CN104574354A - Parking space monitoring method based on edge detection - Google Patents
Parking space monitoring method based on edge detection Download PDFInfo
- Publication number
- CN104574354A CN104574354A CN201410528132.1A CN201410528132A CN104574354A CN 104574354 A CN104574354 A CN 104574354A CN 201410528132 A CN201410528132 A CN 201410528132A CN 104574354 A CN104574354 A CN 104574354A
- Authority
- CN
- China
- Prior art keywords
- parking stall
- monitoring method
- major side
- further characterized
- edge
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
Abstract
The invention provides a parking space msonitoring method based on edge detection. The boundary of each parking space image contains an exposed boundary which is not blocked by any car. Among detected edges, main edges refer to the boundary located at an effective area (ROI) of parking space image and substantially parallel to the exposed boundary. If the main edges satisfy at least one kind of a first preparatory condition: 1) the total number of main edges exceeds the minimum number of first preparatory conditions; 2) the total lengths of the main edges are greater than the minimum lengths of the first preparatory conditions so that one parking space is in the occupied state.
Description
Technical field
The present invention relates to electronic applications, or rather, relate to the monitoring to parking stall.
Background technology
Drivers often feel put about because can not find parking stall.Find the process wastes time on parking stall, increase fuel consume, have negative effect to environment.In order to economize energy, and improve environmental quality, in the urgent need to developing a kind of parking stall monitoring system, it can Real-Time Monitoring parking stall take situation.After acquisition parking stall takies data, driver promptly can find parking stall near destination.
Parking stall law enforcement is an important component part of city management.Existing parking stall enforcement system carries out permanent patrol to realize to parking street and parking lot by parking enforcement personnel.This patrol pattern needs to consume a large amount of manpower and materials.In order to avoid this waste, people wish to utilize above-mentioned parking stall monitoring system to measure in each parking stall down time of to stop.
What the monitoring of above-mentioned parking stall and enforcement system all needed to monitor at any time parking stall takies situation.Because it can monitor several parking stall simultaneously, camera is well suited for this task.Prior art discloses the multiple parking stall monitoring system based on camera.The U.S. Patent application 2013/0265423 of Bernal etc. discloses a kind of parking stall monitoring system based on background relative method: in the effective coverage (ROI) of a parking stall image, is compared by the current image on this parking stall with its background image (image when namely this parking stall does not account for).If there is very big difference, then this parking stall is in and accounts for state.Here, the ROI on a parking stall refers to that the region of image procossing is carried out on this parking stall in monitoring.
The shooting angle of background relative method to camera is very sensitive.At that time time the shooting angle of camera less (camera is liftoff very closely mounted), the result of mistake may be obtained.Figure 1A shows such example.This parking area comprises 4 parking stall A1-A4.The image 50 of parking stall A2 is a parallelogram abcd (b and c point does not show), and it is blocked by the large portion of vehicle 40a that (the parking stall A1 being parked in A2 front, parking stall) is parked in front.At this moment, background relative method can think that parking stall A2 takies mistakenly.
Background relative method is also very sensitive to shade.When parking stall is covered by shade, the result of mistake may be produced.Figure 1B shows an example.Suppose that background image does not have to take during shade at parking stall A2.When monitoring, parking stall A2 is covered by shade 55.At this moment, background relative method can be thought that parking stall A2 is in mistakenly and accounts for state.Except shooting angle and shade, background relative method is also responsive to the state of ground on light intensity and parking stall: the change (as from fine day to cloudy day) of light intensity may lead to errors result; Rainy day, snow sky or have the ground of leaf all likely to lead to errors result.All things considered, the monitoring effect of background relative method is more unstable.
Summary of the invention
Fundamental purpose of the present invention is economize energy, and environmental protect quality.
Another object of the present invention improves the more stable parking stall monitoring system of a kind of monitoring effect.
Another object of the present invention is to provide a kind of to shooting angle, shade, light intensity and the state of ground insensitive parking stall monitoring method.
According to these and other object, the present invention proposes a kind of parking stall monitoring method based on edge detection.Camera is taken a parking area containing several parking stall.An exposure border and at least one Ouluding boundary are contained in the border of each parking stall image.Wherein, expose border not by any occlusion, and the occlusion that Ouluding boundary may be parked by, and expose border and Ouluding boundary not parallel.The image of a processor to shooting carries out image procossing, and utilizes edge-detection algorithm to detect the edge in each parking stall ROI.In all edges detected, major side refers to edge substantially parallel with exposing border in the ROI of parking stall.A parking stall is in and accounts for state, if its major side meet first pre-conditioned at least one: 1) total number of major side more than first preset minimum number; 2) total length of major side presets shortest length more than first.In the present invention, because major side represents the existence of a vehicle, it is otherwise known as edge feature.
On the other hand, a parking stall is in the state of not accounting for, if its secondary edge meet second pre-conditioned at least one: 1) total number of secondary edge more than second preset minimum number; 2) total length of secondary edge presets shortest length more than second.Here, secondary edge refers to edge substantially parallel with Ouluding boundary in the ROI of parking stall.
Relative to background relative method, the Effect on Detecting of edge-detection algorithm is more stable, it to blocking, shade, light intensity and surface state be insensitive.In order to reduce computation burden during image procossing, edge-detection algorithm only detects the edge feature parked cars, but not its shape.
Accompanying drawing explanation
Figure 1A shows a kind of parking stall of being blocked by front vehicles; Figure 1B shows a kind of parking stall covered by shade.
Fig. 2 A is that a kind of edge-detection algorithm that adopts monitors the process flow diagram having accounted for parking stall; Fig. 2 B is presented at a major side (limit pulling over observing) accounted in the ROI of parking stall.
Fig. 3 A is that a kind of edge-detection algorithm that adopts monitors the process flow diagram not accounting for parking stall; Fig. 3 B is presented at a secondary edge (limit pulling over observing) do not accounted in the ROI of parking stall.
Fig. 4 A-Fig. 4 C elaborates several definition of term " substantially parallel ".
Fig. 5 shows a kind of major side (limit pulling over observing) adopting the tailstock first to enter the vehicle of parking mode.
Fig. 6 A shows a kind of major side (sequence stop) accounted in the ROI of parking stall; Fig. 6 B shows a kind of secondary edge (sequence stop) do not accounted in the ROI of parking stall.One
Notice, these accompanying drawings are only synoptic diagrams, and their not to scale (NTS) are drawn.In order to obvious and conveniently, the portion size in figure and structure may zoom in or out.In different embodiments, identical symbol generally represents corresponding or similar structure.
Embodiment
Fig. 2 A is a kind of process flow diagram adopting edge-detection algorithm to monitor the parking stall A2 that a vehicle 40b takies; Fig. 2 B is presented at this and has accounted for the major side 80a-80d detected in the ROI70 of parking stall.In this embodiment, the vehicle (40b etc.) parked in the A1-A4 of parking stall is paralleled parking, i.e. limit pulling over observing.
In the monitoring of parking stall, the photo (step 100) of camera shooting parking area (comprising parking stall A1, A2).In the image 50 of each parking stall (as A2), an exposure border 20 (ad line) and at least one Ouluding boundary 30 (cd line) are contained in its border.Wherein, expose border 20 not by any occlusion (as illustrated in figs), and the occlusion that Ouluding boundary 30 may be parked by (the vehicle 40a as in Figure 1A), and expose border 20 and Ouluding boundary 30 not parallel.The image of processor to shooting carries out image procossing, and utilizes edge-detection algorithm to the edge (step 110) in the ROI70 detecting parking stall A2.In all edges detected, major side 80a-80d refers to edge (step 120) substantially parallel with exposing border 20 in the ROI70 of parking stall, they are otherwise known as the edge feature of vehicle 40a, and correspond respectively to the lower limb of rear anti-crash thick stick, edge, the lower limb of vehicle rear window and the coboundary of vehicle rear window that boot is given prominence to.If major side 80a-80d meet first pre-conditioned at least one (step 130), parking stall A2 is in and accounts for state (step 140): 1) total number of major side more than first preset minimum number; 2) total length of major side presets shortest length more than first.
In fig. 2b, the vehicle (the vehicle 40a be parked in the A1 of parking stall as in Figure 1A) that the shape of the ROI70 of parking lot A2 is parked before ensureing it can not affect the monitoring to parking stall A2.In this embodiment, ROI70 has a hexagon stuvwx shape, it by by least part of exposed edge circle 20 first upwards to the formation that stretches.
Fig. 3 A is that a kind of edge-detection algorithm that adopts monitors the process flow diagram not accounting for parking stall A2; Fig. 3 B is presented at a secondary edge 90a-90b do not accounted in the ROI70 of parking stall.In all edges be detected, secondary edge refers to edge 90a-90b (step 150) substantially parallel with Ouluding boundary 30 in the ROI of parking stall.They correspond respectively to the ab line be separated by parking stall A2 with A3, and rear parks cars the lower limb of 40c car body.If secondary edge meet second pre-conditioned at least one (step 160), parking stall A2 is considered to not account for (step 170): 1) total number of secondary edge more than second preset minimum number; 2) total length of secondary edge presets shortest length more than second.
Fig. 4 A-Fig. 4 C elaborates several definition of term " substantially parallel ".This term is used in major side and secondary edge.In Figure 4 A, if the difference of any tangential angle θ of pixel and the orientation angle β of straight line 20 of curve 80 is in the first scope, curve 80 is substantially parallel with straight line 20.In figure 4b, if the difference of the direction angle alpha of the best-fit line 80* of curve 80 and the orientation angle β of straight line 20 is in the second scope, curve 80 is also substantially parallel with straight line 20.In figure 4 c, if the distance of a translated line 80x of all pixels of curve 80 and straight line 20 is less than a predeterminable range, curve 80 is also substantially parallel with straight line 20.Here, tangential angle and orientation angle are the angles between straight line and x-axis.Clearly, " substantially parallel " can also have other to define.
Fig. 5 shows the major side that a kind of afterbody first enters vehicle.Other vehicle in vehicle 40b* in fig. 2b and parking area A1-A4 all adopts the mode of limit pulling over observing.But, the parking mode that the vehicle 40b* in Fig. 5 does not adopt headstock first to enter, but adopt the parking mode that the tailstock first enters, namely its headstock is towards exposure border 20.In ROI70, major side 80e, 80f are corresponding to the lower limb of buffer arm and the lower limb of front window before vehicle 40b*.These major side 80e, 80f are also the edge features of vehicle 40b*.
Except limit pulling over observing mode (Fig. 2 B, Fig. 3 B and Fig. 5), edge-detection algorithm is also used for monitoring the parking area adopting sequence stop mode.So-called sequence stop, refers to and stops along vehicle body direction (from beginning to end) between car and car.Noticing, stops and adopts sequence stop mode in most of curbside.Fig. 6 A shows an embodiment.In this embodiment, vehicle 40x etc. are parked in parking stall B1, B2 along street along 10 orders.Stop there is vehicle 40x in the B1 of parking stall, its border comprises exposure border 20 (ef line) and Ouluding boundary 30 (eg line).The ROI70 of parking stall B1 is formed by upwards moving to small part exposure border 20.In ROI70, major side 80x, 80y are the edge features of vehicle 40x, and they correspond to the lower limb of vehicle 40x vehicle body and the lower limb of side vehicle window.With Fig. 2 category-B seemingly, if major side 80x, 80y meet first pre-conditioned at least one, parking stall B1 is in and accounts for state: 1) total number of major side more than first preset minimum number; 2) total length of major side presets shortest length more than first.
Fig. 6 B shows the secondary edge 90x in a kind of ROI70 not accounting for parking stall B1.Notice, the vehicle 40y parked at B1 rear, parking stall may affect the monitoring to parking stall B1.In ROI70, secondary edge 90x corresponds to rear and to park cars the vehicle body lower limb of 40y.Similar to Fig. 3 B, if secondary edge 90x meet second pre-conditioned at least one, parking stall B1 is in the state of not accounting for: 1) total number of secondary edge more than second preset minimum number; 2) total length of secondary edge presets shortest length more than second.
Should understand, not away under the prerequisite of the spirit and scope of the present invention, can change form of the present invention and details, this does not hinder them to apply spirit of the present invention.Therefore, except the spirit according to additional claims, the present invention should not be subject to any restriction.
Claims (10)
1. monitor the method on parking stall, it is characterized in that containing following steps:
Take the image on several parking stall, the border of each parking stall image contains one not by the exposure border of any occlusion;
Detect the edge in the effective coverage of each parking stall image;
The major side substantially parallel with exposing border is selected from the edge detected;
Monitor the dead ship condition on each parking stall: if major side meets first pre-conditioned, this parking stall is in and accounts for state.
2. parking stall according to claim 1 monitoring method, is further characterized in that: described first pre-conditioned be that the total number of this major side is greater than the first minimum number.
3. parking stall according to claim 1 monitoring method, is further characterized in that: described first pre-conditioned be that the total length of this major side is greater than the first shortest length.
4. parking stall according to claim 1 monitoring method, is further characterized in that: the difference of the orientation angle on the tangential angle of this major side and this exposure border is less than the first scope.
5. parking stall according to claim 1 monitoring method, is further characterized in that: the difference of the orientation angle of the best-fit line of this major side and the orientation angle on this exposure border is less than the second scope.
6. parking stall according to claim 1 monitoring method, is further characterized in that: the distance of one of all pixels of this major side and this exposure border translated line is less than a predeterminable range.
7. parking stall according to claim 1 monitoring method, is further characterized in that:
Each parking stall is also containing at least one Ouluding boundary, and this Ouluding boundary may by an occlusion;
The secondary edge substantially parallel with Ouluding boundary is selected from the edge detected;
If secondary edge meets second pre-conditioned, this parking stall is in the state of not accounting for.
8. parking stall according to claim 7 monitoring method, is further characterized in that: described second pre-conditioned be that the total number of this secondary edge is greater than the second minimum number.
9. parking stall according to claim 7 monitoring method, is further characterized in that: described second pre-conditioned be that the total length of this secondary edge is greater than the second shortest length.
10. parking stall according to claim 1 monitoring method, is further characterized in that: vehicle adopts limit to keep to the side or sequential system stops.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US201361883122P | 2013-09-26 | 2013-09-26 | |
| US61/883,122 | 2013-09-26 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104574354A true CN104574354A (en) | 2015-04-29 |
| CN104574354B CN104574354B (en) | 2017-10-10 |
Family
ID=53090330
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410528132.1A Active CN104574354B (en) | 2013-09-26 | 2014-09-26 | Parking position monitoring method based on edge detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104574354B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023594A (en) * | 2016-06-13 | 2016-10-12 | 北京精英智通科技股份有限公司 | Parking stall shielding determination method and device and vehicle management system |
| CN107610523A (en) * | 2017-10-18 | 2018-01-19 | 冯迎安 | A kind of parking stall automatic monitoring alarm method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101656023A (en) * | 2009-08-26 | 2010-02-24 | 西安理工大学 | Management method of indoor car park in video monitor mode |
| CN101807352A (en) * | 2010-03-12 | 2010-08-18 | 北京工业大学 | Method for detecting parking stalls on basis of fuzzy pattern recognition |
| CN101894482A (en) * | 2010-07-15 | 2010-11-24 | 东南大学 | Video technology-based roadside vacant parking position wireless network detection system and method |
| CN102306274A (en) * | 2011-06-17 | 2012-01-04 | 东北大学 | Device for identifying parking space state and method |
| WO2012019941A1 (en) * | 2010-08-12 | 2012-02-16 | Valeo Schalter Und Sensoren Gmbh | Method for assisting a driver in controlling a motor vehicle, and driver assistance system |
| CN102663357A (en) * | 2012-03-28 | 2012-09-12 | 北京工业大学 | Color characteristic-based detection algorithm for stall at parking lot |
| CN202929854U (en) * | 2012-10-31 | 2013-05-08 | 福建师范大学 | Stall state detection device based on wireless positioning |
-
2014
- 2014-09-26 CN CN201410528132.1A patent/CN104574354B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101656023A (en) * | 2009-08-26 | 2010-02-24 | 西安理工大学 | Management method of indoor car park in video monitor mode |
| CN101807352A (en) * | 2010-03-12 | 2010-08-18 | 北京工业大学 | Method for detecting parking stalls on basis of fuzzy pattern recognition |
| CN101894482A (en) * | 2010-07-15 | 2010-11-24 | 东南大学 | Video technology-based roadside vacant parking position wireless network detection system and method |
| WO2012019941A1 (en) * | 2010-08-12 | 2012-02-16 | Valeo Schalter Und Sensoren Gmbh | Method for assisting a driver in controlling a motor vehicle, and driver assistance system |
| CN102306274A (en) * | 2011-06-17 | 2012-01-04 | 东北大学 | Device for identifying parking space state and method |
| CN102663357A (en) * | 2012-03-28 | 2012-09-12 | 北京工业大学 | Color characteristic-based detection algorithm for stall at parking lot |
| CN202929854U (en) * | 2012-10-31 | 2013-05-08 | 福建师范大学 | Stall state detection device based on wireless positioning |
Non-Patent Citations (1)
| Title |
|---|
| 金海民: "基于全方位视距技术的停车诱导系统的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023594A (en) * | 2016-06-13 | 2016-10-12 | 北京精英智通科技股份有限公司 | Parking stall shielding determination method and device and vehicle management system |
| CN107610523A (en) * | 2017-10-18 | 2018-01-19 | 冯迎安 | A kind of parking stall automatic monitoring alarm method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104574354B (en) | 2017-10-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8923565B1 (en) | Parked vehicle detection based on edge detection | |
| JP2019029940A (en) | Accretion detector and vehicle system comprising the same | |
| CN110544211B (en) | Method, system, terminal and storage medium for detecting lens attached object | |
| JP4784452B2 (en) | In-vehicle fog determination device | |
| EP3336764A1 (en) | Vehicle monitoring method and apparatus, processor, and image acquisition device | |
| CN108496176B (en) | Method for identifying objects in the surrounding area of a motor vehicle, driver assistance system and motor vehicle | |
| US9508015B2 (en) | Method for evaluating image data of a vehicle camera taking into account information about rain | |
| JP2009266003A (en) | Object recognizing device and object recognizing method | |
| Niksaz | Automatic traffic estimation using image processing | |
| WO2019085930A1 (en) | Method and apparatus for controlling dual-camera apparatus in vehicle | |
| WO2019013252A1 (en) | Vehicle surroundings recognition device | |
| US20170248958A1 (en) | Adjacent lane verification for an automated vehicle | |
| JP2009245042A (en) | Traffic flow measurement device and program | |
| CN104574354A (en) | Parking space monitoring method based on edge detection | |
| CN102923000A (en) | Automobile active backing speed limiting control method based on binocular vision target detection | |
| CN109447093A (en) | Tail-light detection method based on YUV image | |
| KR20080004833A (en) | Apparatus and method for detecting a navigation vehicle in day and night according to luminous state | |
| CN108711283A (en) | The night monitoring to park cars | |
| US20130156337A1 (en) | Method for removing noise of image | |
| CN111753579A (en) | Detection method and device for designated walk-substituting tool | |
| CN111414857B (en) | Front vehicle detection method based on vision multi-feature fusion | |
| CN108609010A (en) | Parking aid, vehicle and parking assist method | |
| CN113689493A (en) | Lens attachment detection method, lens attachment detection device, electronic equipment and storage medium | |
| CN113435250B (en) | Full-time road detection method and device based on focal plane division infrared polarization imaging | |
| US9811744B2 (en) | Fast and robust stop line detector |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 610041 B-36, 6 Yongfeng Road, Chengdu, Sichuan Applicant after: Zhang Guobiao Address before: 610051 5A-001 box 59, Jianshe Road, Sichuan, Chengdu, China Applicant before: Zhang Guobiao |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20170712 Address after: 610041 B-36, 6 Yongfeng Road, hi tech Zone, Sichuan, Chengdu Applicant after: Chengdu Haicun Aipi Technology Co., Ltd. Address before: 610041 B-36, 6 Yongfeng Road, Chengdu, Sichuan Applicant before: Zhang Guobiao |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder | ||
| CP02 | Change in the address of a patent holder |
Address after: No.16, 34th floor, building 2, No.88, Jitai fifth road, high tech Zone, Chengdu, Sichuan, 640041 Patentee after: ChengDu HaiCun IP Technology LLC Address before: B-36, no.6, Yongfeng Road, high tech Zone, Chengdu, Sichuan 610041 Patentee before: ChengDu HaiCun IP Technology LLC |