CN110009929A - A kind of Vehicle berth management method, equipment and system - Google Patents
A kind of Vehicle berth management method, equipment and system Download PDFInfo
- Publication number
- CN110009929A CN110009929A CN201910195918.9A CN201910195918A CN110009929A CN 110009929 A CN110009929 A CN 110009929A CN 201910195918 A CN201910195918 A CN 201910195918A CN 110009929 A CN110009929 A CN 110009929A
- Authority
- CN
- China
- Prior art keywords
- image
- berth
- vehicle
- same
- present
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07B—TICKET-ISSUING APPARATUS; FARE-REGISTERING APPARATUS; FRANKING APPARATUS
- G07B15/00—Arrangements or apparatus for collecting fares, tolls or entrance fees at one or more control points
- G07B15/02—Arrangements or apparatus for collecting fares, tolls or entrance fees at one or more control points taking into account a variable factor such as distance or time, e.g. for passenger transport, parking systems or car rental systems
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Finance (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of Vehicle berth management methods, equipment and system, method therein includes obtaining image collecting device to the image of the region acquisition comprising at least one berth, pass through the judgement to berth occupied state in two images, and for the whether identical judgement of corresponding vehicle in same berth, it can determine simultaneously and the parking beginning and ending time of the vehicle in multiple berths is determined, the management of Vehicle berth can be completed by a small amount of image by the present invention, can greatly reduce the requirement that power is calculated equipment, reduce flow and hardware cost, the time interval of two images is obtained by arbitrarily adjusting, convenient for being balanced in the accuracy rate of parking charge and image uplink time and calculating cost, in addition, also improve the efficiency of more berth management.
Description
Technical field
The present invention relates to field of image recognition, and in particular to a kind of Vehicle berth management method, equipment and system.
Background technique
Intelligent traffic monitoring system is an important development direction of current traffic monitoring industry, relies on computer vision
The picture shot with technologies such as deep learnings to monitoring camera automatically analyzes, and can be applied to act of violating regulations judgement, vehicle
The many aspects such as berth management, Car license recognition, vehicle cab recognition.It wherein, is identification in the prior art in Vehicle berth management aspect
The vehicle's contour in berth is positioned, so that the occupied state in berth is determined by way of vehicle running track is continuously tracked,
To which the down time to vehicle carries out timing, this mode needs to acquire great amount of images to guarantee tracking effect, it usually needs
Multiple image is uploaded within one second, therefore calculates power and flow cost height, and the accuracy of parking timing is poor, especially in Duo Bo
The efficiency of management is lower in the case where position, causes influence to the intelligent management of Vehicle berth.
Summary of the invention
In view of the above technical problems, the present invention provides a kind of Vehicle berth management method, equipment and system.
The technical scheme to solve the above technical problems is that
In a first aspect, the present invention provides a kind of Vehicle berth management method, comprising:
Image collecting device is obtained to the image of the region acquisition comprising at least one berth；
Judge the occupied state in each berth in image；
If at least one berth is occupied state in a upper image and present image, judge to occupy same berth
Whether corresponding vehicle is same vehicle；
Wherein, if the corresponding vehicle for occupying same berth is same vehicle, next image is waited, to judge next figure
The occupied state in each berth as in；
If the corresponding vehicle for occupying same berth is not same vehicle, the time that will acquire an image is used as upper one
The parking deadline that the vehicle in the berth is occupied in image, the time that will acquire present image is used as to be occupied in present image
The parking initial time of the vehicle in the berth.
The beneficial effects of the present invention are: only by the judgement of berth occupied state in two images, and for same
The whether identical judgement of corresponding vehicle in berth, can determine simultaneously to parking beginning and ending time of the vehicle in multiple berths into
Row determines, can greatly reduce the requirement for calculating equipment power, reduces flow and hardware cost, obtains two width by arbitrarily adjusting
The time interval of image, convenient for being balanced in the accuracy rate of parking charge and image uplink time and calculating cost, in addition,
Also improve the efficiency of more berth management.
Based on the above technical solution, the present invention can also be improved as follows.
Further, this method further include:
If all berths are free state in a upper image and present image, next image is waited, to sentence
The occupied state in each berth in disconnected next image.
Further, this method further include:
If at least one berth is free state in a upper image, and same berth occupies in present image
State then will acquire parking initial time of the time as the vehicle for occupying the berth in present image of present image.
Further, this method further include:
If at least one berth is occupied state in a upper image, and same berth is vacant in present image
State then will acquire parking deadline of the time of an image as the vehicle for occupying the berth in a upper image.
Further, the opposite in berth is arranged in image collecting device.
Further, after obtaining image collecting device to the image of the region acquisition comprising at least one berth, the party
Method further include:
Identify that the profile of object in simultaneously tag image, object include at least vehicle；
Before whether the corresponding vehicle for judging to occupy same berth is same vehicle, further includes:
The profile for the vehicle for occupying same berth in a upper image is placed on identical position in present image, is judged
Occupied in a upper image in the profile and present image of the vehicle in same berth the lap of the profile of other objects with
Whether the ratio that the profile of the vehicle in same berth is occupied in a upper image is lower than preset ratio, occupies if so then execute judgement same
Whether the corresponding vehicle in one berth is same vehicle, does not otherwise execute and wait next image, every in next image to judge
The occupied state in a berth.
Further, whether the corresponding vehicle for judging to occupy same berth is same vehicle, is specifically included:
Previous image and present image are overlapped；
By the input of superimposed image by convolutional neural networks model trained in advance, the phase for occupying same berth is judged
Answer whether vehicle is same vehicle, wherein convolutional neural networks model be based on include distracter sample image it is trained
It arrives.
Further, previous image and present image are overlapped, are specifically included:
3 channel RGB datas of previous image and present image are overlapped, 6 channels of superimposed image are obtained
RGB-RGB data, by 6 channel RGB-RGB data of image after superposition judge to occupy same berth corresponding vehicle whether be
Same vehicle.
Further, the training method of convolutional neural networks model are as follows:
It will include the same vehicle of distracter in the progress of two images of the same area same period and different time
Superposition is used as positive sample, will include that the different vehicle of distracter is overlapped in two images of the same area as negative sample
This, is based on multiple positive samples and negative sample training convolutional neural networks model.
Further, the input of superimposed image is judged to occupy same by convolutional neural networks model trained in advance
Whether the corresponding vehicle in berth is same vehicle, is specifically included:
Based on the phase by occupying same berth in the superimposed image of convolutional neural networks model extraction trained in advance
Answer the vehicle characteristics of vehicle；
Calculate the vehicle characteristics similarity of corresponding vehicle；
It whether is same vehicle according to the corresponding vehicle that similarity judges to occupy same berth.
Further, whether it is same vehicle according to the corresponding vehicle that similarity judges to occupy same berth, specifically includes:
When similarity is more than the upper limit of preset range, determine that occupying the corresponding vehicle in same berth is same vehicle；
When similarity is lower than the lower limit of preset range, determine that occupying the corresponding vehicle in same berth is not same vehicle.
Further, this method further include:
Charging is carried out according to the parking initial time for the same vehicle for occupying same berth and parking deadline.
Second aspect, the present invention also provides a kind of Vehicle berth management equipments, comprising:
Module is obtained, for obtaining image collecting device to the image of the region acquisition comprising at least one berth；
First judgment module, for judging the occupied state in each berth in image；
Second judgment module, if being occupied state in a upper image and present image at least one berth,
Whether the corresponding vehicle for judging to occupy same berth is same vehicle；
Wherein, if the corresponding vehicle for occupying same berth is same vehicle, next image is waited, to judge next figure
The occupied state in each berth as in；
If the corresponding vehicle for occupying same berth is not same vehicle, the time that will acquire an image is used as upper one
The parking deadline that the vehicle in the berth is occupied in image, the time that will acquire present image is used as to be occupied in present image
The parking initial time of the vehicle in the berth.
The beneficial effects of the present invention are: only by the judgement of berth occupied state in two images, and for same
The whether identical judgement of corresponding vehicle in berth, can determine simultaneously to parking beginning and ending time of the vehicle in multiple berths into
Row determines, can greatly reduce the requirement for calculating equipment power, reduces flow and hardware cost, obtains two width by arbitrarily adjusting
The time interval of image, convenient for being balanced in the accuracy rate of parking charge and image uplink time and calculating cost, in addition,
Also improve the efficiency of more berth management.
Further, first judgment module is also used to: if all berths are vacant in a upper image and present image
State then waits next image, to judge the occupied state in each berth in next image；
Further, first judgment module is also used to: if at least one berth is free state in a upper image, and
And same berth is occupied state in present image, then will acquire time of present image as occupying the pool in present image
The parking initial time of the vehicle of position；
Further, first judgment module is also used to: if at least one berth is occupied state in a upper image, and
Same berth is free state in present image, then will acquire time of an image as occupying the pool in a upper image
The parking deadline of the vehicle of position；
Further, the opposite in berth is arranged in image collecting device.
Further, the equipment further include:
Identification module, for obtaining module acquisition image collecting device to the region acquisition comprising at least one berth
After image, identify that the profile of object in simultaneously tag image, object include at least vehicle；
Third judgment module, whether the corresponding vehicle for judging to occupy same berth in the second judgment module is same
Before vehicle, the profile for the vehicle for occupying same berth in a upper image is placed on identical position in present image, is judged
Occupied in a upper image in the profile and present image of the vehicle in same berth the lap of the profile of other objects with
Whether the ratio that the profile of the vehicle in same berth is occupied in a upper image is lower than preset ratio, if then the second judgment module is held
Whether the corresponding vehicle that row judges to occupy same berth is same vehicle, does not execute and wait next image, otherwise to judge
The occupied state in each berth in next image.
Further, the second judgment module specifically includes:
Superpositing unit, for previous image and present image to be overlapped；
Judging unit, for, by convolutional neural networks model trained in advance, judgement to account for by the input of superimposed image
It whether is same vehicle with the corresponding vehicle in same berth, wherein convolutional neural networks model is based on including distracter
Sample image training obtains.
Further, superpositing unit is specifically used for:
3 channel RGB datas of previous image and present image are overlapped, 6 channels of superimposed image are obtained
RGB-RGB data, by 6 channel RGB-RGB data of image after superposition judge to occupy same berth corresponding vehicle whether be
Same vehicle.
Further, the second judgment module, specifically further include:
Training unit, for that will include the same vehicle of distracter in the same area, same period and different time
Two images be overlapped as positive sample, will include that the different vehicle of distracter is carried out in two images of the same area
Superposition is used as negative sample, is based on multiple positive samples and negative sample training convolutional neural networks model.
Further, judging unit specifically includes:
Subelement is extracted, for based on by accounting in the superimposed image of convolutional neural networks model extraction trained in advance
With the vehicle characteristics of the corresponding vehicle in same berth；
Computation subunit, for calculating the vehicle characteristics similarity of corresponding vehicle；
Judgment sub-unit, whether the corresponding vehicle for judging to occupy same berth according to similarity is same vehicle.
Further, judgment sub-unit is specifically used for:
When similarity is more than the upper limit of preset range, determine that occupying the corresponding vehicle in same berth is same vehicle；
When similarity is lower than the lower limit of preset range, determine that occupying the corresponding vehicle in same berth is not same vehicle.
Further, the equipment further include:
Accounting module, for according to the parking initial time of the same vehicle for occupying same berth and parking deadline into
Row charging.
Further, which uses server or chip.
The third aspect, the present invention also provides a kind of Vehicle berth management systems, including Vehicle berth management equipment as above
And image collecting device.
Detailed description of the invention
Fig. 1 is a kind of flow chart of Vehicle berth management method provided in an embodiment of the present invention；
Fig. 2 is the flow chart of another Vehicle berth management method provided in an embodiment of the present invention；
Fig. 3 is a kind of structural block diagram of Vehicle berth management equipment provided in an embodiment of the present invention；
Fig. 4 is the structural block diagram of another Vehicle berth management equipment provided in an embodiment of the present invention.
Specific embodiment
The principle and features of the present invention will be described below with reference to the accompanying drawings, and the given examples are served only to explain the present invention, and
It is non-to be used to limit the scope of the invention.
Fig. 1 is a kind of flow chart of Vehicle berth management method provided in an embodiment of the present invention, as shown in Figure 1, this method
Include:
S1, image collecting device is obtained to the image of the region acquisition comprising at least one berth；
Specifically, image collecting device may be provided at the opposite in the region comprising berth, for example, can be arranged by road side
Camera persistently monitor opposite roadside at least one berth, separated in time interception image obtains two width therein
For image as image to be detected, the interception time interval of two images is unsuitable too long.
S2, the occupied state for judging each berth in image；
Specifically, the step can be realized by a variety of existing ways, such as identified in image by object detection algorithm
Berth region in whether there is object, can also be judged by way of neural metwork training, or using object examine
The mode of method of determining and calculating and neural metwork training combination judgement.
If all berths are free state in a upper image and present image, illustrate there is no vehicle in berth, nothing
Parking timing need to be carried out, needs to wait the next image of image acquisition device, to judge each berth in next image
Occupied state.
If at least one berth is free state in a upper image, and same berth occupies in present image
State, a vehicle for illustrating that at least one current berth has just been driven into occupy, and at this moment, will acquire the time conduct of present image
The parking initial time of the vehicle in the berth is occupied in present image.
If at least one berth is occupied state in a upper image, and same berth is vacant in present image
State illustrates that the vehicle at least one current berth has been driven out to berth, at this moment, will acquire the time of an image as upper
The parking deadline of the vehicle in the berth is occupied in one image.
If S3, at least one berth are occupied state in a upper image and present image, judge to occupy same pool
Whether the corresponding vehicle of position is same vehicle；
Specifically, if a berth is occupied state in a upper image and present image, there are two kinds of possibility,
A kind of may be that the vehicle being parked on the berth in a upper image is still parked on same berth still in present image, that is, exist
What is parked on the berth in a upper image and present image is same vehicle, and alternatively possible is to be parked in the berth in a upper image
On vehicle sailed out of, in the gap of Image Acquisition, separately there is a vehicle to drive into the berth, i.e., in a upper image and present image
In park on the berth is not same vehicle, can judge account for by the methods of image recognition algorithm, neural metwork training here
It whether is same vehicle with the corresponding vehicle in same berth, so that distinguishing both the above may.
Wherein, if the corresponding vehicle for occupying same berth is same vehicle, illustrate that the berth is occupied by same vehicle always,
I.e. the down time of vehicle also is continuing to continue, and cannot obtain it and stop deadline, need that next image is waited to continue
Judgement.
If the corresponding vehicle for occupying same berth is not same vehicle, illustrate that the vehicle for occupying the berth has been become
Change, parking deadline of the time of an image as the vehicle for occupying the berth in a upper image can be will acquire, it will
Obtain parking initial time of the time of present image as the vehicle for occupying the berth in present image.
By continuously acquiring image, same vehicle can be obtained in the parking initial time and parking cut-off on berth
Between, so as to carry out parking charge.
A kind of Vehicle berth management method provided in an embodiment of the present invention only passes through berth occupied state in two images
Judgement, and for the whether identical judgement of corresponding vehicle in same berth, can determine simultaneously in multiple berths
The parking beginning and ending time of vehicle is determined, and can greatly reduce the requirement for calculating equipment power, reduces flow and hardware cost,
Such as: it needs to calculate in the prior art in one second several times, calculate once, if algorithm is deployed in local meter within present 1,2 minute
It calculates in unit, then can greatly reduce the requirement for calculating local computing unit power, reduce computing hardware cost；And if will calculate
Method is disposed beyond the clouds, and video camera only presses minute timing and uploads image data, will be greatly reduced the flow for uploading image and cloud
The calculating power of unit reduces flow and calculates the cost of power, in addition, also improving the efficiency of more berth management.
Optionally, as an embodiment of the present invention, as shown in Fig. 2, after step S1, this method further include:
The profile of object, object include at least vehicle in S4, identification and tag image；
Specifically, the step can be realized by object detection algorithm, after identifying the object in image, rectangle frame can be used
It is marked Deng the profile to the object identified.Wherein, it when identifying object, at least needs to identify the vehicle judged in object
?.
Before step S3, further includes:
S5, the profile for the vehicle for occupying same berth in a upper image is placed on identical position in present image,
Judge the lap that the profile of other objects in the profile and present image of the vehicle in same berth is occupied in a upper image
Whether it is lower than preset ratio with the ratio of the profile for the vehicle for occupying same berth in a upper image, is accounted for if so then execute judgement
Whether it is same vehicle with the corresponding vehicle in same berth, does not execute and wait next image, otherwise to judge next image
In each berth occupied state.
Specifically, when the vehicle's contour in present image is by marked object, it, can shadow when being blocked such as pedestrian, vehicular traffic
Ring judge in subsequent step same berth in the upper image of occupancy and present image corresponding vehicle whether be same vehicle standard
The profile of vehicle in previous image on some berth in the step, is placed on identical position in present image by true property,
To replace the profile of the vehicle in present image on the berth, if the wheel of the vehicle in previous image on the berth
The wide lap with the profile of marked object in present image accounts for the profile of the vehicle in previous image on the berth
Ratio is more than preset ratio, then it is more to illustrate that the vehicle's contour in present image is blocked, needs to obtain next image, and continue
Identification judgement is carried out by above-mentioned steps, until meeting the requirement for being lower than preset ratio, i.e. vehicle's contour in present image is hidden
Keep off it is less, so as to the judgement whether being blocked for vehicle's contour in next image.
Optionally, in this embodiment, step S3 is specifically included:
S3.1, previous image and present image are overlapped；
Specifically, can identify simultaneously two images by single treatment process by superimposed image, identifying processing is improved
Efficiency.
In superposition, 3 channel RGB datas of previous image and present image can be overlapped, obtain superimposed figure
6 channel RGB-RGB data of picture carry out subsequent judgement by 6 channel RGB-RGB data of image after superposition.
Such as: the 3 channel RGB data of part of previous image are as follows:
[[204,204,204],
[204,204,204],
[204,204,204]],
……
The 3 channel RGB data of part of present image are as follows:
[[226,226,226],
[226,226,226],
[226,226,226]],
……
6 channel RGB-RGB data of the image being then superimposed are as follows:
[[204,204,204,226,226,226],
[204,204,204,226,226,226],
[204,204,204,226,226,226]],
……
S3.2, convolutional neural networks model trained in advance is passed through into the input of superimposed image, judges to occupy same pool
Whether the corresponding vehicle of position is same vehicle, wherein convolutional neural networks model is based on the sample image for including distracter
What training obtained.
Specifically, in training convolutional neural networks model, need to one input data of convolutional neural networks model and
One corresponding label exports a result by convolutional neural networks, the result of output and label is compared, and reuses anti-
Data are modified to the direction propagated along comparison result, have so far learnt a wheel, by repeatedly learning, continuous corrective networks
Parameter improves the accuracy of convolutional neural networks model output result.After the training for completing convolutional neural networks model, it can incite somebody to action
In data to be tested input model, model can export corresponding judging result.If when prediction obtained input with instructed
The data practiced are similar, the obtained high of result certainty accuracy rate
In the step, can by camera obtain several scenes under include distracter image as sample, then lead to
Cross manually judge the vehicle in two images whether be same vehicle mode, provide the corresponding label of sample, wherein will wrap
Same vehicle containing distracter is overlapped in two images of the same area, same period and different time as just
Sample will include that two images of different vehicle of distracter are superimposed again as negative sample, be based on multiple positive samples and negative sample
This training convolutional neural networks model.Wherein, the label of settable positive sample is 1, and the label of negative sample is 0.
In the step, can be used include under several scenes distracter sample image as input data to convolutional Neural
Network has carried out a large amount of repetition trainings, with improve there are distracter to the two cars in superimposed image whether
The accuracy judged for same vehicle, wherein the quantity of sample image can according to actual needs accuracy of identification setting.
Optionally, in this embodiment, step S3.2 is specifically included:
Same pool is occupied in S3.2.1, the superimposed image of convolutional neural networks model extraction trained in advance based on process
The vehicle characteristics of the corresponding vehicle of position；
S3.2.2, the vehicle characteristics similarity for calculating corresponding vehicle；
It S3.2.3, according to the corresponding vehicle that similarity judges to occupy same berth whether is same vehicle.
Specifically, being spy for 6 channel RGB-RGB data using VGG the or MOBLIE Net etc. in convolutional neural networks
Sign is extracted, and whether the feature of extraction includes but is not limited to vehicle color, wheel hub, vehicle shape, has skylight etc. that can distinguish knowledge
The feature of other vehicle.
After having extracted feature, similarity is calculated using full articulamentum or one of Bayes's classification or SVM, and then sentence
Whether the vehicle in disconnected previous image and present image out is a vehicle, wherein the floating-point values that similarity is 0 to 1,0 represents
Completely dissimilar, 1 representative is completely similar.
Optionally, in this embodiment, step S3.2.3 is specifically included:
When similarity is more than the upper limit of preset range, determine that occupying the corresponding vehicle in same berth is same vehicle；
When similarity is lower than the lower limit of preset range, determine that occupying the corresponding vehicle in same berth is not same vehicle.
Specifically, the upper and lower bound of preset range can be chosen for same numerical value, different numerical value can also be chosen, for example,
The lower and upper limit for choosing preset range are 0.5, at this point, similarity 0.5 can be used as it is uncertain as a result, i.e. fuzzy value；Choosing
The lower and upper limit for taking preset range are respectively 0.4 and 0.6, at this point, the similarity between 0.4~0.6 can be used as it is uncertain
As a result.
The embodiment of the invention also provides a kind of computer readable storage mediums, including instruction, when it is transported on computers
When row, so that computer is executed such as the method in any of the above-described embodiment.
Fig. 3 is a kind of structural block diagram of Vehicle berth management equipment provided in an embodiment of the present invention, each mould in the equipment
The principle of work and power of block is expounded in foregoing teachings, is repeated no more below.
As shown in figure 3, the equipment includes:
Module is obtained, for obtaining image collecting device to the image of the region acquisition comprising at least one berth；
First judgment module, for judging the occupied state in each berth in image；
If all berths are free state in a upper image and present image, next image is waited, to sentence
The occupied state in each berth in disconnected next image；
If at least one berth is free state in a upper image, and same berth occupies in present image
State then will acquire parking initial time of the time as the vehicle for occupying the berth in present image of present image；
If at least one berth is occupied state in a upper image, and same berth is vacant in present image
State then will acquire parking deadline of the time of an image as the vehicle for occupying the berth in a upper image；
Second judgment module, if being occupied state in a upper image and present image at least one berth,
Whether the corresponding vehicle for judging to occupy same berth is same vehicle；
Wherein, if the corresponding vehicle for occupying same berth is same vehicle, next image is waited, to judge next figure
The occupied state in each berth as in；
If the corresponding vehicle for occupying same berth is not same vehicle, the time that will acquire an image is used as upper one
The parking deadline that the vehicle in the berth is occupied in image, the time that will acquire present image is used as to be occupied in present image
The parking initial time of the vehicle in the berth.
A kind of Vehicle berth management equipment provided in an embodiment of the present invention only passes through berth occupied state in two images
Judgement, and for the whether identical judgement of corresponding vehicle in same berth, can determine simultaneously in multiple berths
The parking beginning and ending time of vehicle is determined, and improves the accuracy of parking timing and the efficiency of more berths management.
Optionally, in this embodiment, the opposite in berth is arranged in image collecting device.
Optionally, in as an embodiment of the present invention, as shown in figure 4, the equipment further include:
Identification module, for obtaining module acquisition image collecting device to the region acquisition comprising at least one berth
After image, identify that the profile of object in simultaneously tag image, object include at least vehicle；
Third judgment module, whether the corresponding vehicle for judging to occupy same berth in the second judgment module is same
Before vehicle, the profile for the vehicle for occupying same berth in a upper image is placed on identical position in present image, is judged
Occupied in a upper image in the profile and present image of the vehicle in same berth the lap of the profile of other objects with
Whether the ratio that the profile of the vehicle in same berth is occupied in a upper image is lower than preset ratio, if then the second judgment module is held
Whether the corresponding vehicle that row judges to occupy same berth is same vehicle, does not execute and wait next image, otherwise to judge
The occupied state in each berth in next image.
Optionally, in this embodiment, the second judgment module specifically includes:
Superpositing unit, for previous image and present image to be overlapped；
Judging unit, for, by convolutional neural networks model trained in advance, judgement to account for by the input of superimposed image
It whether is same vehicle with the corresponding vehicle in same berth, wherein convolutional neural networks model is based on including distracter
Sample image training obtains.
Optionally, in this embodiment, superpositing unit is specifically used for:
3 channel RGB datas of previous image and present image are overlapped, 6 channels of superimposed image are obtained
RGB-RGB data, by 6 channel RGB-RGB data of image after superposition judge to occupy same berth corresponding vehicle whether be
Same vehicle.
Optionally, in this embodiment, the second judgment module, specifically further include:
Training unit, for that will include the same vehicle of distracter in the same area, same period and different time
Two images be overlapped as positive sample, will include that the different vehicle of distracter is carried out in two images of the same area
Superposition is used as negative sample, is based on multiple positive samples and negative sample training convolutional neural networks model.
Optionally, in this embodiment, judging unit specifically includes:
Subelement is extracted, for based on by accounting in the superimposed image of convolutional neural networks model extraction trained in advance
With the vehicle characteristics of the corresponding vehicle in same berth；
Computation subunit, for calculating the vehicle characteristics similarity of corresponding vehicle；
Judgment sub-unit, whether the corresponding vehicle for judging to occupy same berth according to similarity is same vehicle.
Optionally, in this embodiment, judgment sub-unit is specifically used for:
When similarity is more than the upper limit of preset range, determine that occupying the corresponding vehicle in same berth is same vehicle；
When similarity is lower than the lower limit of preset range, determine that occupying the corresponding vehicle in same berth is not same vehicle.
Optionally, in this embodiment, the equipment further include:
Accounting module, for according to the parking initial time of the same vehicle for occupying same berth and parking deadline into
Row charging.
Optionally, in this embodiment, which uses server or chip.
The embodiment of the present invention also provides a kind of Vehicle berth management system, including above-mentioned Vehicle berth management equipment and figure
As acquisition device.
Reader should be understood that in the description of this specification reference term " one embodiment ", " is shown " some embodiments "
The description of example ", " specific example " or " some examples " etc. mean specific features described in conjunction with this embodiment or example, structure,
Material or feature are included at least one embodiment or example of the invention.In the present specification, above-mentioned term is shown
The statement of meaning property need not be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described
It may be combined in any suitable manner in any one or more of the embodiments or examples.In addition, without conflicting with each other, this
The technical staff in field can be by the spy of different embodiments or examples described in this specification and different embodiments or examples
Sign is combined.
In addition, term " first ", " second ", " third " are used for description purposes only, it is not understood to indicate or imply phase
To importance.Unless specifically stated otherwise, the opposite step of the component and step that otherwise illustrate in these embodiments, digital table
It is not limit the scope of the invention up to formula and numerical value.
It is apparent to those skilled in the art that for convenience of description and succinctly, the dress of foregoing description
The specific work process with unit is set, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
In several embodiments provided herein, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, the apparatus embodiments described above are merely exemplary, for example, the division of unit, only
A kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or components can combine or
Person is desirably integrated into another system, or some features can be ignored or not executed.
Unit may or may not be physically separated as illustrated by the separation member, shown as a unit
Component may or may not be physical unit, it can and it is in one place, or may be distributed over multiple networks
On unit.It can select some or all of unit therein according to the actual needs to realize the mesh of the embodiment of the present invention
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, is also possible to two or more units and is integrated in one unit.It is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.
It, can if integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product
To be stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention substantially or
Say that all or part of the part that contributes to existing technology or the technical solution can embody in the form of software products
Out, which is stored in a storage medium, including some instructions are used so that a computer equipment
(can be personal computer, server or the network equipment etc.) executes all or part of each embodiment method of the present invention
Step.And storage medium above-mentioned include: USB flash disk, it is mobile hard disk, read-only memory (ROM, Read-Only Memory), random
Access various Jie that can store program code such as memory (RAM, Random Access Memory), magnetic or disk
Matter.
More than, only a specific embodiment of the invention, but scope of protection of the present invention is not limited thereto, and it is any to be familiar with
Those skilled in the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or substitutions,
These modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be wanted with right
Subject to the protection scope asked.
Claims (10)
1. a kind of Vehicle berth management method characterized by comprising
Image collecting device is obtained to the image of the region acquisition comprising at least one berth；
Judge the occupied state in each berth in image；
If at least one berth is occupied state in a upper image and present image, judge to occupy the corresponding of same berth
Whether vehicle is same vehicle；
Wherein, if the corresponding vehicle for occupying same berth is same vehicle, next image is waited, to judge in next image
The occupied state in each berth；
If the corresponding vehicle for occupying same berth is not same vehicle, the time that will acquire an image is used as in a upper image
The parking deadline of the middle vehicle for occupying the berth, the time that will acquire present image is used as occupies the pool in present image
The parking initial time of the vehicle of position.
2. the method according to claim 1, wherein the method also includes:
If all berths are free state in a upper image and present image, next image is waited, under judging
The occupied state in each berth in one image.
3. the method according to claim 1, wherein the method also includes:
If at least one berth is free state in a upper image, and same berth occupies shape in present image
State then will acquire parking initial time of the time as the vehicle for occupying the berth in present image of present image.
4. the method according to claim 1, wherein the method also includes:
If at least one berth is occupied state in a upper image, and same berth is vacant shape in present image
State then will acquire parking deadline of the time of an image as the vehicle for occupying the berth in a upper image.
5. the method according to claim 1, wherein pair in the berth is arranged in described image acquisition device
Face.
6. the method according to claim 1, wherein obtaining image collecting device to including at least one berth
Region acquisition image after, the method also includes:
Identify that the profile of object in simultaneously tag image, the object include at least vehicle；
Before whether the corresponding vehicle for judging to occupy same berth is same vehicle, further includes:
The profile for the vehicle for occupying same berth in a upper image is placed on identical position in present image, is judged upper
Occupied in one image in the profile and present image of the vehicle in same berth the lap of the profile of other objects with upper one
Whether the ratio that the profile of the vehicle in same berth is occupied in image is lower than preset ratio, occupies same pool if so then execute judgement
Whether the corresponding vehicle of position is same vehicle, does not execute and wait next image, otherwise to judge each pool in next image
The occupied state of position.
7. the method according to claim 1, wherein the corresponding vehicle for judging to occupy same berth whether be
Same vehicle, specifically includes:
Previous image and present image are overlapped；
By the input of superimposed image by convolutional neural networks model trained in advance, the corresponding vehicle for occupying same berth is judged
Whether be same vehicle, wherein the convolutional neural networks model be based on include distracter sample image it is trained
It arrives.
8. a kind of Vehicle berth management equipment characterized by comprising
Module is obtained, for obtaining image collecting device to the image of the region acquisition comprising at least one berth；
First judgment module, for judging the occupied state in each berth in image；
Second judgment module judges if being occupied state in a upper image and present image at least one berth
Whether the corresponding vehicle for occupying same berth is same vehicle；
Wherein, if the corresponding vehicle for occupying same berth is same vehicle, next image is waited, to judge in next image
The occupied state in each berth；
If the corresponding vehicle for occupying same berth is not same vehicle, the time that will acquire an image is used as in a upper image
The parking deadline of the middle vehicle for occupying the berth, the time that will acquire present image is used as occupies the pool in present image
The parking initial time of the vehicle of position.
9. equipment according to claim 8, which is characterized in that the equipment uses server or chip.
10. a kind of Vehicle berth management system, which is characterized in that set including Vehicle berth management as claimed in claim 8 or 9
Standby and image collecting device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910195918.9A CN110009929A (en) | 2019-03-15 | 2019-03-15 | A kind of Vehicle berth management method, equipment and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910195918.9A CN110009929A (en) | 2019-03-15 | 2019-03-15 | A kind of Vehicle berth management method, equipment and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110009929A true CN110009929A (en) | 2019-07-12 |
Family
ID=67167139
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910195918.9A Pending CN110009929A (en) | 2019-03-15 | 2019-03-15 | A kind of Vehicle berth management method, equipment and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110009929A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110599796A (en) * | 2019-09-06 | 2019-12-20 | 北京筑梦园科技有限公司 | Parking space state judgment method, server and parking management system |
| CN110852212A (en) * | 2019-10-29 | 2020-02-28 | 上海眼控科技股份有限公司 | Method and device for checking operation object in vehicle detection |
| CN112381150A (en) * | 2020-11-17 | 2021-02-19 | 上海科技大学 | Confrontation sample detection method based on sample robustness difference |
| CN112382101A (en) * | 2020-11-02 | 2021-02-19 | 北京精英路通科技有限公司 | Parking space management method and device, computer equipment and storage medium |
| CN112784794A (en) * | 2021-01-29 | 2021-05-11 | 深圳市捷顺科技实业股份有限公司 | Vehicle parking state detection method and device, electronic equipment and storage medium |
| CN112885108A (en) * | 2020-12-23 | 2021-06-01 | 爱泊车美好科技有限公司 | Vehicle change detection method and system on parking space based on deep learning algorithm |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632376A (en) * | 2013-12-12 | 2014-03-12 | 江苏大学 | Method for suppressing partial occlusion of vehicles by aid of double-level frames |
| CN103778785A (en) * | 2012-10-23 | 2014-05-07 | 南开大学 | Vehicle tracking and detecting method based on parking lot environment video monitoring |
| CN104298976A (en) * | 2014-10-16 | 2015-01-21 | 电子科技大学 | License plate detection method based on convolutional neural network |
| CN106156750A (en) * | 2016-07-26 | 2016-11-23 | 浙江捷尚视觉科技股份有限公司 | A kind of based on convolutional neural networks to scheme to search car method |
| CN106781680A (en) * | 2017-02-20 | 2017-05-31 | 洪志令 | A kind of curb parking intelligent control method based on the detection of image empty parking space |
| CN108009560A (en) * | 2016-11-02 | 2018-05-08 | 广州图普网络科技有限公司 | Commodity image similar categorization decision method and device |
| CN108389396A (en) * | 2018-02-28 | 2018-08-10 | 北京精英智通科技股份有限公司 | A kind of vehicle matching process, device and charge system based on video |
| CN108830188A (en) * | 2018-05-30 | 2018-11-16 | 西安理工大学 | Vehicle checking method based on deep learning |
| KR20180137302A (en) * | 2017-06-16 | 2018-12-27 | 유한회사 한국케이비에프 | Intelligent system for guiding parking area and method using the same |
-
2019
- 2019-03-15 CN CN201910195918.9A patent/CN110009929A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103778785A (en) * | 2012-10-23 | 2014-05-07 | 南开大学 | Vehicle tracking and detecting method based on parking lot environment video monitoring |
| CN103632376A (en) * | 2013-12-12 | 2014-03-12 | 江苏大学 | Method for suppressing partial occlusion of vehicles by aid of double-level frames |
| CN104298976A (en) * | 2014-10-16 | 2015-01-21 | 电子科技大学 | License plate detection method based on convolutional neural network |
| CN106156750A (en) * | 2016-07-26 | 2016-11-23 | 浙江捷尚视觉科技股份有限公司 | A kind of based on convolutional neural networks to scheme to search car method |
| CN108009560A (en) * | 2016-11-02 | 2018-05-08 | 广州图普网络科技有限公司 | Commodity image similar categorization decision method and device |
| CN106781680A (en) * | 2017-02-20 | 2017-05-31 | 洪志令 | A kind of curb parking intelligent control method based on the detection of image empty parking space |
| KR20180137302A (en) * | 2017-06-16 | 2018-12-27 | 유한회사 한국케이비에프 | Intelligent system for guiding parking area and method using the same |
| CN108389396A (en) * | 2018-02-28 | 2018-08-10 | 北京精英智通科技股份有限公司 | A kind of vehicle matching process, device and charge system based on video |
| CN108830188A (en) * | 2018-05-30 | 2018-11-16 | 西安理工大学 | Vehicle checking method based on deep learning |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110599796A (en) * | 2019-09-06 | 2019-12-20 | 北京筑梦园科技有限公司 | Parking space state judgment method, server and parking management system |
| CN110599796B (en) * | 2019-09-06 | 2022-03-04 | 北京筑梦园科技有限公司 | Parking space state judgment method, server and parking management system |
| CN110852212A (en) * | 2019-10-29 | 2020-02-28 | 上海眼控科技股份有限公司 | Method and device for checking operation object in vehicle detection |
| CN112382101A (en) * | 2020-11-02 | 2021-02-19 | 北京精英路通科技有限公司 | Parking space management method and device, computer equipment and storage medium |
| CN112381150A (en) * | 2020-11-17 | 2021-02-19 | 上海科技大学 | Confrontation sample detection method based on sample robustness difference |
| CN112885108A (en) * | 2020-12-23 | 2021-06-01 | 爱泊车美好科技有限公司 | Vehicle change detection method and system on parking space based on deep learning algorithm |
| CN112784794A (en) * | 2021-01-29 | 2021-05-11 | 深圳市捷顺科技实业股份有限公司 | Vehicle parking state detection method and device, electronic equipment and storage medium |
| CN112784794B (en) * | 2021-01-29 | 2024-02-02 | 深圳市捷顺科技实业股份有限公司 | Vehicle parking state detection method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110009929A (en) | A kind of Vehicle berth management method, equipment and system | |
| CN106652445B (en) | A kind of road traffic accident method of discrimination and device | |
| US20140376769A1 (en) | Method for detecting large size and passenger vehicles from fixed cameras | |
| CN110096975B (en) | Parking space state identification method, equipment and system | |
| CN108133484B (en) | Automatic driving processing method and device based on scene segmentation and computing equipment | |
| CN104029680A (en) | Lane departure warning system and method based on monocular camera | |
| CN105488453A (en) | Detection identification method of no-seat-belt-fastening behavior of driver based on image processing | |
| CN103985182A (en) | Automatic public transport passenger flow counting method and system | |
| CN111047874B (en) | Intelligent traffic violation management method and related product | |
| CN108647708A (en) | Driver evaluation's method, apparatus, equipment and storage medium | |
| CN110378243A (en) | A kind of pedestrian detection method and device | |
| CN104680795A (en) | Vehicle type recognition method and device based on partial area characteristic | |
| CN102622575A (en) | Baseline band video monitoring system and monitoring method | |
| CN106340205A (en) | Traffic monitoring method and traffic monitoring apparatus | |
| CN107610477A (en) | A kind of intelligent transportation radio frequency and video information method for amalgamation processing | |
| Hasegawa et al. | Type classification, color estimation, and specific target detection of moving targets on public streets | |
| US11482012B2 (en) | Method for driving assistance and mobile device using the method | |
| CN110232827A (en) | The recognition methods of free flow toll vehicle type, apparatus and system | |
| CN104200670A (en) | Vehicle model identifying method based on vehicle earth induction characteristics | |
| CN113723273A (en) | Vehicle track information determination method and device and computer equipment | |
| Al Nasim et al. | An automated approach for the recognition of bengali license plates | |
| CN113705549B (en) | Road rescue work node determination method and device and related equipment | |
| CN114693722B (en) | Vehicle driving behavior detection method, detection device and detection equipment | |
| CN110020608A (en) | A kind of vehicle identification method, equipment, system and parking charge system | |
| CN113990077A (en) | Vehicle load and characteristic monitoring system and method |
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 |
Application publication date: 20190712 |
|
| RJ01 | Rejection of invention patent application after publication |