CN104000698A - Electric bicycle obstacle avoiding method integrating sparse representation and particle filter - Google Patents
Electric bicycle obstacle avoiding method integrating sparse representation and particle filter Download PDFInfo
- Publication number
- CN104000698A CN104000698A CN201410244469.XA CN201410244469A CN104000698A CN 104000698 A CN104000698 A CN 104000698A CN 201410244469 A CN201410244469 A CN 201410244469A CN 104000698 A CN104000698 A CN 104000698A
- Authority
- CN
- China
- Prior art keywords
- barrier
- moped scooter
- distance
- controlling value
- current controlling
- 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.)
- Granted
Links
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses an electric bicycle obstacle avoiding method integrating sparse representation and particle filter. The method mainly comprises the steps of: synchronously collecting images ahead by two cameras at the front part of an electric bicycle, judging, detecting and ranging obstacles by an imaging processing module through using an image detection and particle filter tracing algorithm. The obstacle ahead and distance to the same are automatically detected; a central controller works out current controlling value of rotating speed of a hub motor in each wheel according to the distance from the electric bicycle to the obstacle sent by the image processing module (2), and if the current controlling value is lower than current actual speed, the central controller (3) reduces the rotating speed of the hub motor (4) to the current controlling value and meanwhile emits a warning sound. Consequently, the electric bicycle can automatically speed down or brake when the obstacle is within the dangerous distance ahead so as to ensure the life or property safety of the user or other people.
Description
Technical field
The present invention relates to a kind of barrier-avoiding method that merges the Moped Scooter of rarefaction representation particle filter
Background technology
Along with the raising of people's rhythm of life, people less and less goes to look after inconvenient old man, the people with disabilitys of trip such as the walking of family is inconvenient, mental status is bad, have defective vision if having time.This class old man, people with disability go on a journey by Moped Scooter conventionally.When old man, people with disability go on a journey by Moped Scooter, due to reasons such as its mental status is bad, weak-eyeds, easily encounter the barrier in the place ahead and fall down, cause own or other people injury.
Summary of the invention
The object of the present invention is to provide a kind of barrier-avoiding method that merges the Moped Scooter of rarefaction representation particle filter, when the method can make Moped Scooter have barrier in non-security distance forwardly, automatically reduction of speed or brake, ensures user or other people life or property safety.
The present invention realizes the technical scheme that its goal of the invention adopts: a kind of barrier-avoiding method that merges the Moped Scooter of rarefaction representation particle filter, comprises the following steps:
A, image acquisition
A round-the-clock infrared camera is respectively installed on the left and right handrail of Moped Scooter, and when Moped Scooter is advanced, two infrared cameras obtain the image of electric boosted front side with identical sample frequency;
B, image correction process
The image that image processing module collects two infrared cameras carries out image correction process;
C, barrier judgment
Image processing module is every 10 frames, and the first two field picture of choosing two infrared cameras carries out images match processing, obtains disparity map, judges whether to exist barrier: as there is barrier, and carry out the operation of D step, otherwise repeat the operation of this step;
D, detection of obstacles
Image processing module is isolated each barrier by dilation erosion algorithm, zone marker algorithm successively, and frame is selected the region of each barrier; And use triangulation model to find range to barrier, and obtain Moped Scooter to the distance of barrier, in real time the distance of barrier is passed to the central controller of Moped Scooter;
Meanwhile, the random number of 10 groups of Gaussian distributed that are 5 by the coordinate stack variance of four jiaos of barrier regions obtains 10 contiguous barrier template regions, then by affine transformation, the image information in 10 barrier template regions is stored in barrier template T;
E, barrier measuring and calculating: image processing module is to the 2-10 two field picture in 10 two field pictures, in this two field picture, centered by previous frame barrier region, construct 50-100 particle that comprises positional information and image information, image information in barrier template T is carried out to particle filter, obtain the maximum particle of probability of observing as the barrier of this frame; And use triangulation model to find range to barrier, and obtain Moped Scooter to the distance of barrier, in real time the distance of barrier is passed to the central controller of Moped Scooter;
F, keep away barrier and control
The Moped Scooter that the central controller of Moped Scooter transmits according to image processing module is to the distance of barrier, calculate the current controlling value of the wheel hub motor rotating speed of each wheel, if current controlling value is lower than current actual speed, the central controller control operation of slowing down, the rotating speed of wheel hub motor is reduced to current controlling value, send prompt tone simultaneously, make the real-time automatic obstacle-avoiding of Moped Scooter;
Repeat the operation of above A-F step.
Compared with prior art, the invention has the beneficial effects as follows
One, the present invention passes through the place ahead image of two photographic head synchronous acquisition of Moped Scooter front portion, and is processed with track algorithm and carried out three-dimensional barrier judgment, detect and find range by image by image processing module.Automatically detect the place ahead barrier and distance thereof, and by the central controller control operation of automatically slowing down, while making Moped Scooter have barrier in non-security distance forwardly, reduction of speed or brake automatically, ensures user or other people life or property safety.
Two, in every 10 two field pictures, only to the 1st two field picture carry out dilation erosion algorithm, zone marker algorithm is isolated each barrier, image information in the barrier template T that 2-10 two field picture is obtained by the 1st two field picture and carry out particle filter matching treatment, well solved the tracking problem of barrier, realized robustness tracking, tracking efficiency is high; Thereby realized the accurate detection of barrier, efficiently followed the tracks of, the complexity of algorithm is low, real-time good.
Two, the use of dual camera can either be carried out the range estimation of 3 dimension spaces more accurately, also makes its field range wider, has avoided image blind area.
The max speed of above-mentioned Moped Scooter is 6km/h;
The Moped Scooter that in step F, the central controller of Moped Scooter transmits according to image processing module is to the distance of barrier, and the concrete grammar of current controlling value that calculates the wheel hub motor rotating speed of each wheel is:
5m< Moped Scooter is to the distance of barrier, the central controller control operation of not slowing down;
3m< Moped Scooter is during to the distance L EssT.LTssT.LT5m of barrier, and the current controlling value of the wheel hub motor rotating speed of each wheel is 5km/h;
0.5m< Moped Scooter is during to the distance L EssT.LTssT.LT3m of barrier, and the current controlling value of the wheel hub motor rotating speed of each wheel is 3km/h;
Moped Scooter is during to the distance L EssT.LTssT.LT0.5m of barrier, and the current controlling value of the wheel hub motor rotating speed of each wheel is 0km/h.
The distance of such barrier and corresponding wheel hub motor rotating speed controlling value, can well guarantee that Moped Scooter can not brush up against barrier, ensures user or other people life or property safety; Little on the impact of normally travelling of Moped Scooter again.
Below in conjunction with the drawings and specific embodiments, the present invention is described in further detail.
Accompanying drawing explanation
Fig. 1 is for adopting the main TV structure schematic diagram of the Moped Scooter of the inventive method.
The specific embodiment
Embodiment
Fig. 1 illustrates, and a kind of specific embodiment of the present invention is that a kind of barrier-avoiding method that merges the Moped Scooter of rarefaction representation particle filter, comprises the following steps:
A, image acquisition
A round-the-clock infrared camera 1 is respectively installed on the left and right handrail of Moped Scooter, and when Moped Scooter is advanced, two infrared cameras 1 obtain the image of electric boosted front side with identical sample frequency;
B, image correction process
The image that 2 pairs of two infrared cameras 1 of image processing module collect carries out image correction process;
C, barrier judgment
Image processing module 2 is every 10 frames, and the first two field picture of choosing wherein carries out images match processing, obtains disparity map, judges whether to exist barrier: as there is barrier, and carry out the operation of D step, otherwise repeat the operation of this step;
D, detection of obstacles
Image processing module 2 is isolated each barrier by dilation erosion algorithm, zone marker algorithm successively, and frame is selected the region of each barrier; And use triangulation model to find range to barrier, and obtain Moped Scooter to the distance of barrier, in real time the distance of barrier is passed to the central controller of Moped Scooter;
Meanwhile, the random number of 10 groups of Gaussian distributed that are 5 by the coordinate stack variance of four jiaos of barrier regions obtains 10 contiguous barrier template regions, then by affine transformation, the image information in 10 barrier template regions is stored in barrier template T;
E, barrier measuring and calculating: the 2-10 two field picture in 2 pairs of 10 two field pictures of image processing module, in this two field picture, centered by previous frame barrier region, construct 50-100 particle that comprises positional information and image information, image information in barrier template T is carried out to particle filter, obtain the maximum particle of probability of observing as the barrier of this frame; And use triangulation model to find range to barrier, and obtain Moped Scooter to the distance of barrier, in real time the distance of barrier is passed to the central controller of Moped Scooter;
F, keep away barrier and control
The Moped Scooter that the central controller 3 of Moped Scooter transmits according to image processing module 2 is to the distance of barrier, calculate the current controlling value of wheel hub motor 4 rotating speeds of each wheel, if current controlling value is lower than current actual speed, central controller 3 control operation of slowing down, the rotating speed of wheel hub motor 4 is reduced to current controlling value, send prompt tone simultaneously, make the real-time automatic obstacle-avoiding of Moped Scooter;
Repeat the operation of above A-F step.
The max speed of the Moped Scooter that this is routine is 6km/h;
The Moped Scooter that in this routine step F, the central controller 3 of Moped Scooter transmits according to image processing module 2 is to the distance of barrier, and the concrete grammar of current controlling value that calculates wheel hub motor 4 rotating speeds of each wheel is:
5m< Moped Scooter is to the distance of barrier, central controller 3 control operation of not slowing down;
3m< Moped Scooter is during to the distance L EssT.LTssT.LT5m of barrier, and the current controlling value of wheel hub motor 4 rotating speeds of each wheel is 5km/h;
0.5m< Moped Scooter is during to the distance L EssT.LTssT.LT3m of barrier, and the current controlling value of wheel hub motor 4 rotating speeds of each wheel is 3km/h;
Moped Scooter is during to the distance L EssT.LTssT.LT0.5m of barrier, and the current controlling value of wheel hub motor 4 rotating speeds of each wheel is 0km/h.
Claims (2)
1. a barrier-avoiding method that merges the Moped Scooter of rarefaction representation particle filter, comprises the following steps:
A, image acquisition
A round-the-clock infrared camera (1) is respectively installed on the left and right handrail of Moped Scooter, and when Moped Scooter is advanced, two infrared cameras (1) obtain the image of electric boosted front side with identical sample frequency;
B, image correction process
The image that image processing module (2) collects two infrared cameras (1) carries out image correction process;
C, barrier judgment
Image processing module (2) is every 10 frames, and the first two field picture of choosing wherein carries out images match processing, obtains disparity map, judges whether to exist barrier: as there is barrier, and carry out the operation of D step, otherwise repeat the operation of this step;
D, detection of obstacles
Image processing module (2) is isolated each barrier by dilation erosion algorithm, zone marker algorithm successively, and frame is selected the region of each barrier; And use triangulation model to find range to barrier, and obtain Moped Scooter to the distance of barrier, in real time the distance of barrier is passed to the central controller of Moped Scooter;
Meanwhile, the random number of 10 groups of Gaussian distributed that are 5 by the coordinate stack variance of four jiaos of barrier regions obtains 10 contiguous barrier template regions, then by affine transformation, the image information in 10 barrier template regions is stored in barrier template T;
E, barrier measuring and calculating: image processing module (2) is to the 2-10 two field picture in 10 two field pictures, in this two field picture, centered by previous frame barrier region, construct 50-100 particle that comprises positional information and image information, image information in barrier template T is carried out to particle filter, obtain the maximum particle of probability of observing as the barrier of this frame; And use triangulation model to find range to barrier, and obtain Moped Scooter to the distance of barrier, in real time the distance of barrier is passed to the central controller of Moped Scooter;
F, keep away barrier and control
The Moped Scooter that the central controller of Moped Scooter (3) transmits according to image processing module (2) is to the distance of barrier, calculate the current controlling value of wheel hub motor (4) rotating speed of each wheel, if current controlling value is lower than current actual speed, central controller (3) control operation of slowing down, the rotating speed of wheel hub motor (4) is reduced to current controlling value, send prompt tone simultaneously, make the real-time automatic obstacle-avoiding of Moped Scooter;
Repeat the operation of above A-F step.
2. a kind of intelligent electric aid vehicle barrier-avoiding method that merges rarefaction representation particle filter according to claim 1, is characterized in that:
The max speed of described Moped Scooter is 6km/h;
The Moped Scooter that in described step F, the central controller of Moped Scooter (3) transmits according to image processing module (2) is to the distance of barrier, and the concrete grammar of current controlling value that calculates wheel hub motor (4) rotating speed of each wheel is:
5m< Moped Scooter is to the distance of barrier, central controller (3) control operation of not slowing down;
3m< Moped Scooter is during to the distance L EssT.LTssT.LT5m of barrier, and the current controlling value of the wheel hub motor of each wheel (4) rotating speed is 5km/h;
0.5m< Moped Scooter is during to the distance L EssT.LTssT.LT3m of barrier, and the current controlling value of the wheel hub motor of each wheel (4) rotating speed is 3km/h;
Moped Scooter is during to the distance L EssT.LTssT.LT0.5m of barrier, and the current controlling value of the wheel hub motor of each wheel (4) rotating speed is 0km/h.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410244469.XA CN104000698B (en) | 2014-06-04 | 2014-06-04 | A kind of barrier-avoiding method of the Moped Scooter merging rarefaction representation particle filter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410244469.XA CN104000698B (en) | 2014-06-04 | 2014-06-04 | A kind of barrier-avoiding method of the Moped Scooter merging rarefaction representation particle filter |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104000698A true CN104000698A (en) | 2014-08-27 |
| CN104000698B CN104000698B (en) | 2016-07-06 |
Family
ID=51361797
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410244469.XA Expired - Fee Related CN104000698B (en) | 2014-06-04 | 2014-06-04 | A kind of barrier-avoiding method of the Moped Scooter merging rarefaction representation particle filter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104000698B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106671850A (en) * | 2016-08-26 | 2017-05-17 | 吉林省农业科学院 | Farm trailer with hydraulic loading bucket and straight-walking guiding and adjustment method |
| CN110806581A (en) * | 2019-10-21 | 2020-02-18 | 边缘智能研究院南京有限公司 | Medical cart anti-collision detection method, device and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020109788A1 (en) * | 2001-02-09 | 2002-08-15 | Morton Roger A. | Method and system for motion image digital processing |
| CN101090482A (en) * | 2006-06-13 | 2007-12-19 | 唐琎 | Driver fatigue monitoring system and method based on image process and information mixing technology |
| JP2008140101A (en) * | 2006-12-01 | 2008-06-19 | Advanced Telecommunication Research Institute International | Unconstrained and real-time hand tracking device using no marker |
| CN101474116A (en) * | 2009-01-01 | 2009-07-08 | 薛明亮 | Electric bicycle |
-
2014
- 2014-06-04 CN CN201410244469.XA patent/CN104000698B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020109788A1 (en) * | 2001-02-09 | 2002-08-15 | Morton Roger A. | Method and system for motion image digital processing |
| CN101090482A (en) * | 2006-06-13 | 2007-12-19 | 唐琎 | Driver fatigue monitoring system and method based on image process and information mixing technology |
| JP2008140101A (en) * | 2006-12-01 | 2008-06-19 | Advanced Telecommunication Research Institute International | Unconstrained and real-time hand tracking device using no marker |
| CN101474116A (en) * | 2009-01-01 | 2009-07-08 | 薛明亮 | Electric bicycle |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106671850A (en) * | 2016-08-26 | 2017-05-17 | 吉林省农业科学院 | Farm trailer with hydraulic loading bucket and straight-walking guiding and adjustment method |
| CN106671850B (en) * | 2016-08-26 | 2023-03-10 | 吉林省农业科学院 | Agricultural trailer with hydraulic loading hopper and guiding walking straight adjusting method |
| CN110806581A (en) * | 2019-10-21 | 2020-02-18 | 边缘智能研究院南京有限公司 | Medical cart anti-collision detection method, device and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104000698B (en) | 2016-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106647776B (en) | Method and device for judging lane changing trend of vehicle and computer storage medium | |
| US7710246B2 (en) | Vehicle driving assist system | |
| US10627228B2 (en) | Object detection device | |
| CN102390370B (en) | Stereoscopic vision based emergency treatment device and method for running vehicles | |
| CN111198371A (en) | Forward-looking obstacle detection system | |
| JP4914233B2 (en) | Outside monitoring device | |
| CN107563256A (en) | Aid in driving information production method and device, DAS (Driver Assistant System) | |
| CN105844222A (en) | System and method for front vehicle collision early warning based on visual sense | |
| CN104509100B (en) | Three-dimensional body detection device and three-dimensional body detection method | |
| US20170043773A1 (en) | Lane recognition apparatus | |
| US20140002658A1 (en) | Overtaking vehicle warning system and overtaking vehicle warning method | |
| JP2012185588A (en) | In-vehicle camera and in-vehicle camera system | |
| CN103978978A (en) | Inversion projection transformation based lane keeping method | |
| JP2008117073A (en) | Interruption vehicle detection device | |
| CN107430774A (en) | Travel identification device and use its travel assist system | |
| CN113657265B (en) | Vehicle distance detection method, system, equipment and medium | |
| CN109389060A (en) | A kind of vehicle week anti-collision warning method of view-based access control model | |
| JP6034396B2 (en) | Vehicle speed control device and vehicle equipped with the same | |
| CN103481842B (en) | A kind of changing method of moving vehicles detection and tracking pattern | |
| CN108627135A (en) | A kind of method and system for the road conditions that are recessed by front | |
| TW201524813A (en) | Vehicle assistance system and vehicle assistance method | |
| CN104000698A (en) | Electric bicycle obstacle avoiding method integrating sparse representation and particle filter | |
| CN105679090B (en) | A kind of night driver driving householder method based on smart mobile phone | |
| CN202345671U (en) | Video camera-based device for vehicle lane departure detection and early warning | |
| CN117197779A (en) | Track traffic foreign matter detection method, device and system based on binocular vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160706 Termination date: 20190604 |