value (loss) of the loss function is, the larger the difference is, and then the training of the deep neural network becomes a process of reducing the loss as much as possible.

Therefore, the road recognition device needs to calculate the difference between the target road type and the marked preset road type after predicting the target road type, so as to obtain a corresponding target loss value.

404. And updating the model parameters of the initial road recognition model based on the target loss value to obtain a target road recognition model.

In this example, after the target loss value is calculated, the road recognition device can continuously adjust and update the model parameters of the initial road recognition model based on the target loss value until the initial road recognition model iteratively converges, so as to train and obtain the target road recognition model.

In this way, the road recognition device can take the traffic flow characteristics, the traffic speed characteristics and the road network characteristics as the input of the target road model after training to obtain the target road recognition model, and then perform the road recognition processing through the target road model, thereby obtaining the target classification result. The road recognition processing is performed on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics based on the target road recognition model, so as to obtain a target classification result, which can be achieved by the following modes: carrying out fusion processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics to obtain target fusion characteristics; and taking the target fusion characteristic as the input of a target road identification model to obtain a target classification result.

It should be noted that the described target road recognition model may include, but is not limited to, a D-linkNet network, etc., and the embodiments of the present application are not limited to the description.

Referring to fig. 6, a schematic structural diagram of a target road recognition model according to an embodiment of the present application is provided. As shown in fig. 6, taking a D-linkNet network as an example of the target road recognition model, the D-linkNet network includes an encoder sub-network, a feature extraction sub-network, and a decoder sub-network. The described encoder subnetworks can include, but are not limited to, a ResNet34 network, and the like. The road recognition device can input the target fusion characteristic into the D-link Net network after carrying out fusion processing on the traffic flow characteristic, the traffic speed characteristic and the road network characteristic to obtain the target fusion characteristic, so as to carry out coding processing on the target fusion characteristic through an encoder sub-network in the D-link Net network to obtain a first output characteristic. Illustratively, the encoder subnetwork first convolves the target fusion feature using a convolution layer with a convolution kernel size of 7×7, a filter number of 64, and a downsampling step size of 2, and then pooling the results of the convolution. Wherein the pooling window is 3×3, and the downsampling step size is 2. And then, inputting the processed characteristics into four coding units containing residual blocks for processing, and further obtaining first output characteristics. Wherein, the Res-blocks included in each coding unit have the block numbers of 3, 4, 6 and 3 respectively.

Then, the road recognition device can perform feature extraction processing on the first output features through a feature extraction sub-network in the D-linkNet network to obtain second output features, wherein the feature extraction sub-network is composed of an expansion convolution and a resident attention module. Illustratively, FIG. 7 shows a schematic diagram of the structure of a feature extraction sub-network in a D-linkNet network. As shown in fig. 7, the feature extraction sub-network may be formed by a hole convolution and convolution block attention module, forming a parallel network structure. Through the cavity convolution layer, the spatial resolution of the features can be kept unchanged, the feeling range of the features of the central part of the network is increased, and detailed information is kept. It should be noted that the hole convolution layer in the feature extraction sub-network used in the D-linkNet network has skip-connection. As shown in fig. 7, the feature extraction sub-network is divided into 5 extraction units, and the following operations are performed, respectively: (1) the first output characteristics input into the characteristic extraction sub-network are directly processed and output; (2) the first output characteristics input into the characteristic extraction sub-network are processed by the hole convolution operation with the expansion rate of 1 and then are processed and output; (3) the first output characteristics input into the characteristic extraction sub-network are sequentially processed through hole convolution operations with expansion rates of 1 and 2 and then processed and output; (4) the first output characteristics input into the characteristic extraction sub-network are processed by hole convolution operation with expansion rates of 1, 2 and 4 and then are processed and output; (5) the first output characteristics input into the characteristic extraction sub-network are processed by hole convolution operations with expansion rates of 1, 2, 4 and 8 and then are processed and output. The processing processes of the five extraction units are connected in parallel, and after the respective outputs are added at the tail end of the characteristic extraction sub-network, the output is used as the integral output of the characteristic extraction sub-network, so as to obtain a second output characteristic.

And finally, after the first output characteristic and the second output characteristic are obtained, the road identification device inputs the first output characteristic and the second output characteristic into a decoder sub-network in the D-link Net network so as to perform decoding processing through the decoder sub-network in the D-link Net network, and a target classification result is obtained.

The above description uses the D-link net network as the target road recognition model as an example, and the embodiment of the present application is not limited specifically.

For example, fig. 8 and fig. 9 each show a comparison diagram of the recognition result of the present application and the recognition result of the conventional scheme. As shown in fig. 8 and 9, compared with the result obtained by road recognition based on only the remote sensing image in the conventional scheme, it can be found that: on the basis of remote sensing image recognition, after the track features (namely the traffic flow features and the traffic speed features) are added, the road is predicted by adopting the target road recognition model through the adaptation and training of the target road recognition model, so that the problem that the corresponding road cannot be recognized due to dense vegetation, objects such as factories and clouds covering the road can be solved greatly, and meanwhile, the problem that the corresponding road cannot be recognized due to the fact that the track is sparse is solved.

The foregoing description of the solution provided by the embodiments of the present application has been mainly presented in terms of a method. It should be understood that, in order to implement the above-described functions, hardware structures and/or software modules corresponding to the respective functions are included. Those of skill in the art will readily appreciate that the various illustrative modules and algorithm steps described in connection with the embodiments disclosed herein may be implemented as hardware or combinations of hardware and computer software. Whether a function is implemented as hardware or computer software driven hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

The embodiment of the application can divide the functional modules of the device according to the method example, for example, each functional module can be divided corresponding to each function, and two or more functions can be integrated in one processing module. The integrated modules may be implemented in hardware or in software functional modules. It should be noted that, in the embodiment of the present application, the division of the modules is schematic, which is merely a logic function division, and other division manners may be implemented in actual implementation.

The following describes the road recognition device in the embodiment of the present application in detail, and fig. 10 is a schematic diagram of an embodiment of the road recognition device provided in the embodiment of the present application. As shown in fig. 10, the road recognition apparatus may include an acquisition unit 1001 and a processing unit 1002.

The obtaining unit 1001 is configured to obtain a traffic flow characteristic of a road area to be identified, where the traffic flow characteristic is used to indicate a traffic flow of a vehicle passing over the road area to be identified, and a traffic speed characteristic is used to indicate a running speed of the vehicle passing over the road area to be identified. An obtaining unit 1001 is configured to obtain a remote sensing image of the road area to be identified. And a processing unit 1002, configured to perform feature extraction processing on the remote sensing image, so as to obtain a road network feature of the road area to be identified. The processing unit 1002 is configured to perform a road recognition process on the traffic flow feature, the traffic speed feature, and the road network feature based on a target road recognition model, to obtain a target classification result, where the target classification result is used to indicate a road type of a road in the to-be-recognized road area.

In some alternative embodiments, the processing unit 1002 is configured to: traversing each track point on the road area to be identified, wherein each track point is used for indicating the positioning position of a vehicle passing through the road area to be identified; and carrying out accumulated summation processing on each track point to acquire the traffic flow characteristics of the road area to be identified.

In other alternative embodiments, the processing unit 1002 is configured to: traversing each track point on the road area to be identified, wherein each track point is used for indicating the positioning position of a vehicle passing through the road area to be identified; and calculating the passing speed characteristic based on the projection speeds of each track point in at least two directions of the road area to be identified.

In other alternative embodiments, the processing unit 1002 is configured to: calculating the projection speed of each track point in a first direction of the road area to be identified and the projection speed of each track point in a second direction of the road area to be identified, wherein the first direction is perpendicular to the second direction, and the first direction and the second direction are any two of the at least two directions; respectively carrying out average cumulative calculation on the projection speed of each track point in the first direction of the road area to be identified and the projection speed of each track point in the second direction of the road area to be identified to obtain a first projection speed and a second projection speed, wherein the first projection speed is used for indicating the passing speed component of the vehicle in the first direction, and the second projection speed is used for indicating the passing speed component of the vehicle in the second direction; and normalizing the first projection speed and the second projection speed to obtain the passing speed characteristics.

In other alternative embodiments, the obtaining unit 1001 is further configured to: and acquiring a traffic flow characteristic representation, a traffic speed characteristic representation and a road network characteristic representation of a training sample, wherein the training sample is a sample marked with a preset road type on a fusion characteristic obtained by fusing the traffic flow characteristic representation, the traffic speed characteristic representation and the road network characteristic representation, and the road network characteristic representation is obtained from a remote sensing image of the training sample. The processing unit 1002 is further configured to obtain a target road type by using the traffic flow characteristic representation, the traffic speed characteristic representation, and the road network characteristic representation as inputs of an initial road identification model; calculating the difference between the target road type and the preset road type to obtain a target loss value; and updating the model parameters of the initial road recognition model based on the target loss value to obtain the target road recognition model.

In other alternative embodiments, the processing unit 1002 is configured to: carrying out fusion processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics to obtain target fusion characteristics; and taking the target fusion characteristic as the input of the target road recognition model to obtain the target classification result.

In other alternative embodiments, the target road identification model includes a D-linkNet network. The processing unit 1002 is configured to: inputting the target fusion characteristic into the D-linkNet network, and performing coding processing on the target fusion characteristic through an encoder sub-network in the D-linkNet network to obtain a first output characteristic; inputting the first output characteristics into a characteristic extraction sub-network in the D-linkNet network to perform characteristic extraction processing to obtain second output characteristics, wherein the characteristic extraction sub-network is composed of a cavity convolution and convolution block attention module; and inputting the first output characteristic and the second output characteristic into a decoder sub-network in the D-linkNet network so as to perform decoding processing through the decoder sub-network in the D-linkNet network, thereby obtaining the target classification result.

In other alternative embodiments, the processing unit 1002 is configured to: taking the traffic flow characteristic as a first channel layer, the traffic speed characteristic as a second channel layer and the road network characteristic as a third channel layer, wherein the first channel layer, the second channel layer and the third channel layer form an RGB channel; modeling the first channel layer, the second channel layer and the third channel layer to obtain target fusion characteristics.

In other alternative embodiments, the processing unit 1002 is configured to: carrying out road identification processing on the remote sensing image based on a preset identification model to obtain a road segmentation map of the road area to be identified; and taking the road segmentation map of the road area to be identified as the road network characteristics of the road area to be identified.

The road identifying device in the embodiment of the present application is described above from the point of view of the modularized functional entity, and the road identifying device in the embodiment of the present application is described below from the point of view of hardware processing. The described road identification means may be a terminal device, a server or the like. Fig. 11 is a schematic structural diagram of a road recognition device according to an embodiment of the present application. The road recognition device can generate relatively large differences due to different configurations or performances. The road identification means may comprise at least one processor 1101, a communication line 1107, a memory 1103 and at least one communication interface 1104.

The processor 1101 may be a general purpose central processing unit (central processing unit, CPU), microprocessor, application specific integrated circuit (server IC), or one or more integrated circuits for controlling the execution of the program of the present application.

Communication line 1107 may include a pathway to transfer information between the aforementioned components.

Communication interface 1104 uses any transceiver-like device for communicating with other devices or communication networks, such as ethernet, radio access network (radio access network, RAN), wireless local area network (wireless local area networks, WLAN), etc.

The memory 1103 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (random access memory, RAM) or other type of dynamic storage device that can store information and instructions, and the memory may be stand alone and be coupled to the processor via a communication line 1107. The memory may also be integrated with the processor.

The memory 1103 is used for storing computer-executable instructions for executing the present application, and is controlled by the processor 1101. The processor 1101 is configured to execute computer-executable instructions stored in the memory 1103, thereby implementing the method provided by the above-described embodiment of the present application.

Alternatively, the computer-executable instructions in the embodiments of the present application may be referred to as application program codes, which are not particularly limited in the embodiments of the present application.

In a specific implementation, as an embodiment, the road identifying means may comprise a plurality of processors, such as processor 1101 and processor 1102 in fig. 11. Each of these processors may be a single-core (single-CPU) processor or may be a multi-core (multi-CPU) processor. A processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (e.g., computer program instructions).

In a specific implementation, as an embodiment, the road identifying apparatus may further include an output device 1105 and an input device 1106. The output device 1105 communicates with the processor 1101 and may display information in a variety of ways. The input device 1106 is in communication with the processor 1101 and may receive input of a target object in a variety of ways. For example, the input device 1106 may be a mouse, a touch screen device, a sensing device, or the like.

The road identifying device may be a general-purpose device or a special-purpose device. In a specific implementation, the road identifying device may be a server, a terminal device, or the like, or a device having a similar structure as in fig. 11. The embodiment of the application is not limited to the type of the road identification device.

Note that the processor 1101 in fig. 11 may cause the road identifying means to execute the method in the method embodiment corresponding to fig. 1 and 4 by calling the computer-executable instructions stored in the memory 1103.

In particular, the functions/implementations of the processing unit 1002 in fig. 10 may be implemented by the processor 1101 in fig. 11 invoking computer executable instructions stored in the memory 1103. The functions/implementation of the acquisition unit 1001 in fig. 10 can be implemented by the communication interface 1104 in fig. 11.

The embodiment of the present application also provides a computer storage medium storing a computer program for electronic data exchange, the computer program causing a computer to execute part or all of the steps of any one of the methods for identifying a road as described in the above method embodiments.

Embodiments of the present application also provide a computer program product comprising a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any one of the road identification methods described in the method embodiments above.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product.

It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.

In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of elements is merely a logical functional division, and there may be additional divisions of actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The above-described embodiments may be implemented in whole or in part by software, hardware, firmware, or any combination thereof, and when implemented in software, may be implemented in whole or in part in the form of a computer program product.

The computer program product includes one or more computer instructions. When the computer-executable instructions are loaded and executed on a computer, the processes or functions in accordance with embodiments of the present application are fully or partially produced. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). Computer readable storage media can be any available media that can be stored by a computer or data storage devices such as servers, data centers, etc. that contain an integration of one or more available media. Usable media may be magnetic media (e.g., floppy disks, hard disks, magnetic tape), optical media (e.g., DVD), or semiconductor media (e.g., SSD)), or the like.

The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of road identification, comprising:

acquiring traffic flow characteristics and traffic speed characteristics of a road area to be identified, wherein the traffic flow characteristics are used for indicating traffic flow of vehicles passing through the road area to be identified, and the traffic speed characteristics are used for indicating running speed of the vehicles passing through the road area to be identified;

acquiring a remote sensing image of the road area to be identified;

performing feature extraction processing on the remote sensing image to obtain road network features of the road area to be identified;

and carrying out road identification processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics based on a target road identification model to obtain a target classification result, wherein the target classification result is used for indicating the road type of the road in the road area to be identified.

2. The method of claim 1, wherein the obtaining traffic flow characteristics of the road area to be identified comprises:

traversing each track point on the road area to be identified, wherein each track point is used for indicating the positioning position of a vehicle passing through the road area to be identified;

and carrying out accumulated summation processing on each track point to acquire the traffic flow characteristics of the road area to be identified.

3. The method according to claim 1 or 2, wherein the acquiring traffic speed features comprises:

and calculating the passing speed characteristic based on the projection speeds of each track point in at least two directions of the road area to be identified.

4. A method according to claim 3, wherein said calculating the traffic speed characteristics based on the projected speeds of each of the track points in at least two directions of the road area to be identified comprises:

calculating the projection speed of each track point in a first direction of the road area to be identified and the projection speed of each track point in a second direction of the road area to be identified, wherein the first direction is perpendicular to the second direction, and the first direction and the second direction are any two of the at least two directions;

Respectively carrying out average cumulative calculation on the projection speed of each track point in the first direction of the road area to be identified and the projection speed of each track point in the second direction of the road area to be identified to obtain a first projection speed and a second projection speed, wherein the first projection speed is used for indicating the passing speed component of the vehicle in the first direction, and the second projection speed is used for indicating the passing speed component of the vehicle in the second direction;

and normalizing the first projection speed and the second projection speed to obtain the passing speed characteristics.

5. The method according to claim 1 or 2, characterized in that the method further comprises:

acquiring a traffic flow characteristic representation, a traffic speed characteristic representation and a road network characteristic representation of a training sample, wherein the training sample is a sample marked with a preset road type on a fusion characteristic obtained by fusing the traffic flow characteristic representation, the traffic speed characteristic representation and the road network characteristic representation, and the road network characteristic representation is obtained by remote sensing images of the training sample;

The traffic flow characteristic representation, the traffic speed characteristic representation and the road network characteristic representation are used as the input of an initial road identification model to obtain a target road type;

calculating the difference between the target road type and the preset road type to obtain a target loss value;

and updating the model parameters of the initial road recognition model based on the target loss value to obtain the target road recognition model.

6. The method according to claim 1 or 2, wherein the performing the road identification process on the traffic flow characteristic, the traffic speed characteristic, and the road network characteristic based on the target road identification model to obtain a target classification result includes:

carrying out fusion processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics to obtain target fusion characteristics;

and taking the target fusion characteristic as the input of the target road recognition model to obtain the target classification result.

7. The method of claim 6, wherein the target road recognition model comprises a D-linkNet network, wherein the obtaining the target classification result using the target fusion feature as an input to the target road recognition model comprises:

Inputting the target fusion characteristic into the D-linkNet network, and performing coding processing on the target fusion characteristic through an encoder sub-network in the D-linkNet network to obtain a first output characteristic;

inputting the first output characteristics into a characteristic extraction sub-network in the D-linkNet network to perform characteristic extraction processing to obtain second output characteristics, wherein the characteristic extraction sub-network is composed of a cavity convolution and convolution block attention module;

and inputting the first output characteristic and the second output characteristic into a decoder sub-network in the D-linkNet network so as to perform decoding processing through the decoder sub-network in the D-linkNet network, thereby obtaining the target classification result.

8. The method of claim 6, wherein the fusing the traffic flow feature, the traffic speed feature, and the road network feature to obtain a target fused feature comprises:

taking the traffic flow characteristic as a first channel layer, the traffic speed characteristic as a second channel layer and the road network characteristic as a third channel layer, wherein the first channel layer, the second channel layer and the third channel layer form an RGB channel;

Modeling the first channel layer, the second channel layer and the third channel layer to obtain target fusion characteristics.

9. The method according to claim 1 or 2, wherein the performing feature extraction processing on the remote sensing image to obtain road network features of the road area to be identified includes:

carrying out road identification processing on the remote sensing image based on a preset identification model to obtain a road segmentation map of the road area to be identified;

and taking the road segmentation map of the road area to be identified as the road network characteristics of the road area to be identified.

10. The method of claim 1 or 2, wherein the remote sensing image comprises a multispectral remote sensing image, a hyperspectral remote sensing image, or a hyperspectral remote sensing image.

11. A road identification device, characterized by comprising:

the system comprises an acquisition unit, a control unit and a control unit, wherein the acquisition unit is used for acquiring traffic flow characteristics and traffic speed characteristics of a road area to be identified, the traffic flow characteristics are used for indicating traffic flow of vehicles passing through the road area to be identified, and the traffic speed characteristics are used for indicating running speeds of vehicles passing through the road area to be identified;

The acquisition unit is used for acquiring the remote sensing image of the road area to be identified;

the processing unit is used for carrying out feature extraction processing on the remote sensing image to obtain road network features of the road area to be identified;

the processing unit is used for carrying out road identification processing on the traffic flow characteristics, the traffic speed characteristics and the road network characteristics based on a target road identification model to obtain a target classification result, wherein the target classification result is used for indicating the road type of the road in the road area to be identified.

12. The road identification device of claim 11, wherein the processing unit is configured to:

13. A road identification device, characterized in that the road identification device comprises: an input/output (I/O) interface, a processor, and a memory, the memory having program instructions stored therein;

The processor is configured to execute program instructions stored in a memory and to perform the method of any one of claims 1 to 10.

14. A computer readable storage medium comprising instructions which, when run on a computer device, cause the computer device to perform the method of any of claims 1 to 10.

15. A computer program product, characterized in that the computer program product comprises instructions which, when run on a computer device or a processor, cause the computer device or the processor to perform the method of any of claims 1 to 10.