CN112863247A - Road identification method, device, equipment and storage medium - Google Patents
Road identification method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN112863247A CN112863247A CN202011607501.8A CN202011607501A CN112863247A CN 112863247 A CN112863247 A CN 112863247A CN 202011607501 A CN202011607501 A CN 202011607501A CN 112863247 A CN112863247 A CN 112863247A
- Authority
- CN
- China
- Prior art keywords
- road
- image
- current
- semantic segmentation
- vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/167—Driving aids for lane monitoring, lane changing, e.g. blind spot detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/95—Computational photography systems, e.g. light-field imaging systems
- H04N23/951—Computational photography systems, e.g. light-field imaging systems by using two or more images to influence resolution, frame rate or aspect ratio
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/40—Extracting pixel data from image sensors by controlling scanning circuits, e.g. by modifying the number of pixels sampled or to be sampled
- H04N25/46—Extracting pixel data from image sensors by controlling scanning circuits, e.g. by modifying the number of pixels sampled or to be sampled by combining or binning pixels
Abstract
The embodiment of the invention provides a road identification method, a device, equipment and a storage medium, wherein a first image of a current road surface acquired by a camera of a first vehicle is obtained; inputting the first image into a pre-trained road type recognition model so as to obtain a prediction result of each pixel point of the first image output by the road type recognition model; dividing each pixel point into a plurality of groups according to the prediction result of each pixel point, wherein the prediction results of the pixel points in the same group are the same, and the prediction results of the pixel points in different groups are different; determining the road type of the current road surface according to the prediction result of a group of pixel points of which the number meets the preset condition; determining a current adhesion coefficient between the first vehicle and the current road surface according to the road type. The method can realize the accurate identification of the road type of any road surface and the adhesion coefficient between the road surface and the vehicle, and has higher identification accuracy and better robustness, thereby improving the safety of the intelligent driving system of the vehicle.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to a road identification method, apparatus, device, and storage medium.
Background
In an intelligent vehicle driving system, the safety of the system can be effectively improved by identifying the current road type and the adhesion coefficient of the corresponding road surface, but the existing road type identification method mostly depends on an acoustic or vibration sensor, so that the system is relatively high in cost, low in precision and poor in robustness, and the safety of the system is low.
Disclosure of Invention
The embodiment of the invention aims to provide a road identification method, a road identification device, road identification equipment and a storage medium, so as to accurately identify the road type of any road surface and the adhesion coefficient between the road surface and a vehicle, and have high identification precision and good robustness, thereby improving the safety of an intelligent driving system of the vehicle. The specific technical scheme is as follows:
a road identification method, comprising:
acquiring a first image of a current road surface acquired by a camera of a first vehicle;
inputting the first image into a pre-trained road type recognition model so as to obtain a prediction result of each pixel point of the first image output by the road type recognition model;
dividing each pixel point into a plurality of groups according to the prediction result of each pixel point, wherein the prediction results of the pixel points in the same group are the same, and the prediction results of the pixel points in different groups are different;
determining the road type of the current road surface according to the prediction result of a group of pixel points of which the number meets the preset condition;
determining a current adhesion coefficient between the first vehicle and the current road surface according to the road type.
Optionally, the determining, according to the road type, a current adhesion coefficient of the first vehicle to the current road surface includes:
and searching a pre-established coefficient table according to the road type so as to determine the current adhesion coefficient between the first vehicle and the current pavement, wherein the road type is one of a wet asphalt road, a dry asphalt road, a wet concrete road, a dry concrete road, a wet soil road, a dry soil road, a snow road, an ice road and a gravel road.
Optionally, the method further includes:
obtaining a first parameter set, wherein the first parameter set comprises at least one of a current weather condition, a humidity of the current road surface, and a tire model of the first vehicle;
the determining a current adhesion coefficient between the first vehicle and the current road surface according to the road type includes:
and searching a pre-established coefficient table according to the road type and the first parameter group so as to determine the current adhesion coefficient between the first vehicle and the current road surface, wherein the road type is one of a wet asphalt road, a dry asphalt road, a wet concrete road, a dry concrete road, a wet soil road, a dry soil road, a snow road, an ice road and a gravel road.
Optionally, the road type identification model is a semantic segmentation model;
the semantic segmentation model is obtained by training in the following way:
obtaining a standard training data set for training;
cutting the standard training data set into an input set, wherein the size of the input set meets the input requirement of a semantic segmentation model to be trained;
and inputting the input set into the semantic segmentation model to be trained, so as to train the semantic segmentation model to be trained and obtain the semantic segmentation model.
Optionally, the inputting the input set into the to-be-trained semantic segmentation model, so as to train the to-be-trained semantic segmentation model, and obtain the semantic segmentation model, including:
inputting the input set into the semantic segmentation model to be trained to obtain an output result of the semantic segmentation model to be trained;
determining training times according to the output result, the loss value in the training process and the result precision;
adjusting model parameters according to the training times until the output result of the semantic segmentation model to be trained meets the requirement;
and solidifying the weight file of the semantic segmentation model to be trained with the output result meeting the requirements in the semantic segmentation model to be trained meeting the requirements to obtain the semantic segmentation model.
Optionally, the first image is an image which is obtained from an image group and meets a preset image condition, wherein the image group includes a plurality of images of the current road surface acquired by the camera.
Optionally, the preset image condition is: the size of the image meets the input requirement of the road type identification model, and the definition of the image is highest;
or, the preset image condition is: the size of the image accords with the input requirement of the road type identification model, and the occupation range of the current road surface in the image is the largest.
A road identifying device comprising: an image obtaining unit, a prediction result obtaining unit, a grouping unit, a road type determining unit, and an adhesion coefficient determining unit;
the image obtaining unit is configured to obtain a first image of the current road surface acquired by a camera of a first vehicle;
the prediction result obtaining unit is configured to input the first image into a pre-trained road type identification model, so as to obtain a prediction result of each pixel point of the first image output by the road type identification model;
the grouping unit is configured to divide the pixels into a plurality of groups according to the prediction results of the pixels, wherein the prediction results of the pixels in the same group are the same, and the prediction results of the pixels in different groups are different;
the road type determining unit is configured to execute a prediction result of a group of pixel points according with the number of the pixel points meeting a preset condition, and determine the road type of the current road surface;
the adhesion coefficient determination unit is configured to perform determining a current adhesion coefficient between the first vehicle and the current road surface according to the road type.
A storage medium storing a program which, when executed by a processor, implements any of the above-described road identification methods.
A road identification device, the device comprising at least one processor, and at least one memory, bus connected to the processor; the processor and the memory complete mutual communication through the bus; the processor is used for calling a program in the memory, and the program is at least used for realizing any one of the road identification methods.
According to the road identification method, the device, the equipment and the storage medium provided by the embodiment of the invention, the first image of the current road surface acquired by the camera of the first vehicle is obtained; inputting the first image into a pre-trained road type recognition model so as to obtain a prediction result of each pixel point of the first image output by the road type recognition model; dividing each pixel point into a plurality of groups according to the prediction result of each pixel point, wherein the prediction results of the pixel points in the same group are the same, and the prediction results of the pixel points in different groups are different; determining the road type of the current road surface according to the prediction result of a group of pixel points of which the number meets the preset condition; determining a current adhesion coefficient between the first vehicle and the current road surface according to the road type. Therefore, the method can accurately identify the road type of any road surface and the adhesion coefficient between the road surface and the vehicle, and has high identification precision and good robustness, thereby improving the safety of the intelligent driving system of the vehicle. Of course, it is not necessary for any product or method of practicing the invention to achieve all of the above-described advantages at the same time.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart of a road identification method according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a road identification device according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a road identification device according to an embodiment of the present invention.
Detailed Description
Under severe weather conditions such as rain and snow, the road adhesion coefficient is sharply reduced, traffic accidents are frequent, and the life and property safety of drivers and passengers is seriously influenced. In order to reduce the accident rate, the intelligent driving system is inoculated.
The inventor of the scheme finds that in an intelligent driving system, the safety performance of the system can be effectively improved by identifying the current road type and the adhesion coefficient of the corresponding road surface, but the existing road type identification method mostly depends on an acoustic or vibration sensor, so that the system is insufficient in safety system due to the fact that the existing road type identification method is high in relative cost, low in precision and poor in robustness. Or, in a special road section, in order to improve driving safety, road administration staff need to check and issue road types regularly, and traditional manual checking is low in timeliness and time-consuming and labor-consuming.
To solve the above problems, the present inventors propose a road recognition method, apparatus, device, and storage medium. According to the scheme, a large amount of devices similar to sensors or radars do not need to be added on the vehicle, and only improvement is needed on the basis of the original vehicle. The vehicle generally has a driving recorder and a driving computer at present, the invention can acquire the image of the road surface through the driving recorder, then send the image to the driving computer, and the driving computer can execute the method provided by the invention, thereby realizing the accurate identification of the road type of any road surface and the adhesion coefficient between the road surface and the vehicle, having higher identification precision and better robustness, and further improving the safety of the vehicle intelligent driving system.
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the present invention provides a road identification method, comprising:
s100, obtaining a first image of the current road surface acquired by a camera of a first vehicle;
optionally, the camera may be a camera of a vehicle data recorder of the first vehicle or a camera of a reverse image, or may be an additional independent camera, which is not limited in the present invention.
Alternatively, the camera described herein may be communicatively coupled to a device that implements the present invention to transmit images to the device that implements the present invention. The equipment for implementing the invention can be a vehicle-mounted running computer or additionally added equipment. Of course, the device for executing the present invention may also be not on the first vehicle, but on the remote server side, that is, the remote server executes the method of the present invention and then sends the result to the first vehicle, which is not limited by the present invention.
Alternatively, the invention is applicable to any road surface, i.e. for any road surface, a first image thereof may be obtained.
Optionally, the present invention does not set any limit to the first image and the manner of obtaining the first image. For example, in combination with the embodiment shown in fig. 1, in some optional embodiments, the first image is an image that meets a preset image condition and is obtained from an image group, where the image group includes a plurality of images of the current road surface acquired by the camera.
Optionally, a plurality of images are collected, and then one of the images which are relatively in accordance with the preset image condition is selected as the first image, so that the accuracy and the recognition efficiency of the method can be effectively improved. Of course, only one image may be collected directly as the first image, which is not limited in the present invention.
Optionally, the preset image condition may be set according to an actual project requirement, which is not limited in the present invention. For example, in combination with the previous embodiment, in some optional embodiments, the preset image condition is: the size of the image meets the input requirement of the road type identification model, and the definition of the image is highest;
or, the preset image condition is: the size of the image accords with the input requirement of the road type identification model, and the occupation range of the current road surface in the image is the largest.
Optionally, the road type identification model used in the present invention may have some limiting conditions for the input first image, for example, certain requirements for the size, definition, pixels, and the like of the first image may be set specifically according to the actual project requirements, which is not limited by the present invention.
Optionally, the higher the definition of the image is, the more beneficial to accurately identifying the road type of the current road surface is, and the further beneficial to improving the safety of the system is, and the invention does not limit the above.
Optionally, the larger the occupied width of the current road surface in the image is, the smaller the occupied width of other elements is, for example, the smaller the occupied width of the background is, which is more beneficial to identifying the road type of the current road surface and further beneficial to improving the safety of the system, and the invention is not limited thereto.
Of course, the preset image condition may also be other conditions, for example, the boundary of the current road surface in the image may be required to be clearest, the geometric rule is complete, and the like, which is not limited in the present invention.
S200, inputting the first image into a pre-trained road type recognition model so as to obtain a prediction result of each pixel point of the first image output by the road type recognition model;
optionally, the road type recognition model may be a model that has undergone a large amount of training and testing, and the output result meets the actual project requirements, which is not limited in the present invention. For example, in conjunction with the embodiment shown in fig. 1, in some alternative embodiments, the road type identification model is a semantic segmentation model;
the semantic segmentation model is obtained by training in the following way:
step 1, obtaining a standard training data set for training;
optionally, images of road types commonly used in the arrangement project can be acquired through equipment such as a camera according to the project requirements, pixel-level labeling is carried out according to different road types and project requirements, and the labeled images are made to generate a standard training data set. For example, images of various road surfaces are acquired through various ways such as shooting and downloading, and the collected images are sorted and classified. For example, the following are: wet asphalt roads, dry asphalt roads, wet concrete roads, dry concrete roads, snow roads, ice roads, gravel roads, and the like. And labeling the corresponding label on the image of the road type with the specific requirement in the project. For example, some roads have no distinct boundaries and may be labeled according to project specific requirements. And processing the marked images to generate different types of images with different colors, and forming a standard training data set.
Step 2, cutting the standard training data set into an input set, wherein the size of the input set meets the input requirement of a semantic segmentation model to be trained;
optionally, based on the characteristics of the training process of the semantic segmentation model, there is generally a certain requirement on the size of the image input to the semantic segmentation model, so that the images in the standard training data set can be respectively cut to form a data set, which is not limited in the present invention.
And 3, inputting the input set into the semantic segmentation model to be trained, so as to train the semantic segmentation model to be trained, and obtaining the semantic segmentation model.
In combination with the above embodiment, in certain alternative embodiments, the step 3 includes:
step 3.1, inputting the input set into the semantic segmentation model to be trained to obtain an output result of the semantic segmentation model to be trained;
step 3.2, determining training times according to the output result, the loss value and the result precision in the training process;
step 3.3, adjusting model parameters according to the training times until the output result of the semantic segmentation model to be trained meets the requirement;
and 3.4, solidifying the weight file of the semantic segmentation model to be trained with the output result meeting the requirement in the semantic segmentation model to be trained meeting the requirement to obtain the semantic segmentation model.
Alternatively, the data set may be input into a semantic segmentation model, and then model training may be performed empirically. And after the model training is finished, outputting a loss function graph, a verification precision graph and a training weight. And predicting a certain number of road images which do not participate in training by using the model and the training weight, and if the prediction result does not reach the expected effect, modifying the model and the parameters until the expected effect is reached. And solidifying the training weight and the model after the expected effect is achieved, and only keeping the input and output tensors. For example, in the training process, the training times can be determined according to indexes such as a loss value and verification accuracy, a model weight file is generated after the training is finished, the effect verification is performed by using the weight file, and if the recognition accuracy does not meet the requirement, the model parameters can be changed or the model can be changed for training again until the effect is satisfactory.
Optionally, the invention does not set any limit to the road type identification model, and any feasible model belongs to the protection scope of the invention.
S300, dividing each pixel into a plurality of groups according to the prediction result of each pixel, wherein the prediction results of the pixels in the same group are the same, and the prediction results of the pixels in different groups are different;
optionally, the prediction result output by the road type identification model is a prediction result for a pixel level, the prediction result needs to be sorted, the number of each predicted road type of all pixel points in the whole image is counted, and the road type with the largest predicted number can be determined as the road type of the current road surface. If the image includes pixels of the background in addition to pixels of the road type, the road type with the largest predicted number is determined as the road type of the current road surface after the pixels of the background prediction label are eliminated.
S400, determining the road type of the current road surface according to the prediction result of a group of pixel points of which the number meets the preset condition;
optionally, the preset condition described herein is associated with the input first image. For example, if the first image is an image that has not undergone background subtraction processing, and considering that the background occupancy range in the first image may be larger than the road occupancy range in an actual situation, the preset condition may be: the number of pixels is the second largest. If the first image is an image that has undergone background subtraction processing, the preset condition may be: the number of pixels is the largest. The invention is not limited in this regard.
S500, determining a current adhesion coefficient between the first vehicle and the current road surface according to the road type.
Optionally, the road types are different in roughness, material, adhesion and the like of the road surface, so that the current adhesion coefficient between the first vehicle and the current road surface can be determined according to the road types.
Optionally, the present invention does not set any limitation to the method for determining the current adhesion coefficient between the first vehicle and the current road surface according to the road type, and any feasible manner falls within the protection scope of the present invention. For example, in combination with the embodiment shown in fig. 1, in some alternative embodiments, the step S500 includes:
and searching a pre-established coefficient table according to the road type so as to determine the current adhesion coefficient between the first vehicle and the current pavement, wherein the road type is one of a wet asphalt road, a dry asphalt road, a wet concrete road, a dry concrete road, a wet soil road, a dry soil road, a snow road, an ice road and a gravel road.
Optionally, the road surfaces of different road types are different in material, roughness and adhesion, so that the roads common in life can be divided into the above 9 types, and certainly, besides the above 9 types, the actual roads can be divided into more types and corresponding adhesion coefficients can be obtained, so as to establish the coefficient table.
Optionally, the coefficient table records at least a correspondence between different road types and their attachment coefficients, which is not limited in the present invention.
Optionally, the actual road is divided into different types for processing, so that the accuracy of the method can be improved, and the method is not limited to this.
In some alternative embodiments, in combination with the embodiment shown in fig. 1, the method further comprises:
obtaining a first parameter set, wherein the first parameter set comprises at least one of a current weather condition, a humidity of the current road surface, and a tire model of the first vehicle;
the step S500 includes:
and searching a pre-established coefficient table according to the road type and the first parameter group so as to determine the current adhesion coefficient between the first vehicle and the current road surface, wherein the road type is one of a wet asphalt road, a dry asphalt road, a wet concrete road, a dry concrete road, a wet soil road, a dry soil road, a snow road, an ice road and a gravel road.
Optionally, in order to further improve the accuracy of the system, in addition to distinguishing the road types for processing, the influence of weather conditions, road humidity, tire models and the like on the adhesion coefficients may be referred to, and a corresponding coefficient table is established, which is not limited in the present invention.
As shown in fig. 2, the present invention provides a road recognition apparatus, comprising: an image obtaining unit 100, a prediction result obtaining unit 200, a grouping unit 300, a road type determining unit 400, and an adhesion coefficient determining unit 500;
the image obtaining unit 100 is configured to perform obtaining a first image of a current road surface acquired by a camera of a first vehicle;
the prediction result obtaining unit 200 is configured to perform inputting the first image into a pre-trained road type recognition model, so as to obtain a prediction result of each pixel point of the first image output by the road type recognition model;
the grouping unit 300 is configured to perform the step of dividing the pixels into a plurality of groups according to the prediction results of the pixels, wherein the prediction results of the pixels in the same group are the same, and the prediction results of the pixels in different groups are different;
the road type determining unit 400 is configured to perform determining the road type of the current road surface according to a prediction result of a group of pixels, the number of which meets a preset condition;
the adhesion coefficient determination unit 500 is configured to perform determining a current adhesion coefficient between the first vehicle and the current road surface according to the road type.
In some alternative embodiments, in combination with the embodiment shown in fig. 2, the apparatus further comprises: a parameter obtaining unit;
the parameter obtaining unit is configured to perform obtaining a first parameter group, wherein the first parameter group includes at least one of a current weather condition, a humidity of the current road surface, and a tire model of the first vehicle;
the road type determination unit is specifically configured to perform a lookup of a pre-established coefficient table according to the road type and the first parameter set, so as to determine a current adhesion coefficient between the first vehicle and the current road surface, wherein the road type is one of a wet asphalt road, a dry asphalt road, a wet concrete road, a dry concrete road, a wet road, a dry soil road, a snow road, an ice road, and a gravel road.
The present invention provides a storage medium storing a program that when executed by a processor implements a road identification method as described in any one of the above.
The road recognition device comprises a processor and a memory, wherein the image obtaining unit, the prediction result obtaining unit, the grouping unit, the road type determining unit, the adhesion coefficient determining unit and the like are stored in the memory as program units, and the program units stored in the memory are executed by the processor to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more than one, the road type of any road surface and the adhesion coefficient between the road surface and the vehicle can be accurately identified by adjusting the kernel parameters, the identification precision is high, the robustness is good, and therefore the safety of the intelligent driving system of the vehicle is improved. An embodiment of the present invention provides a storage medium having a program stored thereon, which when executed by a processor implements the road identification method.
The embodiment of the invention provides a processor, which is used for running a program, wherein the program executes the road identification method during running.
As shown in fig. 3, the embodiment of the present invention provides a road identification device 70, the road identification device 70 includes at least one processor 701, and at least one memory 702 connected to the processor 701, a bus 703; the processor 701 and the memory 702 complete mutual communication through a bus 703; the processor 701 is configured to call program instructions in the memory 702 to perform the above-described road identification method. The road identification device herein may be a server, a PC, a PAD, a mobile phone, etc.
The present application also provides a computer program product adapted to execute, when executed on a data processing device, a program initialized with the steps comprised by the above-mentioned road identification method.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a device includes one or more processors (CPUs), memory, and a bus. The device may also include input/output interfaces, network interfaces, and the like.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip. The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A road identification method, comprising:
acquiring a first image of a current road surface acquired by a camera of a first vehicle;
inputting the first image into a pre-trained road type recognition model so as to obtain a prediction result of each pixel point of the first image output by the road type recognition model;
dividing each pixel point into a plurality of groups according to the prediction result of each pixel point, wherein the prediction results of the pixel points in the same group are the same, and the prediction results of the pixel points in different groups are different;
determining the road type of the current road surface according to the prediction result of a group of pixel points of which the number meets the preset condition;
determining a current adhesion coefficient between the first vehicle and the current road surface according to the road type.
2. The method of claim 1, wherein said determining a current adhesion coefficient of the first vehicle to the current road surface according to the road type comprises:
and searching a pre-established coefficient table according to the road type so as to determine the current adhesion coefficient between the first vehicle and the current pavement, wherein the road type is one of a wet asphalt road, a dry asphalt road, a wet concrete road, a dry concrete road, a wet soil road, a dry soil road, a snow road, an ice road and a gravel road.
3. The method of claim 1, further comprising:
obtaining a first parameter set, wherein the first parameter set comprises at least one of a current weather condition, a humidity of the current road surface, and a tire model of the first vehicle;
the determining a current adhesion coefficient between the first vehicle and the current road surface according to the road type includes:
and searching a pre-established coefficient table according to the road type and the first parameter group so as to determine the current adhesion coefficient between the first vehicle and the current road surface, wherein the road type is one of a wet asphalt road, a dry asphalt road, a wet concrete road, a dry concrete road, a wet soil road, a dry soil road, a snow road, an ice road and a gravel road.
4. The method of claim 1, wherein the road type identification model is a semantic segmentation model;
the semantic segmentation model is obtained by training in the following way:
obtaining a standard training data set for training;
cutting the standard training data set into an input set, wherein the size of the input set meets the input requirement of a semantic segmentation model to be trained;
and inputting the input set into the semantic segmentation model to be trained, so as to train the semantic segmentation model to be trained and obtain the semantic segmentation model.
5. The method according to claim 4, wherein the inputting the input set into the semantic segmentation model to be trained, so as to train the semantic segmentation model to be trained, and obtaining the semantic segmentation model comprises:
inputting the input set into the semantic segmentation model to be trained to obtain an output result of the semantic segmentation model to be trained;
determining training times according to the output result, the loss value in the training process and the result precision;
adjusting model parameters according to the training times until the output result of the semantic segmentation model to be trained meets the requirement;
and solidifying the weight file of the semantic segmentation model to be trained with the output result meeting the requirements in the semantic segmentation model to be trained meeting the requirements to obtain the semantic segmentation model.
6. The method according to claim 1, wherein the first image is an image which meets a preset image condition and is obtained from an image group, wherein the image group comprises a plurality of images of the current road surface collected by the camera.
7. The method according to claim 6, wherein the preset image condition is: the size of the image meets the input requirement of the road type identification model, and the definition of the image is highest;
or, the preset image condition is: the size of the image accords with the input requirement of the road type identification model, and the occupation range of the current road surface in the image is the largest.
8. A road recognition device, comprising: an image obtaining unit, a prediction result obtaining unit, a grouping unit, a road type determining unit, and an adhesion coefficient determining unit;
the image obtaining unit is configured to obtain a first image of the current road surface acquired by a camera of a first vehicle;
the prediction result obtaining unit is configured to input the first image into a pre-trained road type identification model, so as to obtain a prediction result of each pixel point of the first image output by the road type identification model;
the grouping unit is configured to divide the pixels into a plurality of groups according to the prediction results of the pixels, wherein the prediction results of the pixels in the same group are the same, and the prediction results of the pixels in different groups are different;
the road type determining unit is configured to execute a prediction result of a group of pixel points according with the number of the pixel points meeting a preset condition, and determine the road type of the current road surface;
the adhesion coefficient determination unit is configured to perform determining a current adhesion coefficient between the first vehicle and the current road surface according to the road type.
9. A storage medium characterized by storing a program that when executed by a processor implements the road identification method of any one of claims 1 to 7.
10. A road identification device, characterized in that the road identification device comprises at least one processor, and at least one memory, a bus connected with the processor; the processor and the memory complete mutual communication through the bus; the processor is configured to invoke a program in the memory, the program at least being configured to implement the road identification method of any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011607501.8A CN112863247A (en) | 2020-12-30 | 2020-12-30 | Road identification method, device, equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011607501.8A CN112863247A (en) | 2020-12-30 | 2020-12-30 | Road identification method, device, equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112863247A true CN112863247A (en) | 2021-05-28 |
Family
ID=75998471
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011607501.8A Pending CN112863247A (en) | 2020-12-30 | 2020-12-30 | Road identification method, device, equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112863247A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113486748A (en) * | 2021-06-28 | 2021-10-08 | 同济大学 | Method for predicting friction coefficient of automatic driving road surface, electronic device and medium |
CN115100846A (en) * | 2022-05-09 | 2022-09-23 | 山东金宇信息科技集团有限公司 | Method, device and medium for predicting road accident in tunnel |
CN117437608A (en) * | 2023-11-16 | 2024-01-23 | 元橡科技(北京)有限公司 | All-terrain pavement type identification method and system |
CN115100846B (en) * | 2022-05-09 | 2024-04-30 | 山东金宇信息科技集团有限公司 | Method, equipment and medium for predicting road accident in tunnel |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108985194A (en) * | 2018-06-29 | 2018-12-11 | 华南理工大学 | A kind of intelligent vehicle based on image, semantic segmentation can travel the recognition methods in region |
CN109117691A (en) * | 2017-06-23 | 2019-01-01 | 百度在线网络技术(北京)有限公司 | Drivable region detection method, device, equipment and storage medium |
CN110263844A (en) * | 2019-06-18 | 2019-09-20 | 北京中科原动力科技有限公司 | A kind of method of on-line study and real-time estimation pavement state |
CN110502971A (en) * | 2019-07-05 | 2019-11-26 | 江苏大学 | Road vehicle recognition methods and system based on monocular vision |
CN111507989A (en) * | 2020-04-15 | 2020-08-07 | 上海眼控科技股份有限公司 | Training generation method of semantic segmentation model, and vehicle appearance detection method and device |
CN111695418A (en) * | 2020-04-30 | 2020-09-22 | 上汽大众汽车有限公司 | Method and system for safe driving based on road condition detection |
CN111723849A (en) * | 2020-05-26 | 2020-09-29 | 同济大学 | Road adhesion coefficient online estimation method and system based on vehicle-mounted camera |
CN111845709A (en) * | 2020-07-17 | 2020-10-30 | 燕山大学 | Road adhesion coefficient estimation method and system based on multi-information fusion |
-
2020
- 2020-12-30 CN CN202011607501.8A patent/CN112863247A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109117691A (en) * | 2017-06-23 | 2019-01-01 | 百度在线网络技术(北京)有限公司 | Drivable region detection method, device, equipment and storage medium |
CN108985194A (en) * | 2018-06-29 | 2018-12-11 | 华南理工大学 | A kind of intelligent vehicle based on image, semantic segmentation can travel the recognition methods in region |
CN110263844A (en) * | 2019-06-18 | 2019-09-20 | 北京中科原动力科技有限公司 | A kind of method of on-line study and real-time estimation pavement state |
CN110502971A (en) * | 2019-07-05 | 2019-11-26 | 江苏大学 | Road vehicle recognition methods and system based on monocular vision |
CN111507989A (en) * | 2020-04-15 | 2020-08-07 | 上海眼控科技股份有限公司 | Training generation method of semantic segmentation model, and vehicle appearance detection method and device |
CN111695418A (en) * | 2020-04-30 | 2020-09-22 | 上汽大众汽车有限公司 | Method and system for safe driving based on road condition detection |
CN111723849A (en) * | 2020-05-26 | 2020-09-29 | 同济大学 | Road adhesion coefficient online estimation method and system based on vehicle-mounted camera |
CN111845709A (en) * | 2020-07-17 | 2020-10-30 | 燕山大学 | Road adhesion coefficient estimation method and system based on multi-information fusion |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113486748A (en) * | 2021-06-28 | 2021-10-08 | 同济大学 | Method for predicting friction coefficient of automatic driving road surface, electronic device and medium |
CN115100846A (en) * | 2022-05-09 | 2022-09-23 | 山东金宇信息科技集团有限公司 | Method, device and medium for predicting road accident in tunnel |
CN115100846B (en) * | 2022-05-09 | 2024-04-30 | 山东金宇信息科技集团有限公司 | Method, equipment and medium for predicting road accident in tunnel |
CN117437608A (en) * | 2023-11-16 | 2024-01-23 | 元橡科技(北京)有限公司 | All-terrain pavement type identification method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110415516B (en) | Urban traffic flow prediction method and medium based on graph convolution neural network | |
CN109087510B (en) | Traffic monitoring method and device | |
US9365217B2 (en) | Mobile pothole detection system and method | |
CN111179152B (en) | Road identification recognition method and device, medium and terminal | |
CN112863247A (en) | Road identification method, device, equipment and storage medium | |
CN107591005B (en) | Parking area management method, server and system combining dynamic and static detection | |
CN112706764A (en) | Active anti-collision early warning method, device, equipment and storage medium | |
CN110388929B (en) | Navigation map updating method, device and system | |
US20230056115A1 (en) | Method of Collecting Data from Fleet of Vehicles | |
CN111027535A (en) | License plate recognition method and related equipment | |
CN113011255B (en) | Road surface detection method and system based on RGB image and intelligent terminal | |
US20160189323A1 (en) | Risk determination method, risk determination device, risk determination system, and risk output device | |
CN111881243A (en) | Taxi track hotspot area analysis method and system | |
CN114332707A (en) | Method and device for determining equipment effectiveness, storage medium and electronic device | |
CN113496213A (en) | Method, device and system for determining target perception data and storage medium | |
CN109711341B (en) | Virtual lane line identification method and device, equipment and medium | |
CN109495837A (en) | A kind of condition of road surface monitoring method, equipment and computer readable storage medium | |
CN117012006B (en) | Flood disaster early warning method, equipment and medium for urban road | |
CN110264725B (en) | Method and device for determining road section flow | |
CN115273468B (en) | Traffic jam control strategy generation method and device | |
CN108009671B (en) | Vehicle scheduling method and device | |
CN115031754A (en) | Vehicle driving path planning method and device, electronic equipment and storage medium | |
CN106887138B (en) | A kind of traffic congestion sprawling situation method for detecting and system | |
CN111121803B (en) | Method and device for acquiring common stop points of road | |
CN112597960A (en) | Image processing method, image processing device and computer readable storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210528 |
|
RJ01 | Rejection of invention patent application after publication |