CN111259818A - Road sign identification method, system and device - Google Patents
Road sign identification method, system and device Download PDFInfo
- Publication number
- CN111259818A CN111259818A CN202010056634.4A CN202010056634A CN111259818A CN 111259818 A CN111259818 A CN 111259818A CN 202010056634 A CN202010056634 A CN 202010056634A CN 111259818 A CN111259818 A CN 111259818A
- Authority
- CN
- China
- Prior art keywords
- road sign
- neural network
- network model
- convolutional neural
- picture
- 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.)
- Withdrawn
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 51
- 239000013598 vector Substances 0.000 claims abstract description 41
- 238000012549 training Methods 0.000 claims abstract description 11
- 230000000007 visual effect Effects 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 13
- 238000011176 pooling Methods 0.000 claims description 11
- 238000013475 authorization Methods 0.000 claims description 9
- 230000004913 activation Effects 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000008676 import Effects 0.000 claims description 3
- 238000002360 preparation method Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000006467 substitution reaction Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/582—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of traffic signs
-
- 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
-
- 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
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a road sign identification method, a system and a device, comprising the following steps: collecting pictures of the road sign in different environments, and storing the pictures into a data set; creating a convolutional neural network model using the data set; extracting the feature vectors of the road sign pictures in the data set by using the convolutional neural network model, and storing the extracted feature vectors into a feature vector set; training an extreme learning machine model by utilizing the feature vector set; and identifying the road sign picture to be identified by taking the convolutional neural network model and the extreme learning machine model as a combined model. The invention can accurately identify the road signs under various visual field environments, and can broadcast the identification result by voice, thereby improving the driving safety and providing technical support for unmanned driving.
Description
Technical Field
The invention relates to the technical field of traffic driving, in particular to a road sign identification method, a system, a terminal and a storage medium.
Background
The road traffic mark has rich road traffic related information, can provide driving assistance functions such as reminding and indicating for a driver at any time, reduces the driving pressure of the driver and the traffic dispersion pressure of the road to a certain extent, and reduces the occurrence probability of traffic accidents to a certain extent. However, the driver is only relied on to observe various traffic signs while driving continuously and make correct reflection, so that the observation and judgment pressure of the driver is increased to a great extent, and the driving fatigue is accelerated. Therefore, more automatic and intelligent recognition of road traffic identification images is more and more favored by research on automobile driving, and meanwhile, the research in the field is also an important technology for automatic unmanned driving development.
Disclosure of Invention
In view of the above-mentioned deficiencies of the prior art, the present invention provides a road sign recognition method, system and device to solve the above-mentioned technical problems.
In a first aspect, the present invention provides a road sign recognition method, including:
collecting pictures of the road sign in different environments, and storing the pictures into a data set;
creating a convolutional neural network model using the data set;
extracting the feature vectors of the road sign pictures in the data set by using the convolutional neural network model, and storing the extracted feature vectors into a feature vector set;
training an extreme learning machine model by utilizing the feature vector set;
and identifying the road sign picture to be identified by taking the convolutional neural network model and the extreme learning machine model as a combined model.
Further, prior to said storing the picture to the dataset, the method further comprises:
setting the size of the picture to 64 x 64;
marking the picture according to the road sign category of the picture;
dividing the pictures with the same mark into the same mark group;
the picture is converted to numpy type in python.
Further, the creating a convolutional neural network model using the data set includes:
setting the number of convolution and pooling layers of the convolution neural network model to be three;
setting the size of a convolution kernel of the convolution neural network model to be 5 x 5;
setting a pooling method of the convolutional neural network model as a maximum pooling method, and setting two full-connection layers;
setting each layer of activation function of the convolutional neural network model as a relu function;
and setting an output layer of the convolutional neural network model as a softmax function.
Further, the method further comprises:
obtaining interface calling authorization of a navigation system;
calling a data interface of a navigation system;
and importing the identification result of the road sign identification picture into the navigation system through the data interface.
Further, the method further comprises:
calculating the distance between the road sign and the vehicle through the road sign picture collected by the binocular stereo camera;
and converting the road sign recognition result and the distance into sound prompt information by using a voice prompt and outputting the sound prompt information.
In a second aspect, the present invention provides a landmark identifying system, including:
the data preparation unit is configured for acquiring pictures of the road sign in different environments and storing the pictures into a data set;
a first creation unit configured to create a convolutional neural network model using the data set;
the feature extraction unit is configured to extract feature vectors of the road sign pictures in the data set by using the convolutional neural network model, and store the extracted feature vectors into a feature vector set;
the second establishing unit is configured to train the extreme learning machine model by using the feature vector set;
and the identification application unit is configured to identify the road sign picture to be identified by taking the convolutional neural network model and the extreme learning machine model as a combined model.
Further, the system further comprises:
the authorization acquisition unit is configured for acquiring interface calling authorization of the navigation system;
the interface calling unit is used for calling a data interface of the navigation system;
and the result importing unit is configured to import the road sign identification picture identification result into the navigation system through the data interface.
Further, the system further comprises:
the distance calculation unit is configured for calculating the distance between the road sign and the vehicle through the road sign picture collected by the binocular stereo camera;
and the voice broadcasting unit is configured to convert the road sign recognition result and the distance into voice prompt information by using a voice prompt and output the voice prompt information.
In a third aspect, the present invention provides a landmark identifying apparatus, including:
the device comprises an image collector, a processor and a voice prompter, wherein the image collector is arranged on an automobile body, and the visual field direction of the image collector is consistent with the visual field direction of automobile driving; the voice lifter is arranged on the automobile driving dial plate base; the image collector and the voice prompter are both electrically connected with the processor; the image collector is a binocular stereo camera; the processor performs the method provided by the first aspect.
Further, the device also comprises a navigation terminal which is in wireless communication connection with the processor.
The beneficial effect of the invention is that,
the invention provides a road sign recognition method, a system, a terminal and a storage medium, which are characterized in that a convolutional neural network CNN is firstly built, then a large number of various traffic sign images shot under different climatic conditions and different illumination and angles are selected, the various traffic sign images are sent into the CNN for training after being processed, a trained model is stored, then the pictures are sent into the CNN again to obtain a characteristic vector, another extreme learning machine ELM machine learning method is selected for the extracted characteristics for training a classification model, finally a traffic sign recognition model is obtained, and a road sign image to be recognized is input into the recognition model to obtain a road sign recognition result. The invention can accurately identify the road signs under various visual field environments, and can broadcast the identification result by voice, thereby improving the driving safety and providing technical support for unmanned driving.
In addition, the invention has reliable design principle, simple structure and very wide application prospect.
Drawings
In order to more clearly illustrate the embodiments or technical solutions in the prior art of the present invention, the drawings used in the description of the embodiments or prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained based on these drawings without creative efforts.
FIG. 1 is a schematic flow diagram of a method of one embodiment of the invention.
FIG. 2 is a schematic block diagram of a system of one embodiment of the present invention.
Fig. 3 is a schematic structural diagram of an apparatus according to an embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solution of the present invention, the technical solution in the embodiment of the present invention will be clearly and completely described below with reference to the drawings in the embodiment of the present invention, and it is obvious that the described embodiment is only a part of the embodiment of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The following explains key terms appearing in the present invention.
The CNN model, Convolutional Neural Networks (CNN), is one of the most successful specific DNN cases, and is widely used for image recognition.
An ELM model, an Extreme Learning Machine (ELM), is a novel fast Learning algorithm, and for a single hidden layer neural network, the ELM can randomly initialize input weights and biases and obtain corresponding output weights.
FIG. 1 is a schematic flow diagram of a method of one embodiment of the invention. The implementation subject in fig. 1 may be a road sign recognition system.
As shown in fig. 1, the method 100 includes:
and 150, identifying the road sign picture to be identified by taking the convolutional neural network model and the extreme learning machine model as a combined model.
In order to facilitate understanding of the present invention, the principle of the landmark identifying method of the present invention is combined with the landmark identifying process in the embodiment to further describe the landmark identifying method provided by the present invention.
Specifically, the road sign identification method includes:
and S1, collecting pictures of the road signs in different environments, and storing the pictures into a data set.
The data set of the invention mainly has 6 types of images of traffic sign types, and the types are as follows: speed limit marking, indication marking, warning marking, prohibition release marking and other marks. In order to eliminate influences in the aspects of climate, shooting angle and the like and enable the recognition model to have better robustness and stability, 50000 pieces of image data are collected, and the data comprise various traffic sign images shot under different climate conditions, different angles, different illumination conditions, different resolution sizes and different sizes.
In order to train the model and predict more efficiently and conveniently, the invention processes all road traffic sign images into 64 × 64 images, uses 1-6 type labels respectively representing various traffic signs, and makes corresponding label sets and images to be put together. And converting the processed image data into a numpy type in python for storage, and providing for subsequent model training and testing.
And S2, creating a convolutional neural network model by using the data set.
The CNN (convolutional neural network model) is a neural network which is very popular at present, and is composed of a plurality of convolutional layers and pooling layers which are overlapped with each other, each layer contains neurons which are independent of each other, and the application in the field of image processing is very wide. In the invention, a CNN network is used, and the input of the input layer is the preprocessed traffic sign image with the size of 64 × 64. In the network structure of this embodiment there are three convolution and pooling layers, the convolution kernel size is chosen to be 5 x 5, the pooling method is chosen to be the maximum pooling method, followed by two fully connected layers h _ fc1 and h _ fc 2. The activation function used by each layer in the network is a relu function, and in order to prevent overfitting, in the invention, a dropout skill is added into the CNN network, the output layer of the network is processed by selecting a softmax function, the selected loss function is the cross entropy commonly used in multi-classification, and an Adam optimizer is selected for optimization. When the CNN model is trained, the preprocessed images are continuously sent into the network, so that the function loss value is smaller and smaller, the precision is higher and higher, and the training is stopped until the precision is stabilized to a fixed range and does not change any more, so that the well-trained CNN model is obtained. The specific training process of the CNN model is prior art and is not a protection of the present invention, and therefore will not be described in detail.
And S3, extracting the feature vectors of the road sign pictures in the data set by using the convolutional neural network model, and storing the extracted feature vectors into a feature vector set.
In this embodiment, the 256-dimensional feature vector output by the first fully-connected layer h _ fc1 in the CNN network model is stored, and this vector is a more comprehensive deep-level feature vector extracted from the image by the CNN model. These multi-dimensional feature vectors will then be used as training data for the late-phase ELM model. In this embodiment, all data in the data set of step S1 are converted into feature vectors by using the CNN network model trained in step S3, so as to obtain a feature vector set.
And S4, training the extreme learning machine model by using the feature vector set.
The extreme learner model may be expressed assuming that there are N arbitrary samples (X)i,ti) Wherein X isi＝[xi1,xi2,…,xin]T∈Rn，ti＝[ti1,ti2,…,tim]T∈Rm：
Wherein g (x) is an activation function, WiTo input the weights, βiAs output weights, biIs the offset of the ith hidden layer unit. Wi·XjRepresents WiAnd XjThe inner product of (d).
The goal of single-hidden-layer neural network learning is to minimize the error in the output, which can be expressed as
According to (1) and (2), it is possible to obtain:
Hβ＝T(3)
and (3) solving by using the characteristic vector set to obtain a required extreme-speed learning machine model, wherein H is an output matrix of the hidden layer node, β is an output weight, and T is an expected output.
And S5, taking the convolutional neural network model and the extreme learning machine model as a combined model to identify the road sign picture to be identified.
Selecting a traffic sign image to be identified for preprocessing, sending the image into a pre-trained CNN model, continuously performing operations such as convolution, pooling and the like by the CNN, finally taking a multi-dimensional feature vector output by a first full connection in a network as an input sample of an ELM model, performing classification identification by using the ELM model, and finally accurately judging the type of the traffic sign.
S6, establishing communication connection with the navigation terminal, obtaining interface calling authorization of the navigation system, calling a data interface of the navigation system, and importing the road sign identification picture identification result into the navigation system through the data interface. The navigation terminal is a mobile phone or vehicle-mounted navigation, and the navigation system is navigation software such as a high-grade map. And after the road sign recognition result is imported into a navigation system, the navigation system performs information gathering and voice broadcasting.
In other embodiments of the present invention, the distance between the road sign and the vehicle may also be directly calculated from road sign pictures collected by the binocular stereo camera, and the road sign recognition result and the distance may be converted into voice prompt information by using the voice prompt and output.
As shown in fig. 2, the system 200 includes:
the data preparation unit 210 is configured to collect pictures of the road sign in different environments, and store the pictures into a data set;
a first creating unit 220 configured to create a convolutional neural network model using the data set;
a feature extraction unit 230 configured to extract feature vectors of the road sign pictures in the data set by using the convolutional neural network model, and store the extracted feature vectors into a feature vector set;
a second creating unit 240 configured to train the extreme learning machine model using the feature vector set;
and the identification application unit 250 is configured to identify the road sign picture to be identified by taking the convolutional neural network model and the extreme learning machine model as a combined model.
Optionally, as an embodiment of the present invention, the system further includes:
the authorization acquisition unit is configured for acquiring interface calling authorization of the navigation system;
the interface calling unit is used for calling a data interface of the navigation system;
and the result importing unit is configured to import the road sign identification picture identification result into the navigation system through the data interface.
Optionally, as an embodiment of the present invention, the system further includes:
the distance calculation unit is configured for calculating the distance between the road sign and the vehicle through the road sign picture collected by the binocular stereo camera;
and the voice broadcasting unit is configured to convert the road sign recognition result and the distance into voice prompt information by using a voice prompt and output the voice prompt information.
As shown in fig. 3, the present embodiment provides a road sign recognition apparatus, including:
the device comprises an image collector, a processor and a voice prompter, wherein the image collector is arranged on an automobile body, and the visual field direction of the image collector is consistent with the visual field direction of automobile driving; the voice lifter is arranged on the automobile driving dial plate base; the image collector and the voice prompter are both electrically connected with the processor; the image collector is a binocular stereo camera.
Optionally, as an embodiment of the present invention, the apparatus further includes a navigation terminal, and the navigation terminal is connected to the processor in a wireless communication manner.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
Although the present invention has been described in detail by referring to the drawings in connection with the preferred embodiments, the present invention is not limited thereto. Various equivalent modifications or substitutions can be made on the embodiments of the present invention by those skilled in the art without departing from the spirit and scope of the present invention, and these modifications or substitutions are within the scope of the present invention/any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A road sign recognition method is characterized by comprising the following steps:
collecting pictures of the road sign in different environments, and storing the pictures into a data set;
creating a convolutional neural network model using the data set;
extracting the feature vectors of the road sign pictures in the data set by using the convolutional neural network model, and storing the extracted feature vectors into a feature vector set;
training an extreme learning machine model by utilizing the feature vector set;
and identifying the road sign picture to be identified by taking the convolutional neural network model and the extreme learning machine model as a combined model.
2. The method of claim 1, wherein prior to said storing the picture to the dataset, the method further comprises:
setting the size of the picture to 64 x 64;
marking the picture according to the road sign category of the picture;
dividing the pictures with the same mark into the same mark group;
the picture is converted to numpy type in python.
3. The method of claim 1, wherein creating a convolutional neural network model using the data set comprises:
setting the number of convolution and pooling layers of the convolution neural network model to be three;
setting the size of a convolution kernel of the convolution neural network model to be 5 x 5;
setting a pooling method of the convolutional neural network model as a maximum pooling method, and setting two full-connection layers;
setting each layer of activation function of the convolutional neural network model as a relu function;
and setting an output layer of the convolutional neural network model as a softmax function.
4. The method of claim 1, further comprising:
obtaining interface calling authorization of a navigation system;
calling a data interface of a navigation system;
and importing the identification result of the road sign identification picture into the navigation system through the data interface.
5. The method of claim 1, further comprising:
calculating the distance between the road sign and the vehicle through the road sign picture collected by the binocular stereo camera;
and converting the road sign recognition result and the distance into sound prompt information by using a voice prompt and outputting the sound prompt information.
6. A landmark identifying system, comprising:
the data preparation unit is configured for acquiring pictures of the road sign in different environments and storing the pictures into a data set;
a first creation unit configured to create a convolutional neural network model using the data set;
the feature extraction unit is configured to extract feature vectors of the road sign pictures in the data set by using the convolutional neural network model, and store the extracted feature vectors into a feature vector set;
the second establishing unit is configured to train the extreme learning machine model by using the feature vector set;
and the identification application unit is configured to identify the road sign picture to be identified by taking the convolutional neural network model and the extreme learning machine model as a combined model.
7. The system of claim 6, further comprising:
the authorization acquisition unit is configured for acquiring interface calling authorization of the navigation system;
the interface calling unit is used for calling a data interface of the navigation system;
and the result importing unit is configured to import the road sign identification picture identification result into the navigation system through the data interface.
8. The system of claim 6, further comprising:
the distance calculation unit is configured for calculating the distance between the road sign and the vehicle through the road sign picture collected by the binocular stereo camera;
and the voice broadcasting unit is configured to convert the road sign recognition result and the distance into voice prompt information by using a voice prompt and output the voice prompt information.
9. A road sign recognition apparatus, comprising:
the device comprises an image collector, a processor and a voice prompter, wherein the image collector is arranged on an automobile body, and the visual field direction of the image collector is consistent with the visual field direction of automobile driving; the voice lifter is arranged on the automobile driving dial plate base; the image collector and the voice prompter are both electrically connected with the processor; the image collector is a binocular stereo camera; the processor performing the method of any one of claims 1-5.
10. The apparatus of claim 1, further comprising a navigation terminal in wireless communication connection with the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010056634.4A CN111259818A (en) | 2020-01-18 | 2020-01-18 | Road sign identification method, system and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010056634.4A CN111259818A (en) | 2020-01-18 | 2020-01-18 | Road sign identification method, system and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111259818A true CN111259818A (en) | 2020-06-09 |
Family
ID=70948956
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010056634.4A Withdrawn CN111259818A (en) | 2020-01-18 | 2020-01-18 | Road sign identification method, system and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111259818A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112362070A (en) * | 2020-10-15 | 2021-02-12 | 武汉科技大学城市学院 | Vehicle navigation method and device |
| CN114973207A (en) * | 2022-08-01 | 2022-08-30 | 成都航空职业技术学院 | Road sign identification method based on target detection |
-
2020
- 2020-01-18 CN CN202010056634.4A patent/CN111259818A/en not_active Withdrawn
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112362070A (en) * | 2020-10-15 | 2021-02-12 | 武汉科技大学城市学院 | Vehicle navigation method and device |
| CN114973207A (en) * | 2022-08-01 | 2022-08-30 | 成都航空职业技术学院 | Road sign identification method based on target detection |
| CN114973207B (en) * | 2022-08-01 | 2022-10-21 | 成都航空职业技术学院 | Road sign identification method based on target detection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022083784A1 (en) | Road detection method based on internet of vehicles | |
| CN111563508A (en) | Semantic segmentation method based on spatial information fusion | |
| CN106650731B (en) | Robust license plate and vehicle logo recognition method | |
| CN110717493B (en) | License plate recognition method containing stacked characters based on deep learning | |
| CN109740479A (en) | A kind of vehicle recognition methods, device, equipment and readable storage medium storing program for executing again | |
| CN110991377B (en) | Front mesh identification method of automobile safety auxiliary system based on monocular vision neural network | |
| CN110532546B (en) | Automatic alarm issuing method fusing geographic position and text similarity | |
| CN112016467A (en) | Traffic sign recognition model training method, recognition method, system, device and medium | |
| CN111259818A (en) | Road sign identification method, system and device | |
| CN108323209A (en) | Information processing method, system, cloud processing device and computer program product | |
| CN112990065A (en) | Optimized YOLOv5 model-based vehicle classification detection method | |
| CN114913498A (en) | Parallel multi-scale feature aggregation lane line detection method based on key point estimation | |
| CN108304852B (en) | Method and device for determining road section type, storage medium and electronic device | |
| CN110263836B (en) | Bad driving state identification method based on multi-feature convolutional neural network | |
| CN112784754A (en) | Vehicle re-identification method, device, equipment and storage medium | |
| CN115661484A (en) | Lightweight tire pattern classification method and system | |
| CN114596548A (en) | Target detection method, target detection device, computer equipment and computer-readable storage medium | |
| CN116704490B (en) | License plate recognition method, license plate recognition device and computer equipment | |
| CN114022703A (en) | Efficient vehicle fine-grained identification method based on deep learning | |
| CN113052071A (en) | Method and system for rapidly detecting distraction behavior of driver of hazardous chemical substance transport vehicle | |
| CN112364864A (en) | License plate recognition method and device, electronic equipment and storage medium | |
| CN112288702A (en) | Road image detection method based on Internet of vehicles | |
| CN113743359B (en) | Vehicle re-identification method, model training method and related devices | |
| CN115861595A (en) | Multi-scale domain self-adaptive heterogeneous image matching method based on deep learning | |
| CN115273046A (en) | Driver behavior identification method for intelligent video analysis |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200609 |
|
| WW01 | Invention patent application withdrawn after publication |