CN109472184A - The condition detection method in berth, system and its data processing equipment in road - Google Patents
The condition detection method in berth, system and its data processing equipment in road Download PDFInfo
- Publication number
- CN109472184A CN109472184A CN201710806048.5A CN201710806048A CN109472184A CN 109472184 A CN109472184 A CN 109472184A CN 201710806048 A CN201710806048 A CN 201710806048A CN 109472184 A CN109472184 A CN 109472184A
- Authority
- CN
- China
- Prior art keywords
- berth
- image
- type
- standard
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to condition detection method, system and its data processing equipment in berth in a kind of road, the condition detection method in berth includes: the image data for obtaining camera in the road, wherein the monitoring area of the camera covers multiple berth regions；The berth image in each berth is intercepted from described image data, and determines the type of the berth image；According to identified type, modification of dimension is carried out to the berth image by preset rules；Berth image after modification of dimension is pre-processed, and extracts the characteristic information of the berth image；The characteristic information is input in the berth state model of the respective type pre-established, and determines the state in the berth according to the output result of the berth state model.Implement technical solution of the present invention, disposes simple, at low cost.
Description
Technical field
The present invention relates to the fields intelligent transportation (Intelligent Transportation System, ITS), especially relate to
And the condition detection method in berth, system and its data processing equipment in a kind of road.
Background technique
With the motorization progress faster of domestic large- and-medium size cities, the traffic trip demand rapid growth of citizen, trip side
Formula is increasingly sophisticated, and the randomness of vehicle parking and random implementations are commonplace, cause this crowded road grid traffic pressure into
One step increases.Therefore, imperative to vehicle parking especially curb parking progress orderly management.By implementation " with economic hand
Supplemented by Duan Weizhu+administration means " road-surface concrete toll project, the parking behavior of specification citizen, reasonable distribution parking resource draws
Lead citizen and form good parking habit, the perfect collaboration being finally reached between each element of traffic, allow citizen city life more
Add fine.
Road is more complicated in road, and berth state automaticly inspects mode and usually solved by earth magnetism, radio frequency and video in road
Certainly, still, in these existing schemes, earth magnetism needs to dig road and is laid with ground magnetic machine；Radio frequency needs vehicle to be coupled in advance to penetrate
Frequency signal receiver；Video detection is more flexible, and the method for present mainstream is that portal frame is set up among road, or each
One camera is installed beside berth., commonly there is deployment difficulty in above several schemes or cost is excessively high.
Summary of the invention
The technical problem to be solved in the present invention is that being mentioned for above-mentioned deployment difficulty, the defect at high cost of the prior art
For condition detection method, system and its data processing equipment in berth in a kind of road, dispose simple and at low cost.
The technical solution adopted by the present invention to solve the technical problems is: constructing a kind of state-detection side in berth in road
Method, comprising:
Obtain the image data of camera, wherein the monitoring area of the camera covers multiple berth regions；
The berth image in each berth is intercepted from described image data, and determines the type of the berth image；
According to identified type, modification of dimension is carried out to the berth image by preset rules；
Berth image after modification of dimension is pre-processed, and extracts the characteristic information of the berth image；
The characteristic information is input in the berth state model of the respective type pre-established, and according to the berth
The output result of state model determines the state in the berth.
Preferably, berth state model is established according to the following steps:
S11. in training, modification of dimension is carried out by preset rules to the berth image of same type；
S12. the berth image after modification of dimension is pre-processed；
S13. characteristic information is extracted from pretreated berth image；
S14. preset training method is used, the characteristic information and berth state are trained, to obtain respective class
The berth state model of type.
Preferably, the step of determining the type of the berth image include:
Obtain the parameter information of the berth image；
It is determining to be moored with the immediate standard of the berth image according to the standard berth image of the multiple types pre-established
Bit image, and type corresponding to the immediate standard berth image is determined as to the type of the berth image.
Preferably, according to the standard berth image of the multiple types pre-established, determination is closest with the berth image
Standard berth image the step of include:
S21. from the image of multiple standard berths, determine that direction, orientation and direction, the orientation of the berth image are all the same
Standard berth image；
S22. intermediate standard image is set by identified standard berth image；
S23. according to the berth image and each intermediate standard berth image, judge the distance of two berth images
Whether ratio is greater than the first preset value；And/or judge whether the similarity of two berth images is greater than the second preset value；
S24. immediate standard berth is set by the intermediate standard berth image of the condition met in the step S23
Image.
Preferably, in the step S23, judge whether the distance ratio of two berth images is greater than the first preset value
Step includes:
S231. calculate separately the length or width of the berth image and the length of intermediate standard berth image or
Width；
S232. the lenth ratio or width ratio of the berth image Yu intermediate standard berth image are calculated, and is sentenced
Break lenth ratio calculated or whether width ratio is greater than the first preset value；
Alternatively, judging that the step of whether similarity of two berth images is greater than the second preset value includes:
S233. having in effective berth region and the intermediate standard berth image in the berth image is determined respectively
Imitate berth region；
S234. it is placed on the berth image is Chong Die with intermediate standard berth image in the same coordinate system, and really
The overlapping region in both fixed effective berth region；
S235. the area ratio of the overlapping region and the greater in the effective berth region of the two is calculated, and is judged
Whether area ratio calculated is greater than the second preset value.
Preferably, further includes:
Judge whether the current state in berth changes compared to previous moment state；
If changing, the timing node of state change is obtained；
According to the timing node and the state of berth variation front and back, determine vehicle drives into time or the when of being driven out to
Between.
Preferably, the pretreatment includes filtering processing, image gray processing processing, histogram equalization processing；Alternatively, institute
Stating characteristic information includes histogram feature information, LBP characteristic information or corner feature information.
The present invention also constructs a kind of data processing equipment of the condition detecting system in berth in road characterized by comprising
Image collection module, for obtaining the image data of camera, wherein the monitoring area covering of the camera is more
A berth region；
Determination type module for intercepting the berth image in each berth from described image data, and determines the pool
The type of bit image；
Modification of dimension module, for carrying out size to the berth image by preset rules and repairing according to identified type
Change；
Preprocessing module for pre-processing to the berth image after modification of dimension, and extracts the berth image
Characteristic information；
State determining module, for the characteristic information to be input to the berth state model of the respective type pre-established
In, and determine according to the output result of the berth state model state in the berth.
The present invention also constructs a kind of data processing equipment of the condition detecting system in berth in road, including memory and processing
Device, the memory are stored with computer program, and the processor is for executing the computer program stored in the memory
And realize above method.
The present invention also constructs a kind of condition detecting system in berth in road, comprising:
The monitoring area of camera, the camera covers multiple berth regions；
Above-described data processing equipment.
Implement technical solution of the present invention, since a camera can monitor multiple berth regions simultaneously, so, work as reception
To after image data captured by camera, the berth image in each berth is therefrom first isolated, then determine each berth image
Type.Then, the size that berth image is modified according to identified type, pre-processes, and extract when to berth image
After characteristic information, the state in the berth is determined according to berth state model corresponding to identified type, to reach real-time
The purpose for monitoring berth state can find abnormal (barrier, bicycle, the tricycle, pedestrian occurred in berth in berth in time
Deng), convenient for law enforcement and management.Moreover, the condition detection method compares existing several method, dispose simple and environmentally-friendly, at low cost
It is honest and clean.
Detailed description of the invention
In order to illustrate the embodiments of the present invention more clearly, attached drawing needed in describing below to embodiment makees letter
Singly introduce, it should be apparent that, drawings in the following description are only some embodiments of the invention, skill common for this field
For art personnel, without creative efforts, it is also possible to obtain other drawings based on these drawings.Attached drawing
In:
Fig. 1 is the flow chart of the condition detection method embodiment one in berth in road of the present invention；
Fig. 2 is the schematic diagram in berth region in camera and road；
Fig. 3 is the schematic diagram of a frame image captured by camera；
Fig. 4 is the schematic diagram of the cuboid in determining each berth；
Fig. 5 is the schematic diagram according to the intercepted berth image of identified cuboid；
Fig. 6 A to Fig. 6 L is the schematic diagram of the berth image of different shapes intercepted respectively；
Fig. 7 is the schematic diagram in the effective berth region of determining berth image；
Fig. 8 A to Fig. 8 C is the schematic diagram for calculating the similarity of two berth images；
Fig. 9 is the building-block of logic of the data processing equipment of the condition detecting system in berth in road of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The present invention is based on video detection scheme, camera frame is located at beside way, and single camera can detecte multiple
Berth state is mainly determined by camera overlay area.In curb parking practical application, detectable berth state is mainly wrapped
With vehicles and car-free status is included, it is of course also possible to include the detection of abnormality in berth, such as across berth parking, berth quilt
Rubbish piles with, stops the states such as non power driven vehicle in berth.It drives at the same time it can also further detecting vehicle and berth and is driven out to pool
The timing node of position.
Technical solution of the present invention is divided into two processes when implementing: static data is default, real-time detection and exports knot
Fruit.By the default berth state model that each type can be obtained ahead of time of static data, video data is then acquired in real time, is utilized
Preset berth state model carries out the output of berth state.It should be understood that position, height, angle of camera etc. are in two mistakes
Cheng Zhongying is consistent, once the position of camera, height, angle etc. change, it is default should to re-start static data.
Fig. 1 is the flow chart of the condition detection method embodiment one in berth in road of the present invention, berth in the road of the embodiment
Condition detection method the following steps are included:
S10. the image data of camera is obtained, wherein the monitoring area of the camera covers multiple berth regions；
In this step, illustrate first, camera can be erected at beside road or be mounted on existing capital construction facility
On, and a camera can monitor multiple berth regions, such as 4-10 simultaneously.The type of camera can image for machine gun
Head can also be dome camera.Biggish monitoring area in order to obtain, camera are needed according to the environment at lane scene to installation
Height and drift angle be adjusted.In the example shown in Fig. 2,16 berths are set in road altogether, gunlock camera A can be monitored
The region in berth 1,2,9,10,11,12,13, and dome camera B can monitor the area in berth 3,4,5,6,12,13,14
Domain.It should be understood that actual field only need to guarantee that monitoring area covers and need to examine to the no particular/special requirement of the selection of camera types
Survey berth region.
S20. the berth image in each berth is intercepted from described image data, and determines the type of the berth image；
In this step, illustrate first, since camera can monitor multiple berth regions simultaneously, so, camera
It include the image in multiple berths in captured every frame image.Based on this, before state recognition, need first from complete image
The berth image in each berth is individually intercepted in frame and is come out.In addition, since each berth is different with a distance from camera, together
When due to camera types it is inconsistent, image has buckling phenomenon, is to have certain radian, as shown in Figure 3.In view of these
Factor can be using the stereo-picture in the berth as prototype, and according to berth solid area when dividing the berth image in each berth
The maximum boundary in domain intercepts rectangular area.In the example shown in Figure 4, include in the image of camera A shooting berth 1,2,
9,10,11,12,13 berth image, it is first determined the solid region in each berth, specifically: by way of manually demarcating
(such as each vertex of calibration cuboid) determines cuboid corresponding to each berth, and the bottom surface of cuboid is berth region.
Then, rectangular area is taken according to the maximum boundary of the cuboid, so that the image in the rectangular area is completely shown identified
Cuboid.Finally, the rectangular area is intercepted out, the as berth image in the berth, as shown in Figure 5.
In addition, it should be noted that, due to by factors such as camera types, the distance of camera covering, drift angle and height
Influence, the berth image in each berth intercepted out has different shapes, and as shown in Fig. 6 A to 6L, therefore, it is necessary to right
Berth image is classified.In classification, the standard berth image of several types can be predefined, and set type identification.?
In practical application, using scheduled type as reference, selection and the immediate standard berth figure of berth image intercepted
Picture, and using the type of the immediate standard berth image as oneself type.With the increase of data volume, berth image
Type can dynamically increase, and more accurately be matched with improving.
S30. according to identified type, modification of dimension is carried out to the berth image by preset rules；
It in this step, can be according to the standard corresponding to the type after the type of intercepted berth image has been determined
Size carries out modification of dimension to the berth image, for example, making its modified size be equal to identified class by modification of dimension
The error of standard size corresponding to type, or both within a preset range, with the size of unified same type of berth image.
S40. the berth image after modification of dimension is pre-processed, and extracts the characteristic information of the berth image；
In this step, pretreatment is for example including filtering processing, image gray processing processing, histogram equalization processing.Separately
Outside, when extracting characteristic information, histogram spy can be extracted according to calculation methods such as histogram feature, LBP feature, Corner Detections
Reference breath, LBP characteristic information or corner feature information.
S50. the characteristic information is input in the berth state model of the respective type pre-established, and according to described
The output result of berth state model determines the state in the berth.
In this step, illustrate first, it is trained that berth state model is that the standard berth image of the type passes through
The agent model arrived, so, different berth types, obtained agent model is different.In practical applications, according to spy
Reference ceases (such as LBP feature) and corresponding agent model, is verified using SVM support vector machines, obtains the berth image
Berth state (for example, output 1 represent have vehicle；Output 0 is represented without vehicle).
In a specific embodiment, the berth state model of each type can be established according to following steps:
S11. in training, modification of dimension is carried out by berth image of the preset rules to same type；
S12. the berth image after modification of dimension is pre-processed；
S13. characteristic information is extracted from pretreated berth image；
S14. preset training method is used, the characteristic information and berth state are trained, to obtain respective class
The berth state model of type.
In this embodiment, illustrate first, in training, should be instructed respectively for same type of berth image
Practice, therefore, acquiring image data captured by camera, and after having intercepted the berth image in each berth, it should first really
The type of fixed each berth image.Then, it for same type of berth image, first unified image size, then is pre-processed
And feature extraction, wherein pretreated method includes but is not limited to the methods of filtering, image gray processing, histogram equalization, spy
The method that sign is extracted includes but is not limited to the calculation methods such as histogram feature, LBP feature, Corner Detection.Finally, selection is suitable
Training method, in conjunction with the state (having vehicle, without vehicle, pedestrian etc.) of berth image, training obtains the agent model of the type, wherein
Training method includes but is not limited to use the modes such as SVM support vector machines, artificial neural network, deep learning.
When classifying to berth image, can according to camera cover berth region according to distance, drift angle, height,
Berth length-width ratio and other parameters are classified, and can remove interference of the inactive area to image recognition in this way, improve accuracy rate.
Specifically, it is determined that the step of type of berth image can include:
Obtain the parameter information of the berth image；
It is determining to be moored with the immediate standard of the berth image according to the standard berth image of the multiple types pre-established
Bit image, and type corresponding to the immediate standard berth image is determined as to the type of the berth image.
In this embodiment, the parameter about berth image, it should be noted that, it is divided into preset parameter and dynamic parameter.
Wherein, preset parameter includes berth orientation and berth direction, moreover, berth orientation is the ipsilateral or opposite side in camera,
For example, ipsilateral orientation type is 1, the orientation type of opposite side is 2；Berth direction includes direction and endways direction sidewards, for example,
Direction is 1 sidewards, and endways direction is 2.Dynamic parameter include the berth image and standard berth image distance ratio and/
Or similarity.
In a specific embodiment, according to the standard berth image of the multiple types pre-established, the determining and pool
The step of bit image immediate standard berth image includes:
S21. from the image of multiple standard berths, determine that direction, orientation and direction, the orientation of the berth image are all the same
Standard berth image；
S22. intermediate standard image is set by identified standard berth image；
S23. according to the berth image and each intermediate standard berth image, judge the distance of two berth images
Whether ratio is greater than the first preset value；And/or judge whether the similarity of two berth images is greater than the second preset value；
In this step, it should be noted that, either distance ratio or similarity, with related, the example of parking stall selection
Such as, if there is No. 1-4 four parking stalls, No. 1 parking stall and No. 2 parking stall ratios are 0.9；No. 1 parking stall and No. 4 parking stalls be not necessarily 0.9 with
It is interior.
S24. immediate standard berth is set by the intermediate standard berth image of the condition met in the step S23
Image.
Preferably, the step of whether distance ratio of two berth images is greater than the first preset value packet is judged in step S23
It includes:
S231. calculate separately the length or width of the berth image and the length of intermediate standard berth image or
Width calculates the length or width of berth image, that is, calculate the boundary rectangle length of the stereo-picture in berth in this step
Or width, this parameter can indicate the far and near distance in berth；
S232. the lenth ratio or width ratio of the berth image Yu intermediate standard berth image are calculated, and is sentenced
Break lenth ratio calculated or whether width ratio is greater than the first preset value, the range of the first preset value is, for example, 0.9-1.
Preferably, judge that the step of whether similarity of two berth images is greater than the second preset value includes: in step S23
S233. having in effective berth region and the intermediate standard berth image in the berth image is determined respectively
Imitate berth region；
In this step, it should be noted that, at the effective berth region for determining berth image, it can first determine that berth figure
Inactive area (region i.e. other than the cuboid of berth) as in then carries out masking-out melanism to the inactive area and (or goes to carry on the back
Scape), the region after inactive area is got rid of in the image of berth is the effective coverage of the berth image, for example, to shown in fig. 5
After berth image removes inactive area, extracted effective coverage is as shown in Figure 7.Herein it should be noted that, either berth figure
It is by being no less than 4 points as (rectangular area intercepted from the image data of shooting) or the effective coverage of berth image
The face of composition, for example, the effective coverage of a berth image serve as reasons (744,298), (847,357), (727,536), (390,
537), (373,513) } five point compositions face.
A berth image A11 is shown in conjunction with Fig. 8 A and Fig. 8 B, Fig. 8 A, and its effective berth region (dash area) is
A12, Fig. 8 B show an intermediate standard berth image A21, and its effective berth region (dash area) is A22.
S234. it is placed on the berth image is Chong Die with intermediate standard berth image in the same coordinate system, and really
The overlapping region in both fixed effective berth region；
In this step, in conjunction with Fig. 8 C, two berth images shown in Fig. 8 A, Fig. 8 B are put into the same coordinate system, and
And it is maximum to make its overlapping region, alternatively, guaranteeing the overlapping of standard point when same type, at this point, can determine that the effective of the two
The overlapping region in berth region, i.e. overlapping region (diamond shape grid spaces) are A1222.
S235. the area ratio of the overlapping region and the greater in the effective berth region of the two is calculated, and is judged
Whether area ratio calculated is greater than the second preset value.
In this step, when overlapping region has been determined, the area of the overlapping region can be calculated, and then determine two berths
The area of effective berth region the greater in image, it is clear that effective coverage the greater is having for berth image A11 in the embodiment
Imitate berth region A12.Then, ratio is asked to two areas calculated, area ratio can be obtained.The range of second preset value
For example, 0.85-1.
In a specific example, when carrying out berth classification using sorting parameter, berth direction and berth orientation are can be with
It intuitively determines, distance ratio and similarity needs compare to determine.It, can be pre- for the standard berth image of each type
Its parameter and standard is first established, for example, the parameter list that berth type is 1 are as follows:
| Berth direction | Berth orientation | AREA1 | AREA2 | Berth type |
| 1 | 1 | { (744,298) ... } | { (234,248) ... } | 1 |
When berth Class1 is added in new berth image, it is necessary first to berth direction and berth orientation are consistent,
The distance ratio and similarity in two berths are calculated, again to decide whether to meet the requirement of the berth type.For example, following table is listed
Whether several situations that berth type be 1 are met:
About the classification of berth image, finally it should be noted that, if new berth region is increased, in existing pool
All without matched type in the type of position, new berth type can be established according to the berth parameter of oneself at this time.In addition, also
It should be noted that the process of above-mentioned classification is only used when camera is shot for the first time, once shooting figure has been determined by this process
The type of each berth image as in, the type of the berth image of the corresponding region in every frame image of subsequent shooting are all consistent.
Further, in road of the invention berth condition detection method, after berth state has been determined, may also include with
Lower step:
Judge whether the current state in berth changes compared to previous moment state；
If changing, the timing node of state change is obtained；
According to the timing node and the state of berth variation front and back, determine vehicle drives into time or the when of being driven out to
Between.For example, the state in some berth is become having vehicle from no vehicle, then the timing node is to drive into the time；The state in some berth by
There is vehicle to become no vehicle, then the timing node is to be driven out to the time.
The present invention also constructs a kind of condition detecting system in berth in road, which includes taking the photograph for communication connection
As head and data processing equipment, wherein the monitoring area of camera covers multiple berth regions.
Fig. 9 is the building-block of logic of the data processing equipment embodiment one of the condition detecting system in berth in road of the present invention,
The data processing equipment includes: image collection module 10, determination type module 20, modification of dimension module 30, preprocessing module 40
With state determining module 50, wherein image collection module 10 is used to obtain the image data of camera, wherein the camera
Monitoring area cover multiple berth regions；Determination type module 20 is used to intercept the pool in each berth from described image data
Bit image, and determine the type of the berth image；Modification of dimension module 30 is used for according to identified type, by preset rules
Modification of dimension is carried out to the berth image；Preprocessing module 40 is used to pre-process the berth image after modification of dimension,
And extract the characteristic information of the berth image；State determining module 50 is used to for the characteristic information being input to pre-establish
In the berth state model of respective type, and determine according to the output result of the berth state model state in the berth.
The present invention also constructs a kind of data processing equipment of the condition detecting system in berth in road, the data processing equipment packet
Memory and processor are included, the memory is stored with computer program, which is characterized in that the processor is described for executing
The computer program that stores in memory simultaneously realizes above method.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any bun
Change, equivalent replacement, improvement etc., should be included within scope of the presently claimed invention.
Claims (10)
1. the condition detection method in berth in a kind of road characterized by comprising
Obtain the image data of camera, wherein the monitoring area of the camera covers multiple berth regions；
The berth image in each berth is intercepted from described image data, and determines the type of the berth image；
According to identified type, modification of dimension is carried out to the berth image by preset rules；
Berth image after modification of dimension is pre-processed, and extracts the characteristic information of the berth image；
The characteristic information is input in the berth state model of the respective type pre-established, and according to the berth state
The output result of model determines the state in the berth.
2. the condition detection method in berth in road according to claim 1, which is characterized in that berth state model is under
Column step is established:
S11. in training, modification of dimension is carried out by preset rules to the berth image of same type；
S12. the berth image after modification of dimension is pre-processed；
S13. characteristic information is extracted from pretreated berth image；
S14. preset training method is used, the characteristic information and berth state are trained, to obtain respective type
Berth state model.
3. the condition detection method in berth in road according to claim 2, which is characterized in that determine the berth image
The step of type includes:
Obtain the parameter information of the berth image；
According to the standard berth image of the multiple types pre-established, the determining and immediate standard berth figure of the berth image
Picture, and type corresponding to the immediate standard berth image is determined as to the type of the berth image.
4. the condition detection method in berth in road according to claim 3, which is characterized in that multiple according to what is pre-established
The standard berth image of type, determine standard berth image immediate with the berth image the step of include:
S21. from the image of multiple standard berths, direction, the orientation of direction, orientation and the berth image mark all the same are determined
Quasi- berth image；
S22. intermediate standard image is set by identified standard berth image；
S23. according to the berth image and each intermediate standard berth image, judge the distance ratio of two berth images
Whether the first preset value is greater than；And/or judge whether the similarity of two berth images is greater than the second preset value；
S24. immediate standard berth figure is set by the intermediate standard berth image of the condition met in the step S23
Picture.
5. the condition detection method in berth in road according to claim 4, which is characterized in that in the step S23, sentence
The step of whether distance ratio of disconnected two berth images is greater than the first preset value include:
S231. the length or width of the berth image and the length or width of intermediate standard berth image are calculated separately；
S232. the lenth ratio or width ratio of the berth image Yu intermediate standard berth image are calculated, and judges institute
Whether the lenth ratio or width ratio of calculating are greater than the first preset value；
Alternatively, judging that the step of whether similarity of two berth images is greater than the second preset value includes:
S233. the effective berth region in the berth image and effective pool in the image of the intermediate standard berth are determined respectively
Position region；
S234. it is placed on the berth image is Chong Die with intermediate standard berth image in the same coordinate system, and determines two
The overlapping region in the effective berth region of person；
S235. the area ratio of the overlapping region and the greater in the effective berth region of the two is calculated, and judges to be counted
Whether the area ratio of calculation is greater than the second preset value.
6. the condition detection method in berth in road according to claim 1, which is characterized in that further include:
Judge whether the current state in berth changes compared to previous moment state；
If changing, the timing node of state change is obtained；
According to the timing node and the state of berth variation front and back, determine vehicle drives into time or the time of being driven out to.
7. the condition detection method in berth in road according to claim 1, which is characterized in that the pretreatment includes filtering
Processing, image gray processing processing, histogram equalization processing；Alternatively, the characteristic information includes histogram feature information, LBP
Characteristic information or corner feature information.
8. the data processing equipment of the condition detecting system in berth in a kind of road characterized by comprising
Image collection module, for obtaining the image data of camera, wherein the monitoring area of the camera covers multiple pools
Position region；
Determination type module for intercepting the berth image in each berth from described image data, and determines the berth figure
The type of picture；
Modification of dimension module, for carrying out modification of dimension to the berth image by preset rules according to identified type；
Preprocessing module for pre-processing to the berth image after modification of dimension, and extracts the feature of the berth image
Information；
State determining module, for the characteristic information being input in the berth state model of the respective type pre-established,
And the state in the berth is determined according to the output result of the berth state model.
9. the data processing equipment of the condition detecting system in berth in a kind of road, including memory and processor, the memory
It is stored with computer program, which is characterized in that the processor is used to execute the computer program stored in the memory simultaneously
Realize method described in any one of claim 1-7.
10. the condition detecting system in berth in a kind of road characterized by comprising
The monitoring area of camera, the camera covers multiple berth regions；
Data processing equipment described in claim 8 or 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710806048.5A CN109472184A (en) | 2017-09-08 | 2017-09-08 | The condition detection method in berth, system and its data processing equipment in road |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710806048.5A CN109472184A (en) | 2017-09-08 | 2017-09-08 | The condition detection method in berth, system and its data processing equipment in road |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109472184A true CN109472184A (en) | 2019-03-15 |
Family
ID=65657833
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710806048.5A Pending CN109472184A (en) | 2017-09-08 | 2017-09-08 | The condition detection method in berth, system and its data processing equipment in road |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109472184A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110126819A (en) * | 2019-05-10 | 2019-08-16 | 北京长城华冠汽车技术开发有限公司 | Pass through park control method and the system of recognition of face |
| CN111429469A (en) * | 2019-04-17 | 2020-07-17 | 杭州海康威视数字技术股份有限公司 | Parking position determining method and device, electronic equipment and storage medium |
| CN114005299A (en) * | 2021-10-29 | 2022-02-01 | 中汽创智科技有限公司 | Parking control method, system, equipment and storage medium |
| CN114040104A (en) * | 2021-11-16 | 2022-02-11 | 北京筑梦园科技有限公司 | Equipment debugging method and device and parking management system |
| CN114283371A (en) * | 2022-03-07 | 2022-04-05 | 深圳海润游艇码头工程有限公司 | Intelligent monitoring method and device for yacht wharf berthing platform |
| CN115082571A (en) * | 2022-07-20 | 2022-09-20 | 深圳云游四海信息科技有限公司 | Anomaly detection method and system for in-road parking camera |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101414411A (en) * | 2007-10-17 | 2009-04-22 | 财团法人工业技术研究院 | Image type vacancy detection system and method |
| CN102110376A (en) * | 2011-02-18 | 2011-06-29 | 汤一平 | Roadside parking space detection device based on computer vision |
| CN103093643A (en) * | 2013-01-09 | 2013-05-08 | 东南大学 | Public parking lot berth quantity confirming method |
| CN103473950A (en) * | 2012-06-06 | 2013-12-25 | 刘鉵 | Parking lot parking space monitoring method |
| US20140376769A1 (en) * | 2013-06-20 | 2014-12-25 | Xerox Corporation | Method for detecting large size and passenger vehicles from fixed cameras |
| CN105160322A (en) * | 2015-09-07 | 2015-12-16 | 北京航空航天大学 | Outdoor parking lot non-occupied parking stall identification method based on aerial photography images |
| CN105390021A (en) * | 2015-11-16 | 2016-03-09 | 北京蓝卡科技股份有限公司 | Parking spot state detection method and parking spot state detection device |
| US20160275376A1 (en) * | 2015-03-20 | 2016-09-22 | Netra, Inc. | Object detection and classification |
| CN106469518A (en) * | 2015-08-21 | 2017-03-01 | 中国电信股份有限公司 | The method of the intelligent parking guiding based on video tracking identification, server and system |
| CN106504583A (en) * | 2017-01-05 | 2017-03-15 | 陕西理工学院 | The cloud monitoring method of parking position state, cloud server and cloud monitoring system |
| CN106846891A (en) * | 2017-03-02 | 2017-06-13 | 浙江大学 | A kind of Public Parking berth multistep forecasting method decomposed based on sequence |
| CN106991838A (en) * | 2016-01-20 | 2017-07-28 | 上海海拉电子有限公司 | A kind of parking spot tracking system |
| CN109087523A (en) * | 2018-07-27 | 2018-12-25 | 上海与德通讯技术有限公司 | A kind of parking method and server |
-
2017
- 2017-09-08 CN CN201710806048.5A patent/CN109472184A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101414411A (en) * | 2007-10-17 | 2009-04-22 | 财团法人工业技术研究院 | Image type vacancy detection system and method |
| CN102110376A (en) * | 2011-02-18 | 2011-06-29 | 汤一平 | Roadside parking space detection device based on computer vision |
| CN103473950A (en) * | 2012-06-06 | 2013-12-25 | 刘鉵 | Parking lot parking space monitoring method |
| CN103093643A (en) * | 2013-01-09 | 2013-05-08 | 东南大学 | Public parking lot berth quantity confirming method |
| US20140376769A1 (en) * | 2013-06-20 | 2014-12-25 | Xerox Corporation | Method for detecting large size and passenger vehicles from fixed cameras |
| US20160275376A1 (en) * | 2015-03-20 | 2016-09-22 | Netra, Inc. | Object detection and classification |
| CN106469518A (en) * | 2015-08-21 | 2017-03-01 | 中国电信股份有限公司 | The method of the intelligent parking guiding based on video tracking identification, server and system |
| CN105160322A (en) * | 2015-09-07 | 2015-12-16 | 北京航空航天大学 | Outdoor parking lot non-occupied parking stall identification method based on aerial photography images |
| CN105390021A (en) * | 2015-11-16 | 2016-03-09 | 北京蓝卡科技股份有限公司 | Parking spot state detection method and parking spot state detection device |
| CN106991838A (en) * | 2016-01-20 | 2017-07-28 | 上海海拉电子有限公司 | A kind of parking spot tracking system |
| CN106504583A (en) * | 2017-01-05 | 2017-03-15 | 陕西理工学院 | The cloud monitoring method of parking position state, cloud server and cloud monitoring system |
| CN106846891A (en) * | 2017-03-02 | 2017-06-13 | 浙江大学 | A kind of Public Parking berth multistep forecasting method decomposed based on sequence |
| CN109087523A (en) * | 2018-07-27 | 2018-12-25 | 上海与德通讯技术有限公司 | A kind of parking method and server |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111429469A (en) * | 2019-04-17 | 2020-07-17 | 杭州海康威视数字技术股份有限公司 | Parking position determining method and device, electronic equipment and storage medium |
| CN111429469B (en) * | 2019-04-17 | 2023-11-03 | 杭州海康威视数字技术股份有限公司 | Berth position determining method and device, electronic equipment and storage medium |
| CN110126819A (en) * | 2019-05-10 | 2019-08-16 | 北京长城华冠汽车技术开发有限公司 | Pass through park control method and the system of recognition of face |
| CN114005299A (en) * | 2021-10-29 | 2022-02-01 | 中汽创智科技有限公司 | Parking control method, system, equipment and storage medium |
| CN114040104A (en) * | 2021-11-16 | 2022-02-11 | 北京筑梦园科技有限公司 | Equipment debugging method and device and parking management system |
| CN114283371A (en) * | 2022-03-07 | 2022-04-05 | 深圳海润游艇码头工程有限公司 | Intelligent monitoring method and device for yacht wharf berthing platform |
| CN115082571A (en) * | 2022-07-20 | 2022-09-20 | 深圳云游四海信息科技有限公司 | Anomaly detection method and system for in-road parking camera |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109472184A (en) | The condition detection method in berth, system and its data processing equipment in road | |
| US10846874B2 (en) | Method and apparatus for processing point cloud data and storage medium | |
| Yu et al. | Semiautomated extraction of street light poles from mobile LiDAR point-clouds | |
| CN106373426B (en) | Parking stall based on computer vision and violation road occupation for parking monitoring method | |
| US10388153B1 (en) | Enhanced traffic detection by fusing multiple sensor data | |
| CN112216049B (en) | Construction warning area monitoring and early warning system and method based on image recognition | |
| CN103824070B (en) | A kind of rapid pedestrian detection method based on computer vision | |
| Guan et al. | Using mobile LiDAR data for rapidly updating road markings | |
| CN102542289B (en) | Pedestrian volume statistical method based on plurality of Gaussian counting models | |
| CN102708356B (en) | Automatic license plate positioning and recognition method based on complex background | |
| US9373043B2 (en) | Method and apparatus for detecting road partition | |
| CN102999753B (en) | License plate locating method | |
| CN104036262B (en) | A kind of method and system of LPR car plates screening identification | |
| CN104951775A (en) | Video technology based secure and smart recognition method for railway crossing protection zone | |
| CN102867417A (en) | Taxi anti-forgery system and taxi anti-forgery method | |
| KR20160109761A (en) | Method and System for Recognition/Tracking Construction Equipment and Workers Using Construction-Site-Customized Image Processing | |
| CN102915433A (en) | Character combination-based license plate positioning and identifying method | |
| CN113095301B (en) | Road occupation operation monitoring method, system and server | |
| Tian et al. | Fast cyclist detection by cascaded detector and geometric constraint | |
| CN110033622A (en) | Violation snap-shooting based on unmanned plane aerial photography technology occupies Emergency Vehicle Lane method | |
| Zhang et al. | Rapid inspection of pavement markings using mobile LiDAR point clouds | |
| CN106339657A (en) | Straw incineration monitoring method and device based on monitoring video | |
| CN104915678A (en) | Method and device for detecting target object in power transmission line | |
| CN109255279A (en) | A kind of method and system of road traffic sign detection identification | |
| CN110490150A (en) | A kind of automatic auditing system of picture violating the regulations and method based on vehicle retrieval |
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: 20190315 |