CN113065466A - Traffic light detection system for driving training based on deep learning - Google Patents
Traffic light detection system for driving training based on deep learning Download PDFInfo
- Publication number
- CN113065466A CN113065466A CN202110355934.7A CN202110355934A CN113065466A CN 113065466 A CN113065466 A CN 113065466A CN 202110355934 A CN202110355934 A CN 202110355934A CN 113065466 A CN113065466 A CN 113065466A
- Authority
- CN
- China
- Prior art keywords
- identification
- module
- image
- signal
- controller
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 58
- 238000013135 deep learning Methods 0.000 title claims abstract description 22
- 238000007781 pre-processing Methods 0.000 claims abstract description 19
- 238000010276 construction Methods 0.000 claims abstract description 7
- 238000005457 optimization Methods 0.000 claims abstract description 7
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 230000007547 defect Effects 0.000 abstract description 5
- 239000003086 colorant Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
-
- 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
- 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
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/095—Traffic lights
Abstract
The invention relates to traffic light detection, in particular to a deep learning-based traffic light detection system for driving training, which comprises a controller, wherein the controller receives a training image acquisition module and a detection image which are respectively sent by a detection image acquisition module through an image preprocessing module, is connected with a first image characteristic acquisition module for carrying out characteristic acquisition on the training image, is connected with an identification model construction module for constructing a signal identification model, is connected with an identification result checking module for receiving and checking an identification result of the signal identification model, and is connected with an identification model optimization module for optimizing the signal identification model according to the checking result; the technical scheme provided by the invention can effectively overcome the defect that the state of the traffic signal lamp cannot be accurately and effectively identified in the prior art.
Description
Technical Field
The invention relates to traffic light detection, in particular to a traffic light detection system for driving training based on deep learning.
Background
The identification of the traffic signal light refers to identifying the state of the traffic signal light on the basis of accurately positioning the traffic signal light, for example, for the most common traffic signal light in the form of a traffic light, the identification of the traffic signal light specifically refers to determining the indicating state (for example, allowing to pass, forbidding to pass, etc.) of the traffic signal light by identifying the bright and dark state (for example, the bright and dark state of a red light, a green light, a yellow light, etc.) of the traffic signal light. The identification of the traffic signal lamp can be used for judging the passing state of the traffic intersection, and has important significance in the aspects of automatic driving, navigation prompt, driving training and the like.
At present, signal lamp identification in driving training mainly relies on deep learning, signal lamp images at traffic intersections are obtained through fixed-point mounted cameras, color images are input into a neural network model to be subjected to deep learning to obtain states of traffic signal lamps, and broadcasting is carried out on all vehicles. However, the contour of the traffic signal lamp in the color image shot by the camera is often blurred, so that the accuracy of positioning the traffic signal lamp is reduced, and the state of the traffic signal lamp cannot be accurately identified.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects in the prior art, the invention provides a traffic light detection system for driving training based on deep learning, which can effectively overcome the defect that the state of a traffic signal light cannot be accurately and effectively identified in the prior art.
(II) technical scheme
In order to achieve the purpose, the invention is realized by the following technical scheme:
a traffic light detection system for driving training based on deep learning comprises a controller, wherein the controller receives a training image and a detection image which are respectively sent by a training image acquisition module and a detection image acquisition module through an image preprocessing module, is connected with a first image characteristic acquisition module used for carrying out characteristic acquisition on the training image, is connected with an identification model construction module used for constructing a signal identification model, is connected with an identification result checking module used for receiving and checking an identification result of the signal identification model, and is connected with an identification model optimization module used for optimizing the signal identification model according to the checking result;
the controller is connected with a second image characteristic acquisition module used for carrying out characteristic acquisition on the detected image, the controller is connected with an identification region judgment module used for judging an identification region in the detected image according to the image characteristics, and the controller is connected with an identification accuracy judgment module used for judging the identification accuracy of the optimized signal identification model;
the controller is connected with a standard template acquisition module used for acquiring standard images of signal lamps, the controller is connected with an identification area acquisition module used for acquiring identification areas in detection images, the controller also comprises a contrast analysis module used for performing contrast analysis on the standard images of the signal lamps and the identification areas in the detection images, the controller is connected with a color identification module used for performing color identification on the identification areas in the detection images, and the controller is connected with a comprehensive judgment module used for comprehensively judging the states of the signal lamps according to the comprehensive color identification results of the contrast analysis results.
Preferably, the first image feature acquisition module receives a training image which is sent by the image preprocessing module after preprocessing, and sends the image features extracted from the training image to the signal identification model which is constructed by the identification model construction module, and the identification result checking module checks the identification result of the signal identification model.
Preferably, the recognition result checking module receives an input result of the signal lamp state in the training image manually, and checks the input result with the recognition result of the signal recognition model;
and when the identification result checking module judges that the identification result of the signal identification model is inaccurate, the identification model optimization module optimizes the signal identification model by a random gradient descent method.
Preferably, after the identification area determination module determines the identification area in the detection image, the identification area image in the detection image is sent to the optimized signal identification model, and the signal identification model identifies the state of the signal lamp in the identification area image.
Preferably, when the recognition accuracy judging module judges that the recognition result of the optimized signal recognition model on the state of the signal lamp in the recognition area image is wrong, the controller starts the standard template collecting module, the recognition area collecting module and the color recognition module.
Preferably, the standard template acquisition module acquires standard images of various types of signal lamps from a standard image library, the identification region acquisition module receives identification regions in the detection images sent by the identification region judgment module, the comparison analysis module analyzes and matches the identification regions in the detection images in the standard images of various types of signal lamps, and sends the matched standard images of the signal lamps to the comprehensive judgment module.
Preferably, the traffic light state identification device further comprises an identification result output module connected with the controller and used for outputting a signal light state identification result, when the identification result checking module judges that the identification result of the signal identification model is accurate, the identification result output module directly outputs the identification result of the signal identification model, otherwise, the identification result output module outputs the comprehensive judgment result of the comprehensive judgment module on the signal light state.
Preferably, the training image acquisition module randomly acquires training images from a training image library and sends the training images to the image preprocessing module, and the detection image acquisition module sends the acquired detection images containing signal lamps to the image preprocessing module.
Preferably, the image preprocessing module performs image denoising and image enhancement processing on the training image and the detection image.
(III) advantageous effects
Compared with the prior art, the traffic light detection system for driving training based on deep learning can effectively train the signal recognition model, fully optimize the signal recognition model according to the accuracy of the signal light recognition result, effectively judge the recognition area in the detection image and provide guarantee for accurately recognizing the state of the signal light; when the signal lamp state recognition result is wrong, the signal lamp state can be recognized more accurately by matching with various signal lamp standard images and recognizing colors, so that the defect that the accuracy of a deep learning network model is low in the initial use stage is effectively overcome.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
FIG. 1 is a schematic diagram of the system of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. It is to be understood that the embodiments described are only a few 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.
A traffic light detection system for driving training based on deep learning is disclosed, as shown in figure 1, and comprises a controller, wherein the controller receives a training image acquisition module and a detection image which are respectively sent by a detection image acquisition module through an image preprocessing module, the controller is connected with a first image characteristic acquisition module used for carrying out characteristic acquisition on the training image, the controller is connected with an identification model construction module used for constructing a signal identification model, the controller is connected with an identification result checking module used for receiving and checking an identification result of the signal identification model, and the controller is connected with an identification model optimization module used for optimizing the signal identification model according to the checking result.
The first image characteristic acquisition module receives a training image which is sent by the image preprocessing module after preprocessing, and sends the image characteristics extracted from the training image to the signal identification model which is constructed by the identification model construction module, and the identification result checking module checks the identification result of the signal identification model.
The recognition result checking module receives an input result of the signal lamp state in the training image manually and checks the input result with the recognition result of the signal recognition model;
and when the identification result checking module judges that the identification result of the signal identification model is inaccurate, the identification model optimization module optimizes the signal identification model by a random gradient descent method.
In the technical scheme, the training image library stores training images of the outline of the signal lamp only containing the signal lamp body, and the judgment result of the state of the signal lamp in the training images is printed manually at the specified position in the training images.
The controller is connected with a second image characteristic acquisition module used for carrying out characteristic acquisition on the detected image, the controller is connected with an identification region judgment module used for judging an identification region in the detected image according to the image characteristics, and the controller is connected with an identification accuracy judgment module used for judging the identification accuracy of the optimized signal identification model.
After the identification area judging module judges the identification area in the detection image, the identification area image in the detection image is sent to the optimized signal identification model, and the signal identification model identifies the state of the signal lamp in the identification area image.
In the technical scheme, the training image acquisition module randomly acquires training images from a training image library and sends the training images to the image preprocessing module, the detection image acquisition module sends the detection images containing signal lamps to the image preprocessing module, and the image preprocessing module performs image denoising and image enhancement on the training images and the detection images.
The controller is connected with a standard template acquisition module used for acquiring standard images of the signal lamp, the controller is connected with an identification area acquisition module used for acquiring identification areas in the detection images, the controller also comprises a contrast analysis module used for comparing and analyzing the standard images of the signal lamp and the identification areas in the detection images, the controller is connected with a color identification module used for carrying out color identification on the identification areas in the detection images, and the controller is connected with a comprehensive judgment module used for comprehensively judging the state of the signal lamp according to the comprehensive color identification result of the contrast analysis result.
And when the recognition accuracy judging module judges that the recognition result of the optimized signal recognition model on the state of the signal lamp in the recognition area image is wrong, the controller starts the standard template collecting module, the recognition area collecting module and the color recognition module.
The standard template acquisition module acquires standard images of various types of signal lamps from a standard image library, the identification region acquisition module receives identification regions in the detection images sent by the identification region judgment module, the comparison analysis module analyzes and matches the identification regions in the detection images in the standard images of various types of signal lamps, and sends the matched standard images of the signal lamps to the comprehensive judgment module.
When the signal lamp state recognition result is wrong, the signal lamp state can be recognized more accurately by matching with various signal lamp standard images and recognizing colors, so that the defect that the accuracy of a deep learning network model is low in the initial use stage is effectively overcome.
The comprehensive judgment module comprehensively judges the signal lamp state according to the comparison and analysis result and the comprehensive color recognition result, and can input the optimized signal recognition model together with the corresponding detection image to perform model training so as to continuously improve the accuracy of the deep learning network model for signal lamp state recognition at the initial use stage.
In the technical scheme, the signal lamp state identification device further comprises an identification result output module connected with the controller and used for outputting a signal lamp state identification result, when the identification result checking module judges that the identification result of the signal identification model is accurate, the identification result output module directly outputs the identification result of the signal identification model, otherwise, the identification result output module outputs the comprehensive judgment result of the comprehensive judgment module on the signal lamp state.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (9)
1. The utility model provides a traffic lights detecting system for driving training based on deep learning which characterized in that: the device comprises a controller, wherein the controller receives a training image and a detection image which are respectively sent by a training image acquisition module and a detection image acquisition module through an image preprocessing module, the controller is connected with a first image characteristic acquisition module for carrying out characteristic acquisition on the training image, the controller is connected with an identification model construction module for constructing a signal identification model, the controller is connected with an identification result checking module for receiving and checking an identification result of the signal identification model, and the controller is connected with an identification model optimization module for optimizing the signal identification model according to the checking result;
the controller is connected with a second image characteristic acquisition module used for carrying out characteristic acquisition on the detected image, the controller is connected with an identification region judgment module used for judging an identification region in the detected image according to the image characteristics, and the controller is connected with an identification accuracy judgment module used for judging the identification accuracy of the optimized signal identification model;
the controller is connected with a standard template acquisition module used for acquiring standard images of signal lamps, the controller is connected with an identification area acquisition module used for acquiring identification areas in detection images, the controller also comprises a contrast analysis module used for performing contrast analysis on the standard images of the signal lamps and the identification areas in the detection images, the controller is connected with a color identification module used for performing color identification on the identification areas in the detection images, and the controller is connected with a comprehensive judgment module used for comprehensively judging the states of the signal lamps according to the comprehensive color identification results of the contrast analysis results.
2. The deep learning-based traffic light detection system for driving training according to claim 1, characterized in that: the first image characteristic acquisition module receives a training image which is sent by the image preprocessing module after preprocessing, and sends image characteristics extracted from the training image to the signal identification model which is constructed by the identification model construction module, and the identification result checking module checks the identification result of the signal identification model.
3. The deep learning-based traffic light detection system for driving training according to claim 2, wherein: the recognition result checking module receives an input result of the signal lamp state in the training image manually and checks the input result with the recognition result of the signal recognition model;
and when the identification result checking module judges that the identification result of the signal identification model is inaccurate, the identification model optimization module optimizes the signal identification model by a random gradient descent method.
4. The deep learning based traffic light detection system for driving training according to claim 3, wherein: and after the identification area judging module judges the identification area in the detection image, the identification area image in the detection image is sent to the optimized signal identification model, and the signal identification model identifies the state of the signal lamp in the identification area image.
5. The deep learning based traffic light detection system for driving training according to claim 4, wherein: and when the recognition accuracy judging module judges that the recognition result of the optimized signal recognition model on the state of the signal lamp in the recognition area image is wrong, the controller starts the standard template collecting module, the recognition area collecting module and the color recognition module.
6. The deep learning based traffic light detection system for driving training according to claim 5, wherein: the standard template acquisition module acquires standard images of various types of signal lamps from a standard image library, the identification region acquisition module receives identification regions in detection images sent by the identification region judgment module, the comparison analysis module analyzes and matches the identification regions in the detection images in the standard images of various types of signal lamps, and sends the matched standard images of the signal lamps to the comprehensive judgment module.
7. The deep learning-based traffic light detection system for driving training according to claim 3 or 6, wherein: the signal lamp state identification device comprises a controller, and is characterized by further comprising an identification result output module connected with the controller and used for outputting a signal lamp state identification result, when the identification result checking module judges that the identification result of the signal identification model is accurate, the identification result output module directly outputs the identification result of the signal identification model, otherwise, the identification result output module outputs the comprehensive judgment result of the comprehensive judgment module on the signal lamp state.
8. The deep learning-based traffic light detection system for driving training according to claim 2 or 4, wherein: the training image acquisition module randomly acquires training images from a training image library and sends the training images to the image preprocessing module, and the detection image acquisition module sends the acquired detection images containing signal lamps to the image preprocessing module.
9. The deep learning based traffic light detection system for driving training according to claim 8, wherein: the image preprocessing module carries out image noise reduction and image enhancement processing on the training images and the detection images.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110355934.7A CN113065466A (en) | 2021-04-01 | 2021-04-01 | Traffic light detection system for driving training based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110355934.7A CN113065466A (en) | 2021-04-01 | 2021-04-01 | Traffic light detection system for driving training based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113065466A true CN113065466A (en) | 2021-07-02 |
Family
ID=76565390
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110355934.7A Pending CN113065466A (en) | 2021-04-01 | 2021-04-01 | Traffic light detection system for driving training based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113065466A (en) |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006099718A (en) * | 2004-08-30 | 2006-04-13 | Toyama Prefecture | Personal authentication apparatus |
| JP2010112889A (en) * | 2008-11-07 | 2010-05-20 | Asia Air Survey Co Ltd | Recognition system, recognition method, and recognition program |
| CN102176287A (en) * | 2011-02-28 | 2011-09-07 | 无锡中星微电子有限公司 | Traffic signal lamp identifying system and method |
| KR101599039B1 (en) * | 2015-11-03 | 2016-03-02 | (주)코렉스시스템 | Recognition system for signal lamp and intelligent apparatus of safety crossing for crosswalk using the same |
| CN106650641A (en) * | 2016-12-05 | 2017-05-10 | 北京文安智能技术股份有限公司 | Traffic light positioning and identification method, device and system |
| CN107993470A (en) * | 2016-10-26 | 2018-05-04 | 天津职业技术师范大学 | Count down traffic signal lamp condition detection method and the monitoring system based on the method |
| CN108804983A (en) * | 2017-05-03 | 2018-11-13 | 腾讯科技（深圳）有限公司 | Traffic signal light condition recognition methods, device, vehicle-mounted control terminal and motor vehicle |
| CN109002037A (en) * | 2018-06-27 | 2018-12-14 | 中国人民解放军国防科技大学 | Multi-robot collaborative path following method based on deep learning |
| CN109101561A (en) * | 2018-07-13 | 2018-12-28 | 深圳红酒世界电商股份有限公司 | A kind of wine mark method for distinguishing |
| CN109614940A (en) * | 2018-12-14 | 2019-04-12 | 长沙致天信息科技有限责任公司 | A kind of the switch state monitoring method and relevant apparatus of deck lid |
| CN109657622A (en) * | 2018-12-21 | 2019-04-19 | 广东工业大学 | A kind of detection of traffic lights and recognition methods, device and equipment |
| CN110069986A (en) * | 2019-03-13 | 2019-07-30 | 北京联合大学 | A kind of traffic lights recognition methods and system based on mixed model |
| CN110399895A (en) * | 2019-03-27 | 2019-11-01 | 上海灏领科技有限公司 | The method and apparatus of image recognition |
| CN111582216A (en) * | 2020-05-15 | 2020-08-25 | 安徽师范大学 | Unmanned vehicle-mounted traffic signal lamp identification system and method |
| CN111612094A (en) * | 2020-05-30 | 2020-09-01 | 郑州大学 | Speed false detection and correction method, equipment and computer readable storage medium |
| 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 |
| CN111723614A (en) * | 2019-03-20 | 2020-09-29 | 北京四维图新科技股份有限公司 | Traffic signal lamp identification method and device |
| CN212724025U (en) * | 2020-06-02 | 2021-03-16 | 深圳市豪恩汽车电子装备股份有限公司 | Traffic signal lamp identification system |
-
2021
- 2021-04-01 CN CN202110355934.7A patent/CN113065466A/en active Pending
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006099718A (en) * | 2004-08-30 | 2006-04-13 | Toyama Prefecture | Personal authentication apparatus |
| JP2010112889A (en) * | 2008-11-07 | 2010-05-20 | Asia Air Survey Co Ltd | Recognition system, recognition method, and recognition program |
| CN102176287A (en) * | 2011-02-28 | 2011-09-07 | 无锡中星微电子有限公司 | Traffic signal lamp identifying system and method |
| KR101599039B1 (en) * | 2015-11-03 | 2016-03-02 | (주)코렉스시스템 | Recognition system for signal lamp and intelligent apparatus of safety crossing for crosswalk using the same |
| CN107993470A (en) * | 2016-10-26 | 2018-05-04 | 天津职业技术师范大学 | Count down traffic signal lamp condition detection method and the monitoring system based on the method |
| CN106650641A (en) * | 2016-12-05 | 2017-05-10 | 北京文安智能技术股份有限公司 | Traffic light positioning and identification method, device and system |
| CN108804983A (en) * | 2017-05-03 | 2018-11-13 | 腾讯科技（深圳）有限公司 | Traffic signal light condition recognition methods, device, vehicle-mounted control terminal and motor vehicle |
| CN109002037A (en) * | 2018-06-27 | 2018-12-14 | 中国人民解放军国防科技大学 | Multi-robot collaborative path following method based on deep learning |
| CN109101561A (en) * | 2018-07-13 | 2018-12-28 | 深圳红酒世界电商股份有限公司 | A kind of wine mark method for distinguishing |
| CN109614940A (en) * | 2018-12-14 | 2019-04-12 | 长沙致天信息科技有限责任公司 | A kind of the switch state monitoring method and relevant apparatus of deck lid |
| CN109657622A (en) * | 2018-12-21 | 2019-04-19 | 广东工业大学 | A kind of detection of traffic lights and recognition methods, device and equipment |
| CN110069986A (en) * | 2019-03-13 | 2019-07-30 | 北京联合大学 | A kind of traffic lights recognition methods and system based on mixed model |
| CN111723614A (en) * | 2019-03-20 | 2020-09-29 | 北京四维图新科技股份有限公司 | Traffic signal lamp identification method and device |
| CN110399895A (en) * | 2019-03-27 | 2019-11-01 | 上海灏领科技有限公司 | The method and apparatus of image recognition |
| CN111582216A (en) * | 2020-05-15 | 2020-08-25 | 安徽师范大学 | Unmanned vehicle-mounted traffic signal lamp identification system and method |
| CN111612094A (en) * | 2020-05-30 | 2020-09-01 | 郑州大学 | Speed false detection and correction method, equipment and computer readable storage medium |
| 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 |
| CN212724025U (en) * | 2020-06-02 | 2021-03-16 | 深圳市豪恩汽车电子装备股份有限公司 | Traffic signal lamp identification system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11887064B2 (en) | Deep learning-based system and method for automatically determining degree of damage to each area of vehicle | |
| CN106680287B (en) | Visual detection method for step defects of bearing rivet | |
| De Charette et al. | Real time visual traffic lights recognition based on spot light detection and adaptive traffic lights templates | |
| TWI497422B (en) | A system and method for recognizing license plate image | |
| EP2575077A2 (en) | Road sign detecting method and road sign detecting apparatus | |
| JPH07302304A (en) | Apparatus for analysis of character of image | |
| CN110991221B (en) | Dynamic traffic red light running recognition method based on deep learning | |
| CN106355180B (en) | A kind of license plate locating method combined based on color with edge feature | |
| US10726277B2 (en) | Lane line detection method | |
| CN112330646A (en) | Motor car bottom abnormity detection method based on two-dimensional image | |
| CN106803087A (en) | A kind of car number automatic identification method and system | |
| CN105787475A (en) | Traffic sign detection and identification method under complex environment | |
| CN114820626B (en) | Intelligent detection method for automobile front face part configuration | |
| CN114266743A (en) | FPC defect detection method, system and storage medium based on HSV and CNN | |
| CN104680195A (en) | Method for automatically recognizing vehicle colors in road intersection video and picture | |
| CN105512662A (en) | Detection method and apparatus for unlicensed vehicle | |
| CN111444843A (en) | Multimode driver and vehicle illegal behavior monitoring method and system | |
| CN111680605A (en) | Traffic signal lamp identification system and method | |
| CN113065466A (en) | Traffic light detection system for driving training based on deep learning | |
| KR20190119801A (en) | Vehicle Headlight Alignment Calibration and Classification, Inspection of Vehicle Headlight Defects | |
| CN111160080A (en) | Intelligent automobile appearance consistency detection system based on image recognition | |
| CN116128853A (en) | Production line assembly detection method, system, computer and readable storage medium | |
| CN212724025U (en) | Traffic signal lamp identification system | |
| CN110119769A (en) | A kind of detection method for early warning based on multi-modal vehicle characteristics | |
| CN113771753A (en) | Auxiliary driving method and system for intersection green light reminding |
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 |