A road recognition method and device, electronic equipment and storage medium

Abstract

The application relates to the technical field of artificial intelligence, in particular to a road recognition method and device, an electronic device and a storage medium, which can be applied to a vehicle-mounted scene, and based on candidate through traffic flow and deviated traffic flow corresponding to each historical time period of a to-be-recognized road, a deviated traffic parameter of the to-be-recognized road in a corresponding historical time period is determined, based on candidate traffic speed and candidate through traffic flow corresponding to each historical time period of the to-be-recognized road, a traffic impedance parameter of the to-be-recognized road is determined, and based on each traffic deviated parameter and the traffic impedance parameter, a traffic difficulty category of the to-be-recognized road is obtained. In this way, the road is recognized through the candidate through traffic flow, the deviated traffic parameter and the candidate traffic speed of the to-be-recognized road, so that the efficiency and accuracy of road recognition can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and particularly relates to a road recognition method and device, an electronic device and a storage medium. BACKGROUND

[0002] At present, with the development of navigation technology, more and more navigation applications have emerged, which greatly facilitates daily travel activities. In the process of planning a navigation road, if the road has the problems of narrow width and uneven paving, it will bring poor driving experience, therefore, it is necessary to determine the traffic difficulty category of the road.

[0003] In the related art, when determining the traffic difficulty category of a road, a neural network model is usually used to analyze an image containing the road, so as to determine the traffic difficulty category of the road.

[0004] However, since the width and paving conditions of the road are updated at a relatively fast speed, and the image samples of the road with high traffic difficulty used for training the neural network model are difficult to obtain, the recognition accuracy and efficiency of the road with high traffic difficulty are not high through the method in the related art. SUMMARY

[0005] The embodiments of the present application provide a road recognition method, device, electronic device and storage medium to improve the recognition accuracy and efficiency of the road with high traffic difficulty.

[0006] The specific technical solutions provided by the embodiments of the present application are as follows:

[0007] The embodiments of the present application provide a road recognition method, comprising:

[0008] Based on the candidate through traffic flow and the deviation traffic flow of the to-be-recognized road in each historical time period, a deviation traffic parameter of the to-be-recognized road in the corresponding historical time period is determined, wherein the deviation traffic parameter represents the proportion of objects that do not drive through the to-be-recognized road;

[0009] Based on the candidate traffic speed and the candidate through traffic flow of the to-be-recognized road in each historical time period, a traffic impedance parameter of the to-be-recognized road is determined, wherein the traffic impedance parameter represents the influence degree of the change mode of the candidate through traffic flow on the change mode of the candidate traffic speed;

[0010] Based on each traffic deviation parameter and the traffic impedance parameter, a traffic difficulty category of the to-be-recognized road is obtained.

[0011] Optionally, the determining the traffic impedance parameter of the to-be-identified road based on the candidate traffic speeds and the candidate throughout traffic flows of the to-be-identified road in the respective historical time periods comprises:

[0012] determining a target traffic speed satisfying a speed value condition and a free throughout traffic flow corresponding to the target traffic speed from the candidate traffic speeds of the to-be-identified road in the respective historical time periods;

[0013] determining a target throughout traffic flow satisfying a traffic value condition and a saturated traffic speed corresponding to the target throughout traffic flow from the candidate throughout traffic flows of the to-be-identified road in the respective historical time periods;

[0014] determining the traffic impedance parameter of the to-be-identified road based on the target traffic speed, the free throughout traffic flow, the target throughout traffic flow and the saturated traffic speed.

[0015] Optionally, the determining the traffic impedance parameter of the to-be-identified road based on the target traffic speed, the free throughout traffic flow, the target throughout traffic flow and the saturated traffic speed comprises:

[0016] determining a first speed difference between the target traffic speed and the saturated traffic speed, and determining a first traffic difference between the free throughout traffic flow and the target throughout traffic flow;

[0017] obtaining the traffic impedance parameter of the to-be-identified road based on a ratio between the first speed difference and the first traffic difference.

[0018] Optionally, before the obtaining the traffic difficulty category of the to-be-identified road based on the traffic deviation parameter and the traffic impedance parameter, the method further comprises:

[0019] obtaining a to-be-processed traffic speed and a to-be-processed throughout traffic flow corresponding to each target time period associated with the respective historical time periods;

[0020] determining a smooth impedance parameter of the to-be-identified road based on the free throughout traffic flow, the saturated traffic speed, the to-be-processed traffic speed and the to-be-processed throughout traffic flow.

[0021] Optionally, the determining the smooth impedance parameter of the to-be-identified road based on the free throughout traffic flow, the saturated traffic speed, the to-be-processed traffic speed and the to-be-processed throughout traffic flow comprises:

[0022] determining an average traffic speed between the to-be-processed traffic speeds greater than the saturated traffic speed from the to-be-processed traffic speeds;

[0023] determining an average through traffic flow between each of the to-be-processed through traffic speeds and each of the to-be-processed through traffic flows corresponding to the to-be-processed through traffic speeds;

[0024] determining a smooth impedance parameter of the to-be-identified road based on the average through traffic speed, the average through traffic flow, the saturated through traffic speed and the target through traffic flow.

[0025] Optionally, the determining of the smooth impedance parameter of the to-be-identified road based on the average through traffic speed, the average through traffic flow, the saturated through traffic speed and the target through traffic flow comprises:

[0026] determining a second speed difference between the average through traffic speed and the saturated through traffic speed, and determining a second flow difference between the average through traffic flow and the target through traffic flow;

[0027] obtaining the smooth impedance parameter of the to-be-identified road based on a ratio between the second speed difference and the second flow difference.

[0028] Optionally, the determining of the deviation traffic parameter of the to-be-identified road in each historical time period based on the candidate through traffic flow and the deviation traffic flow corresponding to each historical time period of the to-be-identified road comprises:

[0029] for each historical time period, the following operations are performed respectively:

[0030] obtaining a total through traffic flow corresponding to each traffic direction in the to-be-identified road in a historical time period based on a sum of the candidate through traffic flow and the candidate deviation traffic flow corresponding to each traffic direction in the to-be-identified road in the historical time period;

[0031] determining a candidate traffic parameter corresponding to each traffic direction in the to-be-identified road in the historical time period based on a ratio between the candidate deviation traffic flow and the total through traffic flow corresponding to each traffic direction;

[0032] determining a target traffic parameter meeting a preset parameter condition from the candidate traffic parameters as the deviation traffic parameter of the historical time period.

[0033] Optionally, the obtaining of the traffic difficulty category of the to-be-identified road based on the traffic deviation parameters and the traffic impedance parameter comprises:

[0034] obtaining a road feature set of each other road associated with the to-be-identified road;

[0035] determining the traffic difficulty category of the to-be-identified road based on the road feature set, the traffic deviation parameters, the traffic impedance parameter and the smooth impedance parameter.

[0036] Optionally, the determining of the traffic difficulty category of the to-be-identified road based on the feature set, the respective traffic deviation parameters and the traffic impedance parameter comprises:

[0037] Based on the trained road identification model, the respective candidate traffic speeds, the respective traffic deviation parameters, the traffic impedance parameter and the feature set are taken as input parameters to determine the traffic difficulty category corresponding to the to-be-identified road.

[0038] The embodiment of the present application also provides a road identification device, comprising:

[0039] The first processing module is configured to determine, based on the respective candidate through traffic flow and the deviation traffic flow of the to-be-identified road in each historical time period, the deviation traffic parameter of the to-be-identified road in the corresponding historical time period, wherein the deviation traffic parameter represents the proportion of objects that do not travel through the to-be-identified road.

[0040] The second processing module is configured to determine, based on the respective candidate traffic speed and the candidate through traffic flow of the to-be-identified road in each historical time period, the traffic impedance parameter of the to-be-identified road, wherein the traffic impedance parameter represents the influence degree of the change mode of the candidate through traffic flow on the change mode of the candidate traffic speed.

[0041] The identification module is configured to obtain the traffic difficulty category of the to-be-identified road based on the respective traffic deviation parameters and the traffic impedance parameter.

[0042] Optionally, the second processing module is further configured to:

[0043] determine, from the respective candidate traffic speeds of the to-be-identified road in each historical time period, a target traffic speed that satisfies a speed value condition and a free through traffic flow corresponding to the target traffic speed;

[0044] determine, from the respective candidate through traffic flows of the to-be-identified road in each historical time period, a target through traffic flow that satisfies a flow value condition and a saturated traffic speed corresponding to the target through traffic flow;

[0045] determine the traffic impedance parameter of the to-be-identified road based on the target traffic speed, the free through traffic flow, the target through traffic flow and the saturated traffic speed.

[0046] Optionally, when the traffic impedance parameter of the to-be-identified road is determined based on the target traffic speed, the free through traffic flow, the target through traffic flow and the saturated traffic speed, the second processing module is further configured to:

[0047] determining a first speed difference between the target speed and the saturation speed, and determining a first flow difference between the free flow volume and the target flow volume;

[0048] obtaining the impedance parameter of the to-be-identified road based on a ratio between the first speed difference and the first flow difference.

[0049] Optionally, before the second processing module obtains the traffic difficulty category of the to-be-identified road based on the traffic deviation parameter and the impedance parameter, the second processing module is further configured to:

[0050] obtaining a to-be-processed traffic speed and a to-be-processed flow volume corresponding to each target time period associated with each historical time period;

[0051] determining a smooth impedance parameter of the to-be-identified road based on the free flow volume, the saturation speed, the to-be-processed traffic speed and the to-be-processed flow volume.

[0052] Optionally, when the second processing module determines the smooth impedance parameter of the to-be-identified road based on the free flow volume, the saturation speed, the to-be-processed traffic speed and the to-be-processed flow volume, the second processing module is further configured to:

[0053] determining an average traffic speed between the to-be-processed traffic speeds greater than the saturation speed from the to-be-processed traffic speeds;

[0054] determining an average flow volume between the to-be-processed flow volumes corresponding to the to-be-processed traffic speeds;

[0055] determining a smooth impedance parameter of the to-be-identified road based on the average traffic speed, the average flow volume, the saturation speed and the target flow volume.

[0056] Optionally, when the second processing module determines the smooth impedance parameter of the to-be-identified road based on the average traffic speed, the average flow volume, the saturation speed and the target flow volume, the second processing module is further configured to:

[0057] determining a second speed difference between the average traffic speed and the saturation speed, and determining a second flow difference between the average flow volume and the target flow volume;

[0058] obtaining the smooth impedance parameter of the to-be-identified road based on a ratio between the second speed difference and the second flow difference.

[0059] Optionally, the first processing module is further configured to:

[0060] For each historical time period, the following operations are performed respectively:

[0061] Based on the sum of the candidate through traffic flow and the candidate deviation traffic flow corresponding to each traffic direction of the to-be-identified road in one historical time period, a total traffic flow corresponding to each traffic direction is obtained respectively;

[0062] Based on the ratio between each candidate deviation traffic flow and the corresponding total traffic flow, a candidate traffic parameter corresponding to each traffic direction of the to-be-identified road in the one historical time period is determined;

[0063] From each candidate traffic parameter, a target traffic parameter meeting a preset parameter condition is determined as a deviation traffic parameter of the one historical time period.

[0064] Optionally, the identification module is further configured to:

[0065] Obtain a road feature set of each other road associated with the to-be-identified road;

[0066] Based on the road feature set, each traffic deviation parameter, the traffic impedance parameter and the smoothing impedance parameter, a traffic difficulty category of the to-be-identified road is determined.

[0067] Optionally, when the identification module determines the traffic difficulty category of the to-be-identified road based on the feature set, each traffic deviation parameter and the traffic impedance parameter, the identification module is further configured to:

[0068] Based on the trained road identification model, the to-be-identified road corresponding traffic difficulty category is determined by taking the candidate traffic speed, each traffic deviation parameter, the traffic impedance parameter and the feature set as input parameters.

[0069] In one aspect, an electronic device is provided, which includes a processor and a memory, wherein the memory stores program code, and when the program code is executed by the processor, the processor performs the steps of any one of the above road identification methods.

[0070] In one aspect, a computer storage medium is provided, which stores computer instructions, and when the computer instructions are run on a computer, the computer performs the steps of any one of the above road identification methods.

[0071] In an aspect, an embodiment of the present application provides a computer program product, which comprises computer instructions stored in a computer readable storage medium; when a processor of an electronic device reads the computer instructions from the computer readable storage medium, the processor executes the computer instructions, so that the electronic device performs the steps of any one of the road recognition methods described above.

[0072] Thanks to the technical solutions described above, the embodiments of the present application have at least the following technical effects:

[0073] Based on the candidate through traffic flow and the deviation traffic flow of the to-be-recognized road in each historical time period, the deviation traffic parameter of the to-be-recognized road in the corresponding historical time period is determined, based on the candidate traffic speed and the candidate through traffic flow of the to-be-recognized road in each historical time period, the traffic impedance parameter of the to-be-recognized road is determined, and based on the traffic deviation parameter and the traffic impedance parameter, the traffic difficulty category of the to-be-recognized road is obtained. In this way, without recognizing the image of the to-be-recognized road, the traffic difficulty category of the to-be-recognized road can be determined by using the candidate through traffic flow, the deviation traffic parameter and the candidate traffic speed of the to-be-recognized road. Since the traffic flow, speed and density feature data of the to-be-recognized road are large in amount and easy to obtain, the traffic difficulty category classification by the method in the embodiments of the present application has low recognition cost and good real-time performance, which not only improves the efficiency of road recognition, but also improves the accuracy of road recognition. BRIEF DESCRIPTION OF DRAWINGS

[0074] The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serve to explain the present application, but do not constitute improper limitations on the present application. In the drawings:

[0075] Figure 1 The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serve to explain the present application, but do not constitute improper limitations on the present application. In the drawings:

[0076] Figure 2A The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serve to explain the present application, but do not constitute improper limitations on the present application. In the drawings:

[0077] Figure 2B The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serve to explain the present application, but do not constitute improper limitations on the present application. In the drawings:

[0078] Figure 2C The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serve to explain the present application, but do not constitute improper limitations on the present application. In the drawings:

[0079] Figure 2D The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serve to explain the present application, but do not constitute improper limitations on the present application. In the drawings:

[0080] Figure 2E The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serve to explain the present application, but do not constitute improper limitations on the present application. In the drawings:

[0081] Figure 2F Fig. 1 is a schematic diagram of a road network in an embodiment of the present application;

[0082] Figure 3 Fig. 2 is an example diagram of a road recognition method in an embodiment of the present application;

[0083] Figure 4 Fig. 3 is a structural schematic diagram of a road recognition device in an embodiment of the present application;

[0084] Figure 5 Fig. 4 is a hardware component structural schematic diagram of an electronic device to which an embodiment of the present application is applied. DETAILED DESCRIPTION

[0085] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0086] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments. Based on the embodiments described in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0087] The terms “first”, “second”, and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0088] Some terms in the embodiments of the present application are explained below to facilitate understanding by those of ordinary skill in the art.

[0089] Through traffic flow: represents the number of trajectories that enter from one end of a road and exit from the other end within a historical time period a. Specifically, it is the sum of through traffic flows in each direction of the road i to be recognized.

[0090] For example, assuming that the road to be recognized contains two traffic directions, the through traffic flow is the sum of the through traffic flows in the two traffic directions.

[0091] Deviation traffic flow: represents the number of trajectories that turn around before, at the head, in the middle and at the tail of the road i, in the historical time period a, specifically, the sum of the U-turn traffic flow of the to-be-identified road i in each traffic direction.

[0092] For example, assuming that the to-be-identified road contains two traffic directions, the deviation traffic flow is the sum of the U-turn traffic flow of the two traffic directions.

[0093] Traffic speed: represents the average speed value of all trajectory points passing through the road i in the historical time period a, specifically, the average traffic speed of the to-be-identified road i in each traffic direction.

[0094] For example, assuming that the to-be-identified road contains two traffic directions, the traffic speed is the average traffic speed of the two traffic directions.

[0095] Traffic impedance parameter: represents the degree of influence of the change mode of the candidate through traffic flow on the change mode of the candidate traffic speed.

[0096] Traffic deviation parameter: represents the proportion of objects that do not drive through the road i.

[0097] The design idea of the embodiments of the present application will be briefly introduced as follows:

[0098] Artificial intelligence (AI) is to use digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology of computer science, which tries to understand the essence of intelligence and produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence is to design and implement principles and methods of various intelligent machines, so that machines have the functions of perception, reasoning and decision-making.

[0099] Artificial intelligence technology is a comprehensive discipline, involving a wide range of fields, both hardware and software technologies. Artificial intelligence basic technologies generally include technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning, etc.

[0100] Computer Vision (CV) Computer vision is a science that studies how to make machines "see". More specifically, it refers to the use of cameras and computers to replace human eyes to identify and measure targets, and further process images to make them more suitable for human observation or transmission to instruments for detection. As a scientific discipline, computer vision researches related theories and technologies, aiming to establish artificial intelligence systems that can obtain information from images or multi-dimensional data. Computer vision technology usually includes image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, video content / behavior recognition, three-dimensional object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous localization and mapping, and other technologies. It also includes common face recognition, fingerprint recognition, and other biometric identification technologies.

[0101] Key technologies of Speech Technology include Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) as well as voiceprint recognition technology. Enabling computers to hear, see, speak, and feel is the future direction of human-computer interaction, with voice being one of the most promising ways of human-computer interaction in the future.

[0102] Nature Language processing (NLP) is an important direction in the field of computer science and artificial intelligence. It studies various theories and methods that enable effective communication between humans and computers using natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Therefore, research in this field will involve natural language, which is the language used in daily life, so it is closely related to the study of linguistics. Natural language processing technology usually includes text processing, semantic understanding, machine translation, robot question and answer, knowledge graph, and other technologies.

[0103] Machine Learning (ML) is a multidisciplinary subject that involves probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, and other disciplines. It is dedicated to studying how computers can simulate or implement human learning behavior to acquire new knowledge or skills, and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental approach to making computers intelligent, and its applications are widespread in various fields of artificial intelligence. Machine learning and deep learning usually include artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and rule-based learning.

[0104] Autonomous driving technology usually includes high-precision maps, environmental perception, behavior decision-making, road planning, motion control, and other technologies. Autonomous driving technology has broad application prospects.

[0105] With the research and progress of artificial intelligence technology, artificial intelligence technology is researched and applied in many fields, such as common smart home, smart wearable device, virtual assistant, smart speaker, smart marketing, unmanned vehicle, autonomous vehicle, unmanned aerial vehicle, robot, smart medical treatment, smart customer service, etc. It is believed that with the development of technology, artificial intelligence technology will be applied in more fields and play an increasingly important role.

[0106] At present, with the development of navigation technology, more and more navigation applications have emerged, which greatly facilitates daily travel activities. In the process of planning the navigation road, if the road is a high traffic difficulty road, it will bring poor driving experience, for example, the road has a narrow width and uneven paving condition. Therefore, identifying the high traffic difficulty road and determining the traffic difficulty category of the road plays an important role in improving the driving experience.

[0107] In the related art, the way to determine the traffic difficulty category of the road can be divided into the following two kinds: the first way: prior knowledge, specifically, based on the prior knowledge in the navigation field, the traffic difficulty category is classified; the second way: based on the neural network model to analyze the image containing the road, and determine the traffic difficulty category of the road.

[0108] However, if the first way is used to identify the traffic difficulty category of the road, the efficiency is low, and if the second way is used to identify the traffic difficulty category of the road, since the width and paving condition of the road change quickly, the picture data that can directly reflect the width and paving condition of the road often updates slowly and is difficult to obtain, resulting in that the width and paving data of the road often cannot be updated in time. Therefore, the recognition accuracy and recognition efficiency of the second way for identifying the traffic difficulty category of the road are not high.

[0109] In order to solve the above problems, a road recognition method is provided in the embodiment of the present application. The deviation traffic parameter of the to-be-recognized road in the corresponding historical time period is determined based on the candidate through traffic flow and the deviation traffic flow of the to-be-recognized road in each historical time period respectively. The traffic impedance parameter of the to-be-recognized road is determined based on the candidate traffic speed and the candidate through traffic flow of the to-be-recognized road in each historical time period respectively. The traffic difficulty category of the to-be-recognized road is obtained based on the traffic deviation parameter and the traffic impedance parameter. In this way, the density, speed and flow of the vehicle trajectory on the to-be-recognized road are directly used for road recognition, which not only improves the recognition accuracy of the traffic difficulty category, but also improves the recognition efficiency of the traffic difficulty category.

[0110] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to explain and illustrate the present application, and are not used to limit the present application, and the embodiments and features of the embodiments can be combined with each other without conflict.

[0111] Referring to Figure 1 As shown in the application scenario schematic diagram in the embodiments of the present application. The application scenario schematic diagram includes a navigation terminal 110 and a server 120. The navigation terminal 110 and the server 120 can communicate through a communication network.

[0112] The navigation terminal 110 pre-installs a target application with a navigation function. The function of the target application is not limited to road navigation. The target application can be a pre-installed client application, a web application, a small program, etc. The navigation terminal 110 can include one or more processors, a memory, an I / O interface for interacting with the server 120, and a display screen, etc. The navigation terminal 110 includes but is not limited to a mobile phone, a computer, a smart voice interaction device, a smart home appliance, a vehicle terminal, an aircraft, etc.

[0113] The server is a background server corresponding to the target application, and provides services for the target application. The server 120 can include one or more processors, a memory, and an I / O interface for interacting with the navigation terminal 110, etc. The server 120 can be a standalone physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content distribution networks (CDN), and big data and artificial intelligence platforms, etc. Basic cloud computing services. The navigation terminal 110 and the server 120 can be directly or indirectly connected through wired or wireless communication, and the embodiments of the present application do not limit this.

[0114] Among them, the road recognition method in the embodiments of the present application can be performed on the navigation terminal 110 or on the server 120. When the road recognition is performed by the navigation terminal 110, the navigation terminal 110 determines the deviation traffic parameter of the to-be-recognized road in the corresponding historical time period based on the candidate through traffic and the deviation traffic of the to-be-recognized road in each historical time period, determines the traffic impedance parameter of the to-be-recognized road based on the candidate traffic speed and the candidate through traffic of the to-be-recognized road in each historical time period, obtains the traffic difficulty category of the to-be-recognized road based on each traffic deviation parameter and traffic impedance parameter, and performs navigation road planning based on the traffic difficulty category.

[0115] When the road recognition is performed by the server 120, the server 120 determines the deviation traffic parameter of the to-be-recognized road in the corresponding historical time period based on the candidate through traffic volume and the deviation traffic volume of the to-be-recognized road in each historical time period, determines the traffic impedance parameter of the to-be-recognized road based on the candidate traffic speed and the candidate through traffic volume of the to-be-recognized road in each historical time period, obtains the traffic difficulty category of the to-be-recognized road based on the traffic deviation parameter and the traffic impedance parameter, and sends the obtained traffic difficulty category to the navigation terminal 110, so that the navigation terminal 110 performs navigation road planning based on the traffic difficulty category.

[0116] The scheme provided in the embodiments of the present application relates to road recognition and other technologies of artificial intelligence, and is specifically explained as follows:

[0117] The road recognition method in the embodiments of the present application will be described below with reference to the accompanying drawings. The road recognition method in the embodiments of the present application can be applied to the navigation terminal 110 or the server 120 shown in FIG. 1, and the road recognition method is described below with reference to the navigation terminal 110 shown in FIG. 1. Figure 1 Figure 2A FIG. 2 shows a flowchart of the road recognition method in the embodiments of the present application, and the specific road recognition process is as follows:

[0118] S20: Determine the deviation traffic parameter of the to-be-recognized road in the corresponding historical time period based on the candidate through traffic volume and the deviation traffic volume of the to-be-recognized road in each historical time period.

[0119] The deviation traffic parameter represents the proportion of objects that do not drive through the to-be-recognized road.

[0120] In the embodiments of the present application, since the to-be-recognized road corresponds to the candidate through traffic volume and the deviation traffic volume in each historical time period, when the deviation traffic parameter is calculated, the candidate through traffic volume and the deviation traffic volume corresponding to each historical time period of the to-be-recognized road are first obtained, and then the deviation traffic parameter corresponding to the to-be-recognized road in the corresponding historical time period is determined based on each candidate through traffic volume and the corresponding deviation traffic volume.

[0121] The historical time period can be an hour, a day, a week, a month, half a year, etc.

[0122] Optionally, in the embodiments of the present application, a possible implementation manner for determining the deviation traffic parameter is provided, and the process of determining the deviation traffic parameter in the embodiments of the present application is introduced below by taking any one historical time period (hereinafter referred to as historical time period a) as an example as follows:

[0123] S201: Obtain the total traffic volume corresponding to the corresponding traffic direction based on the sum of the candidate through traffic volume and the candidate deviation traffic volume corresponding to each traffic direction of the to-be-recognized road in the historical time period a.​

[0124] In the embodiments of the present application, the candidate through traffic flow and the candidate deviated traffic flow corresponding to each traffic direction of the to-be-identified road in the historical time period a are obtained, and the following operations are performed for each traffic direction: the candidate through traffic flow corresponding to one traffic direction is added to the candidate deviated traffic flow to obtain the sum of the candidate through traffic flow and the candidate deviated traffic flow, and the sum of the calculated candidate through traffic flow and the candidate deviated traffic flow is the total traffic flow corresponding to the traffic direction.

[0125] For example, assuming that the candidate through traffic flow is 200 and the candidate deviated traffic flow is 50, the total traffic flow corresponding to the traffic direction d of the to-be-identified road i is 200+50=250.

[0126] S202: Determine the candidate traffic parameter corresponding to each traffic direction of the to-be-identified road in a historical time period based on the ratio between each candidate deviated traffic flow and the corresponding total traffic flow.

[0127] In the embodiments of the present application, after obtaining the total traffic flow corresponding to each traffic direction, the ratio between each candidate deviated traffic flow and the corresponding total traffic flow is calculated, and the calculated ratio is taken as the candidate traffic parameter corresponding to the corresponding traffic direction of the to-be-identified road in the historical time period a. The candidate deviated traffic flow and the total traffic flow are negatively correlated.

[0128] In the embodiments of the present application, the candidate traffic parameter corresponding to one traffic direction d of the to-be-identified road i can be represented as:

[0129]

[0130] wherein, is the candidate traffic parameter corresponding to one traffic direction d of the to-be-identified road i, i is the to-be-identified road, d is the traffic direction, and the value is (0, 1), is the candidate through traffic flow of the to-be-identified road i in the traffic direction d, is the candidate deviated traffic flow of the to-be-identified road i in the traffic direction d.

[0131] Therefore, the candidate traffic parameter in the embodiments of the present application characterizes that the candidate deviated traffic flow of the traffic direction d of the to-be-identified road i accounts for the proportion of the total traffic flow (the candidate through traffic flow + the candidate deviated traffic flow ) of the traffic direction d.

[0132] ​S203: From the candidate passage parameters, determine the target passage parameter that meets the preset parameter conditions, and use it as a deviation passage parameter for a historical time period.

[0133] In this embodiment, it is determined whether each candidate traffic parameter meets the preset parameter conditions. Then, the target traffic parameter that meets the preset parameter conditions is determined from each candidate traffic parameter. The determined target traffic parameter is used as the deviation traffic parameter of the road to be identified in the historical time period. The deviation traffic parameter of each historical time period can be determined. Based on the deviation traffic parameter, the road to be identified is classified, which can improve the accuracy of road classification.

[0134] In this embodiment, the preset parameter condition can be determined by selecting the candidate passage parameter with the largest value from among all candidate passage parameters. Therefore, the target passage parameter can be expressed as:

[0135]

[0136] in, The target traffic parameters are the deviation traffic parameters of the road i to be identified. Characterizes the processing method for determining the maximum value.

[0137] For example, see Figure 2B The diagram shown is an example of determining deviation traffic parameters in an embodiment of this application. Assuming one end of the road i to be identified is A and the other end is B, the traffic directions d of the road i to be identified are determined to be 0 and 1, respectively. 0 represents a traffic direction from A to B, and 1 represents a traffic direction from B to A. The candidate deviation traffic flow corresponding to traffic direction 0 is... The candidate through traffic flow is 50. If the value is 200, then the total traffic flow corresponding to direction 0 is: 200 + 50 = 250, based on candidate deviation traffic flow. Total traffic flow is The candidate traffic parameters for road i to be identified in traffic direction 0 are determined as follows: Candidate deviation traffic flow corresponding to road i in traffic direction 1 The candidate through traffic flow is 200. If the value is 400, then the total traffic flow corresponding to traffic direction 1 is: 200 + 400 = 600, based on candidate deviation traffic flow. Total traffic flow is The candidate traffic parameters for road i to be identified in traffic direction 1 are determined as follows: Then, from all the candidate passage parameters, the candidate passage parameter with the largest value is determined as the candidate passage parameter corresponding to passage direction 1, that is, the deviation passage parameter is 0.33.

[0138] The preset parameter conditions in this embodiment can also calculate the average value of each candidate passage parameter. Therefore, the target passage parameter can be expressed as:

[0139]

[0140] For example, assuming the candidate passage parameter for passage direction 0 is 0.21 and the candidate passage parameter for passage direction 1 is 0.56, then the target passage parameter is... .

[0141] S21: Determine the traffic impedance parameters of the road to be identified based on the candidate traffic speed and candidate through traffic flow corresponding to each historical time period.

[0142] Among them, the passage impedance parameter characterizes the degree of influence of the change mode of candidate through passage flow on the change mode of candidate passage speed.

[0143] In this embodiment of the application, candidate through traffic flow and candidate traffic speed corresponding to each historical time period of the road to be identified are obtained, and based on each candidate through traffic flow and each candidate traffic speed, combined with a preset traffic impedance parameter calculation method, the traffic impedance parameter corresponding to the road to be identified is determined.

[0144] Optionally, in this application embodiment, a possible implementation method for determining the passage impedance parameter is provided, specifically including:

[0145] S211: Determine the target traffic speed that meets the speed value conditions and the free through traffic flow corresponding to the target traffic speed from the candidate traffic speeds corresponding to each historical time period of the road to be identified.

[0146] In this embodiment of the application, candidate traffic speeds and candidate through traffic volumes corresponding to the road to be identified in each historical time period are obtained. Based on the speed values ​​corresponding to each candidate traffic speed, a target traffic speed that meets the preset speed value condition is determined from each candidate traffic speed. Furthermore, since each candidate traffic speed corresponds one-to-one with a candidate through traffic volume, the free through traffic volume corresponding to the target traffic speed is obtained.

[0147] In this embodiment of the application, the speed value condition can be the maximum speed value, that is, from each candidate traffic speed, the candidate traffic speed with the maximum speed value is determined as the target traffic speed, and the free through traffic flow corresponding to the target traffic speed is determined.

[0148] For example, see Figure 2C As shown, this is a schematic diagram of the flow-speed relationship in an embodiment of this application. Figure 2CAs shown in the parabola, the candidate through traffic flow and the candidate traffic speed generally present a quadratic function relationship, that is, at a certain candidate traffic speed, the candidate through traffic flow supported by the to-be-identified road is the maximum. Figure 2C The point in the figure represents the candidate through traffic flow and the candidate traffic speed of the to-be-identified road in different historical time periods, wherein the candidate traffic speed with the maximum speed value is 23.5 km / h, that is, 23.5 km / h is taken as the target traffic speed, and the candidate through traffic flow corresponding to 23.5 km / h is determined as 16 veh / d, that is, 16 veh / d is taken as the free through traffic flow.

[0149] S212: determining the target through traffic flow satisfying the flow value condition and the saturated traffic speed corresponding to the target through traffic flow from the candidate through traffic flows of the to-be-identified road in the historical time periods.

[0150] In the embodiment of the present application, the target through traffic flow satisfying the preset flow value condition is determined from the candidate through traffic flows based on the flow value corresponding to each candidate through traffic flow, and since each candidate through traffic flow corresponds to a candidate traffic speed, the saturated traffic speed corresponding to the target through traffic flow is obtained.

[0151] For example, as shown in the figure, Figure 2C the candidate through traffic flow with the maximum flow value is 48 veh / d, that is, 48 veh / d is taken as the target through traffic flow, and the candidate traffic speed corresponding to 48 veh / d is determined as 16.5 km / h, that is, 16.5 km / h is taken as the saturated traffic speed.

[0152] S213: determining the traffic impedance parameter of the to-be-identified road based on the target traffic speed, the free through traffic flow, the target through traffic flow and the saturated traffic speed.

[0153] In the embodiment of the present application, after the target traffic speed, the free through traffic flow, the target through traffic flow and the saturated traffic speed are determined, the traffic impedance parameter of the to-be-identified road is determined based on the target traffic speed, the free through traffic flow, the target through traffic flow and the saturated traffic speed in combination with the preset parameter calculation mode, so that since the traffic impedance parameter can reflect the difficulty degree of the to-be-identified road, the classification of the to-be-identified road based on the traffic impedance parameter can improve the accuracy of the classification.

[0154] Specifically, in the embodiment of the present application, a possible implementation manner is provided for determining the traffic impedance parameter based on the target traffic speed, the free through traffic flow, the target through traffic flow and the saturated traffic speed, which can reflect the difficulty degree of the to-be-identified road, and specifically includes:

[0155] S2131: determine a first speed difference between the target traffic speed and the saturated traffic speed, and determine a first flow difference between the free-through traffic flow and the target-through traffic flow.

[0156] In the embodiments of the present application, the first speed difference is obtained by subtracting the saturated traffic speed from the target traffic speed, and the first flow difference is obtained by subtracting the target-through traffic flow from the free-through traffic flow.

[0157] S2132: obtain the traffic impedance parameter of the to-be-identified road based on the ratio between the first speed difference and the first flow difference.

[0158] In the embodiments of the present application, the ratio between the first speed difference and the first flow difference is calculated, and the calculated ratio is taken as the traffic impedance parameter of the to-be-identified road.

[0159] In the embodiments of the present application, the traffic impedance parameter can be represented as:

[0160]

[0161] The saturated flow corresponding state in each historical time period is , and the target traffic speed corresponding state is , The target traffic speed is The saturated traffic speed corresponding to the target-through traffic flow is The free-through traffic flow corresponding to the target traffic speed is The target-through traffic flow is

[0162] For example, referring to FIG. 1, which is an example diagram for determining the traffic impedance parameter in the embodiments of the present application, the saturated flow corresponding state in each historical time period is Figure 2D =(16.5km / h, 48veh / d), and the target traffic speed corresponding state is = (23.5km / h, 16.5veh / d), so the traffic impedance parameter is =0.22.

[0163] Further, in the embodiments of the present application, when the value of the target traffic speed is large, the calculated traffic impedance parameter is also large, therefore, in order to improve the accuracy of road classification, in the embodiments of the present application, the smooth impedance parameter can also be calculated by combining the states of the multiple target time periods in the history in which the to-be-processed traffic speed is greater than the saturated traffic speed, and the smooth impedance parameter is taken as the feature of the to-be-identified road. The process of determining the smooth impedance parameter in the embodiments of the present application will be described in detail below, which specifically includes:​

[0164] A1: Obtain the to-be-processed traffic speed and the to-be-processed through traffic volume corresponding to each target time period associated with each historical time period.

[0165] In the embodiments of the present application, each historical time period is associated with a plurality of target time periods, and the to-be-processed traffic speed and the to-be-processed through traffic volume corresponding to each target time period are obtained.

[0166] For example, the to-be-processed traffic speed and the to-be-processed through traffic volume corresponding to N target time periods before the historical time period are obtained, and the to-be-processed traffic speed and the to-be-processed through traffic volume corresponding to N target time periods after the historical time period are obtained.

[0167] A2: Determine the smooth impedance parameter of the to-be-identified road based on the free through traffic volume, the saturated traffic speed, the to-be-processed traffic speed, and the to-be-processed through traffic volume.

[0168] In the embodiments of the present application, after obtaining the to-be-processed traffic speed and the to-be-processed through traffic volume, the smooth impedance parameter of the to-be-identified road can be determined based on the free through traffic volume, the saturated traffic speed, the to-be-processed traffic speed, and the to-be-processed through traffic volume, and in combination with a preset smooth impedance parameter calculation method.

[0169] Optionally, in the embodiments of the present application, a possible implementation for calculating the smooth impedance parameter is provided, which specifically includes:

[0170] A21: Determine the average traffic speed between the to-be-processed traffic speeds greater than the saturated traffic speed from the to-be-processed traffic speeds.

[0171] In the embodiments of the present application, it is determined whether each to-be-processed traffic speed is greater than the saturated traffic speed, and the average traffic speed in each target time period in which the to-be-processed traffic speed is greater than the saturated traffic speed is determined from the to-be-processed traffic speeds.

[0172] A22: Determine the average through traffic volume between the to-be-processed through traffic volumes corresponding to each to-be-processed traffic speed.

[0173] In the embodiments of the present application, it is determined whether each to-be-processed traffic speed is greater than the saturated traffic speed, and the average traffic speed in each target time period in which the to-be-processed traffic speed is greater than the saturated traffic speed is determined from the to-be-processed traffic speeds.

[0174] For example, the to-be-processed traffic speed in the target time period b1 is 45 km / h, the to-be-processed through traffic is 15 ved / h, the to-be-processed traffic speed in the target time period b2 is 35 km / h, the to-be-processed through traffic is 21 ved / h, the to-be-processed traffic speed in the target time period b3 is 25 km / h, the to-be-processed through traffic is 34 ved / h, the to-be-processed traffic speed in the target time period b4 is 23 km / h, the to-be-processed through traffic is 9 ved / h, and the saturation traffic speed is 30 km / h. It is determined that the target time periods in which the to-be-processed traffic speed is greater than the saturation traffic speed are b1 and b2, the average traffic speed between b1 and b2 is calculated as , and the average through traffic between b1 and b2 is calculated as .

[0175] A23: Based on the average traffic speed, the average through traffic, the saturation traffic speed, and the target through traffic, the smooth impedance parameter of the to-be-identified road is determined.

[0176] In the embodiments of the present application, after the average traffic speed and the average through traffic are obtained, the smooth impedance parameter of the to-be-identified road is determined based on the average traffic speed, the average through traffic, the saturation traffic speed, and the target through traffic.

[0177] Optionally, in the embodiments of the present application, a possible way of calculating the smooth impedance parameter based on the average traffic parameter, the average through traffic, the saturation traffic speed, and the target through traffic is provided, which specifically includes:

[0178] A231: A second speed difference between the average traffic speed and the saturation traffic speed is determined, and a second flow difference between the average through traffic and the target through traffic is determined.

[0179] In the embodiments of the present application, the average traffic speed is subtracted from the saturation traffic speed to obtain the second speed difference, and the average through traffic is subtracted from the target through traffic to obtain the second flow difference.

[0180] A232: Based on the ratio between the second speed difference and the second flow difference, the smooth impedance parameter of the to-be-identified road is obtained.

[0181] In the embodiments of the present application, after the second speed difference and the second flow difference are obtained, the ratio between the second speed difference and the second flow difference is calculated, and the calculated ratio is taken as the smooth impedance parameter of the to-be-identified road.

[0182] In the embodiments of the present application, the smooth impedance parameter can be represented as:

[0183]

[0184] wherein, denotes a to-be-processed traffic speed of the to-be-identified road i in a t-th target time period in history, denotes an average traffic speed of all target time periods in history in which all to-be-processed speeds are greater than the saturated traffic speed, denotes an average through traffic flow of all target time periods in history in which all to-be-processed speeds are greater than the saturated traffic speed.

[0185] It should be noted that the traffic impedance parameter and the smooth impedance parameter in the embodiments of the present application measure the decrease of the traffic speed caused by the increase of the unit through traffic flow of the to-be-identified road. The narrower and the poorer the road width is, the more sensitive the traffic speed is to the change of the through traffic flow, and the greater the traffic impedance parameter and the smooth impedance parameter are. Therefore, the traffic impedance parameter and the smooth impedance parameter are used to measure the difficulty of the road.

[0186] S22: obtaining a traffic difficulty category of the to-be-identified road based on the traffic deviation parameters and the traffic impedance parameter.

[0187] In the embodiments of the present application, the to-be-identified road is classified based on the traffic deviation parameters and the traffic impedance parameter, and the traffic difficulty category corresponding to the to-be-identified road is determined.

[0188] Optionally, in the embodiments of the present application, when the traffic difficulty category is determined, the to-be-identified road can be classified based on the traffic deviation parameters, the traffic impedance parameter, and the candidate traffic speeds, and the traffic difficulty category corresponding to the to-be-identified road is determined.

[0189] Optionally, in the embodiments of the present application, the road feature set of other roads associated with the to-be-identified road can also be obtained, and the traffic difficulty category corresponding to the to-be-identified road is determined in combination with the road feature set. The process of determining the traffic difficulty category in the embodiments of the present application is described below, which specifically includes:

[0190] S221: obtaining the road feature set of each other road associated with the to-be-identified road.

[0191] In the embodiments of the present application, the road feature set of each other road associated with the to-be-identified road is obtained.

[0192] The road feature set at least includes one or any combination of the following: the traffic deviation parameters, the traffic impedance parameter, the candidate traffic speeds, and the smooth impedance parameter.

[0193] It should be noted that, in the embodiments of the present application, the other roads associated with the to-be-identified road are: other roads having a relevant endpoint with the to-be-identified road in the road network, wherein the other roads having the same endpoint as the to-be-identified road are the first-order neighbors of the to-be-identified road, the first-order neighbors of the first-order neighbors are the second-order neighbors of the to-be-identified road, that is, the other roads having the same endpoint as the first-order neighbors are the second-order neighbors.

[0194] S222: Determine the traffic difficulty category of the to-be-identified road based on the road feature set, the traffic deviation parameters, the traffic impedance parameter, and the smooth impedance parameter.

[0195] In the embodiments of the present application, after obtaining the road feature set, the traffic difficulty category of the to-be-identified road is identified based on the road feature set, the traffic deviation parameters, the traffic impedance parameter, and the smooth impedance parameter, and the traffic difficulty category of the to-be-identified road is obtained.

[0196] In this way, in the embodiments of the present application, the correlation between adjacent roads in terms of difficulty to walk can be reflected, and the classification accuracy can be significantly improved by fusing the features of adjacent roads.

[0197] Optionally, in the embodiments of the present application, the traffic difficulty category of the to-be-identified road can be determined based on the trained road identification model, and the process of determining the traffic difficulty category based on the road identification model in the embodiments of the present application will be described below, which specifically includes:

[0198] Based on the trained road identification model, the traffic difficulty category corresponding to the to-be-identified road is determined by taking the candidate traffic speed, the traffic deviation parameters, the traffic impedance parameter, and the road feature set as input parameters.

[0199] In the embodiments of the present application, the candidate traffic speed, the traffic deviation parameters, the traffic impedance parameter, and the road feature set are input into the trained road identification model to classify the to-be-identified road, and the traffic difficulty category corresponding to the to-be-identified road is output.

[0200] First, the process of training the road identification model in the embodiments of the present application will be described, and the structure of the road identification model is shown in FIG. 1. Figure 2E

[0201] First, based on the candidate traffic speed of the to-be-identified road i , the candidate through traffic flow , the candidate deviation traffic flow , the traffic deviation parameters , the traffic impedance parameter , and the smooth impedance parameter , the road features are constructed, and thus for the to-be-identified road i, the feature vector can be represented as: ​

[0202]

[0203] In this embodiment, the traffic difficulty category can be set to Level 0 (difficulty level), i.e., ordinary roads, Level 1 (difficulty level), Level 2 (difficulty level), ..., Level C (difficulty level). From the entire road set, road samples are extracted using stratified sampling based on factors such as traffic difficulty category, geographical region, and traffic volume to obtain their unique road characteristics. .

[0204] In a road network, other roads sharing the same nodes as a given road are its first-order neighbors, and the first-order neighbors of a first-order neighbor are its second-order neighbors. For each road sample, the set of road features of all its neighbors is obtained. .

[0205] The neighbor can be a first-order, second-order, multi-order, or a combination thereof, and this application does not impose any restrictions on this.

[0206] For example, see Figure 2F The diagram shown is a road network diagram in an embodiment of this application. The two ends of the road i to be identified are A and B, respectively. The roads with the same nodes as node A are n1, n2, n3, and n4, which are first-order neighbors of the road i to be identified. The road with the same nodes as road n4 is n5, so road n5 is a second-order neighbor of the road i to be identified. The road with the same nodes as road n2 is n6, so road n6 is a second-order neighbor of the road i to be identified.

[0207] The difficulty category label for each road sample was determined by manual calibration. For road sample i, its own feature vector The feature set of neighbors The defined difficulty category labels are: Then the road sample can be represented as { , , }, all road samples { , , } constitutes the training sample dataset D.

[0208] Road samples are continuously extracted from the training sample dataset D in batches. For each road sample in each batch, first... and The road features are then fed into the same feature transformation network to obtain the transformed road features. and the converted road feature set In this way, the converted road features and the converted road feature set of all road samples in the same data batch can be obtained.

[0209] The feature conversion neural network can be a fully connected network, a convolutional neural network, a recurrent neural network, or a combination thereof.

[0210] Secondly, the converted road features and the converted road feature set are input into the neighbor aggregation network together, and the converted road features are further extracted and fused to obtain neighbor aggregation features In this way, the neighbor aggregation features of all road samples in the same data batch can be obtained. The neighbor aggregation network can be a fully connected network, a recurrent neural network, an Attention network, a self-Attention network, or a combination thereof.

[0211] Then, the converted road features

[0212] are fused with the obtained neighbor aggregation features to obtain summary features .

[0213] The feature fusion method can be vector element-wise addition, vector element-wise multiplication, and vector splicing. In this way, the summary features of all road samples in the same data batch can be obtained.

[0214] Then, the summary features are input into the road classification network for classification of the passing difficulty category. In this way, the passing difficulty category of all road samples in the same data batch can be obtained. Cross-entropy is used as the loss function, and the cross-entropy loss is calculated using the output passing difficulty category and the known label of all road samples in the same data batch. The gradient descent method is used to train the road recognition model.

[0215] The road classification network can be a fully connected network, a convolutional neural network, a recurrent neural network, or a combination thereof.

[0216] Finally, sample batches are continuously extracted from the training sample data set D, and the above steps are repeated until the end condition is reached, i.e., the accuracy and the loss function reach a certain level, then the training is stopped, and the trained road recognition model is output.

[0217] ​The following describes the process of classifying the roads to be identified using the road recognition model in this embodiment:

[0218] First, the road features of the road i to be identified are input into a feature transformation network for feature transformation to obtain the transformed road features. Furthermore, the road feature set is input into a feature transformation network for feature transformation to obtain the transformed road feature set. .

[0219] The feature transformation network can be a fully connected network, a convolutional neural network, a recurrent neural network, or a combination thereof, and this application does not impose any restrictions on this.

[0220] Secondly, the converted road features and the transformed set of road features The common input neighbor aggregation network is based on road features. Achieve the set of road features Further extraction and fusion are performed to obtain neighbor aggregation features. .

[0221] The neighbor aggregation network can be a fully connected network, a recurrent neural network, an attention network, a self-attention network, or a combination thereof, and this application does not impose any restrictions on this.

[0222] Then, road features Aggregate features with the obtained neighbors By merging, we can obtain the summarized features. .

[0223] The fusion method can be vector element-wise addition, vector element-wise multiplication, or vector concatenation, and this application does not impose any restrictions on this method.

[0224] Finally, the summarized features will be used. The input is fed into a road classification network to classify the road and obtain the road classification result of the road to be identified, i.

[0225] The road classification network can be a fully connected network, a convolutional neural network, a recurrent neural network, or a combination thereof, and this application does not impose any restrictions on this.

[0226] Based on the above embodiments, the road recognition process in this application embodiment will be illustrated below with a specific example. (See reference...) Figure 3 The diagram shown is an example of a road recognition method in an embodiment of this application, specifically including:

[0227] The candidate through traffic flow, the candidate deviation traffic flow and the candidate traffic speed of the to-be-identified road i in each historical time period are obtained, the deviation traffic parameter of the to-be-identified road i in the corresponding historical time period is determined based on the candidate through traffic flow and the candidate deviation traffic flow, the traffic impedance parameter of the to-be-identified road i is determined based on the candidate through traffic flow and the candidate traffic speed, the smooth impedance parameter of the to-be-identified road i is determined based on the to-be-processed traffic speed and the to-be-processed through traffic flow corresponding to each target time period associated with each historical time period, the road feature set of each other road associated with the to-be-identified road is obtained, and the road feature set, the candidate traffic deviation parameter, the smooth impedance parameter, the traffic impedance parameter and the candidate traffic speed are input into the trained road identification model to determine that the traffic difficulty category of the to-be-identified road is difficult to walk level 1.

[0228] Based on the same inventive concept as the above-mentioned method embodiments of the present application, the present embodiment also provides a road identification device. The principle of solving the problem of the device is similar to the above-mentioned method embodiments, so the implementation of the device can be referred to the implementation of the above-mentioned method, and the repeated parts will not be described here.

[0229] Reference Figure 4 As shown in the structure schematic diagram of a road identification device in the present embodiment, the device comprises a first processing module 400, a second processing module 410 and an identification module 420.

[0230] The first processing module 400 is configured to determine the deviation traffic parameter of the to-be-identified road in the corresponding historical time period based on the candidate through traffic flow and the deviation traffic flow corresponding to each historical time period of the to-be-identified road, wherein the deviation traffic parameter represents the proportion of objects that do not drive through the to-be-identified road.

[0231] The second processing module 410 is configured to determine the traffic impedance parameter of the to-be-identified road based on the candidate traffic speed and the candidate through traffic flow corresponding to each historical time period of the to-be-identified road, wherein the traffic impedance parameter represents the influence degree of the change mode of the candidate through traffic flow on the change mode of the candidate traffic speed.

[0232] The identification module 420 is configured to obtain the traffic difficulty category of the to-be-identified road based on the traffic deviation parameter and the traffic impedance parameter.

[0233] Optionally, the second processing module 410 is further configured to:

[0234] determine the target traffic speed satisfying the speed value condition and the free through traffic flow corresponding to the target traffic speed from the candidate traffic speed corresponding to each historical time period of the to-be-identified road.

[0235] determining a target through traffic flow satisfying a flow value condition target from the candidate through traffic flows corresponding to the historical time periods, and a saturated traffic speed corresponding to the target through traffic flow;

[0236] determining a traffic impedance parameter of the to-be-identified road based on the target traffic speed, the free through traffic flow, the target through traffic flow, and the saturated traffic speed.

[0237] Optionally, when the traffic impedance parameter of the to-be-identified road is determined based on the target traffic speed, the free through traffic flow, the target through traffic flow, and the saturated traffic speed, the second processing module 410 is further configured to:

[0238] determining a first speed difference between the target traffic speed and the saturated traffic speed, and determining a first flow difference between the free through traffic flow and the target through traffic flow;

[0239] obtaining the traffic impedance parameter of the to-be-identified road based on a ratio between the first speed difference and the first flow difference.

[0240] Optionally, before the traffic difficulty category of the to-be-identified road is obtained based on the traffic deviation parameter and the traffic impedance parameter, the second processing module 410 is further configured to:

[0241] obtaining a to-be-processed traffic speed and a to-be-processed through traffic flow corresponding to each target time period associated with the historical time periods;

[0242] determining a smooth impedance parameter of the to-be-identified road based on the free through traffic flow, the saturated traffic speed, the to-be-processed traffic speeds, and the to-be-processed through traffic flows.

[0243] Optionally, when the smooth impedance parameter of the to-be-identified road is determined based on the free through traffic flow, the saturated traffic speed, the to-be-processed traffic speeds, and the to-be-processed through traffic flows, the second processing module 410 is further configured to:

[0244] determining an average traffic speed between the to-be-processed traffic speeds greater than the saturated traffic speed from the to-be-processed traffic speeds;

[0245] determining an average through traffic flow between the to-be-processed through traffic flows corresponding to the to-be-processed traffic speeds;

[0246] determining the smooth impedance parameter of the to-be-identified road based on the average traffic speed, the average through traffic flow, the saturated traffic speed, and the target through traffic flow.

[0247] Optionally, when determining the smooth impedance parameter of the to-be-identified road based on the average passing speed, the average through traffic flow, the saturated passing speed and the target through traffic flow, the second processing module 410 is further configured to:

[0248] determine a second speed difference between the average passing speed and the saturated passing speed, and determine a second flow difference between the average through traffic flow and the target through traffic flow;

[0249] obtain the smooth impedance parameter of the to-be-identified road based on a ratio between the second speed difference and the second flow difference.

[0250] Optionally, the first processing module 400 is further configured to:

[0251] for each historical time period, perform the following operations respectively:

[0252] obtain a total traffic flow corresponding to each passing direction based on a sum of the candidate through traffic flow and the candidate deviation traffic flow corresponding to each passing direction of the to-be-identified road in one historical time period;

[0253] determine a candidate passing parameter corresponding to each passing direction of the to-be-identified road in the one historical time period based on a ratio between the candidate deviation traffic flow and the corresponding total traffic flow;

[0254] determine a target passing parameter meeting a preset parameter condition from the candidate passing parameters as a deviation traffic parameter of the one historical time period.

[0255] Optionally, the identification module 420 is further configured to:

[0256] obtain a road feature set of each other road associated with the to-be-identified road;

[0257] determine a passing difficulty category of the to-be-identified road based on the road feature set, each passing deviation parameter, the passing impedance parameter and the smooth impedance parameter.

[0258] Optionally, when determining the passing difficulty category of the to-be-identified road based on the feature set, each passing deviation parameter and the passing impedance parameter, the identification module 420 is further configured to:

[0259] determine the passing difficulty category corresponding to the to-be-identified road based on a trained road identification model, taking the candidate passing speed, each passing deviation parameter, the passing impedance parameter and the feature set as input parameters.

[0260] Those skilled in the art can understand that each aspect of the present application can be implemented as a system, a method or a program product. Therefore, each aspect of the present application can be specifically implemented as a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining software and hardware aspects, which can be collectively referred to as "circuitry", "module" or "system" here.

[0261] In some possible implementation, the road recognition apparatus according to the present application can at least include a processor and a memory. Wherein, the memory stores program codes, when the program codes are executed by the processor, the processor executes the steps in the road recognition method according to various exemplary embodiments of the present application described in the specification. For example, the processor can execute the steps as shown in Figure 2A .

[0262] Based on the same inventive concept as the above method embodiments, the present embodiment also provides an electronic device. In an embodiment, the electronic device can be, for example, a server 120 as shown in Figure 1 , in which the structure of the electronic device can be as shown in Figure 5 , including a memory 501, a communication module 503 and one or more processors 502.

[0263] The memory 501 is used to store computer programs executed by the processor 502. The memory 501 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, programs required for running instant messaging functions and the like; the data storage area can store various instant messaging information and operation instruction sets and the like.

[0264] The memory 501 can be a volatile memory, for example, a random access memory (RAM); the memory 501 can also be a non-volatile memory, for example, a read-only memory, a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD); or the memory 501 is any other medium capable of carrying or storing desired program codes in the form of instructions or data structures and capable of being accessed by a computer, but not limited to this. The memory 501 can be a combination of the above memories.

[0265] The processor 502 can include one or more central processing units (CPUs), or digital processing units, etc. The processor 502 is configured to implement the road recognition method described above when invoking the computer program stored in the memory 501.

[0266] The communication module 503 is configured to communicate with the navigation terminal and other servers.

[0267] The specific connection medium between the memory 501, the communication module 503 and the processor 502 is not limited in the embodiments of the present application. In the embodiments of the present application, the memory 501 and the processor 502 are connected through the bus 504, and the bus 504 is described by a thick line in the embodiments of the present application. The connection mode between other components is only schematically described, and is not limited. The bus 504 can be divided into an address bus, a data bus, a control bus, etc. For the convenience of description, only one thick line is used to describe the bus 504 in the embodiments of the present application, but it is not limited that there is only one bus or only one type of bus. Figure 5 Figure 5 Figure 5

[0268] The memory 501 stores a computer storage medium, and the computer storage medium stores computer executable instructions. The computer executable instructions are used to implement the road recognition method of the embodiments of the present application. The processor 502 is configured to execute the road recognition method described above, as shown in FIG. 6. Figure 2A

[0269] In some possible implementation manners, each aspect of the road recognition method provided by the present application can also be implemented in the form of a program product, which includes program codes. When the program product runs on a computer device, the program codes are used to make the computer device execute the steps in the road recognition method according to various exemplary embodiments of the present application described in the specification, for example, the steps shown in FIG. 6. Figure 2A

[0270] The program product can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0271] ​​​​​The program product of the embodiments of the present application can take a portable compact disc read only memory (CD-ROM) and include a program code, and can be run on a computing device. However, the program product of the present application is not limited thereto, and in the present document, a readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with a command execution system, device, or apparatus.

[0272] The readable signal medium can include a data signal propagated in a baseband or as a part of a carrier wave, in which a readable program code is borne. Such a propagated data signal can take on many forms, including but not limited to electro-magnetic signal, optical signal, or any suitable combination thereof. The readable signal medium can also be any readable medium that can send, propagate, or transfer a program for use by or in connection with a command execution system, apparatus, or device.

[0273] The program code contained in the readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, and the like, or any suitable combination thereof.

[0274] The program code for performing the operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++, and the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server. In the latter scenario, the remote computing device can be connected to the user's computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computing device, such as through the Internet using an Internet Service Provider.

[0275] It should be noted that although several units or sub-units of the apparatus are mentioned in the above detailed description, such division is merely exemplary and not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided into units for embodiment.

[0276] Moreover, although the operations of the method(s) herein can be described in a particular, sequential order, this is not intended to be a requirement or a limitation. Rather, additional steps can be provided before, after, or in between the described steps, and the method(s) can be implemented without some or all of the described steps. Further, steps can be executed in an order other than that described.

[0277] Those skilled in the art will appreciate that embodiments of the present application can be devised for a method, a system, or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer readable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0278] While the preferred embodiments of the application have been described, additional variations and modifications can be employed by those skilled in the art without departing from the spirit and scope of the application. Therefore, it should be understood that the appended claims are intended to cover all such modifications and variations as falling within the scope of the application. Accordingly, the application is intended to embrace all such alterations, modifications, and variations that fall within the scope of the appended claims. In addition, while a particular feature of the application can have been disclosed with respect to only one of several embodiments, such feature can be combined with one or more other features of the same or different embodiments as can be desired and advantageous for any given or

[0279] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A road recognition method, characterized in that, include: Based on the candidate through traffic flow and deviation traffic flow corresponding to the road to be identified in each historical time period, the deviation traffic parameter of the road to be identified in the corresponding historical time period is determined, wherein the deviation traffic parameter represents the proportion of objects that do not travel through the road to be identified. Based on the candidate traffic speed and candidate through traffic flow corresponding to the road to be identified in each historical time period, the traffic impedance parameter of the road to be identified is determined, wherein the traffic impedance parameter characterizes the degree of influence of the change mode of the candidate through traffic flow on the change mode of the candidate speed; Based on the traffic deviation parameters and the traffic impedance parameters, the traffic difficulty category of the road to be identified is obtained.

2. The method as described in claim 1, characterized in that, The process of determining the traffic impedance parameters of the road to be identified based on the candidate traffic speed and candidate through traffic volume corresponding to each historical time period includes: From the candidate traffic speeds corresponding to each historical time period, determine the target traffic speed that meets the speed value condition, and the free through traffic flow corresponding to the target speed; From the candidate through traffic flows corresponding to each historical time period, determine the target through traffic flow that meets the traffic flow value condition, and the saturation traffic speed corresponding to the target through traffic flow; Based on the target traffic speed, the free through traffic flow, the target through traffic flow, and the saturation traffic speed, the traffic impedance parameters of the road to be identified are determined.

3. The method as described in claim 2, characterized in that, The process of determining the traffic impedance parameters of the road to be identified based on the target traffic speed, the free through traffic flow, the target through traffic flow, and the saturation traffic speed includes: Determine a first speed difference between the target traffic speed and the saturation traffic speed, and determine a first flow difference between the free-through traffic flow and the target through traffic flow; The traffic impedance parameters of the road to be identified are obtained based on the ratio between the first speed difference and the first flow rate difference.

4. The method as described in claim 2, characterized in that, Before obtaining the traffic difficulty category of the road to be identified based on the traffic deviation parameter and the traffic impedance parameter, the method further includes: Obtain the processing traffic speed and processing through traffic flow corresponding to each target time period associated with each historical time period; Based on the free through traffic flow, the saturated traffic speed, each traffic speed to be processed, and each through traffic flow to be processed, the smooth impedance parameters of the road to be identified are determined.

5. The method as described in claim 4, characterized in that, The process of determining the smooth impedance parameters of the road to be identified based on the free through traffic flow, the saturated traffic speed, each traffic speed to be processed, and each through traffic flow to be processed includes: From the various traffic speeds to be processed, determine the average traffic speed among the traffic speeds that are greater than the saturation traffic speed; Determine the average through traffic flow rate among the through traffic flows corresponding to each through traffic speed to be processed; The smooth impedance parameters of the road to be identified are determined based on the average traffic speed, the average through traffic flow, the saturation traffic speed, and the target through traffic flow.

6. The method as described in claim 5, characterized in that, Based on the average traffic speed, the average through traffic flow, the saturation traffic speed, and the target through traffic flow, the smooth impedance parameters of the road to be identified are determined, including: Determine a second speed difference between the average traffic speed and the saturation traffic speed, and determine a second flow difference between the average through traffic flow and the target through traffic flow; The smooth impedance parameters of the road to be identified are obtained based on the ratio between the second speed difference and the second flow rate difference.

7. The method according to any one of claims 1-6, characterized in that, The step of determining the deviation traffic parameters of the road to be identified in the corresponding historical time period based on the candidate through traffic flow and deviation traffic flow of the road to be identified in each historical time period includes: Perform the following operations for each historical time period: Based on the sum of candidate through traffic flow and candidate deviation traffic flow for each traffic direction within a historical time period, the total traffic flow for each traffic direction is obtained. Based on the ratio between each candidate deviation traffic flow and the corresponding total traffic flow, the candidate traffic parameters corresponding to each traffic direction of the road to be identified within the historical time period are determined. From the candidate passage parameters, the target passage parameter that meets the preset parameter conditions is determined as the deviation passage parameter for the aforementioned historical time period.

8. The method according to any one of claims 4-6, characterized in that, The method of obtaining the traffic difficulty category of the road to be identified based on each traffic deviation parameter and the traffic impedance parameter includes: Obtain the set of road features of each other road associated with the road to be identified; Based on the road feature set, each traffic deviation parameter, the traffic impedance parameter, and the smoothing impedance parameter, the traffic difficulty category of the road to be identified is determined.

9. The method as described in claim 8, characterized in that, The process of determining the traffic difficulty category of the road to be identified based on the feature set, each traffic deviation parameter, and the traffic impedance parameter includes: Based on the trained road recognition model, the traffic difficulty category corresponding to the road to be identified is determined by using the candidate traffic speed, traffic deviation parameters, traffic impedance parameters and feature set as input parameters.

10. A road recognition device, characterized in that, include: The first processing module is used to determine the deviation traffic parameters of the road to be identified in the corresponding historical time period based on the candidate through traffic flow and deviation traffic flow corresponding to the road to be identified in each historical time period. The deviation traffic parameters represent the proportion of objects that do not travel through the road to be identified. The second processing module is used to determine the traffic impedance parameter of the road to be identified based on the candidate traffic speed and candidate through traffic flow corresponding to each historical time period, wherein the traffic impedance parameter characterizes the degree of influence of the change mode of the candidate through traffic flow on the change mode of the candidate speed. An identification model is used to obtain the traffic difficulty category of the road to be identified based on each traffic deviation parameter and the traffic impedance parameter.

11. The apparatus as claimed in claim 10, characterized in that, The second processing module is also used for: From the candidate traffic speeds corresponding to each historical time period, determine the target traffic speed that meets the speed value condition, and the free through traffic flow corresponding to the target speed; From the candidate through traffic flows corresponding to each historical time period, determine the target through traffic flow that meets the traffic flow value condition, and the saturation traffic speed corresponding to the target through traffic flow; Based on the target traffic speed, the free through traffic flow, the target through traffic flow, and the saturation traffic speed, the traffic impedance parameters of the road to be identified are determined.

12. The apparatus as claimed in claim 10 or 11, characterized in that, The first processing module is also used for: Perform the following operations for each historical time period: Based on the sum of candidate through traffic flow and candidate deviation traffic flow for each traffic direction within a historical time period, the total traffic flow for each traffic direction is obtained. Based on the ratio between each candidate deviation traffic flow and the corresponding total traffic flow, the candidate traffic parameters corresponding to each traffic direction of the road to be identified within the historical time period are determined. From the candidate passage parameters, the target passage parameter that meets the preset parameter conditions is determined as the deviation passage parameter for the aforementioned historical time period.

13. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores program code that, when executed by the processor, causes the processor to perform the steps of any of the methods described in claims 1-9.

14. A computer-readable storage medium, characterized in that, It includes program code that, when run on an electronic device, causes the electronic device to perform the steps of any of the methods described in claims 1-9.

15. A computer program product, characterized in that, It includes computer instructions stored in a computer-readable storage medium; when a processor of an electronic device reads the computer instructions from the computer-readable storage medium, the processor executes the computer instructions, causing the electronic device to perform the steps of any of the methods described in claims 1-9.

Images