CN102622897A - Video-based dynamic vehicle queue length estimation method - Google Patents
Video-based dynamic vehicle queue length estimation method Download PDFInfo
- Publication number
- CN102622897A CN102622897A CN2012100996530A CN201210099653A CN102622897A CN 102622897 A CN102622897 A CN 102622897A CN 2012100996530 A CN2012100996530 A CN 2012100996530A CN 201210099653 A CN201210099653 A CN 201210099653A CN 102622897 A CN102622897 A CN 102622897A
- Authority
- CN
- China
- Prior art keywords
- queue length
- vehicle queue
- video
- vehicle
- highway section
- 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
Images
Abstract
The invention discloses a video-based dynamic vehicle queue length estimation method. According to the method, on the basis that the vehicle queue length at a certain moment is obtained exactly by an image recognition technology, a dynamic vehicle queue length estimation model at several future moments is built by combining the flow of vehicles which enter a road section, a timing parameter of a signal lamp is dynamically adjusted by using the estimated vehicle queue length, an estimation scale of road crowdedness degree or time when the road section is about to be crowded with the vehicles is given, and a reference and an evidence are supplied to forced control over traffic overflow.
Description
Technical field
What the present invention relates to is intelligent transportation field, specifically is a kind of vehicle queue length method for dynamic estimation based on video.
Background technology
Because the control action of traffic lights; The vehicle queue behavior must take place in the urban traffic network under the high saturation state; And accurately obtain and estimate that vehicle queue length is to carry out the assessment of congested in traffic degree; Intersection signal lamp control timing, the important prerequisite of the pressure control of traffic overflow etc.
Find through retrieval: Li Zhe etc., based on the vehicle queue length image detecting system of DSP, computer utility research, 2005,22 (11): 229-230; Qi Hongsheng, Wang Dianhai, signal controlling intersection vehicle queue length, Jilin University's journal (engineering version), 2009,39 (06): 1457-1462; Yao Ronghan, vehicle queue model investigation, Jilin University's PhD dissertation; Huang Lei etc., method and device that vehicle queue length detects, patent of invention, the applying date: 2010.06.28, open day: 2011.01.05; Yang Yonghui etc. are based on the vehicle queue length detection of video, 2011,28 (3): 1037-1041; Li Qingwu etc., the single-frame images pick-up unit and the method for work thereof of intersection vehicle queue length, the applying date: 2007.12.26; Open day: 2009.7.1; Above-mentioned technology all is to utilize video detection technology to obtain the vehicle queue length of the intersection in a certain moment, and the vehicle queue in a plurality of moment in future is not estimated.Find in addition: Dai Leilei etc., saturation signal crossing queue length prediction, Jilin University's journal (engineering version), 2008,38 (06): 1287-1290; This technology is to utilize the adaptive weighting exponential smoothing, and the real-time traffic in import track is predicted, has set up the queue length forecast model that is the basis with the fixed number queuing theory.The present invention then is that the rear side from vehicle queue sets up the vehicle queue length that high-definition camera obtains a certain moment on the road; Set up the following estimation model of vehicle queue length constantly to the speed of queuing behind the vehicle flowrate in combination inflow highway section and the vehicle again, obviously different with this method.
Summary of the invention
The objective of the invention is to propose a kind of vehicle queue length method for dynamic estimation; Be accurately to obtain on the basis of a certain moment vehicle queue length in the existing image recognition technology of utilizing; In conjunction with flowing into the road section traffic volume flow, set up the vehicle queue dynamic estimation model in following several moment, utilize the vehicle queue length of estimating to remove dynamically to adjust the timing parameter of signal lamp; Or provide the opinion rating of the road degree of crowding, or provide time that vehicle queue will overflow etc. on the highway section.
To achieve these goals, the present invention adopts following technical scheme:
A kind of vehicle queue length method for dynamic estimation based on video, the performing step of this method is following:
Step 1: at first set up high-definition camera at the crossing, the upper reaches in vehicle pass-through highway section;
Step 2: the vehicle queue situation of uninterruptedly taking the highway section with high-definition camera;
Step 3: read the state of crossing controller, judge whether the red signal of downstream road junction is opened,, continue step 4 as opening, otherwise, get back to step 2;
Step 4: every with image-recognizing method identification at a distance from Δ T vehicle queue length constantly, utilize vehicle queue length dynamic pre-estimating model to estimate following N Δ T vehicle queue length constantly simultaneously, wherein: N is a natural number, N=1,2,3,4 ...;
Step 5: read the state of crossing controller, judge whether the green light signals of downstream road junction opens, as opening, then this time queue length dynamic estimation process finishes, otherwise, get back to step 4;
Step 6: utilize step 4 to estimate that the queue length of coming out provides foundation for the belisha beacon timing, or judge whether the vehicle queue spillover can take place, or be used for estimating the degree of crowding in highway section.
In the said step 3 and 5; Read the state of crossing controller; Be meant that camera chain and crossing controller are to communicate with one another through COM port; In estimation procedure, camera chain ceaselessly reads the signal timing dial information of crossing controller, thereby can know the signal condition (signal timing dial information refers to the lights state of signal) at crossing, upstream and downstream.
In the said step 4, the time interval of Δ T is set according to road section length by the user, and wherein one of preferred value is 3 seconds or 5 seconds or 7 seconds.
In the said step 4, said dynamic pre-estimating is meant that after the downstream red signal is opened every separated Δ T time is just estimated N queue length once, after the downstream green light signals is opened, just finishes
Said dynamic pre-estimating model is following:
L(t+N)＝L(t)+{[Q
in(t)-Q
out(t)]*N*A?T*L
v}/n
Wherein, L (t) and L (t+N) are respectively current time and N queue length constantly afterwards; Q
In(t) and Q
Out(t) be the magnitude of traffic flow that current time got into and flowed out the highway section respectively, L
vBe the length (rule of thumb get, generally get 5-7 rice) of a car, n is the quantity in queuing track on the highway section.
Beneficial effect of the present invention: utilize video information to come the queue length of vehicle on the Dynamic Recognition highway section; Can bring following benefit for urban transportation: 1. make the crossing controller come the dynamically timing parameter of adjustment signal lamp, make signal lamp timing length to be consistent with actual road conditions according to the vehicle queue length of discerning and estimating; 2. utilize the vehicle queue length of estimating to be used for carrying out the evaluation of the road degree of crowding; 3. according to the vehicle queue length of estimating, can estimate the time that vehicle queue will overflow on the highway section, controlling for the pressure of traffic overflow provides reference and foundation.
Description of drawings
Fig. 1 is the decorating position figure of video camera;
Fig. 2 is a schematic flow sheet of the present invention.
Embodiment
Below in conjunction with accompanying drawing and embodiment the present invention is described further.
Be illustrated in figure 1 as the decorating position figure of video camera, two crossings among the figure, the left side is a downstream road junction, and the right is crossing, the upper reaches, and the position of video camera is erected at crossing, the upper reaches, the rear of the highway section vehicle queue that just is taken.
Be illustrated in figure 2 as schematic flow sheet of the present invention, after system brought into operation, video camera was taken facing to the highway section; After crossing, downstream signal etc. reddens, every at a distance from the Δ T time, to length of the queuing vehicle identification on the highway section; And the utilization prediction model, estimate following N (N=1,2; 3,4 ...) individual Δ T vehicle queue length constantly.
Concrete performing step of the present invention is following:
Step 1: at first set up high-definition camera at the crossing, the upper reaches in vehicle pass-through highway section;
Step 2: the vehicle queue situation of uninterruptedly taking the highway section with high-definition camera;
Step 3: read the state of crossing controller, judge whether the red signal of downstream road junction is opened,, continue step 4 as opening, otherwise, get back to step 2;
Step 4: every with image-recognizing method identification at a distance from Δ T vehicle queue length constantly, utilize the vehicle queue length dynamic pre-estimating model of having set up to estimate following N Δ T vehicle queue length constantly again;
Step 5: read the state of crossing controller, judge whether the green light signals of downstream road junction opens,, finish as opening, otherwise, get back to step 4;
In the said step 3 and 5; Read the state of crossing controller; Be meant that camera chain and crossing controller are to communicate with one another through com port, camera chain can read the timing information of crossing controller, thereby whether the signal that can judge the crossing, upstream and downstream is red light;
In the said step 4, the time interval of Δ T is set according to road section length by the user, can be 3 seconds, also can be not wait in 5 seconds, 7 seconds;
Image recognition technology in the said step 4 will utilize ripe queue length to obtain technology, and document sees reference: Yang Yonghui etc., and based on the vehicle queue length detection of video, 2011,28 (3): 1037-1041:
In the said step 4, the dynamic pre-estimating model is following:
L(t+N)＝L(t)+{[Qin(t)-Qout(t)]*N*ΔT*Lv}/n
Wherein, L (t) and L (t+N) are respectively current time and N queue length constantly afterwards; Qin (t) and Qout (t) are respectively the magnitudes of traffic flow that current time got into and flowed out the highway section, and Lv is the length after the car standardization, generally gets 5-7 rice, and n is the quantity in queuing track on the highway section;
The dynamic pre-estimating of said step 4 is meant that after the downstream red signal is opened every separated Δ T time is just estimated N queue length once, after the downstream green light signals is opened, just finishes.
Though the above-mentioned accompanying drawing specific embodiments of the invention that combines is described; But be not restriction to protection domain of the present invention; One of ordinary skill in the art should be understood that; On the basis of technical scheme of the present invention, those skilled in the art need not pay various modifications that creative work can make or distortion still in protection scope of the present invention.
Claims (4)
1. vehicle queue length method for dynamic estimation based on video is characterized in that the performing step of this method is following:
Step 1: at first set up high-definition camera at the crossing, the upper reaches in vehicle pass-through highway section;
Step 2: the vehicle queue situation of uninterruptedly taking the highway section with high-definition camera;
Step 3: read the state of crossing controller, judge whether the red signal of downstream road junction is opened,, continue step 4 as opening, otherwise, get back to step 2;
Step 4: every with image-recognizing method identification at a distance from Δ T vehicle queue length constantly, utilize vehicle queue length dynamic pre-estimating model to estimate following N Δ T vehicle queue length constantly simultaneously, wherein: N is a natural number, N=1,2,3,4,
Step 5: read the state of crossing controller, judge whether the green light signals of downstream road junction opens, as opening, then the dynamic pre-estimating vehicle queue length finishes, otherwise, get back to step 4;
Step 6: utilize step 4 to estimate that the queue length of coming out provides foundation for the belisha beacon timing, or judge whether the vehicle queue spillover can take place, or be used for estimating the degree of crowding in highway section.
2. the vehicle queue length method for dynamic estimation based on video as claimed in claim 1; It is characterized in that; In the said step 3 and 5, read the state of crossing controller, be meant that camera chain and crossing controller are to communicate with one another through com port; Camera chain reads the timing information of crossing controller, thereby whether the signal that can judge the crossing, upstream and downstream is red signal.
3. the vehicle queue length method for dynamic estimation based on video as claimed in claim 1 is characterized in that in the said step 4, the time interval of Δ T is set according to road section length by the user, and wherein one of preferred value is 3 seconds or 5 seconds or 7 seconds.
4. the vehicle queue length method for dynamic estimation based on video as claimed in claim 1; It is characterized in that in the said step 4, said dynamic pre-estimating is meant after the downstream red signal is opened; Every separated Δ T time is just estimated N queue length once, after the downstream green light signals is opened, just finishes
Said dynamic pre-estimating model is following:
L(t+N)＝L(t)+{[Q
in(t)-Q
out(t)]*N*ΔT*L
v}/n
Wherein, L (t) and L (t+N) are respectively current time and N queue length constantly afterwards; Q
In(t) and Q
Out(t) be the magnitude of traffic flow that current time got into and flowed out the highway section respectively, L
vBe the length of a car, n is the quantity in queuing track on the highway section.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210099653.0A CN102622897B (en) | 2012-04-07 | 2012-04-07 | Video-based dynamic vehicle queue length estimation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210099653.0A CN102622897B (en) | 2012-04-07 | 2012-04-07 | Video-based dynamic vehicle queue length estimation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102622897A true CN102622897A (en) | 2012-08-01 |
| CN102622897B CN102622897B (en) | 2014-04-30 |
Family
ID=46562788
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210099653.0A Expired - Fee Related CN102622897B (en) | 2012-04-07 | 2012-04-07 | Video-based dynamic vehicle queue length estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102622897B (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103268706A (en) * | 2013-04-18 | 2013-08-28 | 同济大学 | Method for detecting vehicle queue length based on local variance |
| CN103871258A (en) * | 2014-03-07 | 2014-06-18 | 北京航空航天大学 | Signal control method for preventing dead lock of intersection |
| CN103927876A (en) * | 2014-05-08 | 2014-07-16 | 山东大学 | Traffic overflow occurrence time predicating method based on video detection |
| CN104157152A (en) * | 2014-08-13 | 2014-11-19 | 安徽科力信息产业有限责任公司 | Traffic signal optimization control method for vehicle queuing overflow state at road intersections |
| CN104751627A (en) * | 2013-12-31 | 2015-07-01 | 西门子公司 | Traffic condition parameter determining method and device |
| CN104994360A (en) * | 2015-08-03 | 2015-10-21 | 北京旷视科技有限公司 | Video monitoring method and video monitoring system |
| CN105513366A (en) * | 2016-02-19 | 2016-04-20 | 上海果路交通科技有限公司 | Method for judging traffic state of road intersection |
| CN107170247A (en) * | 2017-06-06 | 2017-09-15 | 青岛海信网络科技股份有限公司 | One kind determines intersection queue length method and device |
| CN108629990A (en) * | 2018-06-14 | 2018-10-09 | 重庆同济同枥信息技术有限公司 | A kind of real-time dynamic timing method and system based on multi-source data |
| CN109000668A (en) * | 2018-05-25 | 2018-12-14 | 上海汽车集团股份有限公司 | Real-time intelligent air navigation aid based on car networking |
| CN109147318A (en) * | 2018-08-03 | 2019-01-04 | 上海市政工程设计研究总院（集团）有限公司 | A kind of mode of transportation comfort level judgement system for the evacuation of transport hub passenger flow |
| CN109147363A (en) * | 2018-09-10 | 2019-01-04 | 吉林大学 | Traffic intelligent guides system and bootstrap technique |
| CN109147353A (en) * | 2017-08-04 | 2019-01-04 | 张长树 | A kind of intelligent transportation command system |
| CN110111590A (en) * | 2019-06-04 | 2019-08-09 | 南京慧尔视智能科技有限公司 | A kind of vehicle dynamic queue length detection method |
| CN110751829A (en) * | 2019-09-26 | 2020-02-04 | 同济大学 | Vehicle queuing dissipation time prediction method based on image self-learning |
| CN110807913A (en) * | 2018-08-06 | 2020-02-18 | 北京嘀嘀无限科技发展有限公司 | System and method for determining traffic condition |
| CN111554111A (en) * | 2020-04-21 | 2020-08-18 | 河北万方中天科技有限公司 | Signal timing optimization method and device based on multi-source data fusion and terminal |
| CN111932916A (en) * | 2020-07-31 | 2020-11-13 | 长沙理工大学 | Control method and control system for dynamic emergency lane of urban road |
| WO2022116361A1 (en) * | 2020-12-01 | 2022-06-09 | 山东交通学院 | Traffic light control method and system based on urban trunk line vehicle queuing length |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004205234A (en) * | 2002-12-24 | 2004-07-22 | Daimler Chrysler Ag | Method and device for driving support system for vehicle |
| JP2008108033A (en) * | 2006-10-25 | 2008-05-08 | Sumitomo Electric Ind Ltd | Traffic signal control analysis device |
| CN101469985A (en) * | 2007-12-26 | 2009-07-01 | 河海大学常州校区 | Single-frame image detection apparatus for vehicle queue length at road junction and its working method |
| CN101593431A (en) * | 2008-05-26 | 2009-12-02 | 奥城同立科技开发（北京）有限公司 | The method of automatically regulating vehicle traffic condition at crossing |
-
2012
- 2012-04-07 CN CN201210099653.0A patent/CN102622897B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004205234A (en) * | 2002-12-24 | 2004-07-22 | Daimler Chrysler Ag | Method and device for driving support system for vehicle |
| JP2008108033A (en) * | 2006-10-25 | 2008-05-08 | Sumitomo Electric Ind Ltd | Traffic signal control analysis device |
| CN101469985A (en) * | 2007-12-26 | 2009-07-01 | 河海大学常州校区 | Single-frame image detection apparatus for vehicle queue length at road junction and its working method |
| CN101593431A (en) * | 2008-05-26 | 2009-12-02 | 奥城同立科技开发（北京）有限公司 | The method of automatically regulating vehicle traffic condition at crossing |
Non-Patent Citations (2)
| Title |
|---|
| 荣建等: "信号交叉口排队长度动态计算方法研究", 《中国公路学报》 * |
| 贺晓峰: "基于视频图像处理的车辆排队长度检测", 《交通与计算机》 * |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103268706A (en) * | 2013-04-18 | 2013-08-28 | 同济大学 | Method for detecting vehicle queue length based on local variance |
| CN103268706B (en) * | 2013-04-18 | 2015-02-18 | 同济大学 | Method for detecting vehicle queue length based on local variance |
| CN104751627A (en) * | 2013-12-31 | 2015-07-01 | 西门子公司 | Traffic condition parameter determining method and device |
| CN103871258A (en) * | 2014-03-07 | 2014-06-18 | 北京航空航天大学 | Signal control method for preventing dead lock of intersection |
| CN103871258B (en) * | 2014-03-07 | 2016-08-17 | 北京航空航天大学 | A kind of signal control method of crossing anti-locking |
| CN103927876B (en) * | 2014-05-08 | 2016-01-20 | 山东大学 | Based on the traffic overflow time of origin Forecasting Methodology that video detects |
| CN103927876A (en) * | 2014-05-08 | 2014-07-16 | 山东大学 | Traffic overflow occurrence time predicating method based on video detection |
| CN104157152A (en) * | 2014-08-13 | 2014-11-19 | 安徽科力信息产业有限责任公司 | Traffic signal optimization control method for vehicle queuing overflow state at road intersections |
| CN104994360A (en) * | 2015-08-03 | 2015-10-21 | 北京旷视科技有限公司 | Video monitoring method and video monitoring system |
| CN105513366A (en) * | 2016-02-19 | 2016-04-20 | 上海果路交通科技有限公司 | Method for judging traffic state of road intersection |
| CN107170247A (en) * | 2017-06-06 | 2017-09-15 | 青岛海信网络科技股份有限公司 | One kind determines intersection queue length method and device |
| CN107170247B (en) * | 2017-06-06 | 2020-10-30 | 青岛海信网络科技股份有限公司 | Method and device for determining queuing length of intersection |
| CN109147353A (en) * | 2017-08-04 | 2019-01-04 | 张长树 | A kind of intelligent transportation command system |
| CN109000668A (en) * | 2018-05-25 | 2018-12-14 | 上海汽车集团股份有限公司 | Real-time intelligent air navigation aid based on car networking |
| CN108629990A (en) * | 2018-06-14 | 2018-10-09 | 重庆同济同枥信息技术有限公司 | A kind of real-time dynamic timing method and system based on multi-source data |
| CN109147318A (en) * | 2018-08-03 | 2019-01-04 | 上海市政工程设计研究总院（集团）有限公司 | A kind of mode of transportation comfort level judgement system for the evacuation of transport hub passenger flow |
| CN110807913A (en) * | 2018-08-06 | 2020-02-18 | 北京嘀嘀无限科技发展有限公司 | System and method for determining traffic condition |
| CN110807913B (en) * | 2018-08-06 | 2021-08-10 | 北京嘀嘀无限科技发展有限公司 | System and method for determining traffic condition |
| CN109147363A (en) * | 2018-09-10 | 2019-01-04 | 吉林大学 | Traffic intelligent guides system and bootstrap technique |
| CN110111590A (en) * | 2019-06-04 | 2019-08-09 | 南京慧尔视智能科技有限公司 | A kind of vehicle dynamic queue length detection method |
| CN110751829A (en) * | 2019-09-26 | 2020-02-04 | 同济大学 | Vehicle queuing dissipation time prediction method based on image self-learning |
| CN110751829B (en) * | 2019-09-26 | 2021-06-04 | 同济大学 | Vehicle queuing dissipation time prediction method based on image self-learning |
| CN111554111A (en) * | 2020-04-21 | 2020-08-18 | 河北万方中天科技有限公司 | Signal timing optimization method and device based on multi-source data fusion and terminal |
| CN111932916A (en) * | 2020-07-31 | 2020-11-13 | 长沙理工大学 | Control method and control system for dynamic emergency lane of urban road |
| WO2022116361A1 (en) * | 2020-12-01 | 2022-06-09 | 山东交通学院 | Traffic light control method and system based on urban trunk line vehicle queuing length |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102622897B (en) | 2014-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102622897B (en) | Video-based dynamic vehicle queue length estimation method | |
| CN102280036B (en) | Bus rapid transit signal prior timing method under trunk line coordination control | |
| CN103280113B (en) | Self-adaptive intersection signal control method | |
| CN103559795B (en) | Multi-strategy and multi-object self-adaptation traffic control method | |
| CN104103180B (en) | City expressway enters ring road and main line synergistic signal control system and method | |
| CN107730886B (en) | Dynamic optimization method for traffic signals at urban intersections in Internet of vehicles environment | |
| CN105844923B (en) | A kind of induction type pedestrian's street crossing signal control method and system | |
| CN103680157B (en) | A kind of vehicle queue's overflow pre-judging method towards city bottleneck road | |
| CN103680163B (en) | Based on the region bus signals mode of priority of public transport demand | |
| CN102568223B (en) | Signal crossing right turn special lane real-time dynamic control method | |
| CN105321357B (en) | The setting system and method that a kind of isolated intersection same-phase is turned right with pedestrian signal | |
| EP3631616A1 (en) | Road traffic control system, method, and electronic device | |
| CN104575034B (en) | A kind of single-point intersection signal timing dial parameter optimization method based on bayonet socket data | |
| CN108615376A (en) | A kind of integrative design intersection schemes evaluation method based on video detection | |
| CN103236164B (en) | Vehicle controlling method for guaranteeing public transport vehicle prior passing | |
| CN104575038A (en) | Intersection signal control method considering priority of multiple buses | |
| CN108510762A (en) | A kind of multi-thread confluence intelligent signal lamp optimal control method in through street | |
| CN106710245A (en) | Ramp control method for multiple lanes of expressway based on density | |
| CN106683447B (en) | A kind of traffic lamp control method and device | |
| CN106469507A (en) | Dynamic green wave system and method automatically adjusted based on Real-Time Traffic Volume data | |
| CN104505101A (en) | Real-time audio comparison method | |
| CN105405301A (en) | Right-turn signal induction control method for eliminating straight-right-turn convergence conflict | |
| CN105006158A (en) | Single road intersection traffic signal control method based on real-time traffic information | |
| CN103268707A (en) | Signal regulating method for pedestrian crossing road section of bus prior passage | |
| CN103679399A (en) | Station passenger flow linkage management and control method |
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 |
Granted publication date: 20140430 Termination date: 20170407 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |