CN114550143A - Scene recognition method and device during driving of unmanned vehicle - Google Patents
Scene recognition method and device during driving of unmanned vehicle Download PDFInfo
- Publication number
- CN114550143A CN114550143A CN202210455326.8A CN202210455326A CN114550143A CN 114550143 A CN114550143 A CN 114550143A CN 202210455326 A CN202210455326 A CN 202210455326A CN 114550143 A CN114550143 A CN 114550143A
- Authority
- CN
- China
- Prior art keywords
- scene
- unmanned vehicle
- data
- picture
- model
- 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
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Abstract
The disclosure relates to the technical field of unmanned driving, and provides a scene recognition method and device for unmanned vehicle driving. The method comprises the following steps: recognizing a scene picture acquired during the driving of the unmanned vehicle through a scene recognition model to obtain perception data of the unmanned vehicle, wherein the scene recognition model is arranged at a cloud end, and the scene recognition model is trained, learns and stores a corresponding relation between the scene picture and the perception data; correcting the sensing data by using a remote monitoring platform to obtain corrected data; uploading the correction data to a cloud end so as to retrain the scene recognition model by using the correction data; and recognizing the scene pictures acquired in the subsequent driving of the unmanned vehicle by using the retrained scene recognition model. By adopting the technical means, the problem of low accuracy in recognition of various scenes in the driving process of the unmanned vehicle in the prior art is solved.
Description
Technical Field
The disclosure relates to the technical field of unmanned driving, in particular to a scene recognition method and device in the driving of an unmanned vehicle.
Background
At present, in the field of unmanned vehicle unmanned driving, in order to realize intelligent driving of an unmanned vehicle, the unmanned vehicle needs to identify various scenes encountered in the driving of the unmanned vehicle by virtue of the unmanned vehicle or related services, and corresponding operations are adopted according to the identified scenes. For example, when an unmanned vehicle encounters a red light, the vehicle should stop and wait until a green light is lit to pass through the intersection. However, at present, the unmanned vehicle can only recognize specific or simple scenes, and the accuracy of the unmanned vehicle for recognizing most scenes is low, so that the unmanned vehicle can only drive in specific or simple traffic areas.
In the course of implementing the disclosed concept, the inventors found that there are at least the following technical problems in the related art: the problem of low accuracy exists in the recognition of various scenes in the driving process of the unmanned vehicle.
Disclosure of Invention
In view of this, the embodiments of the present disclosure provide a method and an apparatus for identifying scenes during driving of an unmanned vehicle, an electronic device, and a computer-readable storage medium, so as to solve the problem in the prior art that the accuracy rate is low in identifying various scenes during driving of an unmanned vehicle.
In a first aspect of the embodiments of the present disclosure, a method for identifying a scene during driving of an unmanned vehicle is provided, including: recognizing a scene picture acquired during the driving of the unmanned vehicle through a scene recognition model to obtain perception data of the unmanned vehicle, wherein the scene recognition model is arranged at a cloud end, and the scene recognition model is trained, learns and stores a corresponding relation between the scene picture and the perception data; correcting the sensing data by using a remote monitoring platform to obtain corrected data; uploading the correction data to a cloud end so as to retrain the scene recognition model by using the correction data; and recognizing the scene pictures acquired in the subsequent driving of the unmanned vehicle by using the retrained scene recognition model.
In a second aspect of the embodiments of the present disclosure, there is provided a scene recognition apparatus in unmanned vehicle driving, including: the first recognition module is configured to recognize a scene picture acquired during driving of the unmanned vehicle through a scene recognition model to obtain perception data of the unmanned vehicle, wherein the scene recognition model is arranged at a cloud end, and the scene recognition model is trained and learns and stores a corresponding relation between the scene picture and the perception data; the correction module is configured to perform correction processing on the sensing data by using the remote monitoring platform to obtain corrected data; the training module is configured to upload the correction data to the cloud so as to retrain the scene recognition model by using the correction data; and the second identification module is configured to identify the scene picture acquired in the subsequent driving of the unmanned vehicle by using the retrained scene identification model.
In a third aspect of the embodiments of the present disclosure, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above method when executing the computer program.
In a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, which stores a computer program, which when executed by a processor, implements the steps of the above-mentioned method.
Compared with the prior art, the embodiment of the disclosure has the following beneficial effects: recognizing a scene picture acquired during the driving of the unmanned vehicle through a scene recognition model to obtain perception data of the unmanned vehicle, wherein the scene recognition model is arranged at a cloud end, and the scene recognition model is trained, learns and stores a corresponding relation between the scene picture and the perception data; correcting the sensing data by using a remote monitoring platform to obtain corrected data; uploading the correction data to a cloud end so as to retrain the scene recognition model by using the correction data; and recognizing the scene pictures acquired in the subsequent driving of the unmanned vehicle by using the retrained scene recognition model. By adopting the technical means, the problem of low accuracy in recognition of various scenes in the driving process of the unmanned vehicle in the prior art can be solved, and the accuracy of recognition of various scenes in the driving process of the unmanned vehicle is further improved.
Drawings
To more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings needed for the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without inventive efforts.
FIG. 1 is a scenario diagram of an application scenario of an embodiment of the present disclosure;
fig. 2 is a schematic flowchart of a scene recognition method in the driving of an unmanned vehicle according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a scene recognition device in an unmanned vehicle driving according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of an electronic device provided in an embodiment of the present disclosure.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the disclosed embodiments. However, it will be apparent to one skilled in the art that the present disclosure may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present disclosure with unnecessary detail.
A scene recognition method and apparatus in unmanned vehicle driving according to an embodiment of the present disclosure will be described in detail below with reference to the accompanying drawings.
Fig. 1 is a scene schematic diagram of an application scenario of an embodiment of the present disclosure. The application scenario may include terminal devices 1 and 3, unmanned vehicle 2, server 4, and network 5.
The devices 1 and 3 may be hardware or software. When the terminal devices 1 and 3 are hardware, they may be various electronic devices having a display screen and supporting communication with the server 4, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like; when the terminal devices 1 and 3 are software, they may be installed in the electronic device as above. The terminal devices 1 and 3 may be implemented as a plurality of software or software modules, or may be implemented as a single software or software module, which is not limited by the embodiments of the present disclosure. Further, various applications, such as a data processing application, an instant messaging tool, social platform software, a search-type application, a shopping-type application, and the like, may be installed on the terminal devices 1 and 3.
The server 4 may be a server providing various services, for example, a backend server receiving a request sent by a terminal device establishing a communication connection with the server, and the backend server may receive and analyze the request sent by the terminal device and generate a processing result. The server 4 may be one server, may also be a server cluster composed of a plurality of servers, or may also be a cloud computing service center, which is not limited in this disclosure.
The server 4 may be hardware or software. When the server 4 is hardware, it may be various electronic devices that provide various services to the terminal devices 1 and 3, and the unmanned vehicle 2. When the server 4 is software, it may be a plurality of software or software modules that provide various services for the terminal devices 1 and 3 and the unmanned vehicle 2, or may be a single software or software module that provides various services for the terminal devices 1 and 3 and the unmanned vehicle 2, which is not limited by the embodiment of the present disclosure.
The network 5 may be a wired network connected by a coaxial cable, a twisted pair and an optical fiber, or may be a wireless network that can interconnect various Communication devices without wiring, for example, Bluetooth (Bluetooth), Near Field Communication (NFC), Infrared (Infrared), and the like, which is not limited in the embodiment of the present disclosure.
The user can establish a communication connection with the server 4 via the terminal devices 1 and 3, and the unmanned vehicle 2 via the network 5 to receive or transmit information or the like. It should be noted that the specific types, numbers and combinations of the terminal devices 1 and 3, the unmanned vehicles 2, the server 4 and the network 5 may be adjusted according to the actual requirements of the application scenarios, and the embodiment of the present disclosure does not limit this.
Fig. 2 is a schematic flowchart of a scene recognition method in the driving of an unmanned vehicle according to an embodiment of the present disclosure. The scene recognition method in the unmanned vehicle driving of fig. 2 may be performed by the terminal device of fig. 1, or the unmanned vehicle or the server. As shown in fig. 2, the method for recognizing a scene while an unmanned vehicle is traveling includes:
s201, identifying a scene picture acquired during the driving of the unmanned vehicle through a scene identification model to obtain perception data of the unmanned vehicle, wherein the scene identification model is arranged at a cloud end and is trained, and a corresponding relation between the scene picture and the perception data is learned and stored;
s202, correcting the sensing data by using a remote monitoring platform to obtain corrected data;
s203, uploading the correction data to a cloud end so as to retrain the scene recognition model by using the correction data;
and S204, recognizing the scene picture acquired in the subsequent driving of the unmanned vehicle by using the retrained scene recognition model.
The scene recognition model and the later labeling model can be any common neural network model, for example, the models can be a fast-rcnn model (the labeling model and the scene recognition model can be one model or two models). The model training method in the embodiments of the present disclosure may be deep learning. The perception data can be understood as a scene picture plus a label corresponding to the scene picture. For example, a scene picture acquired by an unmanned vehicle before passing through a school has a school indicator, and the perception data corresponding to the scene picture is a school door (the school door can be understood as a label of the scene picture).
The correction processing is to correct the wrong sensing data, for example, a scene picture at the school doorway is recognized as a bus stop, and at this time, a mistake is recognized (that is, the sensing data is wrong), and the sensing data is corrected, and the obtained correction data is the scene picture and a label at the school doorway.
According to the method and the device, after the scene picture with the error identification is corrected, the scene identification model is retrained by using the correction data, so that the accuracy of the scene picture identification by the scene identification model can be improved.
According to the technical scheme provided by the embodiment of the disclosure, scene pictures acquired during driving of the unmanned vehicle are identified through a scene identification model to obtain perception data of the unmanned vehicle, wherein the scene identification model is arranged at the cloud end and is trained, and the corresponding relation between the scene pictures and the perception data is learned and stored; correcting the sensing data by using a remote monitoring platform to obtain corrected data; uploading the correction data to a cloud end so as to retrain the scene recognition model by using the correction data; and recognizing the scene pictures acquired in the subsequent driving of the unmanned vehicle by using the retrained scene recognition model. By adopting the technical means, the problem of low accuracy in recognition of various scenes in the driving process of the unmanned vehicle in the prior art can be solved, and the accuracy of recognition of various scenes in the driving process of the unmanned vehicle is further improved.
In step 202, the sensing data is corrected by using the remote monitoring platform, so as to obtain corrected data, which includes: receiving a correction instruction corresponding to the sensing data by using a remote monitoring platform, and correcting the sensing data according to the correction instruction to obtain corrected data; or the perception data is corrected by utilizing a first labeling model arranged on the remote monitoring platform, and corrected data are obtained.
And receiving a correction instruction sent by a user by using the remote monitoring platform, and correcting the perception data according to the correction instruction, namely correcting the wrong perception data by a person. The first labeling model is used for correcting the perception data, so that the labor intensity is reduced, and the correction has better real-time performance and efficiency. The first annotation model is a weak model, and the first annotation model is different from the scene recognition model in that, for example, the annotation model is trained with only 200 pictures, while the scene recognition model should use thousands of pictures to ensure accuracy. The second annotation model is similar to the first annotation model, and is a weak model, except that the second annotation model is trained using the features of the picture.
Before the first labeling model arranged on the remote monitoring platform is used for correcting the perception data and the corrected data is obtained, the method further comprises the following steps: acquiring a first data set, wherein the first data set comprises a plurality of scene pictures; labeling the first data set to obtain a label corresponding to each scene picture in the first data set; and training the first labeling model by using the first data set after the labeling processing.
The first data set is smaller in size than the second data set, for example, the first data set has hundreds of scene pictures, and the second data set has thousands of scene pictures. The first labeling model is trained, and deep learning can be performed.
Before step 201 is executed, that is, before the scene picture acquired while the unmanned vehicle is running is identified by the scene identification model to obtain the perception data of the unmanned vehicle, the method further includes: acquiring a second data set, wherein the second data set comprises a plurality of scene pictures; extracting picture characteristics of each scene picture in the second data set; inputting the picture characteristics of each scene picture in the second data set into a second labeling model, and outputting a label corresponding to each scene picture, wherein the second labeling model is trained and learns and stores the corresponding relation between the picture characteristics and the labels; and taking the picture characteristics of each scene picture as the input of the scene recognition model, taking the label of each scene picture as the output of the scene recognition model, and training the scene recognition model.
The corrective treatment can be understood as a re-labeling. The picture characteristics of each scene picture in the second data set processed by the second labeling model carry corresponding labels. In order to further improve the accuracy of the scene recognition model, the embodiment of the present disclosure takes the picture characteristics of each scene picture as the input of the scene recognition model, takes the label of each scene picture as the output of the scene recognition model, and trains the scene recognition model (the above training of the first labeling model is to train the scene recognition model by taking each scene picture as the input of the scene recognition model and taking the label of each scene picture as the output of the scene recognition model). Of course, it is obvious that training the first labeling model may also be a training method using the embodiments of the present disclosure.
Extracting picture features of each scene picture in the second data set, including: extracting color features of each scene picture in the second data set by using a color histogram method, wherein the picture features comprise: color features and shape features; and extracting the shape feature of each scene picture in the second data set by utilizing a Fourier shape descriptor method.
Because signs exist in many traffic scenes, and the signs express different information through different colors and shapes or numbers of patterns on the signs, the color features and the shape features of each scene picture are extracted, and the color features and the shape features of each scene picture are utilized to improve the accuracy of picture identification. For example, if the speed limit sign is red and the upper number is 60 km/hour, it means that the maximum speed of the vehicle in the scene is 60 km/hour. The color histogram and the fourier shape descriptor are common techniques for extracting image features, and are not described herein again. It should be noted that, the color feature of the scene picture may be extracted, and a method for extracting a color set, a color moment, a color aggregation vector, and a color correlation diagram may also be used, and a method for extracting a shape feature of the scene picture may also be used, which is an extraction method using a boundary feature method, a geometric parameter method, and a shape invariant moment method.
In step 201, identifying a scene picture acquired during driving of the unmanned vehicle by using a scene identification model to obtain perception data of the unmanned vehicle, including: acquiring a scene picture of an unmanned vehicle in running in real time; extracting the color characteristics of the scene picture by using a color histogram method; extracting shape features of the scene picture by utilizing a Fourier shape descriptor method; and inputting the color characteristics and the shape characteristics of the scene picture into the trained scene recognition model, and outputting perception data.
In step 204, recognizing the scene picture acquired in the subsequent driving of the unmanned vehicle by using the retrained scene recognition model, including: carrying out model distillation treatment on the retrained scene recognition model, and downloading a result of the model distillation treatment to the unmanned vehicle so as to recognize scene pictures acquired in the subsequent driving of the unmanned vehicle; or carrying out model pruning on the retrained scene recognition model, and downloading the result of the model pruning to the unmanned vehicle so as to recognize the scene picture acquired in the subsequent driving of the unmanned vehicle.
In order to enable the scene recognition service provided by the scene recognition model to be more stable and avoid the problem of scene recognition service interruption caused by network disconnection, the embodiment of the disclosure obtains the model with the smaller model scale corresponding to the scene recognition model through model distillation processing or model pruning processing, downloads the model with the smaller model scale to the unmanned vehicle, and provides the scene recognition service for the unmanned vehicle by using the model with the smaller model scale.
All the above optional technical solutions may be combined arbitrarily to form optional embodiments of the present application, and are not described herein again.
The following are embodiments of the disclosed apparatus that may be used to perform embodiments of the disclosed methods. For details not disclosed in the embodiments of the apparatus of the present disclosure, refer to the embodiments of the method of the present disclosure.
Fig. 3 is a schematic diagram of a scene recognition device during unmanned vehicle driving according to an embodiment of the present disclosure. As shown in fig. 3, the scene recognition device during unmanned vehicle driving includes:
the first identification module 301 is configured to identify a scene picture acquired during driving of the unmanned vehicle through a scene identification model to obtain perception data of the unmanned vehicle, wherein the scene identification model is arranged at a cloud end, and the scene identification model is trained and learns and stores a corresponding relationship between the scene picture and the perception data;
the correcting module 302 is configured to perform correcting processing on the sensing data by using the remote monitoring platform to obtain corrected data;
a training module 303 configured to upload the rectification data to the cloud for retraining the scene recognition model using the rectification data;
and the second identification module 304 is configured to identify the scene pictures acquired in the subsequent driving of the unmanned vehicle by using the retrained scene identification model.
The scene recognition model and the later labeling model can be any common neural network model, for example, the models can be a fast-rcnn model (the labeling model and the scene recognition model can be one model or two models). The model training method in the embodiments of the present disclosure may be deep learning. The perception data can be understood as a scene picture plus a label corresponding to the scene picture. For example, a scene picture acquired by an unmanned vehicle before passing through a school has a school indicator, and the perception data corresponding to the scene picture is a school door (the school door can be understood as a label of the scene picture).
The correction processing is to correct the wrong sensing data, for example, a scene picture at the school doorway is recognized as a bus stop, and at this time, a mistake is recognized (that is, the sensing data is wrong), and the sensing data is corrected, and the obtained correction data is the scene picture and a label at the school doorway.
According to the method and the device, after the scene picture with the wrong recognition is corrected, the scene recognition model is retrained by using the correction data, so that the accuracy of the scene picture recognized by the scene recognition model can be improved.
According to the technical scheme provided by the embodiment of the disclosure, scene pictures acquired during the driving of the unmanned vehicle are identified through a scene identification model to obtain perception data of the unmanned vehicle, wherein the scene identification model is arranged at the cloud end, and the scene identification model is trained, learns and stores the corresponding relation between the scene pictures and the perception data; correcting the sensing data by using a remote monitoring platform to obtain corrected data; uploading the correction data to a cloud end so as to retrain the scene recognition model by using the correction data; and recognizing the scene pictures acquired in the subsequent driving of the unmanned vehicle by using the retrained scene recognition model. By adopting the technical means, the problem of low accuracy in recognition of various scenes in the driving process of the unmanned vehicle in the prior art can be solved, and the accuracy of recognition of various scenes in the driving process of the unmanned vehicle is further improved.
Optionally, the correcting module 302 is further configured to receive a correction instruction corresponding to the sensing data by using the remote monitoring platform, and perform correction processing on the sensing data according to the correction instruction to obtain corrected data; or the perception data is corrected by utilizing a first labeling model arranged on the remote monitoring platform, and corrected data are obtained.
And receiving a correction instruction sent by a user by using the remote monitoring platform, and correcting the perception data according to the correction instruction, namely correcting the wrong perception data by a person. The first labeling model is used for correcting the perception data, so that the labor intensity is reduced, and the correction has better real-time performance and efficiency. The first annotation model is a weak model, and the first annotation model is different from the scene recognition model in that, for example, the annotation model is trained with only 200 pictures, while the scene recognition model should use thousands of pictures to ensure accuracy. The second annotation model is similar to the first annotation model, and is a weak model, except that the second annotation model is trained using the features of the picture.
Optionally, the rectification module 302 is further configured to acquire a first data set, wherein the first data set includes a plurality of scene pictures; labeling the first data set to obtain a label corresponding to each scene picture in the first data set; and training the first labeling model by using the first data set after the labeling processing.
The first data set is smaller in size than the second data set, for example, the first data set has hundreds of scene pictures, and the second data set has thousands of scene pictures. The first labeling model is trained, and deep learning can be performed.
Optionally, the first identifying module 301 is further configured to obtain a second data set, where the second data set includes a plurality of scene pictures; extracting picture characteristics of each scene picture in the second data set; inputting the picture characteristics of each scene picture in the second data set into a second labeling model, and outputting a label corresponding to each scene picture, wherein the second labeling model is trained and learns and stores the corresponding relation between the picture characteristics and the labels; and taking the picture characteristics of each scene picture as the input of the scene recognition model, taking the label of each scene picture as the output of the scene recognition model, and training the scene recognition model.
The corrective treatment can be understood as a re-labeling. The picture characteristics of each scene picture in the second data set processed by the second labeling model carry corresponding labels. In order to further improve the accuracy of the scene recognition model, the embodiment of the present disclosure takes the picture characteristics of each scene picture as the input of the scene recognition model, takes the label of each scene picture as the output of the scene recognition model, and trains the scene recognition model (the above training of the first labeling model is to train the scene recognition model by taking each scene picture as the input of the scene recognition model and taking the label of each scene picture as the output of the scene recognition model). Of course, it is obvious that training the first labeling model may also be a training method using the embodiments of the present disclosure.
Optionally, the first identification module 301 is further configured to extract a color feature of each scene picture in the second data set by using a color histogram method, where the picture feature includes: color features and shape features; and extracting the shape feature of each scene picture in the second data set by utilizing a Fourier shape descriptor method.
Because signs exist in many traffic scenes, and the signs express different information through different colors and shapes or numbers of patterns on the signs, the color features and the shape features of each scene picture are extracted, and the color features and the shape features of each scene picture are utilized to improve the accuracy of picture identification. For example, if the speed limit sign is red and the upper number is 60 km/hour, it means that the maximum speed of the vehicle in the scene is 60 km/hour. The color histogram and the fourier shape descriptor are common techniques for extracting image features, and are not described herein again. It should be noted that, the color feature of the scene picture may be extracted, and a method for extracting a color set, a color moment, a color aggregation vector, and a color correlation diagram may also be used, and a method for extracting a shape feature of the scene picture may also be used, which is an extraction method using a boundary feature method, a geometric parameter method, and a shape invariant moment method.
Optionally, the first identification module 301 is further configured to obtain a scene picture of the unmanned vehicle in the driving process in real time; extracting the color characteristics of the scene picture by using a color histogram method; extracting shape features of the scene picture by utilizing a Fourier shape descriptor method; and inputting the color characteristics and the shape characteristics of the scene picture into the trained scene recognition model, and outputting perception data.
Optionally, the second identification module 304 is further configured to perform model distillation processing on the retrained scene identification model, and download a result of the model distillation processing to the unmanned vehicle, so as to identify a scene picture acquired during subsequent driving of the unmanned vehicle; or carrying out model pruning on the retrained scene recognition model, and downloading the result of the model pruning to the unmanned vehicle so as to recognize the scene picture acquired in the subsequent driving of the unmanned vehicle.
In order to enable the scene recognition service provided by the scene recognition model to be more stable and avoid the problem of scene recognition service interruption caused by network disconnection, the embodiment of the disclosure obtains the model with the smaller model scale corresponding to the scene recognition model through model distillation processing or model pruning processing, downloads the model with the smaller model scale to the unmanned vehicle, and provides the scene recognition service for the unmanned vehicle by using the model with the smaller model scale.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present disclosure.
Fig. 4 is a schematic diagram of an electronic device 4 provided by the embodiment of the present disclosure. As shown in fig. 4, the electronic apparatus 4 of this embodiment includes: a processor 401, a memory 402 and a computer program 403 stored in the memory 402 and executable on the processor 401. The steps in the various method embodiments described above are implemented when the processor 401 executes the computer program 403. Alternatively, the processor 401 implements the functions of the respective modules/units in the above-described respective apparatus embodiments when executing the computer program 403.
Illustratively, the computer program 403 may be partitioned into one or more modules/units, which are stored in the memory 402 and executed by the processor 401 to accomplish the present disclosure. One or more modules/units may be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution of the computer program 403 in the electronic device 4.
The electronic device 4 may be a desktop computer, a notebook, a palm computer, a cloud server, or other electronic devices. The electronic device 4 may include, but is not limited to, a processor 401 and a memory 402. Those skilled in the art will appreciate that fig. 4 is merely an example of the electronic device 4, and does not constitute a limitation of the electronic device 4, and may include more or less components than those shown, or combine certain components, or different components, e.g., the electronic device may also include input-output devices, network access devices, buses, etc.
The Processor 401 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The storage 402 may be an internal storage unit of the electronic device 4, for example, a hard disk or a memory of the electronic device 4. The memory 402 may also be an external storage device of the electronic device 4, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the electronic device 4. Further, the memory 402 may also include both internal storage units of the electronic device 4 and external storage devices. The memory 402 is used for storing computer programs and other programs and data required by the electronic device. The memory 402 may also be used to temporarily store data that has been output or is to be output.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules, so as to perform all or part of the functions described above. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the technical solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.
In the embodiments provided in the present disclosure, it should be understood that the disclosed apparatus/electronic device and method may be implemented in other ways. For example, the above-described apparatus/electronic device embodiments are merely illustrative, and for example, a module or a unit may be divided into only one logical function, and may be implemented in other ways, and multiple units or components may be combined or integrated into another system, or some features may be omitted or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, the present disclosure may implement all or part of the flow of the method in the above embodiments, and may also be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of the above methods and embodiments. The computer program may comprise computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain suitable additions or additions that may be required in accordance with legislative and patent practices within the jurisdiction, for example, in some jurisdictions, computer readable media may not include electrical carrier signals or telecommunications signals in accordance with legislative and patent practices.
The above examples are only intended to illustrate the technical solutions of the present disclosure, not to limit them; although the present disclosure has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present disclosure, and are intended to be included within the scope of the present disclosure.
Claims (10)
1. A scene recognition method in the driving of an unmanned vehicle is characterized by comprising the following steps:
recognizing a scene picture acquired during the driving of the unmanned vehicle through a scene recognition model to obtain perception data of the unmanned vehicle, wherein the scene recognition model is arranged at a cloud end, and the scene recognition model is trained, learns and stores a corresponding relation between the scene picture and the perception data;
correcting the perception data by using a remote monitoring platform to obtain corrected data;
uploading the correction data to the cloud end so as to retrain the scene recognition model by using the correction data;
and recognizing the scene pictures acquired in the subsequent driving of the unmanned vehicle by using the retrained scene recognition model.
2. The method of claim 1, wherein the performing a corrective action on the sensory data using the remote monitoring platform to obtain corrective data comprises:
receiving a correction instruction corresponding to the perception data by using the remote monitoring platform, and performing correction processing on the perception data according to the correction instruction to obtain correction data; or
And carrying out correction processing on the perception data by utilizing a first labeling model arranged on the remote monitoring platform to obtain the correction data.
3. The method according to claim 2, wherein before the performing the correction processing on the perception data by using the first labeling model set on the remote monitoring platform to obtain the correction data, the method further comprises:
acquiring a first data set, wherein the first data set comprises a plurality of scene pictures;
labeling the first data set to obtain a label corresponding to each scene picture in the first data set;
and training the first labeling model by using the first data set after the labeling processing.
4. The method according to claim 1, wherein before the scene picture acquired while the unmanned vehicle is traveling is identified by the scene identification model to obtain the perception data of the unmanned vehicle, the method further comprises:
acquiring a second data set, wherein the second data set comprises a plurality of scene pictures;
extracting picture characteristics of each scene picture in the second data set;
inputting the picture characteristics of each scene picture in the second data set into a second labeling model, and outputting a label corresponding to each scene picture, wherein the second labeling model is trained, learns and stores the corresponding relationship between the picture characteristics and the labels;
and taking the picture characteristics of each scene picture as the input of the scene recognition model, taking the label of each scene picture as the output of the scene recognition model, and training the scene recognition model.
5. The method of claim 4, wherein the extracting the picture feature of each scene picture in the second data set comprises:
extracting color features of each scene picture in the second data set by using a color histogram method, wherein the picture features comprise: the color feature and shape feature;
extracting the shape feature of each scene picture in the second data set by using a Fourier shape descriptor method.
6. The method according to claim 1, wherein the identifying the scene picture acquired while the unmanned vehicle is running through the scene identification model to obtain the perception data of the unmanned vehicle comprises:
acquiring a scene picture of the unmanned vehicle in the driving process in real time;
extracting the color characteristics of the scene picture by using a color histogram method;
extracting shape features of the scene picture by utilizing a Fourier shape descriptor method;
and inputting the color characteristic and the shape characteristic of the scene picture into the trained scene recognition model, and outputting the perception data.
7. The method according to claim 1, wherein the recognizing the scene picture acquired in the subsequent driving of the unmanned vehicle by using the retrained scene recognition model comprises:
carrying out model distillation treatment on the scene recognition model after retraining, and downloading the result of the model distillation treatment to the unmanned vehicle so as to recognize scene pictures acquired in the subsequent driving of the unmanned vehicle; or
And carrying out model pruning on the scene recognition model after retraining, and downloading a result of the model pruning to the unmanned vehicle so as to recognize scene pictures acquired in the subsequent driving of the unmanned vehicle.
8. A scene recognition device in an unmanned vehicle, characterized by comprising:
the first recognition module is configured to recognize a scene picture acquired during driving of the unmanned vehicle through a scene recognition model to obtain perception data of the unmanned vehicle, wherein the scene recognition model is arranged at a cloud end, and the scene recognition model is trained, learns and saves a corresponding relation between the scene picture and the perception data;
the correction module is configured to perform correction processing on the perception data by using a remote monitoring platform to obtain corrected data;
a training module configured to upload the remediation data to the cloud for retraining the scene recognition model with the remediation data;
and the second identification module is configured to identify the scene picture acquired in the subsequent driving of the unmanned vehicle by using the retrained scene identification model.
9. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210455326.8A CN114550143A (en) | 2022-04-28 | 2022-04-28 | Scene recognition method and device during driving of unmanned vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210455326.8A CN114550143A (en) | 2022-04-28 | 2022-04-28 | Scene recognition method and device during driving of unmanned vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114550143A true CN114550143A (en) | 2022-05-27 |
Family
ID=81667113
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210455326.8A Pending CN114550143A (en) | 2022-04-28 | 2022-04-28 | Scene recognition method and device during driving of unmanned vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114550143A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116434041A (en) * | 2022-12-05 | 2023-07-14 | 北京百度网讯科技有限公司 | Mining method, device and equipment for error perception data and automatic driving vehicle |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107609502A (en) * | 2017-09-05 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | Method and apparatus for controlling automatic driving vehicle |
| US20200226377A1 (en) * | 2020-03-25 | 2020-07-16 | Intel Corporation | Robust object detection and classification using static-based cameras and events-based cameras. |
| CN111723746A (en) * | 2020-06-22 | 2020-09-29 | 江苏云从曦和人工智能有限公司 | Scene recognition model generation method, system, platform, device and medium |
| CN111796663A (en) * | 2019-04-09 | 2020-10-20 | Oppo广东移动通信有限公司 | Scene recognition model updating method and device, storage medium and electronic equipment |
| CN111950723A (en) * | 2019-05-16 | 2020-11-17 | 武汉Tcl集团工业研究院有限公司 | Neural network model training method, image processing method, device and terminal equipment |
| CN113705554A (en) * | 2021-08-13 | 2021-11-26 | 北京百度网讯科技有限公司 | Training method, device and equipment of image recognition model and storage medium |
| CN113723159A (en) * | 2021-02-26 | 2021-11-30 | 腾讯科技（深圳）有限公司 | Scene recognition model training method, scene recognition method and model training device |
-
2022
- 2022-04-28 CN CN202210455326.8A patent/CN114550143A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107609502A (en) * | 2017-09-05 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | Method and apparatus for controlling automatic driving vehicle |
| CN111796663A (en) * | 2019-04-09 | 2020-10-20 | Oppo广东移动通信有限公司 | Scene recognition model updating method and device, storage medium and electronic equipment |
| CN111950723A (en) * | 2019-05-16 | 2020-11-17 | 武汉Tcl集团工业研究院有限公司 | Neural network model training method, image processing method, device and terminal equipment |
| US20200226377A1 (en) * | 2020-03-25 | 2020-07-16 | Intel Corporation | Robust object detection and classification using static-based cameras and events-based cameras. |
| CN111723746A (en) * | 2020-06-22 | 2020-09-29 | 江苏云从曦和人工智能有限公司 | Scene recognition model generation method, system, platform, device and medium |
| CN113723159A (en) * | 2021-02-26 | 2021-11-30 | 腾讯科技（深圳）有限公司 | Scene recognition model training method, scene recognition method and model training device |
| CN113705554A (en) * | 2021-08-13 | 2021-11-26 | 北京百度网讯科技有限公司 | Training method, device and equipment of image recognition model and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116434041A (en) * | 2022-12-05 | 2023-07-14 | 北京百度网讯科技有限公司 | Mining method, device and equipment for error perception data and automatic driving vehicle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107909065B (en) | Method and device for detecting face occlusion | |
| CN110020620B (en) | Face recognition method, device and equipment under large posture | |
| CN108491805B (en) | Identity authentication method and device | |
| CN108427939B (en) | Model generation method and device | |
| CN109902659B (en) | Method and apparatus for processing human body image | |
| US20190087647A1 (en) | Method and apparatus for facial recognition | |
| US11126882B2 (en) | Method and device for license plate positioning | |
| CN109002820B (en) | License plate recognition method and device and related equipment | |
| US20190087683A1 (en) | Method and apparatus for outputting information | |
| CN110321958A (en) | Training method, the video similarity of neural network model determine method | |
| US20190197328A1 (en) | Method and apparatus for outputting information | |
| CN112052850A (en) | License plate recognition method and device, electronic equipment and storage medium | |
| CN111126481A (en) | Training method and device of neural network model | |
| CN109726481B (en) | Auxiliary method and device for robot construction and terminal equipment | |
| CN112396005A (en) | Biological characteristic image recognition method and device, electronic equipment and readable storage medium | |
| CN108319888A (en) | The recognition methods of video type and device, terminal | |
| CN111767831B (en) | Method, apparatus, device and storage medium for processing image | |
| CN107729928A (en) | Information acquisition method and device | |
| CN108170751B (en) | Method and apparatus for handling image | |
| CN114550143A (en) | Scene recognition method and device during driving of unmanned vehicle | |
| CN108304852B (en) | Method and device for determining road section type, storage medium and electronic device | |
| CN110400560B (en) | Data processing method and device, storage medium and electronic device | |
| CN112329666A (en) | Face recognition method and device, electronic equipment and storage medium | |
| CN112418089A (en) | Gesture recognition method and device and terminal | |
| CN111353470B (en) | Image processing method and device, readable medium and electronic equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220527 |