CN104134067A - Road vehicle monitoring system based on intelligent visual Internet of Things - Google Patents
Road vehicle monitoring system based on intelligent visual Internet of Things Download PDFInfo
- Publication number
- CN104134067A CN104134067A CN201410319408.5A CN201410319408A CN104134067A CN 104134067 A CN104134067 A CN 104134067A CN 201410319408 A CN201410319408 A CN 201410319408A CN 104134067 A CN104134067 A CN 104134067A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- car
- intelligent
- monitoring system
- system based
- 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
Landscapes
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a road vehicle monitoring system based on an intelligent visual Internet of Things. Through a method that a visual label of a vehicle in a city road is extracted and a visual label library of the vehicle is established, an intelligent monitoring function of road vehicle track is realized. The road vehicle monitoring system mainly consists of a plurality of intelligent visual sensing nodes, a central server and a central database, wherein the intelligent visual sensing nodes are used for extracting the license plate number, the vehicle logo, the vehicle type and the vehicle body color of a road vehicle as a visual label, sending the visual label to the central server to be analyzed and processed, and storing the analyzed and processed visual label into the central database for PC client software to carry out vehicle monitoring and processing. The intelligent visual sensing nodes mainly take an embedded processor Cortex-A8 as a core, a USB high definition camera is externally connected, and a vehicle label extraction algorithm based on an Open CV (Computer Vision) library is operated. The invention exhibits strong stability and instantaneity and high application value.
Description
Technical field
The invention belongs to technical field of image processing, relate to the intelligent control method of road vehicle whereabouts, specifically a kind of supervisory system that realizes road vehicle by extracting the vision label of vehicle in urban road.
Background technology
Intelligent vision Internet of Things (IVIOT) is the important component part of generation information technology, is also the upgraded version of Internet of Things.Intelligent vision Internet of Things is to analyze perception people, car, thing by vision sensor, communication, intelligent vision, by the agreement of agreement, any object is connected with internet, carry out message exchange and communicate by letter, realize with this kind of intelligent network that Intelligent Recognition, location to object are followed the tracks of and monitored in real time." the intelligent vision Internet of Things " built by terminal users such as public place management, intelligent building, traffic control, school, hospital, prison, greenhouse, finance, military affairs, community, individual video equipment, can realize unified monitoring, management and scheduling to social resources.Therefore, intelligent vision technology of Internet of things is with a wide range of applications.
Traditional road vehicles monitoring system need to all be installed each car GPS or RFID radio-frequency card, so not only waste time and energy, and this scheme requires every car owner all to install, implement very difficult, also keep away simultaneously unavoidable car owner deliberately close or damage positioning equipment with escape follow the tracks of may.The vision label that road vehicles monitoring system based on intelligent vision Internet of Things can be contactless, interference-free obtains road vehicle, and and central server carry out exchanges data analysis, the whereabouts track of monitoring road vehicle that can be real-time.There are to very large effect in traffic department and public security department etc., there is very important theory value and realistic meaning.
Summary of the invention
The present invention is directed to the demand of road vehicles monitoring system, designed a kind of road vehicles monitoring system based on intelligent vision Internet of Things, implement easily, monitoring is stable.
Technical solution of the present invention is to provide a kind of road vehicles monitoring system based on intelligent vision Internet of Things, it is characterized in that: comprising:
Multiple intelligent vision sensing nodes that formed by video camera and flush bonding processor Cortex-A8,
Be used for carrying out with intelligent vision sensing node the central server of information interaction,
For storing the central database of vision label data,
Intelligent sensor node is connected with central server by Ethernet,
PC client computer can be connected with central server by Ethernet, and the traffic video of addressable each intelligent sensor node stream.
The aforesaid road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: described intelligent vision sensing node is laid many as much as possible on road.
The aforesaid road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: described flush bonding processor Cortex-A8 is connected with video camera by USB interface, carries out the transmission of video flowing.
The aforesaid road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: described flush bonding processor Cortex-A8 loads linux system, and operational vehicle vision tag extraction algorithm.
The aforesaid road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: described intelligent sensor node carries out the transmission of video flowing by the project MJPG-streamer that increases income and increase special Socket thread carrying out the transmission of vehicle vision label.
The aforesaid road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: described vehicle vision tag extraction algorithm, comprises number-plate number recognizer, car mark recognizer, vehicle targets and body color recognizer.
The aforesaid road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: described number-plate number recognizer, and step is:
1) use V4L2 to drive interface to obtain video flowing;
2) use Haar-like feature detection and be partitioned into the general location of vehicle;
3) algorithm that uses discrete wavelet analysis, morphological dilations processing and colouring information etc. to combine carries out car plate location;
4) using the eliminating of morphology connected domain and car plate prior imformation to combine carries out cutting apart of characters on license plate;
5) use CNN convolutional neural networks sorter to identify characters on license plate;
The aforesaid road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: described car mark recognizer, and step is:
1) according to the car plate position of aforementioned location, get upwardly extending 5 car plate height, width is car plate width, using this region as car target coarse positioning result;
2) by after the canny rim detection of this region, carry out horizontal projection, get that wherein one section of continuous intensive low frequency region is as car target up-and-down boundary, width is car plate width, this region is as the final positioning result of car target;
3) in this region, use car mark template to carry out sliding window template matches, each class car mark template all can be returned to a matching value, gets the car mark template of optimum matching as final recognition result.
The aforesaid road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: described vehicle targets, and step is:
1) use Haar-like feature detection and be partitioned into the cardinal principle estimated position of vehicle;
2) search connected domain and calculate its minimum boundary rectangle;
3) pass through camera calibration, calculate the actual (tube) length roomy little (unit is rice) of minimum boundary rectangle, the vehicle that minimum boundary rectangle length is less than to 6 meters is judged to be compact car, otherwise be judged to be large car (owing to having a lot of vehicles by automotive type segmentation, being only categorized as two kinds of vehicles here: compact car and large car).
The aforesaid road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: described body color recognizer, and step is:
1) get a rectangular area directly over aforementioned car cursor position of having good positioning, length is car plate length, and width is the twice of car plate height;
2) by this region by RGB model conversion to HSV model;
3) according to the H of this each pixel in region, S, V component is determined color classification, and counts the pixel number of each color classification, gets the recognition result of total maximum color classification as body color.
The aforesaid road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: described intelligent sensor node carries out the transmission of video flowing by the project MJPG-streamer that increases income, and step is:
1) transplant and increase income project MJPG-streamer to flush bonding processor Cortex-A8;
2) call two plug-in units of input_uvc and output_http, camera video stream is sent to central server and PC client computer by http agreement.
Intelligent vision sensing node extracts the number-plate number, Che Biao, vehicle, body color of road vehicle as vision label, and will be by way of place, by way of moment etc. as adeditive attribute record, then data are sent to central server analysis and processing together, and be stored in central database, carry out vehicle monitoring processing for PC client software.
The beneficial effect that the present invention reaches:
Road vehicles monitoring system based on intelligent vision Internet of Things of the present invention, by extracting the vision label of vehicle in urban road, and sets up the method for vehicle vision tag library, realizes the intelligent monitoring function of road vehicle whereabouts.System is mainly made up of multiple intelligent vision sensing nodes, central server and central database.Intelligent vision sensing node extracts the number-plate number, Che Biao, vehicle, body color of road vehicle as vision label, and will be by way of place, by way of moment etc. as adeditive attribute record, then data are sent to central server analysis and processing together, and be stored in central database, carry out vehicle monitoring processing for PC client software.This intelligent vision sensing node is mainly taking flush bonding processor Cortex-A8 as core, and therefore circumscribed USB high-definition camera, and the vehicle tag extraction algorithm of operation based on OpenCV vision storehouse have ensured real-time and the stability of video flow processing.The present invention has stronger stability and real-time, has higher using value.
Brief description of the drawings
Fig. 1 is the structural representation that the present invention is based on the road vehicles monitoring system of intelligent vision Internet of Things;
Fig. 2 is intelligent vision sensing node scheme of installation;
Fig. 3 is driven and is gathered traffic video flow step schematic diagram by V4L2;
Fig. 4 is that MJPG-streamer carries out video flowing and vision label transmission flow schematic diagram.
Embodiment
The specific embodiment of the present invention is done to further detailed description below.
As shown in Figure 1, the hardware components that the present invention is based on the road vehicles monitoring system of intelligent vision Internet of Things comprises: multiple intelligent vision sensing nodes that are made up of video camera and flush bonding processor Cortex-A8, video camera is used for gathering traffic video stream, and flush bonding processor Cortex-A8 extracts vehicle vision label from the traffic video stream of camera acquisition; Central server, for carrying out exchanges data and analysis with intelligent vision sensing node; Central database, for store car vision label data.
As shown in Figure 2, intelligent vision sensing node is installed at a certain distance one on urban road, ensures the enough close of node.The installation of video camera need to be overhead approximately 6 meters, adopts the camera lens of 36 millimeters, and approximately 25 meters of shooting distances are far away.Secondary light source is separately vertical rod preferably, from vehicle detection position not too away from.Ensureing that brightness of image is not too low, and in the not high situation of picture noise, shutter speed fast (being not less than 1/1000) as much as possible, can ensure higher picture quality like this.
It is as follows that the software of the road vehicles monitoring system based on intelligent vision Internet of Things of the present invention is realized concrete steps:
1) transplant OpenCV vision storehouse and the linux system of video streaming project MJPG-streamer to flush bonding processor of increasing income;
2) use V4L2 to drive and gather traffic video stream, concrete steps as shown in Figure 3;
3) license plate number recognizer, its concrete steps are as follows:
A. use Haar-like feature detection and be partitioned into the general location of vehicle;
B. the main flow process in the location of car plate comprises four parts: vehicle image is carried out haar wavelet decomposition by Part I, obtain high frequency imaging and low-frequency image, and carry out the sampling of 1/2 step-length, retain the high frequency imaging (know through experiment, license plate area is comparatively highlighted obvious in this image) of vertical direction; It is the minimum pixel value of 15% that Part II is set adaptive threshold, with the filtering overwhelming majority's useless noise with obtain license plate candidate area; Part III carries out horizontal expansion to bianry image, license plate candidate area is filled full; Part IV passes through to judge the colouring information of each license plate candidate area, and then finally determines license plate area, as car plate positioning result.
C. characters on license plate cut apart the method that main employing morphology connected domain is got rid of and car plate prior imformation combines, specific implementation is: first license plate area is carried out to connected component labeling, delete wherein non-compliant region, such as the too large or too little region of area and the too large or too little region of depth-width ratio, then by the prior imformation of characters on license plate spacing, the character zone of likely by mistake deleting is recovered, ensure that the character zone number retaining is 7, and meet the prior imformation of characters on license plate, finally taking the minimum boundary rectangle of each character as last segmentation result;
What d. the identification of characters on license plate adopted is CNN convolutional neural networks sorter: CNN sorter is the training method based on gradient and backpropagation.First by all training character pictures and corresponding label input CNN network, if can not get the output of expecting, the backpropagation flow process of steering error signal at output layer.By the cycle alternation of these two processes, carry out error function Gradient Descent strategy in weight vector space, make error function reach minimum value to have completed the training of sorter.
4) car mark recognizer, its concrete steps are as follows:
A. according to the car plate position of aforementioned location, get upwardly extending 5 car plate height, width is car plate width, using this region as car target coarse positioning result;
B. by after the canny rim detection of this region, carry out horizontal projection, get that wherein one section of continuous intensive low frequency region is as car target up-and-down boundary, width is car plate width, and this region is as the final positioning result of car target;
C. in this region, use car mark template to carry out sliding window template matches, each class car mark template all can be returned to a matching value, gets the car mark template of optimum matching as final recognition result.
5) vehicle targets, its concrete steps are as follows:
A. use Haar-like feature detection and be partitioned into the general location of vehicle;
B. search connected domain and calculate its minimum boundary rectangle;
C. pass through camera calibration, calculate the actual (tube) length roomy little (unit is rice) of minimum boundary rectangle, the vehicle that minimum boundary rectangle length is less than to 6 meters is judged to be compact car, otherwise be judged to be large car (owing to having a lot of vehicles by automotive type segmentation, being only categorized as two kinds of vehicles here: compact car and large car).
6) body color recognizer, its concrete steps are as follows:
A. get a rectangular area directly over aforementioned car cursor position of having good positioning, length is car plate length, and width is the twice of car plate height;
B. by this region by RGB model conversion to HSV model;
C. according to the H of this each pixel in region, S, V component is determined color classification, and counts the pixel number of each color classification, gets color classification that sum the is maximum recognition result as body color.
7) using vehicle license plate number, Che Biao, vehicle and body color as identification label, vehicle is current to be recorded as its adeditive attribute by way of place with by way of the moment, and on the basis of MJPG-streamer, increase an exclusively Socket data transmission line journey, send data to central server analysis and central database processing.
8) call input_uvc and two plug-in units of output_http of MJPG-streamer, camera video stream is sent to central server and PC client computer by http agreement, idiographic flow as shown in Figure 4.
Claims (10)
1. the road vehicles monitoring system based on intelligent vision Internet of Things, is characterized in that: comprising:
Multiple intelligent vision sensing nodes that formed by video camera and flush bonding processor Cortex-A8,
Be connected with intelligent sensor node by Ethernet and carry out the central server of information interaction,
For storing the central database of vision label data,
Be connected, access the PC client computer of the traffic video stream of each intelligent sensor node with central server by Ethernet.
2. the road vehicles monitoring system based on intelligent vision Internet of Things according to claim 1, is characterized in that: described flush bonding processor Cortex-A8 is connected with video camera by USB interface.
3. the road vehicles monitoring system based on intelligent vision Internet of Things according to claim 1, is characterized in that: described flush bonding processor Cortex-A8 loads linux system, and operational vehicle vision tag extraction algorithm.
4. the road vehicles monitoring system based on intelligent vision Internet of Things according to claim 1, is characterized in that: described intelligent sensor node carries out the transmission of video flowing by the project MJPG-streamer that increases income and increase special Socket thread carrying out the transmission of vehicle vision label.
5. the road vehicles monitoring system based on intelligent vision Internet of Things according to claim 3, it is characterized in that: described vehicle vision tag extraction algorithm, comprises number-plate number recognizer, car mark recognizer, vehicle targets and body color recognizer.
6. the road vehicles monitoring system based on intelligent vision Internet of Things according to claim 5, is characterized in that: described number-plate number recognizer, and step is:
1) use V4L2 to drive interface to obtain video flowing;
2) use Haar-like feature detection and be partitioned into the general location of vehicle;
3) algorithm that uses discrete wavelet analysis, morphological dilations processing and colouring information to combine carries out car plate location;
4) using the eliminating of morphology connected domain and car plate prior imformation to combine carries out cutting apart of characters on license plate;
5) use CNN convolutional neural networks sorter to identify characters on license plate.
7. the road vehicles monitoring system based on intelligent vision Internet of Things according to claim 5, is characterized in that: described car mark recognizer, and step is:
1) according to the car plate position of location, get upwardly extending 5 car plate height, width is car plate width, using this region as car target coarse positioning result;
2) by after the canny rim detection of this region, carry out horizontal projection, get that wherein one section of continuous intensive low frequency region is as car target up-and-down boundary, width is car plate width, this region is as the final positioning result of car target;
3) in this region, use car mark template to carry out sliding window template matches, each class car mark template is all returned to a matching value, gets the car mark template of matching value maximum as final recognition result.
8. the road vehicles monitoring system based on intelligent vision Internet of Things according to claim 5, is characterized in that: described vehicle targets, and step is:
1) use Haar-like feature detection and be partitioned into the estimated position of vehicle;
2) search connected domain and calculate its minimum boundary rectangle;
3) pass through camera calibration, the actual (tube) length that calculates minimum boundary rectangle is roomy little, and the vehicle that minimum boundary rectangle length is less than to 6 meters is judged to be compact car, otherwise is judged to be large car.
9. the road vehicles monitoring system based on intelligent vision Internet of Things according to claim 5, is characterized in that: described body color recognizer, and step is:
1) get a rectangular area directly over the car cursor position of having good positioning, length is car plate length, and width is the twice of car plate height;
2) by this region by RGB model conversion to HSV model;
3) determine color classification according to the H of this each pixel in region, S, V component, and count the pixel number of each color classification, get the recognition result of total maximum color classification as body color.
10. the road vehicles monitoring system based on intelligent vision Internet of Things according to claim 4, is characterized in that: described intelligent sensor node carries out the transmission of video flowing by the project MJPG-streamer that increases income, and step is:
1) transplant and increase income project MJPG-streamer to flush bonding processor Cortex-A8;
2) call two plug-in units of input_uvc and output_http, camera video stream is sent to central server and PC client computer by http agreement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410319408.5A CN104134067A (en) | 2014-07-07 | 2014-07-07 | Road vehicle monitoring system based on intelligent visual Internet of Things |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410319408.5A CN104134067A (en) | 2014-07-07 | 2014-07-07 | Road vehicle monitoring system based on intelligent visual Internet of Things |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104134067A true CN104134067A (en) | 2014-11-05 |
Family
ID=51806741
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410319408.5A Pending CN104134067A (en) | 2014-07-07 | 2014-07-07 | Road vehicle monitoring system based on intelligent visual Internet of Things |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104134067A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354568A (en) * | 2015-08-24 | 2016-02-24 | 西安电子科技大学 | Convolutional neural network based vehicle logo identification method |
| CN105488534A (en) * | 2015-12-04 | 2016-04-13 | 中国科学院深圳先进技术研究院 | Method, device and system for deeply analyzing traffic scene |
| CN105898225A (en) * | 2016-04-26 | 2016-08-24 | 中国科学技术大学 | Real-time camera monitoring method based Openwrt router |
| CN106326893A (en) * | 2016-08-25 | 2017-01-11 | 安徽水滴科技有限责任公司 | Vehicle color recognition method based on area discrimination |
| CN106600722A (en) * | 2016-11-14 | 2017-04-26 | 南京积图网络科技有限公司 | Toll service device, method and system |
| CN107729801A (en) * | 2017-07-11 | 2018-02-23 | 银江股份有限公司 | A kind of vehicle color identifying system based on multitask depth convolutional neural networks |
| CN108960005A (en) * | 2017-05-19 | 2018-12-07 | 内蒙古大学 | The foundation and display methods, system of subjects visual label in a kind of intelligent vision Internet of Things |
| CN108985276A (en) * | 2018-08-21 | 2018-12-11 | 盯盯拍（深圳）技术股份有限公司 | Vision AI algorithmic system design method, vision AI algorithmic system design device and vision AI algorithmic system |
| CN112465910A (en) * | 2020-11-26 | 2021-03-09 | 成都新希望金融信息有限公司 | Target shooting distance obtaining method and device, storage medium and electronic equipment |
-
2014
- 2014-07-07 CN CN201410319408.5A patent/CN104134067A/en active Pending
Non-Patent Citations (3)
| Title |
|---|
| 杜宇人 等: "基于车辆轮廓定位匹配的车型识别方法", 《扬州大学学报（自然科学版）》 * |
| 程海粟 等: "基于Cortex-A8处理器的车牌识别系统设计", 《计算机与现代化》 * |
| 黎衡: "常见小型车车标识别", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354568A (en) * | 2015-08-24 | 2016-02-24 | 西安电子科技大学 | Convolutional neural network based vehicle logo identification method |
| CN105488534A (en) * | 2015-12-04 | 2016-04-13 | 中国科学院深圳先进技术研究院 | Method, device and system for deeply analyzing traffic scene |
| CN105898225A (en) * | 2016-04-26 | 2016-08-24 | 中国科学技术大学 | Real-time camera monitoring method based Openwrt router |
| CN105898225B (en) * | 2016-04-26 | 2019-02-01 | 中国科学技术大学 | A kind of real-time photography head monitoring method based on Openwrt router |
| CN106326893A (en) * | 2016-08-25 | 2017-01-11 | 安徽水滴科技有限责任公司 | Vehicle color recognition method based on area discrimination |
| CN106600722A (en) * | 2016-11-14 | 2017-04-26 | 南京积图网络科技有限公司 | Toll service device, method and system |
| CN108960005A (en) * | 2017-05-19 | 2018-12-07 | 内蒙古大学 | The foundation and display methods, system of subjects visual label in a kind of intelligent vision Internet of Things |
| CN107729801A (en) * | 2017-07-11 | 2018-02-23 | 银江股份有限公司 | A kind of vehicle color identifying system based on multitask depth convolutional neural networks |
| CN107729801B (en) * | 2017-07-11 | 2020-12-18 | 银江股份有限公司 | Vehicle color recognition system based on multitask deep convolution neural network |
| CN108985276A (en) * | 2018-08-21 | 2018-12-11 | 盯盯拍（深圳）技术股份有限公司 | Vision AI algorithmic system design method, vision AI algorithmic system design device and vision AI algorithmic system |
| CN112465910A (en) * | 2020-11-26 | 2021-03-09 | 成都新希望金融信息有限公司 | Target shooting distance obtaining method and device, storage medium and electronic equipment |
| CN112465910B (en) * | 2020-11-26 | 2021-12-28 | 成都新希望金融信息有限公司 | Target shooting distance obtaining method and device, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104134067A (en) | Road vehicle monitoring system based on intelligent visual Internet of Things | |
| US11816899B2 (en) | Methods and systems for determining object activity within a region of interest | |
| KR102122859B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| CN104303193B (en) | Target classification based on cluster | |
| CN103069434B (en) | For the method and system of multi-mode video case index | |
| CN103383733B (en) | A kind of track based on half machine learning video detecting method | |
| US12056589B2 (en) | Methods and systems for accurately recognizing vehicle license plates | |
| CN116824859B (en) | Intelligent traffic big data analysis system based on Internet of things | |
| CN111241343A (en) | Road information monitoring and analyzing detection method and intelligent traffic control system | |
| CN104134068B (en) | Monitoring vehicle feature representation and classification method based on sparse coding | |
| CN104239309A (en) | Video analysis retrieval service side, system and method | |
| Abidin et al. | A systematic review of machine-vision-based smart parking systems | |
| CN102902960A (en) | Leave-behind object detection method based on Gaussian modelling and target contour | |
| Magrini et al. | Computer vision on embedded sensors for traffic flow monitoring | |
| CN112651293A (en) | Video detection method for road illegal stall setting event | |
| CN104159088A (en) | System and method of remote monitoring of intelligent vehicle | |
| Sun et al. | Exploiting deeply supervised inception networks for automatically detecting traffic congestion on freeway in China using ultra-low frame rate videos | |
| Kumar et al. | Resource efficient edge computing infrastructure for video surveillance | |
| CN108965804A (en) | A kind of video structural technology for city security protection | |
| Li et al. | Intelligent transportation video tracking technology based on computer and image processing technology | |
| Zheng et al. | A deep learning–based approach for moving vehicle counting and short-term traffic prediction from video images | |
| Ua-Areemitr et al. | Low-cost road traffic state estimation system using time-spatial image processing | |
| Sun et al. | Automated human use mapping of social infrastructure by deep learning methods applied to smart city camera systems | |
| CN116311166A (en) | Traffic obstacle recognition method and device and electronic equipment | |
| Vu et al. | Traffic incident recognition using empirical deep convolutional neural networks model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20141105 |
|
| RJ01 | Rejection of invention patent application after publication |