CN109711285A - Training, test method and the device of identification model - Google Patents
Training, test method and the device of identification model Download PDFInfo
- Publication number
- CN109711285A CN109711285A CN201811508096.7A CN201811508096A CN109711285A CN 109711285 A CN109711285 A CN 109711285A CN 201811508096 A CN201811508096 A CN 201811508096A CN 109711285 A CN109711285 A CN 109711285A
- Authority
- CN
- China
- Prior art keywords
- identified
- image
- identification model
- sensor
- training
- 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.)
- Granted
Links
Landscapes
- Image Analysis (AREA)
Abstract
The present invention proposes a kind of training of identification model, test method and device, wherein, method includes: that control object to be identified is moved along setting track, in object to be identified along setting track moving process, the sensor that control fixed bit installs carries out Image Acquisition to object to be identified, according to the mark of object to be identified, acquired image is labeled, model training is carried out to identification model using the image after mark, the sensor installed by controlling fixed bit, Image Acquisition is carried out to the object to be identified moved on setting track, and acquired image is labeled according to the identification information of known object to be identified, it does not need to be marked manually after manually identifying acquired image one by one, reduce cost of labor, the efficiency of image labeling is improved simultaneously, solve use in the prior art based on depth The technical issues of identification model of habit needs to expend a large amount of artificial acquisition image and carries out image labeling, and efficiency is lower, higher cost.
Description
Technical field
The present invention relates to Traffic Sign Recognition technical field more particularly to a kind of training of identification model, test method and
Device.
Background technique
Traffic Sign Recognition is one of pre-requisite abilities of intelligent vehicle, and Traffic Sign Recognition is mainly by being mounted on vehicle
On video camera acquisition road on road signs information, the identification indicated by model trained in advance, so as to
Important traffic information (such as speed limit, no parking) is transmitted to the control system of intelligent vehicle in time, to control vehicle
Safe operation, it is therefore, accurate to identify efficiently and in real time traffic sign, there is critical role in vehicle drive.
In the prior art, using the Traffic Sign Recognition algorithm model of deep learning, which needs a large amount of tools
Someone's fortification have first marked the raw video image of desired value as training data, are trained to the model, then using warp
Cross the test that trained model carries out model.But traffic mark board is many kinds of, and acquisition and mark difficulty are larger, needs to expend big
Manpower, financial resources and the material resources of amount, and it is not able to satisfy the demand of diversified training data and test data.
Summary of the invention
The present invention is directed to solve at least some of the technical problems in related technologies.
For this purpose, the first purpose of this invention is to propose a kind of training method of identification model, by controlling fixed bit
The sensor installed carries out Image Acquisition to the object to be identified moved on setting track, and according to known to be identified
The identification information of object is labeled acquired image, does not need to carry out hand after manually identifying acquired image one by one
Dynamic mark, reduces cost of labor, while improving the efficiency of image labeling.
Second object of the present invention is to propose a kind of test method of identification model.
Third object of the present invention is to propose a kind of training device of identification model.
Fourth object of the present invention is to propose a kind of test device of identification model.
5th purpose of the invention is to propose a kind of computer equipment.
6th purpose of the invention is to propose a kind of non-transitorycomputer readable storage medium.
In order to achieve the above object, first aspect present invention embodiment proposes a kind of training method of identification model, comprising:
Control object to be identified is moved along setting track；
In the object to be identified along the moving process of the setting track, the sensor that control fixed bit installs is to institute
It states object to be identified and carries out Image Acquisition；
According to the mark of the object to be identified, acquired image is labeled；
Model training is carried out to identification model using the image after mark.
In order to achieve the above object, second aspect of the present invention embodiment proposes a kind of test method of identification model, comprising:
Control object to be identified is moved along setting track；
In the object to be identified along the moving process of the setting track, the sensor that control fixed bit installs is to institute
It states object to be identified and carries out Image Acquisition；
Image input identification model will be collected, the object identity of the identification model output is obtained；
Whether matched according to the object identity with the object to be identified, the identification model is tested.
In order to achieve the above object, third aspect present invention embodiment proposes a kind of training device of identification model, the device
Include:
Control module is moved for controlling object to be identified along setting track；
Acquisition module, for along the moving process of the setting track, control fixed bit to be installed in the object to be identified
The sensor set carries out Image Acquisition to the object to be identified；
First labeling module is labeled acquired image for the mark according to the object to be identified；
Training module, for carrying out model training to identification model using the image after mark.
In order to achieve the above object, fourth aspect present invention embodiment proposes a kind of test device of identification model, the device
Include:
Mobile module is moved for controlling object to be identified along setting track；
Processing module, for along the moving process of the setting track, control fixed bit to be installed in the object to be identified
The sensor set carries out Image Acquisition to the object to be identified；
Input module obtains the object identity of the identification model output for that will collect image input identification model；
First test module, for whether being matched according to the object identity with the object to be identified, to the identification
Model is tested.
In order to achieve the above object, fourth aspect present invention embodiment proposes a kind of computer equipment, including memory, processing
Device and storage on a memory and the computer program that can run on a processor, when the processor executes described program, reality
Now training method as described in relation to the first aspect, alternatively, realizing the test method as described in second aspect.
In order to achieve the above object, fourth aspect present invention embodiment proposes a kind of non-transitory computer-readable storage medium
Matter is stored thereon with computer program, when which is executed by processor, realizes training method as described in relation to the first aspect, or
Person realizes the test method as described in second aspect.
Technical solution provided by the embodiment of the present invention may include it is following the utility model has the advantages that
Control object to be identified is moved along setting track, and in object to be identified along setting track moving process, control is solid
It positions the sensor installed and Image Acquisition is carried out to object to be identified, according to the mark of object to be identified, to collected figure
As being labeled, model training is carried out to identification model using the image after mark, the sensing installed by controlling fixed bit
Device carries out Image Acquisition to the object to be identified moved on setting track, and is believed according to the mark of known object to be identified
Breath is labeled acquired image, does not need to be marked manually after manually identifying acquired image one by one, reduces
Cost of labor, while improving the efficiency of image labeling.
Detailed description of the invention
Above-mentioned and/or additional aspect and advantage of the invention will become from the following description of the accompanying drawings of embodiments
Obviously and it is readily appreciated that, in which:
Fig. 1 is a kind of flow diagram of the training method of identification model provided by the embodiment of the present invention；
Fig. 2 is the flow diagram of another kind identification model training method provided by the embodiment of the present invention；
Fig. 3 is a kind of flow diagram of the test method of identification model provided by the embodiment of the present invention；
Fig. 4 is the flow diagram of the test method of another kind identification model provided by the embodiment of the present invention；
Fig. 5 is a kind of structural schematic diagram of the training device of identification model provided in an embodiment of the present invention；
Fig. 6 is a kind of structural schematic diagram of the test device of identification model provided in an embodiment of the present invention；And
Fig. 7 shows the block diagram for being suitable for the exemplary computer device for being used to realize the application embodiment.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to is used to explain the present invention, and is not considered as limiting the invention.
Below with reference to the accompanying drawings training, test method and the device of the identification model of the embodiment of the present invention are described.
Fig. 1 is a kind of flow diagram of the training method of identification model provided by the embodiment of the present invention.
As shown in Figure 1, method includes the following steps:
Step 101, control object to be identified is moved along setting track.
As a kind of possible implementation, in the both sides of the road in specific place, a set of cableway, cableway are built along road
On be provided with driving device, which can drive object to be identified, along cableway constraint track with setting speed and side
To movement.
Step 102, in object to be identified along setting track moving process, the sensor that control fixed bit installs is treated
Identify that object carries out Image Acquisition.
In the embodiment of the present invention, sensor is the sensor with image collecting function, for example, imaging sensor, the biography
Sensor can be set in video camera, also can be set in monitoring camera, be not construed as limiting in the present embodiment.
Specifically, there is the sensor of image collecting function in the specific position fixed setting of road, in object to be identified
Along cableway in preset speed, direction moving process, sensor that control fixed bit installs to object to be identified into
Row Image Acquisition, acquires multiple images.
Optionally, object to be identified can also carry out it is appropriate block and/or be stained processing etc., pass through the various states of setting
Under object to be identified, and the image of the object to be identified under above-mentioned various states is acquired, as training sample, for subsequent
Identification model can be improved to the recognition capability of object to be identified under abnormal scene in model training.
Step 103, according to the mark of object to be identified, acquired image is labeled.
Specifically, the mark of object to be identified is obtained, for example, object to be identified is traffic marking board, then traffic marking board
Mark then include traffic marking board type, meaning etc., e.g., the type of traffic marking board is that speed limit category shows board, meaning
For 60 kilometers of 30 kilometers/hour of speed limit or speed limit/per hour etc., wherein the mark of object to be identified is to preset
Good parameter is realized through known identification information according to the mark of object to be identified to acquired image automatic marking,
It does not need to be marked manually again after manually identifying acquired image one by one, reduces cost of labor, meanwhile, it improves and adopts
Collect the efficiency of image labeling.
Step 104, model training is carried out to identification model using the image after mark.
Specifically, the acquisition image after mark is trained as identification model of the training sample to images to be recognized,
So that identification model can learn to obtain the corresponding relationship between input picture and object identity, i.e., the model that training is completed can
To identify the corresponding mark of object to be identified for including in current image to be collected.
In the identification model training method of the present embodiment, control object to be identified is moved along setting track, to be identified right
Image Acquisition is carried out to object to be identified as along setting track moving process, controlling the sensor that fixed bit installs, according to
The mark of object to be identified, is labeled acquired image, carries out model instruction to identification model using the image after mark
Practice, the sensor installed by controlling fixed bit, Image Acquisition is carried out to the object to be identified moved on setting track, and
Acquired image is labeled according to the identification information of known object to be identified, is not needed manually to acquired image
It is marked manually after identifying one by one, reduces cost of labor, while improving the efficiency of image labeling.
It is described in a upper embodiment, by carrying out Image Acquisition to the object to be identified moved along desired guiding trajectory, due to
The identification information of object to be identified is known information, can use identification information and realizes to the automatic marking for acquiring image, leads to
The image for crossing mark is trained identification model, and the method for the embodiment of the present invention is mainly used in intelligent vehicle to traffic marking
In the identification scene of board, can the relative position of intelligent vehicle and traffic marking board accurately identify also the operation peace that will affect vehicle
Entirely, for example, the traffic marking board of instruction road surface ahead failure, for this purpose, being based on a upper embodiment, the embodiment of the invention provides another
A kind of training method of identification model, is obtained by calculation the relative position information of traffic signboard and intelligent vehicle, will be opposite
Location information and identification information are used as markup information, are labeled to acquired image, using the image after mark as instruction
Practice sample to be trained identification model, to improve the fineness of model identification.
Fig. 2 is the flow diagram of another kind identification model training method provided by the embodiment of the present invention, such as Fig. 2 institute
Show, this method may comprise steps of:
Step 201, control object to be identified is moved along setting track.
Specifically, the step 101 being referred in an embodiment, principle is identical, and details are not described herein again.
In the embodiment of the present invention, object to be identified be traffic marking board, traffic marking board be it is multiple, it is all kinds of comprising being labeled with
The traffic marking board of traffic sign, and/or, the traffic marking board under various occlusion states, and/or, it is various to be stained under state
Traffic marking board.By the traffic marking board being arranged under various states, and acquire traffic signboard under above-mentioned various states
Image is used for subsequent model training, identification model can be improved to traffic marking board under abnormal scene as training sample
Recognition capability.
Step 202, in object to be identified along setting track moving process, the sensor that control fixed bit installs is treated
Identify that object carries out Image Acquisition.
Specifically, the sensor that fixed bit installs, for example, in intelligent vehicle actual motion scene, in intelligent vehicle
Image capture module, and according in practical intelligent vehicle Run-time scenario, the driving requirements of intelligent vehicle, by sensor with preset
Highly, pitch angle, and rotation angle etc., are fixed on the specific position of test road so that sensor can collect it is default
The image of traffic marking board in range wraps in acquired image since the identification information of traffic signboard is known information
The traffic signboard contained has determined corresponding identification information.
Step 203, the position that object to be identified is moved to along setting track when sensor acquires image is determined.
Specifically, it is determined that at the time of sensor acquires image, and the setting speed mobile according to driving device, it is calculated
The moving distance of driving device can be determined according to the moving distance and moving direction of driving device when sensor acquires image
The position that driving device is moved to along setting track.
It should be noted that when to obtain acquisition image, object to be identified is moved to along setting track in the present embodiment
Therefore position calculates, driving device can at the uniform velocity be moved with the speed of setting, but be compared simultaneously in the present embodiment to simplify
It is not construed as limiting.
Step 204, the position that object to be identified is moved to along setting track when acquiring image according to sensor, determines sensing
The relative positional relationship of device and object to be identified, and obtain acquired image under a variety of relative positional relationships.
Specifically, the position of sensor is fixed, and object to be identified is along setting track when acquiring image according to sensor
The position being moved to can determine the relative positional relationship of sensor and object to be identified, wherein the relative positional relationship includes
Relative altitude, relative distance and relative bearing etc. between object to be identified and the sensor for acquiring image.In turn, to be identified
In object moving process, at the time of controlling sensor multi collect image, and record when acquiring image every time, it can obtain a variety of
Acquired image under relative positional relationship.
Step 205, using sensor and object to be identified relative positional relationship and object to be identified mark, to adopting
The image collected is labeled.
Specifically, after collecting multiple images, according to each corresponding sensor of acquisition image and object to be identified
Relative positional relationship, and the predetermined identification information of object to be identified currently acquired carries out acquired image
Mark, the sample as identification model training.
Step 206, model training is carried out to identification model using the image after mark.
Specifically, the mark of object to be identified in image is labelled in sample image in advance, and when acquisition image
The relative positional relationship of Vehicles and Traffic Signs, input identification model are trained, and model is allowed to learn to be acquired
The model can be applied in actual vehicle Run-time scenario, to traffic mark by the corresponding relationship of the image and object to be identified that arrive
Will carries out intelligent recognition, to indicate that vehicle is run.
In the training method of the identification model of the embodiment of the present invention, in object to be identified along setting track moving process,
According to the mobile setting speed of driving device and direction, driving device is in the position of cableway when can calculate determining sensor acquisition image
Set, so that it is determined that object to be identified is along the position that is moved to of setting track, and then can determine sensor that fixed bit installs with
The relative position of object to be identified, the mark of relationship and object to be identified in acquisition image, schemes acquisition depending on the relative position
As being labeled, using the image after mark as training sample, realize by the identification information of preset object to be identified with
Calculate determining relative positional relationship when acquiring image, acquisition image be labeled, manpower is saved, reduce manpower at
This, and the annotating efficiency of image is improved, sufficient amount of training sample can be provided, for model training so as to improve mould
The accuracy of type training.
Above described embodiment describes in model training scene, how according to known identification information and the phase being calculated
Intelligent dimension is carried out to image of the location information to acquisition, training sample is obtained, to the method that model is trained, and according to
The identification information known obtains image after being labeled to image, is also used as test sample, the identification for completing to training
Model is tested, for this purpose, providing a kind of test method of identification model in the present embodiment, is applied in test scene, is led to
Control object to be identified is crossed along setting track moving process, the image of object to be identified is acquired, will include known object mark
The images to be recognized of knowledge inputs identification model and is identified, carry out the training effect of test model as test sample.
Fig. 3 is a kind of flow diagram of the test method of identification model provided by the embodiment of the present invention, such as Fig. 3 institute
Show, the method includes the steps of:
Step 301, control object to be identified is moved along setting track.
In the embodiment of the present invention, object to be identified is traffic sign, traffic sign be it is multiple, comprising being labeled with all kinds of traffic
The traffic marking board of mark, and/or, the traffic marking board under various occlusion states, and/or, the various traffic being stained under state
Nameplate.
As a kind of possible implementation, in the both sides of the road in specific place, a set of cableway, cableway are built along road
On be provided with driving device, traffic sign can be hung on the drive means, which can drive object to be identified, along rope
The track of road constraint is mobile with setting speed and direction.
Step 302, in object to be identified along setting track moving process, the sensor that control fixed bit installs is treated
Identify that object carries out Image Acquisition.
In the embodiment of the present invention, sensor is the sensor with image collecting function, for example, imaging sensor, the biography
Sensor can be set in video camera, also can be set in monitoring camera, in the present embodiment and be not construed as limiting.
Specifically, there is the sensor of image collecting function in the specific position fixed setting of cableway, in object to be identified
Along cableway in preset speed, direction moving process, the sensor that control fixed bit installs is between the preset time
Every, or with preset time point to object to be identified carry out Image Acquisition, acquire multiple images.
Step 303, acquired image is inputted into identification model, obtains the object identity of identification model output.
Wherein, object identity includes title, the type of object and the meaning etc. of object.
Specifically, by acquired image in step 302, the identification model that training is completed is inputted, identification model is pre-
It first passes through training study and obtains the corresponding relationship between image and the mark of object, according to the image of input, i.e., the exportable figure
The mark for the object for including as in.
Step 304, whether matched according to object identity with object to be identified, identification model is tested.
Specifically, it is determined that object identity and the object to be identified of model output identify whether to match, if matching illustrates mould
The effect of type training is preferable, and the mark of the object to be identified in image can be recognized accurately by model, if mismatching, explanation
The effect of model training is poor, needs to re-start training to model, for example, increasing the quantity etc. of training sample.
In the test method of the identification model of the embodiment of the present invention, control object to be identified along setting track move, to
Identify object along setting track moving process, the sensor that installs of control fixed bit, to move on setting track to
Identify object carry out Image Acquisition, since the identification information of object to be identified is known parameter, by it is collected included to
It identifies that the image of the identification information of object as test sample, is input to identification model, obtains the object mark of identification model output
Know, determines whether object identity matches with object to be identified, the recognition accuracy of identification model is tested, realizes and is not required to
Manually test sample is labeled, reduce cost of labor.
Based on a upper embodiment, the embodiment of the invention provides the signals of the process of the test method of another identification model
Figure, further illustrate, during being acquired to image, not only in available image object to be identified identification information,
The relative positional relationship that determining object to be identified and sensor can also be calculated, because identifying field to traffic marking board in intelligent vehicle
Under scape, distance of the traffic marking board apart from vehicle also has important indicative function to the operation of intelligent vehicle, for example, before instruction
Side's the traffic marking board that no through traffic, thus can using the image comprising identification information and relative positional relationship as test sample,
The recognition accuracy of identification model is tested.Fig. 4 is the test of another kind identification model provided by the embodiment of the present invention
The flow diagram of method, as shown in figure 4, this method includes following step:
Step 401, control object to be identified is moved along setting track.
Step 402, in object to be identified along setting track moving process, the sensor that control fixed bit installs is treated
Identify that object carries out Image Acquisition.
Specifically, step 401- step 402 is referred to the step 301 in an embodiment and 302, and principle is identical, this
Place repeats no more.
Step 403, the position that object to be identified is moved to along setting track when sensor acquires image is determined.
Step 404, the position that object to be identified is moved to along setting track when acquiring image according to sensor, determines sensing
The relative positional relationship of device and object to be identified, and obtain acquired image under a variety of relative positional relationships.
Specifically, step 403- step 404 is referred to the step 203- step 204 in Fig. 2 corresponding embodiment, principle phase
Together, details are not described herein again.
Step 405, acquired image is inputted into identification model, obtains the test position and object mark of identification model output
Know.
Specifically, the object to be identified for including in acquired image has been determined by calculation in above-mentioned steps,
Carry out Image Acquisition when, object to be identified and acquire image sensor between relative positional relationship, the relative positional relationship
Relative altitude, relative distance and relative bearing etc. between sensor comprising object to be identified and acquisition image, simultaneously as
The mark of object to be identified be it is known, using acquired image as test sample input identification model, obtain identification model
The corresponding test position of the image and object identity of output.
Step 406, identification model is tested.
As a kind of possible implementation, the corresponding test position of acquisition image of identification model output and acquisition are schemed
The relative positional relationship of picture is compared, if the relative positional relationship matching of the test position of model output and acquisition image,
Illustrate that identification model is accurately fixed higher, otherwise, illustrates that the recognition accuracy of identification model is poor, need to model further progress
Training, to improve the recognition accuracy of model.
As alternatively possible implementation, by the corresponding test position of acquisition image of identification model output and acquisition
The relative positional relationship of image is compared, and the mark of the object identity of identification model output and object to be identified is compared
It is right, if model output test position and acquisition image relative positional relationship matching, and model output object identity with to
The mark of identification object also matches, then illustrates that identification model is accurately fixed higher, otherwise, illustrate the recognition accuracy of identification model not
It is enough, it needs to the training of model further progress, to improve the recognition accuracy of model.
In the test method of the identification model of the embodiment of the present invention, in object to be identified along setting track moving process,
According to the mobile setting speed of driving device and direction, driving device is in the position of cableway when can calculate determining sensor acquisition image
Set, so that it is determined that object to be identified is along the position that is moved to of setting track, and then can determine sensor that fixed bit installs with
The relative position of object to be identified, by it is collected included the identification information of object to be identified and the image of relative positional relationship
As test sample, it is input to identification model, the object identity and test position of identification model output is obtained, determines object identity
Whether matched with object to be identified, and determine whether test position matches with relative positional relationship, determines the knowledge of identification model
Other accuracy realizes and does not need manually to be labeled test sample, reduces cost of labor.
In order to realize above-described embodiment, the present invention also proposes a kind of training device of identification model.
Fig. 5 is a kind of structural schematic diagram of the training device of identification model provided in an embodiment of the present invention.
As shown in figure 5, the device includes: control module 51, acquisition module 52, the first labeling module 53 and training module
54。
Control module 51 is moved for controlling object to be identified along setting track.
Acquisition module 52, for the biography that along setting track moving process, control fixed bit is installed in object to be identified
Sensor carries out Image Acquisition to object to be identified.
First labeling module 53 is labeled acquired image for the mark according to object to be identified.
Training module 54, for carrying out model training to identification model using the image after mark.
Further, in a kind of possible implementation of the embodiment of the present invention, the device further include: first determines mould
Block, obtains module and the second labeling module at the second determining module.
As a kind of possible implementation, driving device is for driving object to be identified, along the rail of cableway constraint
Mark is mobile with setting speed and direction.
First determining module, for determining the biography according to the mobile setting speed of the driving device and direction
The driving device is in the position of the cableway when sensor acquisition image；According to the driving device in the position of the cableway,
Determine the position that the object to be identified is moved to along the setting track.
Second determining module, the object to be identified is along the setting track when for according to sensor acquisition image
The position being moved to determines the relative positional relationship of the sensor Yu the object to be identified；
Module is obtained, for obtaining acquired image under a variety of relative positional relationships.
Second labeling module, for the relative positional relationship using the sensor and the object to be identified, to acquisition
To image be labeled.
As a kind of possible implementation, object to be identified is traffic marking board, and the traffic marking board is multiple, packet
The traffic marking board for being labeled with all kinds of traffic signs is included, and/or, the traffic marking board under various occlusion states, and/or, it is various
The traffic marking board being stained under state.
It should be noted that the aforementioned device that the embodiment is also applied for the explanation of training method embodiment, this
Place repeats no more.
In the training device of the identification model of the embodiment of the present invention, in object to be identified along setting track moving process,
According to the mobile setting speed of driving device and direction, driving device is in the position of cableway when can calculate determining sensor acquisition image
Set, so that it is determined that object to be identified is along the position that is moved to of setting track, and then can determine sensor that fixed bit installs with
The relative position of object to be identified, the mark of relationship and object to be identified in acquisition image, schemes acquisition depending on the relative position
As being labeled, using the image after mark as training sample, realize by the identification information of preset object to be identified with
Calculate determining relative positional relationship when acquiring image, acquisition image be labeled, manpower is saved, reduce manpower at
This, and the annotating efficiency of image is improved, sufficient amount of training sample can be provided, for model training so as to improve mould
The accuracy of type training.
In order to realize above-described embodiment, the present invention also proposes a kind of test device of identification model.
Fig. 6 is a kind of structural schematic diagram of the test device of identification model provided in an embodiment of the present invention.
As shown in fig. 6, the device includes: mobile module 61, processing module 62, input module 63 and the first test module
64。
Mobile module 61 is moved for controlling object to be identified along setting track.
Processing module 62, for the biography that along setting track moving process, control fixed bit is installed in object to be identified
Sensor carries out Image Acquisition to object to be identified.
Input module 63 obtains the object identity of identification model output for that will collect image input identification model.
Whether the first test module 64 surveys identification model for being matched according to object identity with object to be identified
Examination.
Further, in a kind of possible implementation of the embodiment of the present invention, the device further include: first determines mould
Block, obtains module and the second test module at the second determining module.
As a kind of possible implementation, driving device is constrained for driving the object to be identified along the cableway
Track it is mobile with setting speed and direction.
First determining module, for determining the biography according to the mobile setting speed of the driving device and direction
The driving device is in the position of the cableway when sensor acquisition image；According to the driving device in the position of the cableway,
Determine the position that the object to be identified is moved to along the setting track.
Second determining module, the object to be identified is along the setting track when for according to sensor acquisition image
The position being moved to determines the relative positional relationship of the sensor Yu the object to be identified.
Module is obtained, for obtaining acquired image under a variety of relative positional relationships.
Second test module, for obtaining the test position of the identification model output；According to the test position and institute
It states whether relative positional relationship matches, the identification model is tested.
As a kind of possible implementation, object to be identified is traffic marking board, and the traffic marking board is multiple, packet
The traffic marking board for being labeled with all kinds of traffic signs is included, and/or, the traffic marking board under various occlusion states, and/or, it is various
The traffic marking board being stained under state.
It should be noted that the aforementioned device that the embodiment is also applied for the explanation of test method embodiment, this
Place repeats no more.
In the test device of the identification model of the embodiment of the present invention, in object to be identified along setting track moving process,
According to the mobile setting speed of driving device and direction, driving device is in the position of cableway when can calculate determining sensor acquisition image
Set, so that it is determined that object to be identified is along the position that is moved to of setting track, and then can determine sensor that fixed bit installs with
The relative position of object to be identified, by it is collected included the identification information of object to be identified and the image of relative positional relationship
As test sample, it is input to identification model, the object identity and test position of identification model output is obtained, determines object identity
Whether matched with object to be identified, and determine whether test position matches with relative positional relationship, determines the knowledge of identification model
Other accuracy realizes and does not need manually to be labeled test sample, reduces cost of labor.
In order to realize above-described embodiment, the embodiment of the present invention proposes a kind of computer equipment, including memory, processor
And the computer program that can be run on a memory and on a processor is stored, when the processor executes described program, realize
Training method as described in aforementioned training method embodiment, alternatively, realizing the test side as described in aforementioned test method embodiment
Method.
In order to realize above-described embodiment, the embodiment of the present invention proposes a kind of non-transitorycomputer readable storage medium,
It is stored thereon with computer program, when which is executed by processor, realizes the training as described in aforementioned training method embodiment
Method, alternatively, realizing the test method as described in aforementioned test method embodiment.
Fig. 7 shows the block diagram for being suitable for the exemplary computer device for being used to realize the application embodiment.What Fig. 7 was shown
Computer equipment 12 is only an example, should not function to the embodiment of the present application and use scope bring any restrictions.
As shown in fig. 7, computer equipment 12 is showed in the form of universal computing device.The component of computer equipment 12 can be with
Including but not limited to: one or more processor or processing unit 16, system storage 28 connect different system components
The bus 18 of (including system storage 28 and processing unit 16).
Bus 18 indicates one of a few class bus structures or a variety of, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using any bus structures in a variety of bus structures.It lifts
For example, these architectures include but is not limited to industry standard architecture (Industry Standard
Architecture；Hereinafter referred to as: ISA) bus, microchannel architecture (Micro Channel Architecture；Below
Referred to as: MAC) bus, enhanced isa bus, Video Electronics Standards Association (Video Electronics Standards
Association；Hereinafter referred to as: VESA) local bus and peripheral component interconnection (Peripheral Component
Interconnection；Hereinafter referred to as: PCI) bus.
Computer equipment 12 typically comprises a variety of computer system readable media.These media can be it is any can be by
The usable medium that computer equipment 12 accesses, including volatile and non-volatile media, moveable and immovable medium.
Memory 28 may include the computer system readable media of form of volatile memory, such as random access memory
Device (Random Access Memory；Hereinafter referred to as: RAM) 30 and/or cache memory 32.Computer equipment 12 can be with
It further comprise other removable/nonremovable, volatile/non-volatile computer system storage mediums.Only as an example,
Storage system 34 can be used for reading and writing immovable, non-volatile magnetic media, and (Fig. 7 do not show, commonly referred to as " hard drive
Device ").Although being not shown in Fig. 7, the disk for reading and writing to removable non-volatile magnetic disk (such as " floppy disk ") can be provided and driven
Dynamic device, and to removable anonvolatile optical disk (such as: compact disc read-only memory (Compact Disc Read Only
Memory；Hereinafter referred to as: CD-ROM), digital multi CD-ROM (Digital Video Disc Read Only
Memory；Hereinafter referred to as: DVD-ROM) or other optical mediums) read-write CD drive.In these cases, each driving
Device can be connected by one or more data media interfaces with bus 18.Memory 28 may include that at least one program produces
Product, the program product have one group of (for example, at least one) program module, and it is each that these program modules are configured to perform the application
The function of embodiment.
Program/utility 40 with one group of (at least one) program module 42 can store in such as memory 28
In, such program module 42 include but is not limited to operating system, one or more application program, other program modules and
It may include the realization of network environment in program data, each of these examples or certain combination.Program module 42 is usual
Execute the function and/or method in embodiments described herein.
Computer equipment 12 can also be with one or more external equipments 14 (such as keyboard, sensing equipment, display 24
Deng) communication, can also be enabled a user to one or more equipment interact with the computer equipment 12 communicate, and/or with make
The computer equipment 12 any equipment (such as network interface card, the modulatedemodulate that can be communicated with one or more of the other calculating equipment
Adjust device etc.) communication.This communication can be carried out by input/output (I/O) interface 22.Also, computer equipment 12 may be used also
To pass through network adapter 20 and one or more network (such as local area network (Local Area Network；Hereinafter referred to as:
LAN), wide area network (Wide Area Network；Hereinafter referred to as: WAN) and/or public network, for example, internet) communication.Such as figure
Shown, network adapter 20 is communicated by bus 18 with other modules of computer equipment 12.It should be understood that although not showing in figure
Out, other hardware and/or software module can be used in conjunction with computer equipment 12, including but not limited to: microcode, device drives
Device, redundant processing unit, external disk drive array, RAID system, tape drive and data backup storage system etc..
Processing unit 16 by the program that is stored in system storage 28 of operation, thereby executing various function application and
Data processing, such as realize the method referred in previous embodiment.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
It can be combined in any suitable manner in one or more embodiment or examples.In addition, without conflicting with each other, the skill of this field
Art personnel can tie the feature of different embodiments or examples described in this specification and different embodiments or examples
It closes and combines.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include at least one this feature.In the description of the present invention, the meaning of " plurality " is at least two, such as two, three
It is a etc., unless otherwise specifically defined.
Any process described otherwise above or method description are construed as in flow chart or herein, and expression includes
It is one or more for realizing custom logic function or process the step of executable instruction code module, segment or portion
Point, and the range of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discussed suitable
Sequence, including according to related function by it is basic simultaneously in the way of or in the opposite order, Lai Zhihang function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
Expression or logic and/or step described otherwise above herein in flow charts, for example, being considered use
In the order list for the executable instruction for realizing logic function, may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (such as computer based system, including the system of processor or other can be held from instruction
The instruction fetch of row system, device or equipment and the system executed instruction) it uses, or combine these instruction execution systems, device or set
It is standby and use.For the purpose of this specification, " computer-readable medium ", which can be, any may include, stores, communicates, propagates or pass
Defeated program is for instruction execution system, device or equipment or the dress used in conjunction with these instruction execution systems, device or equipment
It sets.The more specific example (non-exhaustive list) of computer-readable medium include the following: there is the electricity of one or more wirings
Interconnecting piece (electronic device), portable computer diskette box (magnetic device), random access memory (RAM), read-only memory
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium can even is that the paper that can print described program on it or other are suitable
Medium, because can then be edited, be interpreted or when necessary with it for example by carrying out optical scanner to paper or other media
His suitable method is handled electronically to obtain described program, is then stored in computer storage.
It should be appreciated that each section of the invention can be realized with hardware, software, firmware or their combination.Above-mentioned
In embodiment, software that multiple steps or method can be executed in memory and by suitable instruction execution system with storage
Or firmware is realized.Such as, if realized with hardware in another embodiment, following skill well known in the art can be used
Any one of art or their combination are realized: have for data-signal is realized the logic gates of logic function from
Logic circuit is dissipated, the specific integrated circuit with suitable combinational logic gate circuit, programmable gate array (PGA), scene can compile
Journey gate array (FPGA) etc..
Those skilled in the art are understood that realize all or part of step that above-described embodiment method carries
It suddenly is that relevant hardware can be instructed to complete by program, the program can store in a kind of computer-readable storage medium
In matter, which when being executed, includes the steps that one or a combination set of embodiment of the method.
It, can also be in addition, each functional unit in each embodiment of the present invention can integrate in a processing module
It is that each unit physically exists alone, can also be integrated in two or more units in a module.Above-mentioned integrated mould
Block both can take the form of hardware realization, can also be realized in the form of software function module.The integrated module is such as
Fruit is realized and when sold or used as an independent product in the form of software function module, also can store in a computer
In read/write memory medium.
Storage medium mentioned above can be read-only memory, disk or CD etc..Although having been shown and retouching above
The embodiment of the present invention is stated, it is to be understood that above-described embodiment is exemplary, and should not be understood as to limit of the invention
System, those skilled in the art can be changed above-described embodiment, modify, replace and become within the scope of the invention
Type.
Claims (14)
1. a kind of training method of identification model, which is characterized in that the described method comprises the following steps:
Control object to be identified is moved along setting track；
In the object to be identified along the moving process of the setting track, the sensor that installs of control fixed bit to it is described to
Identify that object carries out Image Acquisition；
According to the mark of the object to be identified, acquired image is labeled；
Model training is carried out to identification model using the image after mark.
2. training method according to claim 1, which is characterized in that it is described using mark after image to identification model into
Before row model training, further includes:
The position that the object to be identified is moved to along the setting track when acquiring image according to the sensor, determine described in
The relative positional relationship of sensor and the object to be identified；
Obtain acquired image under a variety of relative positional relationships.
3. training method according to claim 2, which is characterized in that the driving device is described to be identified right for driving
As the track along cableway constraint is moved with setting speed and direction；Described in when the acquisition image according to the sensor
The position that object to be identified is moved to along the setting track, determines the relative position of the sensor Yu the object to be identified
Before relationship, further includes:
According to the mobile setting speed of the driving device and direction, the driving when sensor acquisition image is determined
Device is in the position of the cableway；
According to the driving device in the position of the cableway, determine what the object to be identified was moved to along the setting track
Position.
4. training method according to claim 2, which is characterized in that it is described using mark after image to identification model into
Before row model training, further includes:
Using the relative positional relationship of the sensor and the object to be identified, acquired image is labeled.
5. training method according to claim 1-4, which is characterized in that the object to be identified is traffic marking
Board, the traffic marking board are multiple, the traffic marking board including being labeled with all kinds of traffic signs, and/or, various occlusion states
Under traffic marking board, and/or, the various traffic marking boards being stained under state.
6. a kind of test method of identification model, which is characterized in that the described method comprises the following steps:
Control object to be identified is moved along setting track；
In the object to be identified along the moving process of the setting track, the sensor that installs of control fixed bit to it is described to
Identify that object carries out Image Acquisition；
Acquired image is inputted into identification model, obtains the object identity of the identification model output；
Whether matched according to the object identity with the object to be identified, the identification model is tested.
7. test method according to claim 6, which is characterized in that it is described to collect image input identification model, it obtains
Before the object identity exported to the identification model, further includes:
The position that the object to be identified is moved to along the setting track when acquiring image according to the sensor, determine described in
The relative positional relationship of sensor and the object to be identified；
Obtain acquired image under a variety of relative positional relationships.
8. test method according to claim 7, which is characterized in that the driving device is described to be identified right for driving
As the track along cableway constraint is moved with setting speed and direction；Described in when the acquisition image according to the sensor
The position that object to be identified is moved to along the setting track, determines the relative position of the sensor Yu the object to be identified
Before relationship, further includes:
According to the mobile setting speed of the driving device and direction, the driving when sensor acquisition image is determined
Device is in the position of the cableway；
According to the driving device in the position of the cableway, determine what the object to be identified was moved to along the setting track
Position.
9. test method according to claim 7, which is characterized in that it is described will collect image input identification model it
Afterwards, further includes:
Obtain the test position of the identification model output；
Whether matched according to the test position with the relative positional relationship, the identification model is tested.
10. according to the described in any item training methods of claim 6-9, which is characterized in that the object to be identified is traffic mark
Show that board, the traffic marking board are multiple, the traffic marking board including being labeled with all kinds of traffic signs, and/or, it is various to block shape
Traffic marking board under state, and/or, the various traffic marking boards being stained under state.
11. a kind of training device of identification model characterized by comprising
Control module is moved for controlling object to be identified along setting track；
Acquisition module, for along the moving process of the setting track, control fixed bit to be installed in the object to be identified
Sensor carries out Image Acquisition to the object to be identified；
First labeling module is labeled acquired image for the mark according to the object to be identified；
Training module, for carrying out model training to identification model using the image after mark.
12. a kind of test device of identification model characterized by comprising
Mobile module is moved for controlling object to be identified along setting track；
Processing module, for along the moving process of the setting track, control fixed bit to be installed in the object to be identified
Sensor carries out Image Acquisition to the object to be identified；
Input module obtains the object identity of the identification model output for that will collect image input identification model；
First test module, for whether being matched according to the object identity with the object to be identified, to the identification model
It is tested.
13. a kind of computer equipment, which is characterized in that including memory, processor and store on a memory and can handle
The computer program run on device when the processor executes described program, realizes such as instruction as claimed in any one of claims 1 to 5
Practice method, alternatively, realizing the test method as described in any in claim 6-10.
14. a kind of non-transitorycomputer readable storage medium, is stored thereon with computer program, which is characterized in that the program
Such as training method as claimed in any one of claims 1 to 5 is realized when being executed by processor, alternatively, realizing as in claim 6-10
Any test method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811508096.7A CN109711285B (en) | 2018-12-11 | 2018-12-11 | Training and testing method and device for recognition model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811508096.7A CN109711285B (en) | 2018-12-11 | 2018-12-11 | Training and testing method and device for recognition model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109711285A true CN109711285A (en) | 2019-05-03 |
| CN109711285B CN109711285B (en) | 2021-02-05 |
Family
ID=66256332
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811508096.7A Active CN109711285B (en) | 2018-12-11 | 2018-12-11 | Training and testing method and device for recognition model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109711285B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110288629A (en) * | 2019-06-24 | 2019-09-27 | 湖北亿咖通科技有限公司 | Target detection automatic marking method and device based on moving Object Detection |
| CN110490089A (en) * | 2019-07-29 | 2019-11-22 | 四川省视频电子有限责任公司 | A kind of image recognition technology of satellite receiving equipment |
| CN111325292A (en) * | 2020-03-11 | 2020-06-23 | 中国电子工程设计院有限公司 | Object behavior identification method and device |
| CN111862144A (en) * | 2020-07-01 | 2020-10-30 | 睿视智觉（厦门）科技有限公司 | Method and device for determining object movement track fraction |
| CN112200217A (en) * | 2020-09-09 | 2021-01-08 | 天津津航技术物理研究所 | Identification algorithm evaluation method and system based on infrared image big data |
| CN112987707A (en) * | 2019-11-29 | 2021-06-18 | 北京京东乾石科技有限公司 | Automatic driving control method and device for vehicle |
| CN113312442A (en) * | 2021-07-30 | 2021-08-27 | 景网技术有限公司 | Smart city electronic map generation method and system |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719499A (en) * | 2016-04-21 | 2016-06-29 | 百度在线网络技术（北京）有限公司 | Traffic sign recognition test method and device |
| CN105957145A (en) * | 2016-04-29 | 2016-09-21 | 百度在线网络技术（北京）有限公司 | Road barrier identification method and device |
| CN107451526A (en) * | 2017-06-09 | 2017-12-08 | 蔚来汽车有限公司 | The structure of map and its application |
| CN107577981A (en) * | 2016-07-04 | 2018-01-12 | 高德信息技术有限公司 | A kind of road traffic index identification method and device |
| US20180033144A1 (en) * | 2016-09-21 | 2018-02-01 | Realize, Inc. | Anomaly detection in volumetric images |
| US20180129912A1 (en) * | 2016-11-07 | 2018-05-10 | Nec Laboratories America, Inc. | System and Method for Learning Random-Walk Label Propagation for Weakly-Supervised Semantic Segmentation |
| US20180136332A1 (en) * | 2016-11-15 | 2018-05-17 | Wheego Electric Cars, Inc. | Method and system to annotate objects and determine distances to objects in an image |
| CN108764365A (en) * | 2018-06-06 | 2018-11-06 | 国网福建省电力有限公司厦门供电公司 | A kind of device signboard detection method |
| CN108776808A (en) * | 2018-05-25 | 2018-11-09 | 北京百度网讯科技有限公司 | A kind of method and apparatus for detecting ladle corrosion defect |
| CN109376664A (en) * | 2018-10-29 | 2019-02-22 | 百度在线网络技术（北京）有限公司 | Machine learning training method, device, server and medium |
-
2018
- 2018-12-11 CN CN201811508096.7A patent/CN109711285B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719499A (en) * | 2016-04-21 | 2016-06-29 | 百度在线网络技术（北京）有限公司 | Traffic sign recognition test method and device |
| CN105957145A (en) * | 2016-04-29 | 2016-09-21 | 百度在线网络技术（北京）有限公司 | Road barrier identification method and device |
| CN107577981A (en) * | 2016-07-04 | 2018-01-12 | 高德信息技术有限公司 | A kind of road traffic index identification method and device |
| US20180033144A1 (en) * | 2016-09-21 | 2018-02-01 | Realize, Inc. | Anomaly detection in volumetric images |
| US20180129912A1 (en) * | 2016-11-07 | 2018-05-10 | Nec Laboratories America, Inc. | System and Method for Learning Random-Walk Label Propagation for Weakly-Supervised Semantic Segmentation |
| WO2018085749A1 (en) * | 2016-11-07 | 2018-05-11 | Nec Laboratories America, Inc | System and method for learning random-walk label propagation for weakly-supervised semantic segmentation |
| US20180136332A1 (en) * | 2016-11-15 | 2018-05-17 | Wheego Electric Cars, Inc. | Method and system to annotate objects and determine distances to objects in an image |
| CN107451526A (en) * | 2017-06-09 | 2017-12-08 | 蔚来汽车有限公司 | The structure of map and its application |
| CN108776808A (en) * | 2018-05-25 | 2018-11-09 | 北京百度网讯科技有限公司 | A kind of method and apparatus for detecting ladle corrosion defect |
| CN108764365A (en) * | 2018-06-06 | 2018-11-06 | 国网福建省电力有限公司厦门供电公司 | A kind of device signboard detection method |
| CN109376664A (en) * | 2018-10-29 | 2019-02-22 | 百度在线网络技术（北京）有限公司 | Machine learning training method, device, server and medium |
Non-Patent Citations (2)
| Title |
|---|
| LOTFI ABDI等: "Deep learning traffic sign detection, recognition and augmentation", 《SAC2017: PROCEEDINGS OF THE SYMPOSIUM ON APPLIED COMPUTING》 * |
| 徐岩等: "一种改进的交通标志图像识别算法", 《激光与电子学进展》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110288629A (en) * | 2019-06-24 | 2019-09-27 | 湖北亿咖通科技有限公司 | Target detection automatic marking method and device based on moving Object Detection |
| CN110490089A (en) * | 2019-07-29 | 2019-11-22 | 四川省视频电子有限责任公司 | A kind of image recognition technology of satellite receiving equipment |
| CN110490089B (en) * | 2019-07-29 | 2023-04-07 | 四川省视频电子有限责任公司 | Image identification method of satellite receiving equipment |
| CN112987707A (en) * | 2019-11-29 | 2021-06-18 | 北京京东乾石科技有限公司 | Automatic driving control method and device for vehicle |
| CN111325292A (en) * | 2020-03-11 | 2020-06-23 | 中国电子工程设计院有限公司 | Object behavior identification method and device |
| CN111325292B (en) * | 2020-03-11 | 2023-05-02 | 中国电子工程设计院有限公司 | Object behavior recognition method and device |
| CN111862144A (en) * | 2020-07-01 | 2020-10-30 | 睿视智觉（厦门）科技有限公司 | Method and device for determining object movement track fraction |
| CN112200217A (en) * | 2020-09-09 | 2021-01-08 | 天津津航技术物理研究所 | Identification algorithm evaluation method and system based on infrared image big data |
| CN112200217B (en) * | 2020-09-09 | 2023-06-09 | 天津津航技术物理研究所 | Identification algorithm evaluation method and system based on infrared image big data |
| CN113312442A (en) * | 2021-07-30 | 2021-08-27 | 景网技术有限公司 | Smart city electronic map generation method and system |
| CN113312442B (en) * | 2021-07-30 | 2021-11-09 | 景网技术有限公司 | Smart city electronic map generation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109711285B (en) | 2021-02-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109711285A (en) | Training, test method and the device of identification model | |
| CN109871909A (en) | Image-recognizing method and device | |
| CN110377025A (en) | Sensor aggregation framework for automatic driving vehicle | |
| US20170109458A1 (en) | Testbed for lane boundary detection in virtual driving environment | |
| CN110276287A (en) | Method for detecting parking stalls, device, computer equipment and storage medium | |
| CN109491902A (en) | Interactive testing method, apparatus and system | |
| CN109871895A (en) | The defect inspection method and device of circuit board | |
| CN110045734A (en) | Method, apparatus and computer equipment are determined based on the parameters weighting of path planning | |
| CN109429518A (en) | Automatic Pilot traffic forecast based on map image | |
| CN110040159B (en) | Train parking control method and device and control equipment | |
| CN110599853B (en) | Intelligent teaching system and method for driving school | |
| CN107590301A (en) | Support virtual sensor data generation system and method for the exploitation beneficial to the algorithm of the traveling of railway intersection under the conditions of different weather | |
| CN109703465A (en) | The control method and device of vehicle-mounted imaging sensor | |
| CN109710658A (en) | Automatic Pilot evaluating method, device and equipment | |
| CN109710705A (en) | Map point of interest treating method and apparatus | |
| CN110647886A (en) | Interest point marking method and device, computer equipment and storage medium | |
| WO2019047597A1 (en) | Method and apparatus for recognizing lighting driving scene | |
| CN109977826A (en) | The classification recognition methods of object and device | |
| CN109376664A (en) | Machine learning training method, device, server and medium | |
| CN110084230B (en) | Image-based vehicle body direction detection method and device | |
| CN113497926A (en) | Intelligent correction of visual defects | |
| CN111158479A (en) | VR-based subject-two driving examination simulation exercise method and system | |
| CN109740632A (en) | Similarity model training method and device based on the more measurands of multisensor | |
| CN116311131A (en) | Intelligent driving-up enhancing method, system and device based on multi-view looking around | |
| CN106355329A (en) | Method and system for testing sign aligning ability of driver |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211011 Address after: 105 / F, building 1, No. 10, Shangdi 10th Street, Haidian District, Beijing 100085 Patentee after: Apollo Intelligent Technology (Beijing) Co.,Ltd. Address before: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing Patentee before: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) Co.,Ltd. |
|
| TR01 | Transfer of patent right |