CN113065399A - Traffic sign recognition system based on vehicle-mounted platform - Google Patents
Traffic sign recognition system based on vehicle-mounted platform Download PDFInfo
- Publication number
- CN113065399A CN113065399A CN202110241498.0A CN202110241498A CN113065399A CN 113065399 A CN113065399 A CN 113065399A CN 202110241498 A CN202110241498 A CN 202110241498A CN 113065399 A CN113065399 A CN 113065399A
- Authority
- CN
- China
- Prior art keywords
- traffic
- identification
- module
- traffic sign
- identifier
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013145 classification model Methods 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000010801 machine learning Methods 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 19
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 238000001514 detection method Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 7
- 238000000034 method Methods 0.000 claims description 5
- 238000003066 decision tree Methods 0.000 claims description 4
- 239000013598 vector Substances 0.000 claims description 3
- 206010039203 Road traffic accident Diseases 0.000 abstract description 6
- 230000000694 effects Effects 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/09—Recognition of logos
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Multimedia (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a traffic sign recognition system based on a vehicle-mounted platform, which comprises an acquisition module, a recognition module, an output module and a data processing module, wherein the acquisition module shoots video streams along a road along a vehicle; the identification module is used for identifying the video and the information acquired by the acquisition module and identifying a corresponding traffic identification; the output module is used for outputting the identified traffic identification to the vehicle-mounted platform; the data processing module is used for constructing and storing the existing traffic sign samples, establishing a sample feature set, and obtaining a classification model of the traffic sign through machine learning, so that the recognition module can conveniently recognize the traffic sign according to the existing samples. The traffic signs in the collected files are screened out by extracting the characteristics of the video files collected in the front of the road, and the screened traffic signs are subjected to traffic sign specific classification and content identification, so that the effect of accurately identifying the large categories and the specific contents of the traffic signs in real time is realized, the information of the road in front of a driver is timely reminded, and traffic accidents are avoided.
Description
Technical Field
The invention relates to the technical field of vehicle auxiliary driving, in particular to a traffic sign recognition system based on a vehicle-mounted platform.
Background
Road traffic accidents in China occur frequently, the road traffic accident fatality rate is high, when conditions such as traffic control, road construction transformation and road conditions are complex, drivers are prone to mistakenly enter forbidden areas, overspeed driving or front dangerous areas due to negligence or misjudgment of traffic signs, and serious consequences of traffic accidents are caused.
Disclosure of Invention
The invention aims to provide a traffic sign identification system based on a vehicle-mounted platform so as to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme: the traffic sign recognition system based on the vehicle-mounted platform comprises an acquisition module, a recognition module, an output module and a data processing module,
the acquisition module is used for shooting videos and images along the road so as to acquire traffic identification;
the identification module is used for identifying the video and the information acquired by the acquisition module and identifying a corresponding traffic identification;
the output module is used for outputting the identified traffic identification to the vehicle-mounted platform, so that the intelligent vehicle and a driver can conveniently take corresponding driving measures;
the data processing module is used for constructing and storing the existing traffic sign samples, establishing a sample feature set, obtaining a classification model of the traffic sign through machine learning, and facilitating the recognition module to recognize the traffic sign according to the existing samples.
In a preferred embodiment, the acquisition module is a high dynamic camera with a high definition camera, and the acquisition module is installed at the position of a rearview mirror of a vehicle, and the output of the acquisition module is a video sequence or picture with 640 x 480 pixels.
In a preferred embodiment, the identification module comprises image feature extraction, traffic identification category identification and traffic identification content identification.
In an embodiment, the image feature extraction process includes: decomposing a video sequence into single-frame images, carrying out filtering and denoising treatment on each sample image according to a median filtering algorithm to obtain filtered sample images, extracting the shape contour of the sample images according to an ellipse fitting algorithm and a polygon approximation algorithm, obtaining the contour region color of the shape contour, extracting traffic identification characteristics, and removing redundant characteristics with strong correlation.
In a preferred embodiment, the traffic identification category identification process includes: and identifying the image through the information obtained by extracting the image features by a multilayer decision tree, matching the similarity of the feature vectors and the classification model, and judging to obtain a detection result.
In a preferred embodiment, the detection result includes a red prohibition identifier, a yellow warning identifier and a blue indication identifier, where the red prohibition identifier includes a circular identifier and a triangular identifier, the yellow warning identifier is a triangular identifier, and the blue indication identifier includes a circular identifier and a rectangular identifier.
In an embodiment of the invention, the traffic identification content is identified as a traffic identification subdivision category, the traffic identification area is converted into a hexagonal pyramid model format, the number and the distribution of pixel points of each color in the identification area are counted according to the hue parameters of the pixel points, and the specific content of the identification is determined by combining the preset number and the distribution of the pixels of the specific identification in various traffic identifications.
In an embodiment, the data processing module comprises a neural network learning module and a storage module, wherein the neural network learning module is used for extracting and learning the characteristics of the existing traffic identification and training and learning the identification result of each identification module, so as to enrich the traffic identification classification model; the storage module is used for storing the traffic identification information, and the identification module can read and identify the traffic identification information conveniently.
In a preferred embodiment, the neural network learning module inputs the characteristics of the traffic identifier and outputs classification result information of the traffic identifier, which specifically includes:
the training sample gallery establishing module is used for acquiring a traffic signboard sample image and a non-traffic signboard sample image and establishing a training sample gallery;
the characteristic selection module is used for carrying out characteristic selection on the images in the training sample gallery by using Haar wavelet characteristics, MB-LBP characteristics and SURF characteristics;
and the comprehensive training module is used for inputting the selected features into the three-layer perceptron neural network to carry out comprehensive training of the traffic identification so as to obtain a traffic identification classification model.
Compared with the prior art, the invention has the beneficial effects that: according to the invention, a traffic sign classification model is established through learning of the neural network learning module, the traffic signs in the collected files are screened out by extracting the characteristics of the video files collected in front of the road, and then the screened traffic signs are subjected to traffic sign specific classification and content recognition, so that the effect of accurately recognizing the large categories and the specific content of the traffic signs in real time is realized, the driver is timely reminded of the road information in front, and the occurrence of traffic accidents is avoided.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments 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 invention provides a technical scheme that: the traffic sign recognition system based on the vehicle-mounted platform comprises an acquisition module, a recognition module, an output module and a data processing module,
the acquisition module is fixed on a vehicle, can shoot video streams along the road along the vehicle, and is convenient for acquiring traffic marks in front of the road in time;
the identification module is used for identifying the video and the information acquired by the acquisition module and identifying a corresponding traffic identification;
the output module is used for outputting the identified traffic identification to the vehicle-mounted platform, so that the intelligent vehicle and a driver can conveniently take corresponding driving measures;
the data processing module is used for constructing and storing the existing traffic sign samples, establishing a sample feature set, and obtaining a classification model of the traffic sign through machine learning, so that the recognition module can conveniently recognize the traffic sign according to the existing samples.
The acquisition module is for having the high dynamic camera of high definition camera, and the acquisition module installs in vehicle rearview mirror position department, and the acquisition module output is 640 x 480 pixel video sequence, and the acquisition module can last smooth shooting video 30 minutes at least, and video quality does not receive the electronic mobile device and generate heat, shake the influence of environmental factor such as shake to guarantee that the file is applicable to the analytic processing in later stage.
Further, the identification module comprises image feature extraction, traffic identification category identification and traffic identification content identification.
Further, the image feature extraction process is as follows: decomposing a video sequence into single-frame images, carrying out filtering and denoising processing on each sample image according to a median filtering algorithm to obtain a filtered sample image, extracting the shape contour of the sample image according to an ellipse fitting algorithm and a polygon approximation algorithm, obtaining the contour region color of the shape contour, extracting traffic identification characteristics, and removing redundant characteristics with strong correlation.
Further, the method comprises the steps of firstly carrying out filtering denoising processing on a sample image, then extracting edges of the regions by using an edge detection operator, further using algorithms such as ellipse fitting and polygon (trilateral and quadrilateral) approximation, combining constraint information such as area size and color to realize the detection and confirmation of the shape of the traffic sign, and carrying out rough classification (referring to a classification strategy) by using a multilayer decision tree according to the specific color and shape information of the traffic sign.
Further, the traffic identification category identification process is as follows: and identifying the image through the information obtained by extracting the image features by a multilayer decision tree, matching the similarity of the feature vectors and the classification model, and judging to obtain a detection result.
Further, the detection result comprises a red prohibition identifier, a yellow warning identifier and a blue indication identifier, wherein the red prohibition identifier comprises a circular identifier and a triangular identifier, the yellow warning identifier is a triangular identifier, the blue indication identifier comprises a circular identifier and a rectangular identifier, and the classification of the traffic identifiers is based on reference part 2 of road traffic identifiers and marking identifiers: road traffic signs (GB 5768.2-2009).
The traffic identification content is identified into traffic identification subdivision classification, the traffic identification area is converted into a hexagonal pyramid model format, the number and distribution of pixel points of each color in the identification area are counted according to the tone parameters of the pixel points, and the specific content of the identification is determined by combining the number and distribution of the preset pixels of the specific identification in various traffic identifications.
The data processing module comprises a neural network learning module and a storage module, wherein the neural network learning module is used for extracting and learning the characteristics of the existing traffic identification and training and learning the identification result of each identification module, so that a traffic identification classification model is enriched; the storage module is used for storing the traffic identification information, and the identification module can read and identify the traffic identification information conveniently.
Further, the neural network learning module inputs the characteristics of the traffic identification, and outputs the classification result information of the traffic identification, which specifically includes:
the training sample gallery establishing module is used for acquiring a traffic signboard sample image and a non-traffic signboard sample image and establishing a training sample gallery;
the characteristic selection module is used for selecting the characteristics of the images in the training sample gallery by using the Haar wavelet characteristics, the MB-LBP characteristics and the SURF characteristics;
and the comprehensive training module is used for inputting the selected features into the three-layer perceptron neural network to carry out comprehensive training of the traffic identification so as to obtain a traffic identification classification model.
The basic idea of the learning process is to select or construct an image feature which is beneficial to the target description of the attention type, and map a set of marked image samples to a feature space through a feature extraction algorithm to form a feature sample set; and then, taking the sample set as input, and carrying out supervision training on the corresponding classification model.
In summary, the invention establishes a traffic sign classification model through learning of the neural network learning module, screens out traffic signs in the collected files by extracting the characteristics of the video files collected in front of the road, and then carries out specific classification and content identification on the traffic signs after screening, thereby realizing the effect of accurately identifying the large categories and specific contents of the traffic signs in real time, reminding drivers of road information in front in time and avoiding traffic accidents.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. Traffic sign identification system based on vehicle platform, including collection module, identification module, output module and data processing module, its characterized in that:
the acquisition module is used for shooting videos and images along the road so as to acquire traffic identification;
the identification module is used for identifying the video and the information acquired by the acquisition module and identifying a corresponding traffic identification;
the output module is used for outputting the identified traffic identification to the vehicle-mounted platform, so that the intelligent vehicle and a driver can conveniently take corresponding driving measures;
the data processing module is used for constructing and storing the existing traffic sign samples, establishing a sample feature set, obtaining a classification model of the traffic sign through machine learning, and facilitating the recognition module to recognize the traffic sign according to the existing samples.
2. The vehicle platform based traffic sign recognition system of claim 1, wherein: the acquisition module is a high dynamic camera with a high definition camera, is arranged at the position of a vehicle rearview mirror, and outputs a video sequence or a picture with 640 multiplied by 480 pixels.
3. The vehicle platform based traffic sign recognition system of claim 1, wherein: the identification module comprises image feature extraction, traffic identification category identification and traffic identification content identification.
4. The vehicle platform based traffic sign recognition system of claim 3, wherein: the image feature extraction process comprises the following steps: decomposing a video sequence into single-frame images, carrying out filtering and denoising treatment on each sample image according to a median filtering algorithm to obtain filtered sample images, extracting the shape contour of the sample images according to an ellipse fitting algorithm and a polygon approximation algorithm, obtaining the contour region color of the shape contour, extracting traffic identification characteristics, and removing redundant characteristics with strong correlation.
5. The vehicle platform based traffic sign recognition system of claim 3, wherein: the traffic identification category identification process comprises the following steps: and identifying the image through the information obtained by extracting the image features by a multilayer decision tree, matching the similarity of the feature vectors and the classification model, and judging to obtain a detection result.
6. The vehicle platform based traffic sign recognition system of claim 5, wherein: the detection result comprises a red forbidden identifier, a yellow alarming identifier and a blue indicating identifier, wherein the red forbidden identifier comprises a circular identifier and a triangular identifier, the yellow alarming identifier is a triangular identifier, and the blue indicating identifier comprises a circular identifier and a rectangular identifier.
7. The vehicle platform based traffic sign recognition system of claim 3, wherein: the traffic identification content is identified into traffic identification fine classification, the traffic identification area is converted into a hexagonal pyramid model format, the number and distribution of pixel points of each color in the identification area are counted according to the tone parameters of the pixel points, and the specific content of the identification is determined by combining the preset pixel number and distribution of specific identifications in various traffic identifications.
8. The vehicle platform based traffic sign recognition system of claim 1, wherein: the data processing module comprises a neural network learning module and a storage module, wherein the neural network learning module is used for extracting and learning the characteristics of the existing traffic identification and training and learning the identification result of each identification module, so that a traffic identification classification model is enriched; the storage module is used for storing the traffic identification information, and the identification module can read and identify the traffic identification information conveniently.
9. The vehicle platform based traffic sign recognition system of claim 8, wherein: the neural network learning module inputs the characteristics of the traffic identification and outputs classification result information of the traffic identification, and the neural network learning module specifically comprises the following steps:
the training sample gallery establishing module is used for acquiring a traffic signboard sample image and a non-traffic signboard sample image and establishing a training sample gallery;
the characteristic selection module is used for carrying out characteristic selection on the images in the training sample gallery by using Haar wavelet characteristics, MB-LBP characteristics and SURF characteristics;
and the comprehensive training module is used for inputting the selected features into the three-layer perceptron neural network to carry out comprehensive training of the traffic identification so as to obtain a traffic identification classification model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110241498.0A CN113065399A (en) | 2021-03-04 | 2021-03-04 | Traffic sign recognition system based on vehicle-mounted platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110241498.0A CN113065399A (en) | 2021-03-04 | 2021-03-04 | Traffic sign recognition system based on vehicle-mounted platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113065399A true CN113065399A (en) | 2021-07-02 |
Family
ID=76559738
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110241498.0A Pending CN113065399A (en) | 2021-03-04 | 2021-03-04 | Traffic sign recognition system based on vehicle-mounted platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113065399A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113963543A (en) * | 2021-11-03 | 2022-01-21 | 中国矿业大学 | Method and system for identifying road danger of dangerous goods transportation tank car |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107679508A (en) * | 2017-10-17 | 2018-02-09 | 广州汽车集团股份有限公司 | Road traffic sign detection recognition methods, apparatus and system |
| CN111680605A (en) * | 2020-06-02 | 2020-09-18 | 深圳市豪恩汽车电子装备股份有限公司 | Traffic signal lamp identification system and method |
-
2021
- 2021-03-04 CN CN202110241498.0A patent/CN113065399A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107679508A (en) * | 2017-10-17 | 2018-02-09 | 广州汽车集团股份有限公司 | Road traffic sign detection recognition methods, apparatus and system |
| CN111680605A (en) * | 2020-06-02 | 2020-09-18 | 深圳市豪恩汽车电子装备股份有限公司 | Traffic signal lamp identification system and method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113963543A (en) * | 2021-11-03 | 2022-01-21 | 中国矿业大学 | Method and system for identifying road danger of dangerous goods transportation tank car |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105844257B (en) | The early warning system and method for road sign are missed based on machine vision travelling in fog day | |
| CN106469309B (en) | The method and apparatus of vehicle monitoring, processor, image capture device | |
| Omachi et al. | Detection of traffic light using structural information | |
| CN102375982B (en) | Multi-character characteristic fused license plate positioning method | |
| Saha et al. | License Plate localization from vehicle images: An edge based multi-stage approach | |
| CN103093249B (en) | A kind of taxi identification method based on HD video and system | |
| Sulaiman et al. | Development of automatic vehicle plate detection system | |
| CN103824081A (en) | Method for detecting rapid robustness traffic signs on outdoor bad illumination condition | |
| CN104463241A (en) | Vehicle type recognition method in intelligent transportation monitoring system | |
| CN102163278B (en) | Illegal vehicle intruding detection method for bus lane | |
| CN106570494A (en) | Traffic signal lamp recognition method and device based on convolution neural network | |
| CN108205649B (en) | Method and device for recognizing state of driver for calling and answering | |
| Le et al. | Real time traffic sign detection using color and shape-based features | |
| CN106778646A (en) | Model recognizing method and device based on convolutional neural networks | |
| Jiang et al. | Car plate recognition system | |
| Dehshibi et al. | Persian vehicle license plate recognition using multiclass Adaboost | |
| CN113269161A (en) | Traffic signboard detection method based on deep learning | |
| CN109858459A (en) | System and method based on police vehicle-mounted video element information realization intelligently parsing processing | |
| Do et al. | Speed limit traffic sign detection and recognition based on support vector machines | |
| CN115424217A (en) | AI vision-based intelligent vehicle identification method and device and electronic equipment | |
| CN113065399A (en) | Traffic sign recognition system based on vehicle-mounted platform | |
| Ismail | License plate Recognition for moving vehicles case: At night and under rain condition | |
| Deb et al. | Vehicle license plate detection algorithm based on color space and geometrical properties | |
| Liu et al. | Real-time speed limit sign detection and recognition from image sequences | |
| CN112071079B (en) | Machine vision vehicle high beam detection early warning system based on 5G transmission |
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 |