CN108545021A - A kind of auxiliary driving method and system of identification special objective - Google Patents
A kind of auxiliary driving method and system of identification special objective Download PDFInfo
- Publication number
- CN108545021A CN108545021A CN201810343108.9A CN201810343108A CN108545021A CN 108545021 A CN108545021 A CN 108545021A CN 201810343108 A CN201810343108 A CN 201810343108A CN 108545021 A CN108545021 A CN 108545021A
- Authority
- CN
- China
- Prior art keywords
- special objective
- identification
- special
- objective
- deep learning
- 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
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Q—ARRANGEMENT OF SIGNALLING OR LIGHTING DEVICES, THE MOUNTING OR SUPPORTING THEREOF OR CIRCUITS THEREFOR, FOR VEHICLES IN GENERAL
- B60Q9/00—Arrangement or adaptation of signal devices not provided for in one of main groups B60Q1/00 - B60Q7/00, e.g. haptic signalling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention provides a kind of auxiliary driving method and system of identification special objective, belong to intelligent driving technical field, the present invention is based on the presence that the images steganalysis system of deep learning can identify special objective when driving, and then remind driver, cause driver's note that enhance traffic safety.
Description
Technical field
The present invention relates to the technical field of intelligent driving more particularly to a kind of auxiliary driving method identifying special objective and it is
System.
Background technology
Traffic safety is always social question of common concern, is related to the life security of driver and pedestrian, how to have
Effect ground improves traffic safety, and it is always the direction that people make great efforts to reduce the probability that traffic accident occurs.When driving, it is expert at
Some special objectives in vehicle direction and roadside need driver especially to look after, and are likely to neglect since the attention rate of people is limited
Slightly some special targets, such as old man, animal, special direction board etc..Old man goes across the road and can take action slow, needs driver
It gives someone extra help, and some driving direction boards can indicate next road conditions, these targets can all influence traffic safety.
The algorithm of deep learning reaches its maturity, and is especially showed in images steganalysis excellent, can efficiently identify
Go out the target of calibration, conventional target knows method for distinguishing and is generally divided into three phases：First some times are selected on given image
Then finally classified using trained grader to these extracted region features in the region of choosing.It is as follows respectively：
A) regional choice：Certain part in figure is framed as candidate region using various sizes of sliding window.
B) feature extraction：Extract the relevant visual signature in candidate region.Such as the common Harr features of Face datection；Pedestrian
Detection and general goals detect common HOG features etc..Due to the Morphological Diversity of target, illumination variation diversity, background is more
The factors such as sample so that it is not so easy to design the feature of a robust, however the quality for extracting feature directly influences point
The accuracy of class.
C) grader：It is identified using grader, such as common SVM models.
The algorithm of images steganalysis may be implemented at present can effectively realize the identification mankind of special objective, it is only necessary to
Once according to different target re -training.
Invention content
In order to solve the above technical problems, the present invention proposes a kind of auxiliary driving method of identification special objective.It can be with
The effectively special objectives such as identification nameplate, old man, animal, and remind driver to pay attention to by way of voice broadcast.
A kind of auxiliary driving method of identification special objective, during the algorithm training identification driving based on deep learning
The model of special objective, and the deep learning model that training is completed is applied in DAS (Driver Assistant System), passes through voice broadcast
Mode reminds driver to pay attention to.
Concrete operation step is as follows：
1）Driving real image is acquired by vehicle-mounted camera first, the neural network model of selected digital image identification, it is special to set
Targeted species, and nominal data collection of taking this as a foundation, are trained in server end；
2）The trained deep learning model for special objective identification is identified by the special objective of onboard system
Module is realized；
3）Real-time image acquisition in driving conditions identifies special objective type by special objective identification module, passes to
In-vehicle processor；
4）In-vehicle processor plays voice reminder, indicates special objective kind according to the target type identified, control vehicle audio
Class, and suggest that driver takes operation appropriate.
The invention also provides a kind of DAS (Driver Assistant System)s of identification special objective, include mainly the camera shooting mounted on roof
Head, special objective identification module, in-vehicle processor and sound equipment；Wherein,
Camera：Acquisition driving image；
Special objective identification module：Realize the trained deep learning model for special objective identification；
In-vehicle processor：The neural network model of selected digital image identification, sets special objective type, and calibration number of taking this as a foundation
According to collection, it is trained；According to the target type identified, control vehicle audio plays voice reminder, indicates special objective type,
And suggest that driver takes operation appropriate.
The beneficial effects of the invention are as follows
The images steganalysis in driving conditions is realized according to the design of the algorithm of deep learning, efficiently identifies out and is expert at
The special objective in vehicle direction, and surveillant is reminded by way of voice, play the role of enhancing traffic safety.
Description of the drawings
The system construction drawing of 1 present invention of attached drawing；
The training flow chart of 2 present invention of attached drawing.
Specific implementation mode
More detailed elaboration is carried out to present disclosure below：
Whole system includes camera, special objective identification module, in-vehicle processor and the sound equipment mounted on roof, such as attached drawing
Shown in 1.Deep learning target identification module supports current common images steganalysis model to train to obtain, such as SSD, R-
The process of the neural network models such as CNN, training to application is as shown in Fig. 2, and neural network model, which only needs to identify, manually to be set
Fixed special objective, including：Animal, pupil, nameplate, red light, old man, service area, gas station；Identification as needed
Special objective quantity, need to adjust neural network and finally classify the classification number of layer.
Specific workflow is：
1）Driving real image is acquired by vehicle-mounted camera first, the neural network model of selected digital image identification, it is special to set
Targeted species, and nominal data collection of taking this as a foundation, are trained in server end；
2）The trained deep learning model for special objective identification is identified by the special objective of onboard system
Module is realized；
3）Real-time image acquisition in driving conditions identifies special objective type by special objective identification module, passes to
In-vehicle processor；
4）In-vehicle processor plays voice reminder, indicates special objective kind according to the target type identified, control vehicle audio
Class, and suggest that driver takes operation appropriate, such as chronic, parking etc..
Claims (3)
1. a kind of auxiliary driving method of identification special objective, which is characterized in that
The model of special objective during algorithm training identification driving based on deep learning, and the deep learning that training is completed
Model is applied in auxiliary driving method, reminds driver to pay attention to by way of voice broadcast.
2. according to the method described in claim 1, it is characterized in that,
Concrete operation step is as follows：
1）Driving real image is acquired by vehicle-mounted camera first, the neural network model of selected digital image identification, it is special to set
Targeted species, and nominal data collection of taking this as a foundation, are trained in server end；
2）The trained deep learning model for special objective identification is identified by the special objective of onboard system
Module is realized；
3）Real-time image acquisition in driving conditions identifies special objective type by special objective identification module, passes to
In-vehicle processor；
4）In-vehicle processor plays voice reminder, indicates special objective kind according to the target type identified, control vehicle audio
Class, and suggest that driver takes operation appropriate.
3. a kind of DAS (Driver Assistant System) of identification special objective, which is characterized in that main includes the camera for being mounted on roof, spy
Different target identification module, in-vehicle processor and sound equipment；Wherein,
Camera：Acquisition driving image；
Special objective identification module：Realize the trained deep learning model for special objective identification；
In-vehicle processor：The neural network model of selected digital image identification, sets special objective type, and calibration number of taking this as a foundation
According to collection, it is trained；According to the target type identified, control vehicle audio plays voice reminder, indicates special objective type,
And suggest that driver takes operation appropriate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810343108.9A CN108545021A (en) | 2018-04-17 | 2018-04-17 | A kind of auxiliary driving method and system of identification special objective |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810343108.9A CN108545021A (en) | 2018-04-17 | 2018-04-17 | A kind of auxiliary driving method and system of identification special objective |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108545021A true CN108545021A (en) | 2018-09-18 |
Family
ID=63515383
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810343108.9A Pending CN108545021A (en) | 2018-04-17 | 2018-04-17 | A kind of auxiliary driving method and system of identification special objective |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108545021A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110796883A (en) * | 2019-11-06 | 2020-02-14 | 山东浪潮人工智能研究院有限公司 | Electric bicycle violation reminding method and device based on image recognition |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6560582B1 (en) * | 2000-01-05 | 2003-05-06 | The United States Of America As Represented By The Secretary Of The Navy | Dynamic memory processor |
| CN102866695A (en) * | 2012-09-25 | 2013-01-09 | 浙江吉利汽车研究院有限公司杭州分公司 | Aided driving system on rain or snow days and aided driving method |
| CN103802670A (en) * | 2012-11-07 | 2014-05-21 | 原相科技股份有限公司 | Monitoring and alarming system for vehicles |
| CN105608444A (en) * | 2016-01-27 | 2016-05-25 | 大连楼兰科技股份有限公司 | Wild animal image identification method used for automatic driving |
| CN105930830A (en) * | 2016-05-18 | 2016-09-07 | 大连理工大学 | Road surface traffic sign recognition method based on convolution neural network |
| CN107220643A (en) * | 2017-04-12 | 2017-09-29 | 广东工业大学 | The Traffic Sign Recognition System of deep learning model based on neurological network |
| CN107272490A (en) * | 2017-06-30 | 2017-10-20 | 黄咏奎 | Automobile and family intelligent control system |
| CN107463892A (en) * | 2017-07-27 | 2017-12-12 | 北京大学深圳研究生院 | Pedestrian detection method in a kind of image of combination contextual information and multi-stage characteristics |
| DE102017114122A1 (en) * | 2016-06-28 | 2017-12-28 | Ford Global Technologies, Llc | Detect physical threats that approach a vehicle |
| CN107609602A (en) * | 2017-09-28 | 2018-01-19 | 吉林大学 | A kind of Driving Scene sorting technique based on convolutional neural networks |
| CN107650640A (en) * | 2017-09-20 | 2018-02-02 | 上海工程技术大学 | The method for early warning and early warning system that a kind of automobile parking based on image procossing opens the door |
| US20180101738A1 (en) * | 2016-10-06 | 2018-04-12 | Smr Patents S.A.R.L. | Object detection and classification with fourier fans |
-
2018
- 2018-04-17 CN CN201810343108.9A patent/CN108545021A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6560582B1 (en) * | 2000-01-05 | 2003-05-06 | The United States Of America As Represented By The Secretary Of The Navy | Dynamic memory processor |
| CN102866695A (en) * | 2012-09-25 | 2013-01-09 | 浙江吉利汽车研究院有限公司杭州分公司 | Aided driving system on rain or snow days and aided driving method |
| CN103802670A (en) * | 2012-11-07 | 2014-05-21 | 原相科技股份有限公司 | Monitoring and alarming system for vehicles |
| CN105608444A (en) * | 2016-01-27 | 2016-05-25 | 大连楼兰科技股份有限公司 | Wild animal image identification method used for automatic driving |
| CN105930830A (en) * | 2016-05-18 | 2016-09-07 | 大连理工大学 | Road surface traffic sign recognition method based on convolution neural network |
| DE102017114122A1 (en) * | 2016-06-28 | 2017-12-28 | Ford Global Technologies, Llc | Detect physical threats that approach a vehicle |
| US20180101738A1 (en) * | 2016-10-06 | 2018-04-12 | Smr Patents S.A.R.L. | Object detection and classification with fourier fans |
| CN107220643A (en) * | 2017-04-12 | 2017-09-29 | 广东工业大学 | The Traffic Sign Recognition System of deep learning model based on neurological network |
| CN107272490A (en) * | 2017-06-30 | 2017-10-20 | 黄咏奎 | Automobile and family intelligent control system |
| CN107463892A (en) * | 2017-07-27 | 2017-12-12 | 北京大学深圳研究生院 | Pedestrian detection method in a kind of image of combination contextual information and multi-stage characteristics |
| CN107650640A (en) * | 2017-09-20 | 2018-02-02 | 上海工程技术大学 | The method for early warning and early warning system that a kind of automobile parking based on image procossing opens the door |
| CN107609602A (en) * | 2017-09-28 | 2018-01-19 | 吉林大学 | A kind of Driving Scene sorting technique based on convolutional neural networks |
Non-Patent Citations (1)
| Title |
|---|
| 林付春，张荣芬，刘宇红: "基于深度学习的智能辅助驾驶系统设计", 《贵州大学学报（自然科学版）》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110796883A (en) * | 2019-11-06 | 2020-02-14 | 山东浪潮人工智能研究院有限公司 | Electric bicycle violation reminding method and device based on image recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109637151B (en) | Method for identifying illegal driving of emergency lane on highway | |
| WO2019105342A1 (en) | Method and apparatus for detecting fake license plates of vehicles, readable storage medium, and electronic device | |
| CN110197589B (en) | Deep learning-based red light violation detection method | |
| CN106935035B (en) | Parking offense vehicle real-time detection method based on SSD neural network | |
| CN105844257B (en) | The early warning system and method for road sign are missed based on machine vision travelling in fog day | |
| CN105702048B (en) | Highway front truck illegal road occupation identifying system based on automobile data recorder and method | |
| CN102375982B (en) | Multi-character characteristic fused license plate positioning method | |
| CN103324930B (en) | A kind of registration number character dividing method based on grey level histogram binaryzation | |
| CN109948416A (en) | A kind of illegal occupancy bus zone automatic auditing method based on deep learning | |
| CN109800633A (en) | A kind of illegal judgment method of Manpower Transportation, device and electronic equipment | |
| CN109948418A (en) | A kind of illegal automatic auditing method of violation guiding based on deep learning | |
| CN103824452A (en) | Lightweight peccancy parking detection device based on full view vision | |
| CN103268489A (en) | Motor vehicle plate identification method based on sliding window searching | |
| CN104751634A (en) | Comprehensive application method of expressway tunnel driving image acquisition information | |
| CN110991221B (en) | Dynamic traffic red light running recognition method based on deep learning | |
| CN101261722A (en) | Electronic police background intelligent management and automatic implementation system | |
| CN110930723B (en) | Illegal parking detection implementation method | |
| CN103927548B (en) | Novel vehicle collision avoiding brake behavior detection method | |
| KR102015945B1 (en) | Method for packaging learning images for atonomous vehicle and apparatus thereof | |
| JP2010205160A (en) | Method for notifying speed-limit sign recognition result | |
| CN104537840B (en) | System for detecting illegally operated taxis | |
| CN110378243A (en) | A kind of pedestrian detection method and device | |
| CN109948419A (en) | A kind of illegal parking automatic auditing method based on deep learning | |
| CN105184301A (en) | Method for distinguishing vehicle azimuth by utilizing quadcopter | |
| CN106919939A (en) | A kind of traffic signboard Tracking Recognition method and system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180918 |
|
| RJ01 | Rejection of invention patent application after publication |