CN106980855B - Traffic sign rapid identification and positioning system and method - Google Patents
Traffic sign rapid identification and positioning system and method Download PDFInfo
- Publication number
- CN106980855B CN106980855B CN201710213257.9A CN201710213257A CN106980855B CN 106980855 B CN106980855 B CN 106980855B CN 201710213257 A CN201710213257 A CN 201710213257A CN 106980855 B CN106980855 B CN 106980855B
- Authority
- CN
- China
- Prior art keywords
- traffic sign
- traffic
- positioning
- image
- module
- 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.)
- Active
Links
Images
Classifications
-
- 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
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/245—Aligning, centring, orientation detection or correction of the image by locating a pattern; Special marks for positioning
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a traffic sign rapid identification and positioning system and a method thereof, comprising the following steps: the system comprises a video stream acquisition module, a video stream processing module and a video processing module, wherein the video stream acquisition module at least comprises a high-definition camera and a positioning module and is used for shooting videos along the road and positioning the shot videos; the traffic sign recognition module is used for recognizing the traffic signs in the road video shot by the video stream acquisition module and recording positioning information; and the traffic sign map positioning module is used for positioning the traffic signs identified by the traffic sign identification module on a map. The system also comprises a data post-processing module, wherein the data post-processing module is used for counting the number and the corresponding positions of the traffic signs along the road, and the map comprises video stream key frame interception and an overview map of the positions of the traffic signs on the electronic map. The invention can quickly and accurately position the traffic signs along the road and provides a tool for traffic accident investigation.
Description
Technical Field
The invention relates to a traffic sign rapid identification and positioning system and a method, belonging to the technical field of vehicle auxiliary driving.
Background
The road traffic accidents in China are frequent, and the fatality rate of the road traffic accidents is high. The data show that in 2014, the road traffic accidents reported together in China are 676 thousands of times, 196812 times related to casualties cause 58523 deaths, and the direct property loss is 10.8 million yuan. Wherein, accidents on the combined road section of a curve, a ramp and a bent slope account for more than 20 percent, and the death rate accounts for about 30 percent. The traffic sign facilities on the curved slope section are particularly important, the road condition of a driver can be prompted, and the accident occurrence frequency can be effectively reduced.
However, at present, the condition that the road traffic sign of a part of the curved road section is lost still exists, and serious traffic accidents are frequently caused. Investigation of the types and the number of traffic signs along the accident road section is an important investigation content in the investigation of the serious traffic accidents. In recent years, relevant researchers have fully recognized the importance of road traffic signs in road traffic safety, and have performed a series of analyses and studies on the identification and positioning of traffic signs by means of image processing, remote sensing, global positioning, pattern recognition, and the like.
Chinese patent application CN 102609702 a discloses a method and system for quickly positioning a road sign, which includes an acquisition unit, a segmentation unit, an area acquisition unit and a positioning unit. The method comprises the steps of firstly obtaining a road image, then dividing the obtained road image into an upper part image and a lower part image, then obtaining an effective candidate area of the upper part image by adopting a blue detection model based on an RGB color model, and finally carrying out horizontal long straight line detection on the obtained effective candidate area so as to position a direction mark. Because the patent application adopts the linear feature detection to replace the traditional rectangular feature detection, the time for positioning the road sign can be greatly shortened, and the robustness is greatly improved. The patent application is widely applied to the field of traffic sign identification as a rapid positioning method and a rapid positioning system for road direction signs. However, the patent application aims at a single image rather than a video stream file, manual operation is needed when photographing, the type of the traffic sign cannot be identified, and only the traffic sign can be identified. In addition, the positioning precision is not high by applying an image processing algorithm, and the actual requirement is difficult to meet.
Chinese patent application CN105718860 a provides a positioning method and system based on driving safety map and binocular traffic sign recognition, which primarily positions a driving vehicle by using a positioning system in a high-precision map; simultaneously, collecting images in front of the vehicle, and detecting and identifying the traffic signs in the images; and the coordinates of the traffic signs are identified and obtained on the high-precision map, the distance between the vehicle and the signs is measured, and the positions of the vehicle are calculated by comparing the coordinates of the traffic signs, so that the vehicle positioning is realized. This patent application has increased the collection of road traffic sign on traditional navigation data's basis, adopts the road sign to carry out the auxiliary action to the location of vehicle, through mark coordinate and size that left and right eyes camera discerned, carries out the distance calculation between vehicle and the traffic sign to the spatial position coordinate that has deposited the traffic sign in the high accuracy map calculates the position of vehicle, thereby provides sub-meter level's coordinate positioning, establishes the topological network that can be based on the lane. The precondition of this patent application is that the positioning coordinates of the traffic sign need to be known to confirm the coordinates of the vehicle, and the purpose of positioning the traffic sign is not achieved.
Chinese patent application CN 104361350 a provides a traffic sign recognition system, which is characterized in that: the identification system comprises a high-dynamic camera arranged at the position of a rearview mirror in a vehicle, and is used for identifying traffic marking lines, traffic signal lamps and traffic signs from road environment images and establishing corresponding space-time association models after collecting traffic identification information of a road front pavement; by adopting the structure and the method, the application of the patent combines the time and space relationship to establish the space-time association criterion of the traffic identification result, identifies various traffic identifications in the same image, fuses the identification results of the various traffic identifications to obtain a credible output result, and reduces the influence on the driving of the intelligent vehicle caused by the identification error of the traffic identification. The traffic sign recognition system is mainly used for building traffic sign space correlation information and commanding the intelligent vehicle to run. And the traffic sign can only identify its seven frames and not the specific kind.
The above three patent applications all relate to traffic sign identification, and the patent application CN 102609702 a and the patent application CN 105718860A also relate to positioning. At present, camera-based traffic sign recognition is only applicable to a single still photograph, and is not applicable to traffic sign recognition in a dynamic video stream file; in addition, the current road traffic sign can only identify the outline characteristics of the traffic sign, but cannot accurately distinguish the type of the traffic sign; finally, the positioning of the traffic signs is not accurate. At the present stage, further research is needed on the rapid identification and accurate positioning of the traffic signs so as to meet the requirements of identification and positioning of road traffic signs related to traffic accidents in China.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a system and a method for quickly identifying and positioning traffic signs, which can quickly and accurately identify the traffic signs along the road and position the traffic signs on an electronic map, and have good operability and precision.
According to the technical scheme provided by the invention, the traffic sign rapid identification and positioning system is characterized by comprising the following components:
the system comprises a video stream acquisition module, a video stream processing module and a video processing module, wherein the video stream acquisition module at least comprises a high-definition camera and a positioning module and is used for shooting videos along the road and positioning the shot videos;
the traffic sign recognition module is used for recognizing the traffic signs in the road video shot by the video stream acquisition module and recording positioning information;
and the traffic sign map positioning module is used for positioning the traffic signs identified by the traffic sign identification module on a map.
The system further comprises a data post-processing module, wherein the data post-processing module is used for counting the number of the road section traffic signs and deriving a traffic sign position distribution map.
Furthermore, the video stream acquisition module adopts an electronic mobile device with a high-definition camera and a positioning module.
Further, the working process of the traffic sign recognition module comprises the steps of recognizing the appearance of the traffic sign in the motion camera, recognizing the content of the traffic sign and classifying the recognized traffic sign.
Further, the process of identifying the traffic sign shape in the motion camera is as follows: firstly, establishing a sample image set according to the existing traffic signs, extracting samples of the traffic signs in the sample images, establishing a sample characteristic set, and obtaining a classification model of the traffic signs through machine learning; then, decomposing a video sequence image from the video stream, establishing a scanning window image set, acquiring the characteristics of the traffic sign from the video image, and obtaining a characteristic vector in the image; then, similarity matching is carried out on the feature vectors and the classification models, and a detection result is obtained through judgment; meanwhile, after the confirmed traffic sign feature vector is input into the classification model, the classification model feeds the feature vector back to the sample feature set, and the learning optimization classification model is continuously carried out;
further, the traffic sign map positioning module works as follows: the method comprises the steps of utilizing image frames of recognized traffic signs shot at different positions in a video stream to reconstruct a three-dimensional model which can be measured by the traffic signs in the video, calculating the coordinates from the center position of the three-dimensional traffic signs to the coordinate system of a camera where each image frame is located, combining the positioning information of the shot image frames to obtain the positioning information of the same traffic sign in each image frame, and determining the final positioning information of the traffic signs by utilizing the positioning information based on a least square algorithm.
The traffic sign rapid identification and positioning method is characterized by comprising the following steps:
step S1: shooting videos along a road and positioning the shot videos;
step S2: decomposing the video sequence obtained in the step S1 into single-frame images, stacking the single-frame images, and recording the positioning information corresponding to each stack image; taking out the image at the top of the stack, extracting the characteristic vector in the frame of image, analyzing the characteristic vector with the sample characteristic set in the classification model, detecting whether the frame of image has a traffic sign, if not, giving up the frame of image, and taking out a frame of image from the top of the stack again for repeated analysis; if yes, the traffic sign and the frame image are stacked newly, the stack is used for storing the image, and the new image is continuously taken out from the top of the image stack for analysis until all the frames containing the traffic sign are detected; n frames of traffic sign maps with local positioning information can be obtained, and the identified traffic signs are classified according to standards to provide data for subsequent data statistics; after completing the identification and classification of one traffic sign, adding 1 to the stack ID number for storing the detected traffic sign, and reentering the next detection process;
step S3: after the identification and classification of the traffic signs are completed, a three-dimensional model of the traffic signs can be constructed based on all n frames of images in a stack for storing the same traffic signs, and the positioning information of the traffic signs can be obtained by combining the positioning information of the n frames of images; at this time, the traffic sign should have n positioning information, and the optimal positioning information of the traffic sign is obtained by using the least square method described above.
Furthermore, the electronic map containing the traffic sign distribution can be output by combining the electronic map and the classified traffic sign types.
The invention has the advantages that: the invention can quickly and accurately position the traffic signs along the road and position the traffic signs to the electronic map, and the traffic sign positioning distribution map can visually display various signs along the road, thereby having good flexibility and operability in the investigation of road traffic accidents. The invention comprehensively applies technical methods such as image recognition, feature detection, navigation positioning, multi-view three-dimensional positioning and the like, has good operability and precision, and can provide an effective tool for investigation of traffic sign facilities of the very large traffic accident road.
Drawings
Fig. 1 is a block diagram of a traffic sign fast recognition and positioning system according to the present invention.
Fig. 2 is a schematic diagram of the operation of the traffic sign recognition module.
Fig. 3 is a schematic diagram of the operation of the traffic sign map location module.
Fig. 4 is a schematic diagram of an outputtable traffic sign distribution map.
Fig. 5 is a flow chart of the operation of the traffic sign fast identification and positioning system of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
As shown in fig. 1, the system for rapidly identifying and positioning a traffic sign of the present invention includes a video stream acquisition module, a traffic sign identification module, a traffic sign map positioning module, and a data post-processing module.
The video stream acquisition module is based on the electronic mobile device, installs the APP that the cooperation was used on the electronic mobile device, starts this APP, calls the camera of electronic mobile device from the area at the APP interface, fixes this electronic mobile device on the vehicle, can shoot the video stream on road along the way along the car, when shooting the video, records the locating information of every frame picture camera shooting instant local. The electronic mobile equipment can continuously and smoothly shoot the video for at least 30 minutes, and the video quality is not influenced by environmental factors such as heating and shaking of the electronic mobile equipment, so that the video stream file is ensured to be suitable for later analysis and processing.
The working process of the traffic sign identification module comprises the steps of identifying the appearance of the traffic sign in the motion camera, identifying the content of the traffic sign and classifying the identified traffic sign. The shot video stream is a high-definition video, so that after the traffic sign in the mobile video is identified, the content on the traffic sign in the image can be continuously identified, the identified traffic sign is classified, and for the traffic sign which cannot identify the content, the coarse classification can be carried out according to the outline shape of the identified traffic sign. The classification of road traffic signs is based on reference section 2 of road traffic signs and marking lines: road traffic signs (GB 5768.2-2009).
The traffic sign recognition module integrates an image processing algorithm, can recognize characteristic targets in the moving camera, and is based on the road traffic sign and the marking part 2: road traffic signs (GB5768.2-2009) are classified. The traffic sign recognition module adopts a moving target detection algorithm based on feature classification, the original moving target detection algorithm based on feature classification requires that a camera is fixed and a target to be detected moves. In this coordinate system, the camera is relatively stationary, and the road and roadside features are relatively moving, so a moving object detection algorithm based on feature classification can be used. As shown in fig. 2, the moving object detection based on feature classification includes two processes, i.e., a learning process and a decision process. 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 pattern recognition classifier to finally obtain a trained detection classifier. The basic idea of the decision making process is that all regions possibly containing the target of the attention type in the current image are listed firstly, then a trained detection classifier is used for quantifying the possibility of the target existing in the regions, and finally the output of the classifier is evaluated by using a decision strategy to realize the detection of the target. Two core points existing in the feature classification-based moving object detection are image features and a classification model, wherein the construction of the classification model is closely related to the dimension of a feature vector. In the invention, the traffic sign belongs to image features with smaller dimensions, and mainly comprises features such as a color histogram, a color moment, HOG, LBP and the like, so that a distance measurement decision method can be used, namely, an optimal linear decision threshold value is calculated by utilizing the intra-class distance of the features of a training target sample, and the distance between the features of the image to be decided and the average features of the target sample is compared, so that the detection of the target in a scene is realized.
After the identification of the traffic sign is completed, the positioning information of the traffic sign needs to be determined. The object in the two-dimensional image does not have the positioning information, but in the invention, when the camera shoots the video stream file, the positioning information of the camera is recorded when each frame of picture is shot. As shown in fig. 3, in the video stream, the same traffic sign is captured from different positions (position1, position2, … …, and position) and identified by the traffic sign identification module of the present invention, and the n frames of images all record the local positioning information. Therefore, a three-dimensional reconstruction method based on multiple views can be adopted, a three-dimensional model of the traffic sign in the image is constructed based on the n frames of images, the positioning information of the traffic sign is determined based on the position1, the positions 2, … … and the positions and the distances from the three-dimensional model to the positions 1, 2, … … and positions, the most accurate positioning information of the traffic sign is determined from the n positioning information by adopting a least square method, and a positioning formula 1 is shown.
In the formula (1), PsignFor locating traffic signs, PkAnd the positioning information corresponding to the traffic sign in the single frame image.
And after the identification and the positioning of the traffic sign are completed, the next data analysis and processing work can be carried out. One of the functions of the traffic sign rapid identification and positioning system is to draw a traffic sign distribution electronic map, so that the distribution information of the traffic signs is added to the original electronic map. As shown in fig. 4, the bottom map of the traffic sign distribution electronic map is an electronic map of a certain expressway, and the traffic sign is recognized as a warning sign of 'rain and fog weather slow-down and slow-moving' by combining the traffic sign recognition function and the positioning function of the traffic sign rapid recognition and positioning system of the invention, and the place is located at the K1386+700 section of the expressway. The traffic sign rapid identification and positioning system can complete the distribution and information display of all traffic signs on the road section.
The working flow of the traffic sign rapid identification and positioning system is shown in fig. 5, before the system starts to work, the electronic mobile equipment needs to be fixed at a certain position of a vehicle, so that a camera of the electronic mobile equipment is ensured to be free from shielding, videos along a road can be shot clearly and continuously, the endurance condition and the memory condition of the electronic mobile equipment are checked, and high-definition videos with the duration of more than 30 minutes can be shot and stored. After the preparation work is finished, the electronic mobile equipment is started, the special APP installed on the electronic mobile equipment is opened, the vehicle is started, and the video sequence files along the shooting road are set. And after the video shooting is finished, exporting the video file to a workstation, wherein the workstation is integrated with an image processing algorithm. Firstly, a video sequence can be decomposed into single-frame images, stacking is carried out, and local positioning information corresponding to each stack image is recorded. Taking out the image at the top of the stack, extracting the characteristic vector in the frame of image, analyzing the characteristic vector with the sample characteristic set in the classification model, detecting whether the frame of image has a traffic sign, if not, giving up the frame of image, and taking out a frame of image from the top of the stack again for repeated analysis; if so, the traffic sign and the frame image are newly stacked, the stack is used for storing the image, and the new image is continuously taken out from the top of the image stack for analysis until all the frames containing the traffic sign are detected. The n frames of traffic sign maps with the local positioning information can be obtained, and the identified traffic signs are classified according to the standard to provide data for subsequent data statistics. After completing the identification and classification of a traffic sign, the stack ID number for storing the detected traffic sign is increased by 1, and the next detection process is entered again. After the identification and classification of the traffic signs are completed, a three-dimensional model of the traffic signs can be constructed based on all n frames of images in a stack storing the same traffic signs, and the positioning information of the traffic signs can be obtained by combining the positioning information of the n frames of images. At this time, the traffic sign should have n positioning information, and the optimal positioning information of the traffic sign is obtained by using the least square method described above. And combining the electronic map with the classified traffic sign types to output the electronic map containing the traffic sign distribution. Meanwhile, the traffic signs can be counted, and whether the traffic signs meet the relevant road standards and specifications or not is analyzed.
Claims (6)
1. A traffic sign rapid identification and positioning system is characterized by comprising:
the system comprises a video stream acquisition module, a video stream processing module and a video processing module, wherein the video stream acquisition module at least comprises a high-definition camera and a positioning module and is used for shooting videos along the road and positioning the shot videos;
the traffic sign recognition module is used for recognizing the traffic signs in the road video shot by the video stream acquisition module and recording positioning information;
the traffic sign map positioning module is used for positioning the traffic signs identified by the traffic sign identification module on a map;
the system also comprises a data post-processing module, wherein the data post-processing module is used for counting the number of the road traffic signs and deriving a traffic sign position distribution map;
the working process of the traffic sign map positioning module is as follows: the method comprises the steps of utilizing image frames of recognized traffic signs shot at different positions in a video stream to reconstruct a three-dimensional model which can be measured by the traffic signs in the video, calculating the coordinates from the center position of the three-dimensional traffic signs to the coordinate system of a camera where each image frame is located, combining the positioning information of the shot image frames to obtain the positioning information of the same traffic sign in each image frame, and determining the final positioning information of the traffic signs by utilizing the positioning information based on a least square algorithm.
2. The traffic sign rapid identification and positioning system according to claim 1, wherein: the video stream acquisition module adopts an electronic mobile device with a high-definition camera and a positioning module.
3. The traffic sign rapid identification and positioning system according to claim 1, wherein: the working process of the traffic sign identification module comprises the steps of identifying the appearance of the traffic sign in the motion camera, identifying the content of the traffic sign and classifying the identified traffic sign.
4. A traffic sign rapid identification and positioning system as claimed in claim 3, wherein: the process of identifying the appearance of the traffic sign in the motion camera comprises the following steps: firstly, establishing a sample image set according to the existing traffic signs, extracting samples of the traffic signs in the sample images, establishing a sample characteristic set, and obtaining a classification model of the traffic signs through machine learning; then, decomposing a video sequence image from the video stream, establishing a scanning window image set, acquiring the characteristics of the traffic sign from the video image, and obtaining a characteristic vector in the image; then, similarity matching is carried out on the feature vectors and the classification models, and a detection result is obtained through judgment; meanwhile, after the confirmed traffic sign feature vector is input into the classification model, the classification model feeds the feature vector back to the sample feature set, and the learning optimization classification model is continuously carried out.
5. A traffic sign rapid identification and positioning method is characterized by comprising the following steps:
step S1: shooting videos along a road and positioning the shot videos;
step S2: decomposing the video sequence obtained in the step S1 into single-frame images, stacking the single-frame images, and recording the positioning information corresponding to each stack image; taking out the image at the top of the stack, extracting the characteristic vector in the frame of image, analyzing the characteristic vector with the sample characteristic set in the classification model, detecting whether the frame of image has a traffic sign, if not, giving up the frame of image, and taking out a frame of image from the top of the stack again for repeated analysis; if yes, the traffic sign and the frame image are stacked newly, the stack is used for storing the image, and the new image is continuously taken out from the top of the image stack for analysis until all the frames containing the traffic sign are detected; n frames of traffic sign maps with local positioning information can be obtained, and the identified traffic signs are classified according to standards to provide data for subsequent data statistics; after completing the identification and classification of one traffic sign, adding 1 to the stack ID number for storing the detected traffic sign, and reentering the next detection process;
step S3: after the identification and classification of the traffic signs are completed, a three-dimensional model of the traffic signs can be constructed based on all n frames of images in a stack for storing the same traffic signs, and the positioning information of the traffic signs can be obtained by combining the positioning information of the n frames of images; at this time, the traffic sign should have n positioning information, and the optimal positioning information of the traffic sign is obtained by using the least square method.
6. The traffic sign rapid identification and positioning method according to claim 5, characterized in that: and combining the electronic map with the classified traffic sign types to output the electronic map containing the traffic sign distribution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710213257.9A CN106980855B (en) | 2017-04-01 | 2017-04-01 | Traffic sign rapid identification and positioning system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710213257.9A CN106980855B (en) | 2017-04-01 | 2017-04-01 | Traffic sign rapid identification and positioning system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106980855A CN106980855A (en) | 2017-07-25 |
| CN106980855B true CN106980855B (en) | 2020-04-17 |
Family
ID=59345077
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710213257.9A Active CN106980855B (en) | 2017-04-01 | 2017-04-01 | Traffic sign rapid identification and positioning system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106980855B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109029339B (en) * | 2018-05-09 | 2023-10-13 | 苏州天瞳威视电子科技有限公司 | Traffic sign distance measurement method and device based on vision |
| CN109215487A (en) * | 2018-08-24 | 2019-01-15 | 宽凳（北京）科技有限公司 | A kind of high-precision cartography method based on deep learning |
| CN109829401A (en) * | 2019-01-21 | 2019-05-31 | 深圳市能信安科技股份有限公司 | Traffic sign recognition method and device based on double capture apparatus |
| CN110501018B (en) * | 2019-08-13 | 2021-11-02 | 广东星舆科技有限公司 | Traffic sign information acquisition method for high-precision map production |
| CN110889378B (en) * | 2019-11-28 | 2023-06-09 | 湖南率为控制科技有限公司 | Multi-view fusion traffic sign detection and identification method and system thereof |
| CN112926575A (en) * | 2021-01-22 | 2021-06-08 | 北京嘀嘀无限科技发展有限公司 | Traffic accident recognition method, device, electronic device and medium |
| CN113255578B (en) * | 2021-06-18 | 2022-04-29 | 亿咖通(湖北)技术有限公司 | Traffic identification recognition method and device, electronic equipment and storage medium |
| CN115965926B (en) * | 2023-03-16 | 2023-06-02 | 四川京炜数字科技有限公司 | Vehicle-mounted road sign marking inspection system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010041584A1 (en) * | 2008-10-10 | 2010-04-15 | Kabushiki Kaisha Toshiba | Imaging system and method |
| KR20130106106A (en) * | 2012-03-19 | 2013-09-27 | 현대모비스 주식회사 | Appratus and method for judgment 3 dimension |
| CN103743383A (en) * | 2014-02-11 | 2014-04-23 | 天津市星际空间地理信息工程有限公司 | Automatic extraction method for road information based on point cloud |
| CN104463935A (en) * | 2014-11-11 | 2015-03-25 | 中国电子科技集团公司第二十九研究所 | Lane rebuilding method and system used for traffic accident restoring |
| CN104517317A (en) * | 2015-01-08 | 2015-04-15 | 东华大学 | Three-dimensional reconstruction method of vehicle-borne infrared images |
| US9146125B2 (en) * | 2012-06-05 | 2015-09-29 | Apple Inc. | Navigation application with adaptive display of graphical directional indicators |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345630B (en) * | 2013-06-14 | 2016-08-17 | 合肥工业大学 | A kind of traffic signs localization method based on spherical panoramic video |
| CN103971087B (en) * | 2013-07-12 | 2017-04-19 | 湖南纽思曼导航定位科技有限公司 | Method and device for searching and recognizing traffic signs in real time |
| CN103925927B (en) * | 2014-04-18 | 2016-09-07 | 中国科学院软件研究所 | A kind of traffic mark localization method based on Vehicular video |
| CN103971126B (en) * | 2014-05-12 | 2017-08-08 | 百度在线网络技术（北京）有限公司 | A kind of traffic sign recognition method and device |
| CN105718860B (en) * | 2016-01-15 | 2019-09-10 | 武汉光庭科技有限公司 | Localization method and system based on driving safety map and binocular Traffic Sign Recognition |
| CN105702152A (en) * | 2016-04-28 | 2016-06-22 | 百度在线网络技术（北京）有限公司 | Map generation method and device |
| CN106204800B (en) * | 2016-07-06 | 2018-08-07 | 福州瑞芯微电子股份有限公司 | The method, apparatus and automobile data recorder that automatic traffic landmark identification is reminded |
| CN106326858A (en) * | 2016-08-23 | 2017-01-11 | 北京航空航天大学 | Road traffic sign automatic identification and management system based on deep learning |
| CN106372577A (en) * | 2016-08-23 | 2017-02-01 | 北京航空航天大学 | Deep learning-based traffic sign automatic identifying and marking method |
-
2017
- 2017-04-01 CN CN201710213257.9A patent/CN106980855B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010041584A1 (en) * | 2008-10-10 | 2010-04-15 | Kabushiki Kaisha Toshiba | Imaging system and method |
| KR20130106106A (en) * | 2012-03-19 | 2013-09-27 | 현대모비스 주식회사 | Appratus and method for judgment 3 dimension |
| US9146125B2 (en) * | 2012-06-05 | 2015-09-29 | Apple Inc. | Navigation application with adaptive display of graphical directional indicators |
| CN103743383A (en) * | 2014-02-11 | 2014-04-23 | 天津市星际空间地理信息工程有限公司 | Automatic extraction method for road information based on point cloud |
| CN104463935A (en) * | 2014-11-11 | 2015-03-25 | 中国电子科技集团公司第二十九研究所 | Lane rebuilding method and system used for traffic accident restoring |
| CN104517317A (en) * | 2015-01-08 | 2015-04-15 | 东华大学 | Three-dimensional reconstruction method of vehicle-borne infrared images |
Non-Patent Citations (4)
| Title |
|---|
| Computing object-based saliency in urban scenes using laser sensing;Yipu Zhao et al;《 2012 IEEE International Conference on Robotics and Automation》;20120518;第4436-4443页 * |
| Traffic sign detection, state estimation, and identification using onboard sensors;Anh Vu et al;《16th International IEEE Conference on Intelligent Transportation Systems (ITSC 2013)》;20131009;第875-880页 * |
| 序列图像中交通标志定位检测技术研究;肖学钢;《中国优秀硕士学位论文全文数据库信息科技辑》;20090115;第I138-1052页 * |
| 道路交通标线检测识别与建图方法研究;陈放;《中国优秀硕士学位论文全文数据库信息科技辑》;20120715;第I138-1775页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106980855A (en) | 2017-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106980855B (en) | Traffic sign rapid identification and positioning system and method | |
| CN106709436B (en) | Track traffic panoramic monitoring-oriented cross-camera suspicious pedestrian target tracking system | |
| CN105488454B (en) | Front vehicles detection and ranging based on monocular vision | |
| CN110619279B (en) | Road traffic sign instance segmentation method based on tracking | |
| CN105046196B (en) | Front truck information of vehicles structuring output method based on concatenated convolutional neutral net | |
| De Charette et al. | Real time visual traffic lights recognition based on spot light detection and adaptive traffic lights templates | |
| CN100545867C (en) | Aerial shooting traffic video frequency vehicle rapid checking method | |
| CN111429484B (en) | Multi-target vehicle track real-time construction method based on traffic monitoring video | |
| CN104239867B (en) | License plate locating method and system | |
| CN110298300B (en) | Method for detecting vehicle illegal line pressing | |
| CN104978567B (en) | Vehicle checking method based on scene classification | |
| CN109101924A (en) | A kind of pavement marking recognition methods based on machine learning | |
| CN107016362B (en) | Vehicle weight recognition method and system based on vehicle front windshield pasted mark | |
| CN112825192B (en) | Object identification system and method based on machine learning | |
| Chen | Nighttime vehicle light detection on a moving vehicle using image segmentation and analysis techniques | |
| CN103268489A (en) | Motor vehicle plate identification method based on sliding window searching | |
| CN104102909B (en) | Vehicle characteristics positioning and matching process based on lenticular information | |
| CN111915583B (en) | Vehicle and pedestrian detection method based on vehicle-mounted thermal infrared imager in complex scene | |
| CN104615986A (en) | Method for utilizing multiple detectors to conduct pedestrian detection on video images of scene change | |
| CN111781600A (en) | Vehicle queuing length detection method suitable for signalized intersection scene | |
| CN114170580A (en) | Highway-oriented abnormal event detection method | |
| CN107506753B (en) | Multi-vehicle tracking method for dynamic video monitoring | |
| Fusek et al. | Adaboost for parking lot occupation detection | |
| CN110675442A (en) | Local stereo matching method and system combined with target identification technology | |
| Coronado et al. | Detection and classification of road signs for automatic inventory systems using computer vision |
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 |