CN107610506A - The detection method and system of parking position state - Google Patents
The detection method and system of parking position state Download PDFInfo
- Publication number
- CN107610506A CN107610506A CN201710787304.0A CN201710787304A CN107610506A CN 107610506 A CN107610506 A CN 107610506A CN 201710787304 A CN201710787304 A CN 201710787304A CN 107610506 A CN107610506 A CN 107610506A
- Authority
- CN
- China
- Prior art keywords
- parking stall
- gray level
- parking
- image
- contour
- 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
- Image Analysis (AREA)
- Image Processing (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a kind of detection method and system of parking position state, its detection method includes：Obtain first gray level image in parking lot under without car state；First gray level image is converted into contour images；The area information of each parking stall is obtained according to contour images, area information includes the regional extent and positional information of each parking stall；Obtain second gray level image in parking lot under current state；According to area information, contour detecting is carried out to each parking stall in the second gray level image, each self-corresponding contour area sum in each parking stall is calculated, contour area sum is the area sum of all profiles in the region of some parking stall in the second gray level image；By each contour area sum respectively compared with the discrimination threshold of setting, the idle state of each parking stall in parking lot under current state is obtained.There is the free time in the achievable accurate feedback of the present invention, facilitate car owner to find, save down time, and improve the accuracy and promptness of detection to car owner where parking stall.
Description
Technical field
The invention belongs to technical field of image processing, is related to a kind of parking guidance technology, more particularly to a kind of parking lot car
The detection method and system of position state.
Background technology
China is one of numerous country of world population, and with the fast development of economy, the living standard of people obtains
Great improvement, be no longer satisfied with it is basic have enough to eat and wear, and then do one's utmost to pursue the quality of life.Nowadays, private car almost into
The standard configuration of each family, this not only represents the quality of life, the also very big trip for facilitating people.But work as private car
While rapid popularization, as a populous nation, thing followed vehicle congestion problem is also really troubling.Especially arrive
Working, come off duty, during festivals or holidays, " trip is difficult " is into the common fault in each city.And it is extremely difficult, be in parking stall for should not
The ground of a seat parking is found under the situation asked.And this also exacerbates the congestion of traffic indirectly.
In face of huge parking pressure, while parking efficiency is improved, alleviate traffic problems, intelligent managing system of car parking
Also arise at the historic moment.Wherein idle parking stall detection is a link important in the system, and it can greatly save car owner
The time of parking stall is blindly found in the state of situation is not known.
At present, the method for relevant parking stall detection is broadly divided into two major classes, and one kind is traditional sensor-based physics
Feature detection techniques, another kind of is the modern image detecting technique based on video.Traditional physical features detection technique is main
Using the induction coil detection technique of changes of magnetic field situation, pass through the sound wave parking stall measure technology of reflection echo and utilization
Dynamic weighing detection technique of detector deformation etc..Although this kind of detection method is with low content of technology, due to needing on ground
Under constructed, operating difficulties, thus installation and maintenance cost it is bigger.
And traditional physical features detection technique is compared, modern video images detection technology then efficiently avoid above-mentioned
Problem.Meng Yan, Sun Jun etc. propose the detection algorithm based on sampled point, by customizing sampled point on each parking stall in advance,
Then parking space state is differentiated according to capped sampled point number.Yang Yingjie, Zhang Wenlong etc. are proposed based on Harris and SVM
The detection algorithm that angle point judges, is classified by SVM SVMs to the angle point of extraction, so that it is determined that parking stall, then
Parking space state is judged according to the comparison of angle point number and threshold value in parking stall.Li Yucheng, tight beautiful jasmine etc. are proposed based on parking stall
The detection algorithm of assembled state network model, improved HSI parking stalls image pixel is classified by Bayes classifier, so
The value of the conditional probability obtained afterwards using training in advance, calculate the state transition probability in the network model.Finally according to car
Bit combination state network model, under the meaning of maximum a posteriori probability, accurately and effectively detect the virtual condition of parking stall.
Found during parking space state detects, according to the difference of the shooting angle under actual conditions, or parking stall
The problem of parking stall line is blocked after being occupied be present, actual parking space state detection band can be given to come greatly inconvenience, actual inspection
Certain deviation can also be had by surveying result.
The content of the invention
The purpose of the present invention is to propose to a kind of detection method and system of parking position state, it may apply to parking and lead
Domain is led, it is idle to realize that accurate feedback exists to car owner where parking stall, facilitates car owner to find, saves down time, and improve
The accuracy and promptness of detection.
In order to solve the above-mentioned technical problem, present invention employs following technical proposals：
The present invention proposes a kind of detection method of parking position state, and it comprises the following steps：
Obtain first gray level image in parking lot under without car state；
First gray level image is converted into contour images；
The area information of each parking stall is obtained according to contour images, area information includes regional extent and the position of each parking stall
Confidence ceases；
Obtain second gray level image in parking lot under current state；
According to area information, contour detecting is carried out to each parking stall in the second gray level image, each parking stall is calculated
Each self-corresponding contour area sum, contour area sum are all profiles in the region of some parking stall in the second gray level image
Area sum；
By each contour area sum respectively compared with the discrimination threshold of setting, obtain under current state in parking lot
The idle state of each parking stall.
Preferably, the step of the first gray level image being converted into contour images includes：First gray level image is converted into two
It is worth image, corrosion treatment is carried out to bianry image and obtains corrosion image, and bianry image and corrosion image work is poor, obtain profile
Image.
Preferably, the step of the first gray level image being converted into bianry image includes：First gray level image is subjected to Gauss
Filtering process, and the image obtained after gaussian filtering process is converted into bianry image.
Preferably, the step of obtaining the area information of each parking stall according to contour images includes：According to contour images, utilize
Hough straight-line detections obtain the area information of each parking stall.
Preferably, before the step of carrying out contour detecting to each parking stall in the second gray level image, in addition to：With first
On the basis of gray level image, image registration is carried out to the second gray level image.
Preferably, image registration includes translation processing, rotation processing and/or scaling processing.
Preferably, the calculation formula of each self-corresponding contour area sum in each parking stall is：Wherein, each profile in the region region [i] of i-th of parking stall is countour
[i] [j], its area are countour_area [i] [j], and profile total number is n [i], i.e., 0<=j<=n [i], D [i] are i-th
The contour area sum of individual parking stall.
Preferably, each parking stall is provided with discrimination threshold corresponding with its regional extent size.
Preferably, discrimination threshold expression formula is corresponding to each parking stall：D0[i]=region_area [i] * ratio, its
In, region_area [i] is the regional extent size of i-th parking stall, and ratio is default ratio value, D0[i] is i-th of car
Discrimination threshold corresponding to position.
The present invention proposes a kind of detecting system of parking position state, and it includes：
Image collection module, stop for obtaining first gray level image in parking lot under without car state and obtaining under current state
Second gray level image in parking lot；
Image processing module, it is connected with image collection module, for the first gray level image to be converted into contour images, is used in combination
Contour detecting is carried out inside each parking stall in second gray level image；
Computing module is analyzed, is connected with image processing module, the region for obtaining each parking stall according to contour images is believed
Breath, area information include the regional extent and positional information of each parking stall；Each self-corresponding contoured surface in each parking stall is calculated
Product sum, contour area sum are the area sum of all profiles in the region of some parking stall in the second gray level image；And
Judge module, it is connected with analysis computing module, for discrimination threshold of each contour area sum respectively with setting to be compared
Compared with obtaining the idle state of each parking stall in parking lot under current state.
The present invention is compared to the beneficial effect of prior art：The invention provides a kind of inspection of parking position state
Method and system are surveyed, its detection method comprises the following steps：Obtain first gray level image in parking lot under without car state；By first
Gray level image is converted into contour images；The area information of each parking stall is obtained according to contour images, area information includes each car
The regional extent and positional information of position；Obtain second gray level image in parking lot under current state；According to area information, to second
Each parking stall in gray level image carries out contour detecting, and each self-corresponding contour area sum in each parking stall, profile is calculated
Area sum is the area sum of all profiles in the region of some parking stall in the second gray level image；By each contour area it
Respectively compared with the discrimination threshold of setting, the idle state of each parking stall in parking lot under current state is obtained.It is examined
Examining system includes image collection module, image processing module, analysis computing module and judge module.The present invention efficiently solves reality
The difference of shooting angle in the case of border, or parking stall be occupied after exist parking stall line be blocked caused by detection error ask
Topic.There is the free time in the achievable accurate feedback of the present invention, facilitate car owner to find, save down time, and carry to car owner where parking stall
The high accuracy and promptness of detection.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the detection method for the parking position state that one embodiment of the invention provides；
Fig. 2 is the profile schematic diagram for the parking stall that another embodiment of the present invention provides.
Embodiment
Below in conjunction with accompanying drawing, the technical characteristic above-mentioned and other to the present invention and advantage are clearly and completely described,
Obviously, described embodiment is only the section Example of the present invention, rather than whole embodiments.
As shown in figure 1, a kind of detection method for parking position state that one embodiment of the invention proposes, it includes following
Step：
S1, the first gray level image for obtaining parking lot under without car state.
S2, the first gray level image is converted into contour images.
S3, the area information according to each parking stall of contour images acquisition, area information include the regional extent of each parking stall
And positional information.
S4, the second gray level image for obtaining parking lot under current state.
S5, according to area information, contour detecting is carried out to each parking stall in the second gray level image, each car is calculated
Each self-corresponding contour area sum in position, contour area sum are all wheels in the region of some parking stall in the second gray level image
Wide area sum.The area sum of all profiles refers in the region of some parking stall, can in the regional extent of the parking stall
It is identified as each contour area sum of automobile profile.
S6, by each contour area sum respectively compared with the discrimination threshold of setting, obtain stopping under current state
The idle state of each parking stall in.Idle state can be occupancy or idle condition.
Specifically, each each self-corresponding contour area sum in parking stall obtains in the following manner：In the second gray level image
Middle correspondence finds the regional extent of each parking stall got in the first gray level image.A certain parking stall is selected, calculates the second ash
The area sum of each profile in image in the parking stall regional extent is spent, can now obtain the contour area corresponding to the parking stall
Sum, the contour area sum are the contour areas for the car parked on the parking stall.It is each right that each parking stall can be calculated successively
The contour area sum answered.When in parking stall without car, the contour area sum of this parking stall is zero.Specifically, the first gray level image
The RGB image in the parking lot under without car state can be shot by digital camera, and RGB image is converted into the first gray-scale map
Picture.First gray level image can also be the gray level image in the parking lot under digital camera shooting without car state.
In another embodiment of the invention, the step of the first gray level image being converted into contour images includes：By first
Gray level image is converted into bianry image, and corrosion treatment is carried out to bianry image and obtains corrosion image, and by bianry image with corroding
It is poor that image is made, and obtains contour images.Corrosion treatment can be expressed as detecting image with structural element, and finding out can in image
To put down the region of the structural element.Corrosion is that one kind eliminates boundary point, the process for making border internally shrink.Corrosion treatment can
With for eliminating small and insignificant object.By corrosion treatment, and bianry image and corrosion image work is poor, it can obtain more
For clearly contour images, so as to lift the accuracy of parking space state detection.
In another embodiment of the invention, the step of the first gray level image being converted into bianry image includes：By first
Gray level image carries out gaussian filtering process, and the image obtained after gaussian filtering process is converted into bianry image.Pass through Gauss
Filtering process, noise unnecessary in the first gray level image is can remove, it is clear to obtain so as to obtain clearly bianry image
Contour images, with lifted parking space state detection accuracy.
In another embodiment of the invention, the step of obtaining the area information of each parking stall according to contour images includes：
According to contour images, the area information of each parking stall of Hough straight-line detections acquisition is utilized., can by using Hough straight-line detections
Quickly and accurately obtain the area information of each parking stall.
In another embodiment of the invention, in the second gray level image each parking stall carry out contour detecting the step of it
Before, in addition to：On the basis of the first gray level image, image registration is carried out to the second gray level image.When the numeral for shooting is taken the photograph
When picture is fixed, and digital camera is not present when shaking, it is not necessary to carries out image registration to the second gray level image.Work as number
When word camera is subjected to displacement or shaken, because the parking stall profile of both the first gray level image and the second gray level image can not be right
Standard, now need using the first gray level image as object of reference, image registration is carried out to the second gray level image, to lift parking space state inspection
The accuracy of survey.
In another embodiment of the invention, image registration uses Fourier-Mellin transform, and image registration is included at translation
Reason, rotation processing and/or scaling processing.By using Fourier-Mellin transform, and converted accordingly, can be quickly by
One gray level image and the second gray level image registration.Image registration also can realize image registration using other mapping modes.Image is matched somebody with somebody
Accurate step is specially：Assuming that the second gray level image f2 (x, y) subject to registration and the first gray level image f1 as benchmark image
Amount of zoom a, rotation amount θ between (x, y) be present0, translational movement (x0, y0), then
F2 (x, y)=f1 (a (xcos θ0+ysinθ0)+x0, a (- xsin θ0+ycosθ0)+y0)(1)；
Fourier transformation is carried out to (1), then
F2 (u, v)=a-2exp(-j(ux0+vy0))F1[a-1(ucosθ0+vsinθ0),a-1(-ucosθ0+vsinθ0)] (2)；
For the Fourier transformation exp (- j (ux of translational movement0+vy0), using phase coherent techniques, Fourier is carried out to it
Inverse transformation, it can obtain impulse function δ (x-x0,y-y0).The function has obvious sharp peaks at deviation post, can obtain accordingly
To offset (x0,y0)。
For amount of zoom a and rotation amount θ0, the remainder removed in (2) beyond translational movement is subjected to Log-Polor changes
Change (i.e. polar coordinate transform), it is assumed that F1(u, v)->G (r, θ), then G2(r, θ)=G1(r-lna,θ-θ0), it is seen then that rotation and scaling
Become translation relation again under log-polar, rotation amount a and amount of zoom are can obtain using above-mentioned phase coherent techniques
θ0。
In another embodiment of the invention, the calculation formula of contour area sum is：
Wherein, each profile in the region region [i] of i-th of parking stall is
Countour [i] [j], its area are countour_area [i] [j], and profile total number is n [i], i.e., 0<=j<=n [i], D
[i] is the contour area sum of i-th of parking stall.Due to the complexity that vehicle constructs in itself, when carrying out contour detecting to vehicle,
Many small profiles can be detected.But undeniably, most profile is each structure for belonging to vehicle, that is to say car
The part of itself.Therefore, being approximately the contour area of vehicle in itself with the area sum of each profile in the region of parking stall can
Reduce and differentiate error.
In another embodiment of the invention, each parking stall is provided with discrimination threshold corresponding with its regional extent size.
When judging the idle state of parking stall, there is its specific predetermined threshold on each parking stall.The reason is that as shown in Fig. 2 wherein
Deviation caused by the reason for numeral is parking stall order in figure, and shooting angle and shooting distance etc. are different so that each parking stall exists
Profile and its area in image are all different, so needing to set specific decision threshold according to specific parking space information
Value, to reduce differentiation error.
In another embodiment of the invention, discrimination threshold expression formula is corresponding to each parking stall：D0[i]=region_
Area [i] * ratio, wherein, region_area [i] is the regional extent size of i-th of parking stall, and ratio is default ratio
Value, D0[i] is discrimination threshold corresponding to i-th of parking stall.For example, in order to eliminate some errors and the shadow in actual detection process
Ring, default ratio value ratio is arranged to 85%.Sentence if each contour area sum D [i] in i-th of parking stall is more than
Other threshold value D0[i], then judge that the parking stall is occupied；If D [i] is less than threshold value D0[i], then it is assumed that the parking stall is idle.
One embodiment of the invention proposes a kind of detecting system of parking position state, and it includes：
Image collection module, stop for obtaining first gray level image in parking lot under without car state and obtaining under current state
Second gray level image in parking lot；
Image processing module, it is connected with image collection module, for the first gray level image to be converted into contour images, is used in combination
Contour detecting is carried out inside each parking stall in second gray level image；
Computing module is analyzed, is connected with image processing module, the region for obtaining each parking stall according to contour images is believed
Breath, area information include the regional extent and positional information of each parking stall；Each self-corresponding contoured surface in each parking stall is calculated
Product sum, contour area sum are the area sum of all profiles in the region of some parking stall in the second gray level image；And
Judge module, it is connected with analysis computing module, for discrimination threshold of each contour area sum respectively with setting to be compared
Compared with obtaining the idle state of each parking stall in parking lot under current state.
Particular embodiments described above, the purpose of the present invention, technical scheme and beneficial effect are carried out further
Describe in detail, it will be appreciated that the foregoing is only the specific embodiment of the present invention, the protection being not intended to limit the present invention
Scope.Particularly point out, to those skilled in the art, within the spirit and principles of the invention, that is done any repaiies
Change, equivalent substitution, improvement etc., should be included in the scope of the protection.
Claims (10)
1. a kind of detection method of parking position state, it is characterised in that comprise the following steps：
Obtain first gray level image in parking lot under without car state；
First gray level image is converted into contour images；
The area information of each parking stall is obtained according to the contour images, the area information includes the regional extent of each parking stall
And positional information；
Obtain second gray level image in parking lot under current state；
According to the area information, contour detecting is carried out to each parking stall in second gray level image, is calculated each
Each self-corresponding contour area sum in parking stall, the contour area sum be the second gray level image in some parking stall region in
The area sum of all profiles；
By each contour area sum respectively compared with the discrimination threshold of setting, obtain under the current state in parking lot
The idle state of each parking stall.
2. the detection method of parking position state according to claim 1, it is characterised in that by first gray-scale map
Include as the step of being converted into contour images：First gray level image is converted into bianry image, the bianry image is entered
Row corrosion treatment obtains corrosion image, and the bianry image and corrosion image work is poor, obtains contour images.
3. the detection method of parking position state according to claim 2, it is characterised in that by first gray-scale map
Include as the step of being converted into bianry image：First gray level image is subjected to gaussian filtering process, and by gaussian filtering
The image obtained after reason is converted into bianry image.
4. the detection method of parking position state according to claim 2, it is characterised in that according to the contour images
The step of area information for obtaining each parking stall, includes：According to the contour images, each car is obtained using Hough straight-line detections
The area information of position.
5. the detection method of parking position state according to claim 1, it is characterised in that to second gray-scale map
Before the step of each parking stall as in carries out contour detecting, in addition to：On the basis of first gray level image, to described
Two gray level images carry out image registration.
6. the detection method of parking position state according to claim 5, it is characterised in that described image registration includes
Translation processing, rotation processing and/or scaling processing.
7. the detection method of parking position state according to claim 1, it is characterised in that each parking stall is each
The calculation formula of corresponding contour area sum is：Wherein, the region of i-th of parking stall
Each profile in region [i] is countour [i] [j], and its area is countour_area [i] [j], profile total number
For n [i], i.e., 0<=j<=n [i], D [i] are the contour area sum of i-th of parking stall.
8. the detection method of parking position state according to claim 1, it is characterised in that each parking stall is provided with and it
The discrimination threshold corresponding to regional extent size.
9. the detection method of parking position state according to claim 8, it is characterised in that each parking stall is corresponding
Discrimination threshold expression formula be：D0[i]=region_area [i] * ratio, wherein, region_area [i] is i-th of parking stall
Regional extent size, ratio is default ratio value, D0[i] is discrimination threshold corresponding to i-th of parking stall.
A kind of 10. detecting system of parking position state, it is characterised in that including：
Image collection module, for obtaining first gray level image in parking lot under without car state and obtaining parking lot under current state
The second gray level image；
Image processing module, it is connected with described image acquisition module, for first gray level image to be converted into contour images,
And for contour detecting is carried out in second gray level image inside each parking stall；
Computing module is analyzed, is connected with described image processing module, for obtaining the area of each parking stall according to the contour images
Domain information, the area information include the regional extent and positional information of each parking stall；Each parking stall is calculated each to correspond to
Contour area sum, the contour area sum be the second gray level image in some parking stall region in all profiles face
Product sum；And
Judge module, it is connected with the analysis computing module, for by each contour area sum differentiation threshold with setting respectively
Value is compared, and obtains the idle state of each parking stall in parking lot under the current state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710787304.0A CN107610506A (en) | 2017-09-04 | 2017-09-04 | The detection method and system of parking position state |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710787304.0A CN107610506A (en) | 2017-09-04 | 2017-09-04 | The detection method and system of parking position state |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107610506A true CN107610506A (en) | 2018-01-19 |
Family
ID=61056116
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710787304.0A Pending CN107610506A (en) | 2017-09-04 | 2017-09-04 | The detection method and system of parking position state |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107610506A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108346313A (en) * | 2018-04-19 | 2018-07-31 | 济南浪潮高新科技投资发展有限公司 | A kind of method for detecting vacant parking place and system |
| CN108564814A (en) * | 2018-06-06 | 2018-09-21 | 清华大学苏州汽车研究院（吴江） | A kind of parking position detection method and device based on image |
| CN109508682A (en) * | 2018-11-20 | 2019-03-22 | 成都通甲优博科技有限责任公司 | A kind of detection method on panorama parking stall |
| CN109631896A (en) * | 2018-07-23 | 2019-04-16 | 同济大学 | A kind of parking lot autonomous parking localization method based on vehicle vision and motion information |
| CN109993991A (en) * | 2018-11-30 | 2019-07-09 | 浙江工商大学 | Parking stall condition detection method and system |
| CN110598512A (en) * | 2018-06-13 | 2019-12-20 | 杭州海康威视数字技术股份有限公司 | Parking space detection method and device |
| CN110779533A (en) * | 2019-09-16 | 2020-02-11 | 腾讯科技（深圳）有限公司 | Navigation method, system and electronic equipment |
| CN114613191A (en) * | 2022-03-14 | 2022-06-10 | 江苏云舟通信科技有限公司 | State judgment platform applying distribution density data analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105096298A (en) * | 2014-05-08 | 2015-11-25 | 东北大学 | Grid feature point extraction method based on fast line extraction |
| WO2016071512A1 (en) * | 2014-11-06 | 2016-05-12 | Tomtom International B.V. | Method for estimating the occupancy of a parking lot |
| CN105869141A (en) * | 2015-12-15 | 2016-08-17 | 乐视致新电子科技（天津）有限公司 | Image registration method and apparatus |
| CN105913685A (en) * | 2016-06-25 | 2016-08-31 | 上海大学 | Video surveillance-based carport recognition and intelligent guide method |
| CN106127801A (en) * | 2016-06-16 | 2016-11-16 | 乐视控股（北京）有限公司 | A kind of method and apparatus of moving region detection |
-
2017
- 2017-09-04 CN CN201710787304.0A patent/CN107610506A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105096298A (en) * | 2014-05-08 | 2015-11-25 | 东北大学 | Grid feature point extraction method based on fast line extraction |
| WO2016071512A1 (en) * | 2014-11-06 | 2016-05-12 | Tomtom International B.V. | Method for estimating the occupancy of a parking lot |
| CN105869141A (en) * | 2015-12-15 | 2016-08-17 | 乐视致新电子科技（天津）有限公司 | Image registration method and apparatus |
| CN106127801A (en) * | 2016-06-16 | 2016-11-16 | 乐视控股（北京）有限公司 | A kind of method and apparatus of moving region detection |
| CN105913685A (en) * | 2016-06-25 | 2016-08-31 | 上海大学 | Video surveillance-based carport recognition and intelligent guide method |
Non-Patent Citations (1)
| Title |
|---|
| 陈洁: "《关爱生命 预防事故》", 30 September 2006 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108346313A (en) * | 2018-04-19 | 2018-07-31 | 济南浪潮高新科技投资发展有限公司 | A kind of method for detecting vacant parking place and system |
| CN108564814A (en) * | 2018-06-06 | 2018-09-21 | 清华大学苏州汽车研究院（吴江） | A kind of parking position detection method and device based on image |
| CN108564814B (en) * | 2018-06-06 | 2020-11-17 | 清华大学苏州汽车研究院（吴江） | Image-based parking lot parking space detection method and device |
| CN110598512A (en) * | 2018-06-13 | 2019-12-20 | 杭州海康威视数字技术股份有限公司 | Parking space detection method and device |
| CN110598512B (en) * | 2018-06-13 | 2022-05-06 | 杭州海康威视数字技术股份有限公司 | Parking space detection method and device |
| CN109631896A (en) * | 2018-07-23 | 2019-04-16 | 同济大学 | A kind of parking lot autonomous parking localization method based on vehicle vision and motion information |
| CN109508682A (en) * | 2018-11-20 | 2019-03-22 | 成都通甲优博科技有限责任公司 | A kind of detection method on panorama parking stall |
| CN109993991A (en) * | 2018-11-30 | 2019-07-09 | 浙江工商大学 | Parking stall condition detection method and system |
| CN110779533A (en) * | 2019-09-16 | 2020-02-11 | 腾讯科技（深圳）有限公司 | Navigation method, system and electronic equipment |
| CN110779533B (en) * | 2019-09-16 | 2022-04-12 | 腾讯科技（深圳）有限公司 | Navigation method, system and electronic equipment |
| CN114613191A (en) * | 2022-03-14 | 2022-06-10 | 江苏云舟通信科技有限公司 | State judgment platform applying distribution density data analysis |
| CN114613191B (en) * | 2022-03-14 | 2022-08-12 | 江苏云舟通信科技有限公司 | State judgment platform applying distribution density data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107610506A (en) | The detection method and system of parking position state | |
| US20210073573A1 (en) | Ship identity recognition method based on fusion of ais data and video data | |
| CN106373426B (en) | Parking stall based on computer vision and violation road occupation for parking monitoring method | |
| CN102867414B (en) | Vehicle queue length measurement method based on PTZ (Pan/Tilt/Zoom) camera fast calibration | |
| CN104156731B (en) | Vehicle License Plate Recognition System and method based on artificial neural network | |
| CN109784344A (en) | A kind of non-targeted filtering method of image for ground level mark identification | |
| CN109272489A (en) | Inhibit the method for detecting infrared puniness target with multiple dimensioned local entropy based on background | |
| CN105512720A (en) | Public transport vehicle passenger flow statistical method and system | |
| Li et al. | Nighttime lane markings recognition based on Canny detection and Hough transform | |
| CN112329776B (en) | License plate detection method and device based on improved CenterNet network | |
| CN108446696A (en) | A kind of end-to-end licence plate recognition method based on deep learning | |
| CN108537099A (en) | A kind of licence plate recognition method of complex background | |
| CN109214380B (en) | License plate inclination correction method | |
| CN106778540B (en) | Parking detection is accurately based on the parking event detecting method of background double layer | |
| CN106815583A (en) | A kind of vehicle at night license plate locating method being combined based on MSER and SWT | |
| CN110599497A (en) | Drivable region segmentation method based on deep neural network | |
| CN109886200A (en) | A kind of unmanned lane line detection method based on production confrontation network | |
| CN102610104B (en) | Onboard front vehicle detection method | |
| CN104200213B (en) | One kind being based on multipart vehicle checking method | |
| Wang et al. | License plate location algorithm based on edge detection and morphology | |
| CN105966314B (en) | Lane departure warning method based on double inexpensive cameras | |
| CN108268866B (en) | Vehicle detection method and system | |
| CN108389191A (en) | The detection method in target shadow region in a kind of SAR image | |
| CN112052768A (en) | Urban illegal parking detection method and device based on unmanned aerial vehicle and storage medium | |
| CN114708560B (en) | YOLOX algorithm-based illegal parking detection 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180119 |