CN108875562B - A kind of public transport people flow rate statistical method and system - Google Patents
A kind of public transport people flow rate statistical method and system Download PDFInfo
- Publication number
- CN108875562B CN108875562B CN201810397363.1A CN201810397363A CN108875562B CN 108875562 B CN108875562 B CN 108875562B CN 201810397363 A CN201810397363 A CN 201810397363A CN 108875562 B CN108875562 B CN 108875562B
- Authority
- CN
- China
- Prior art keywords
- local peaking
- characteristic value
- door
- bus
- image
- 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.)
- Expired - Fee Related
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
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30242—Counting objects in image
Abstract
The present invention relates to a kind of public transport people flow rate statistical method and system, comprising: obtains the video image in bus, and extracts the foreground image of each frame in video image;When car door is in shutdown state, the characteristic value of enabling detection zone in each frame foreground image is obtained;Judge whether the characteristic value of enabling detection zone is greater than the first given threshold;Obtain the characteristic value in occupant detection region and this feature value corresponding time point in each frame foreground image;Obtain the characteristic value of shutdown detection zone in each frame foreground image;Judge whether the characteristic value of shutdown detection zone is greater than the second given threshold;It is worth corresponding time point according to the characteristic value and this feature in occupant detection region in each frame video image of acquisition, obtains the number of getting on the bus during this enabling is closed the door on bus.The present invention does not need to carry out the training of large sample size, reduces operand, improves statistical efficiency, improves statistical accuracy, of less demanding to hardware device, reduces costs low.
Description
Technical field
The present invention relates to public transport fields, more particularly to a kind of public transport people flow rate statistical method and system.
Background technique
In order to be more reasonably scheduled to bus, the service and management level in city are promoted, the public affairs of passenger are promoted
Ride experience is handed over, needs to count the flow of the people of bus.Existing public transport people flow rate statistical method probably includes as follows
Two kinds:
(1) special image processing equipment is used
Such as patent CN201310464624 discloses a kind of people flow rate statistical method based on Kinect stereoscopic vision.It should
Method reads depth image using OpenNI, and by Threshold segmentation, de-noising, projection operation, obtains tracking object perspective view,
Pursuit path point set is obtained in tracking object perspective view, and count tracking is carried out to people with this.The method needs Kinect 3D vertical
Body-sensing video camera is as video input, higher cost more demanding to picture pick-up device.
Bus stream of people's statistical system based on depth image is proposed there are also patent CN201610840255.The system is logical
The depth image acquisition equipment for crossing car door mouth obtains the depth image of passenger, and front-end processor is connected with vehicle, obtains vehicle shape
State, control equipment open and close and calculate passenger traffic flow amount.The system needs depth image sensor rather than traditional camera, to taking the photograph
As equipment requirement is higher;Unlatching/the closed state that must could obtain Vehicular door by connecting with Vehicular system simultaneously, increases
Hardware cost.
(2) the training detection of classifier number of people realizes people flow rate statistical in turn
Such as patent CN201710417286 first with support vector machines to head of passenger gradient orientation histogram feature
Learnt to obtain number of people classifier.The setting of down-sampled and interest region is carried out to every frame image of input video, is reused
Detection of classifier number of people target, and multiple target tracking is realized using Hungary Algorithm and core correlation filtering, dummy line is finally set
Complete the automatic counting of passenger traffic flow amount.The method is cut into spy after needing to acquire a large amount of various number of people samples and negative sample
Fixed size just can be carried out classifier training, and the trained time is also longer.And if the classifier accuracy after training is not high,
It also needs to increase sample size and carries out re -training, it is relatively time consuming laborious.Simultaneously when carrying out number of people detection be avoid missing inspection need into
Row multiple scale detecting increases many calculation amounts.
Summary of the invention
Based on this, the object of the present invention is to provide a kind of public transport people flow rate statistical method, has and do not need to carry out greatly
The training of sample size greatly reduces operand, improves statistical efficiency, improves statistical accuracy, of less demanding to hardware device,
Greatly reduce advantage at low cost.
A kind of public transport people flow rate statistical method, includes the following steps:
Step S0: vehicle door status is set to off door state by system initialization;
Step S1: obtaining the video image of present frame in bus, and extracts the prospect in the video image of the present frame
Image;
Step S2: judging vehicle door status, if car door is in door opening state, enters step S5;If car door is in shutdown shape
State then enters step S3.
Step S3: the characteristic value of enabling detection zone in foreground image is obtained;
Step S4: judging whether the characteristic value of enabling detection zone is greater than the first given threshold, if so, by vehicle door status
It is set on door state, and continues step S5;Otherwise, the video image of next frame is obtained, and using the next frame image as working as
The video image of previous frame returns to step S1;
Step S5: the characteristic value and this feature for obtaining occupant detection region in foreground image are worth corresponding time point, and
It stores to dynamic array;
Step S6: the characteristic value of shutdown detection zone in foreground image is obtained;
Step S7: judging whether the characteristic value of shutdown detection zone is greater than the second given threshold, if so, continuing step
S8, and vehicle door status is set to off door state;Otherwise, obtain the video image of next frame, and using the next frame image as
The video image of present frame returns to step S1;
Step S8: being worth corresponding time point according to the characteristic value in the occupant detection region in dynamic array and this feature,
Obtain the number of getting on the bus during this enabling is closed the door on bus;
Step S9: obtaining the video image of next frame, and using the next frame image as the video image of present frame, returns to
Step S1, the number of getting on the bus during being closed the door with the enabling to next time on bus count.
Compared with the prior art, the present invention can be obtained by obtaining video image further according to the pixel on video image
The number for taking away passenger loading during closing the door does not need the training for carrying out large sample size, greatly reduces operand, improve
Statistical efficiency improves statistical accuracy.Further, video image can be obtained by common video camera, to hardware
Equipment requirement is not high, greatly reduces at low cost.
Further, in step sl, it obtains in bus after the video image of present frame, to the current frame video image
Noise is removed by gaussian filtering, then foreground image is obtained by mixture Gaussian background model, to remove image noise, after raising
Continuous statistical accuracy.
Further, in step s 8, the number of getting on the bus during this enabling is closed the door on bus is obtained, including is walked as follows
It is rapid:
Step S81: acquisition time of frame is successive where the characteristic value in the occupant detection region in dynamic array is pressed it
Sequence sequentially stores;
Step S82: according to the characteristic value in the occupant detection region sequentially stored, local peaking and the local peaking are obtained
Corresponding time point;
Step S83: the local peaking that will acquire sequentially is stored by the sequencing at its corresponding time point, and with storage
First local peaking is as current local peaking;
Step S84: the time interval between current local peaking and next local peaking is calculated;
Step S85: judge whether the time interval between current local peaking and next local peaking sets less than third
Determine threshold value, if so, deleting the lesser local peaking of amplitude in current local peaking and next local peaking, retains width
It is worth biggish local peaking, and using the local peaking of the reservation as new current local peaking;Otherwise, retain current local peaks
Value and next local peaking, and using next local peaking as new current local peaking;
Step S86: judging whether current local peaking is the last one local peaking, if so, continuing step S87;It is no
Then, step S84 is returned to;
Step S87: counting the number of the local peaking of reservation, obtains the number of passenger loading during this enabling is closed the door.
The present invention also provides a kind of public transport people flow rate statistical systems, including processor, are adapted for carrying out each instruction;And storage
Equipment is suitable for storing a plurality of instruction, and described instruction is suitable for being loaded and being executed by the processor:
Judge vehicle door status, and when car door is in shutdown state, obtains the feature of enabling detection zone in foreground image
Value, and when the characteristic value of enabling detection zone is greater than the first given threshold, then vehicle door status is set on door state, and obtain
The characteristic value in occupant detection region and this feature value corresponding time point in foreground image are taken, and is stored to dynamic array;It obtains
Take the characteristic value of shutdown detection zone in foreground image;When the characteristic value of shutdown detection zone is greater than the second given threshold, then
Vehicle door status is set to off door state, and according to the characteristic value and this feature value pair in the occupant detection region in dynamic array
The time point answered obtains the number of getting on the bus during this enabling is closed the door on bus;
When car door is in door opening state, then the characteristic value in occupant detection region and the spy in foreground image are directly acquired
Value indicative corresponding time point, and store to dynamic array.
Compared with the prior art, the present invention can be obtained by obtaining video image further according to the pixel on video image
The number for taking away passenger loading during closing the door does not need the training for carrying out large sample size, greatly reduces operand, improve
Statistical efficiency improves statistical accuracy.Further, video image can be obtained by common video camera, to hardware
Equipment requirement is not high, greatly reduces at low cost.
In order to better understand and implement, the invention will now be described in detail with reference to the accompanying drawings.
Detailed description of the invention
Fig. 1 is the flow chart of public transport people flow rate statistical method in the embodiment of the present invention;
Fig. 2 is the process of the method for the number of getting on the bus during obtaining this shutdown of opening the door in the embodiment of the present invention on bus
Figure.
Specific embodiment
Referring to Fig. 1, its flow chart for public transport people flow rate statistical method in the embodiment of the present invention.Public transport flow of the people system
Meter method, includes the following steps:
Step S0: vehicle door status is set to off door state by system initialization;
Step S1: obtaining the video image of present frame in bus, and extracts the prospect in the video image of the present frame
Image.
In one embodiment, it will be used to obtain the video camera riding position of the video image in bus before bus
Car body top on door, the camera video image that taken of passengers is got on the bus vertically downward;Wherein video camera uses normal conventional
Video camera.
In one embodiment, the current frame video image is passed through after the video image of present frame in acquisition bus
Gaussian filtering removes noise, then obtains foreground image by mixture Gaussian background model.Wherein, the foreground image is two-value
Figure characterizes the region of movement as white, and characterizing non-athletic region is black.
Step S2: judging vehicle door status, if car door is in door opening state, enters step S5;If car door is in shutdown shape
State then enters step S3.
Step S3: the characteristic value of enabling detection zone in foreground image is obtained.
Enabling detection zone in the present invention is the region in video image at the enabling of public transport front door, and regional scope can
To adjust according to actual needs, specifically, the region that pixel changes greatly when public transport front door can be opened the door is as enabling detection zone
Domain.
Step S4: judging whether the characteristic value of enabling detection zone is greater than the first given threshold, if so, by vehicle door status
It is set on door state, and continues step S5;Otherwise, the video image of next frame is obtained, and using the next frame image as working as
The video image of previous frame returns to step S1.
Wherein, if the characteristic value of enabling detection zone is greater than the first given threshold, illustrate car door opening, there is passenger or more
Vehicle can start to carry out people flow rate statistical;If the characteristic value of enabling detection zone is less than or equal to the first given threshold, illustrate vehicle
Door is not opened, then without statistics.
In one embodiment, the characteristic value of enabling detection zone is in the enabling detection zone in the foreground image
Characterize the ratio of the pixel of movement and total pixel number of the enabling detection zone.First given threshold is 0.5-
0.9, it is preferred that first given threshold is 0.7.
Step S5: the characteristic value and this feature for obtaining occupant detection region in foreground image are worth corresponding time point, and
It stores to dynamic array.
Occupant detection region in the present invention be needed during passenger loading in video image by region,
Specific regional scope can adjust according to actual needs.
In one embodiment, the characteristic value in occupant detection region is in the occupant detection region in the foreground image
Characterize the ratio of the pixel of movement and total pixel number in the occupant detection region.
Step S6: the characteristic value of shutdown detection zone in foreground image is obtained.
Shutdown detection zone in the present invention is the region in video image at the shutdown of public transport front door, and regional scope can
To adjust according to actual needs, specifically, can be using the public transport front door region that pixel changes greatly at closing time as shutdown detection zone
Domain.
In one embodiment, the characteristic value of shutdown detection zone is in the shutdown detection zone in the foreground image
Characterize the ratio of the pixel of movement and total pixel number of the shutdown detection zone.
Step S7: judging whether the characteristic value of shutdown detection zone is greater than the second given threshold, if so, continuing step
S8, and vehicle door status is set to off door state;Otherwise, obtain the video image of next frame, and using the next frame image as
The video image of present frame returns to step S1.
Wherein, if the characteristic value of shutdown detection zone is greater than the second given threshold, illustrate closing of the door, passenger's stop over
Only get on or off the bus;If the characteristic value of shutdown detection zone is less than or equal to the second given threshold, continue to obtain occupant detection region
Characteristic value.Second given threshold is 0.5-0.9, it is preferred that second given threshold is 0.7.
Step S8: being worth corresponding time point according to the characteristic value in the occupant detection region in dynamic array and this feature,
Obtain the number of getting on the bus during this enabling is closed the door on bus.
Step S9: obtaining the video image of next frame, and using the next frame image as the video image of present frame, returns to
Step S1, the number of getting on the bus during being closed the door with the enabling to next time on bus count.
Referring to Fig. 2, Fig. 2 is the number of getting on the bus during obtaining this shutdown of opening the door in the embodiment of the present invention on bus
The flow chart of method.
Obtain this open the door close the door during on bus get on the bus number when, specifically comprise the following steps:
Step S81: acquisition time of frame is successive where the characteristic value in the occupant detection region in dynamic array is pressed it
Sequence sequentially stores.
Step S82: according to the characteristic value in the occupant detection region sequentially stored, local peaking and the local peaking are obtained
Corresponding time point.
In one embodiment, the method for local peaking is obtained are as follows: the characteristic value in the occupant detection region sequentially stored
In, it will be greater than each 10 characteristic values in its front and back, and be greater than the characteristic value of the 4th given threshold as local peaking.I.e. originally
The characteristic value in the occupant detection region of invention is at least 21, and each 10 characteristic values in front and back do not consider.4th setting
Threshold value is 0.1-0.3, it is preferred that the 4th given threshold is 0.2.
Step S83: the local peaking that will acquire sequentially is stored by the sequencing at its corresponding time point, and with storage
First local peaking is as current local peaking.
Step S84: the time interval between current local peaking and next local peaking is calculated.
Step S85: judge whether the time interval between current local peaking and next local peaking sets less than third
Determine threshold value, if so, deleting the lesser local peaking of amplitude in current local peaking and next local peaking, retains width
It is worth biggish local peaking, and using the local peaking of the reservation as new current local peaking;Otherwise, retain current local peaks
Value and next local peaking, and using next local peaking as new current local peaking.
Wherein, it if the time interval between current local peaking and next local peaking is less than third given threshold, says
The two bright local peakings are the characteristic value of same people, it is only necessary to retain a biggish local peaking to characterize the feature of the people
?.If the time interval between current local peaking and next local peaking is greater than or equal to third given threshold, explanation
The two local peakings are the characteristic value of two people, and therefore, it is necessary to the two are characterized different people local peakings to retain.?
In one embodiment, the third given threshold is 1-3 seconds, it is preferred that the third given threshold is 2 seconds.
Step S86: judging whether current local peaking is the last one local peaking, if so, continuing step S87;It is no
Then, step S84 is returned to.
Step S87: counting the number of the local peaking of reservation, obtains the number of passenger loading during this enabling is closed the door.
Compared with the prior art, the present invention can be obtained by obtaining video image further according to the pixel on video image
The number for taking away passenger loading during closing the door does not need the training for carrying out large sample size, greatly reduces operand, improve
Statistical efficiency improves statistical accuracy.Further, video image can be obtained by common video camera, to hardware
Equipment requirement is not high, greatly reduces at low cost.
The present invention also provides a kind of public transport people flow rate statistical systems, including processor, are adapted for carrying out each instruction;And storage
Equipment is suitable for storing a plurality of instruction, and described instruction is suitable for being loaded and being executed by the processor:
The video image of present frame in bus is obtained, and extracts the foreground image in the video image of the present frame;
Judge vehicle door status, and when car door is in shutdown state, obtains the feature of enabling detection zone in foreground image
Value, and when the characteristic value of enabling detection zone is greater than the first given threshold, then vehicle door status is set on door state, and obtain
The characteristic value in occupant detection region and this feature value corresponding time point in foreground image are taken, and is stored to dynamic array;It obtains
Take the characteristic value of shutdown detection zone in foreground image;When the characteristic value of shutdown detection zone is greater than the second given threshold, then
Vehicle door status is set to off door state, and according to the characteristic value and this feature value pair in the occupant detection region in dynamic array
The time point answered obtains the number of getting on the bus during this enabling is closed the door on bus;
When car door is in door opening state, then the characteristic value in occupant detection region and the spy in foreground image are directly acquired
Value indicative corresponding time point, and store to dynamic array.
In one embodiment, it will be used to obtain the video camera riding position of the video image in bus before bus
Car body top on door, the camera video image that taken of passengers is got on the bus vertically downward;Wherein video camera uses normal conventional
Video camera.
In one embodiment, the current frame video image is passed through after the video image of present frame in acquisition bus
Gaussian filtering removes noise, then obtains foreground image by mixture Gaussian background model.Wherein, the foreground image is two-value
Figure characterizes the region of movement as white, and characterizing non-athletic region is black.
In the present invention, in system initialization, the vehicle door status is set to off door state.
Enabling detection zone in the present invention is the region in video image at the enabling of public transport front door, and regional scope can
To adjust according to actual needs, specifically, the region that pixel changes greatly when public transport front door can be opened the door is as enabling detection zone
Domain.
Wherein, if the characteristic value of enabling detection zone is greater than the first given threshold, illustrate car door opening, there is passenger or more
Vehicle can start to carry out people flow rate statistical;If the characteristic value of enabling detection zone is less than or equal to the first given threshold, illustrate vehicle
Door is not opened, then without statistics.
In one embodiment, the characteristic value of enabling detection zone is in the enabling detection zone in the foreground image
Characterize the ratio of the pixel of movement and total pixel number of the enabling detection zone.First given threshold is 0.5-
0.9, it is preferred that first given threshold is 0.7.
Occupant detection region in the present invention be needed during passenger getting on/off in video image by region,
Its specific regional scope can adjust according to actual needs.
In one embodiment, the characteristic value in occupant detection region is in the occupant detection region in the foreground image
Characterize the ratio of the pixel of movement and total pixel number in the occupant detection region.
Shutdown detection zone in the present invention is the region in video image at the shutdown of public transport front door, and regional scope can
To adjust according to actual needs, specifically, can be using the public transport front door region that pixel changes greatly at closing time as shutdown detection zone
Domain.
In one embodiment, the characteristic value of shutdown detection zone is in the shutdown detection zone in the foreground image
Characterize the ratio of the pixel of movement and total pixel number of the shutdown detection zone.
Wherein, if the characteristic value of shutdown detection zone is greater than the second given threshold, illustrate closing of the door, passenger's stop over
Only get on or off the bus;If the characteristic value of shutdown detection zone is less than or equal to the second given threshold, continue to obtain occupant detection region
Characteristic value.Second given threshold is 0.5-0.9, it is preferred that second given threshold is 0.7.
Obtain this open the door close the door during on bus get on the bus number when, the processor is loaded and is executed:
The sequencing of the acquisition time of frame where the characteristic value in the occupant detection region in dynamic array is pressed it is sequentially
Storage;
According to the characteristic value in the occupant detection region sequentially stored, when obtaining local peaking and the corresponding local peaking
Between point;
The local peaking that will acquire sequentially is stored by the sequencing at its corresponding time point, and with first office of storage
Portion's peak value is as current local peaking;
Calculate the time interval between current local peaking and next local peaking;
Judge whether the time interval between current local peaking and next local peaking is less than third given threshold, if
It is then in current local peaking and next local peaking, to delete the lesser local peaking of amplitude, retain the biggish office of amplitude
Portion's peak value, and using the local peaking of the reservation as new current local peaking;Otherwise, retain current local peaking and next
Local peaking, and using next local peaking as new current local peaking;
When current local peaking is the last one local peaking, the number of the local peaking of reservation is counted, this is obtained
The number of passenger loading during enabling is closed the door.
In one embodiment, the method for local peaking is obtained are as follows: the characteristic value in the occupant detection region sequentially stored
In, it will be greater than each 10 characteristic values in its front and back, and be greater than the characteristic value of the 4th given threshold as local peaking.I.e. originally
The characteristic value in the occupant detection region of invention is at least 21, and each 10 characteristic values in front and back do not consider.4th setting
Threshold value is 0.1-0.3, it is preferred that the 4th given threshold is 0.2.
Wherein, it if the time interval between current local peaking and next local peaking is less than third given threshold, says
The two bright local peakings are the characteristic value of same people, it is only necessary to retain a biggish local peaking to characterize the feature of the people
?.If the time interval between current local peaking and next local peaking is greater than or equal to third given threshold, explanation
The two local peakings are the characteristic value of two people, and therefore, it is necessary to the two are characterized different people local peakings to retain.?
In one embodiment, the third given threshold is 1-3 seconds, it is preferred that the third given threshold is 2 seconds.
Compared with the prior art, the present invention can be obtained by obtaining video image further according to the pixel on video image
The number for taking away passenger loading during closing the door does not need the training for carrying out large sample size, greatly reduces operand, improve
Statistical efficiency improves statistical accuracy.Further, video image can be obtained by common video camera, to hardware
Equipment requirement is not high, greatly reduces at low cost.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to protection of the invention
Range.
Claims (8)
1. a kind of public transport people flow rate statistical method, which comprises the steps of:
Step S0: vehicle door status is set to off door state by system initialization;
Step S1: obtaining the video image of present frame in bus, and extracts the foreground image in the video image of the present frame;
Step S2: judging vehicle door status, if car door is in door opening state, enters step S5;If car door is in shutdown state,
Enter step S3;
Step S3: the characteristic value of enabling detection zone in foreground image is obtained;
Step S4: judging whether the characteristic value of enabling detection zone is greater than the first given threshold, if so, vehicle door status is arranged
For door opening state, and continue step S5;Otherwise, the video image of next frame is obtained, and using the next frame image as present frame
Video image, return to step S1;
Step S5: the characteristic value and this feature for obtaining occupant detection region in foreground image are worth corresponding time point, and store
To dynamic array;
Step S6: the characteristic value of shutdown detection zone in foreground image is obtained;
Step S7: judging whether the characteristic value of shutdown detection zone is greater than the second given threshold, if so, continue step S8, and
Vehicle door status is set to off door state;Otherwise, the video image of next frame is obtained, and using the next frame image as present frame
Video image, return to step S1;
Step S8: corresponding time point is worth according to the characteristic value in the occupant detection region in dynamic array and this feature, is obtained
Number of getting on the bus during this shutdown of opening the door on bus;
Step S9: obtaining the video image of next frame, and using the next frame image as the video image of present frame, returns to step
S1, the number of getting on the bus during being closed the door with the enabling to next time on bus count;
In step s 8, the number of getting on the bus during this enabling is closed the door on bus is obtained, is included the following steps:
Step S81: by the characteristic value in the occupant detection region in dynamic array by the sequencing of the acquisition time of frame where it
Sequentially store;
Step S82: it according to the characteristic value in the occupant detection region sequentially stored, obtains local peaking and the local peaking is corresponding
Time point;
Step S83: the local peaking that will acquire sequentially is stored by the sequencing at its corresponding time point, and with the first of storage
A local peaking is as current local peaking;
Step S84: the time interval between current local peaking and next local peaking is calculated;
Step S85: judge whether the time interval between current local peaking and next local peaking is less than third setting threshold
Value deletes the lesser local peaking of amplitude if so, in current local peaking and next local peaking, retain amplitude compared with
Big local peaking, and using the local peaking of the reservation as new current local peaking;Otherwise, retain current local peaking and
Next local peaking, and using next local peaking as new current local peaking;
Step S86: judging whether current local peaking is the last one local peaking, if so, continuing step S87;Otherwise, it returns
To step S84;
Step S87: counting the number of the local peaking of reservation, obtains the number of passenger loading during this enabling is closed the door.
2. public transport people flow rate statistical method according to claim 1, it is characterised in that: in step sl, obtain bus
After the video image of interior present frame, noise is removed by gaussian filtering to the current frame video image, then carry on the back by mixed Gaussian
Scape model obtains foreground image.
3. public transport people flow rate statistical method according to claim 2, it is characterised in that: the foreground image is binary map,
The region of characterization movement is white, and characterizing non-athletic region is black.
4. public transport people flow rate statistical method according to claim 1, it is characterised in that: detection of opening the door in the foreground image
The characteristic value in region is to characterize the pixel of movement and total pixel of the enabling detection zone in the enabling detection zone
Several ratio.
5. public transport people flow rate statistical method according to claim 1, it is characterised in that: occupant detection in the foreground image
The characteristic value in region is to characterize the pixel of movement and total pixel in the occupant detection region in the occupant detection region
Several ratio.
6. public transport people flow rate statistical method according to claim 1, it is characterised in that: detection of closing the door in the foreground image
The characteristic value in region is to characterize the pixel of movement and total pixel of the shutdown detection zone in the shutdown detection zone
Several ratio.
7. public transport people flow rate statistical method according to claim 1, it is characterised in that: in step S82, obtain local peaks
The method of value are as follows: in the characteristic value in the occupant detection region sequentially stored, it will be greater than each 10 characteristic values in its front and back,
And it is greater than the characteristic value of the 4th given threshold as local peaking.
8. a kind of public transport people flow rate statistical system, it is characterised in that: including processor, be adapted for carrying out each instruction;And storage is set
It is standby, it is suitable for storing a plurality of instruction, described instruction is suitable for being loaded and being executed by the processor:
The video image of present frame in bus is obtained, and extracts the foreground image in the video image of the present frame;
Judge vehicle door status, and when car door is in shutdown state, obtains the characteristic value of enabling detection zone in foreground image, and
When the characteristic value of enabling detection zone is greater than the first given threshold, then vehicle door status is set on door state, and before acquisition
The characteristic value and this feature in occupant detection region are worth corresponding time point in scape image, and store to dynamic array;Before acquisition
The characteristic value of shutdown detection zone in scape image;When the characteristic value of shutdown detection zone is greater than the second given threshold, then by vehicle
Door state is set to off door state, and corresponding according to the characteristic value in the occupant detection region in dynamic array and this feature value
Time point obtains the number of getting on the bus during this enabling is closed the door on bus;
When car door is in door opening state, then the characteristic value in occupant detection region and this feature value in foreground image are directly acquired
Corresponding time point, and store to dynamic array;
When counting the number of bus, the processor is loaded and is executed:
The characteristic value in the occupant detection region in dynamic array is sequentially stored by the sequencing of the acquisition time of frame where it;
According to the characteristic value in the occupant detection region sequentially stored, local peaking and the local peaking corresponding time are obtained
Point;
The local peaking that will acquire sequentially is stored by the sequencing at its corresponding time point, and with first local peaks of storage
Value is used as current local peaking;
Calculate the time interval between current local peaking and next local peaking;
Judge whether the time interval between current local peaking and next local peaking is less than third given threshold, if so,
Then in current local peaking and next local peaking, the lesser local peaking of amplitude is deleted, retains the biggish part of amplitude
Peak value, and using the local peaking of the reservation as new current local peaking;Otherwise, retain current local peaking and next office
Portion's peak value, and using next local peaking as new current local peaking;
When current local peaking is the last one local peaking, the number of the local peaking of reservation is counted, this enabling is obtained
The number of passenger loading during shutdown.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810397363.1A CN108875562B (en) | 2018-04-28 | 2018-04-28 | A kind of public transport people flow rate statistical method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810397363.1A CN108875562B (en) | 2018-04-28 | 2018-04-28 | A kind of public transport people flow rate statistical method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108875562A CN108875562A (en) | 2018-11-23 |
CN108875562B true CN108875562B (en) | 2019-10-11 |
Family
ID=64327238
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810397363.1A Expired - Fee Related CN108875562B (en) | 2018-04-28 | 2018-04-28 | A kind of public transport people flow rate statistical method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108875562B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114511821B (en) * | 2022-04-18 | 2022-07-26 | 深圳市爱深盈通信息技术有限公司 | Statistical method and system for number of persons getting on and off, computer equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104821025A (en) * | 2015-04-29 | 2015-08-05 | 广州运星科技有限公司 | Passenger flow detection method and detection system thereof |
CN105913367A (en) * | 2016-04-07 | 2016-08-31 | 北京晶众智慧交通科技股份有限公司 | Public bus passenger flow volume detection system and method based on face identification and position positioning |
CN106559647A (en) * | 2015-09-28 | 2017-04-05 | 株式会社日立制作所 | Enabling is closed the door detection means and method, artificial abortion's detecting system and the vehicles |
CN107330386A (en) * | 2017-06-21 | 2017-11-07 | 厦门中控智慧信息技术有限公司 | A kind of people flow rate statistical method and terminal device |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101383004A (en) * | 2007-09-06 | 2009-03-11 | 上海遥薇实业有限公司 | Passenger target detecting method combining infrared and visible light images |
CN107403137B (en) * | 2017-06-29 | 2020-01-31 | 山东师范大学 | Dense crowd flow calculation method and device based on video |
-
2018
- 2018-04-28 CN CN201810397363.1A patent/CN108875562B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104821025A (en) * | 2015-04-29 | 2015-08-05 | 广州运星科技有限公司 | Passenger flow detection method and detection system thereof |
CN106559647A (en) * | 2015-09-28 | 2017-04-05 | 株式会社日立制作所 | Enabling is closed the door detection means and method, artificial abortion's detecting system and the vehicles |
CN105913367A (en) * | 2016-04-07 | 2016-08-31 | 北京晶众智慧交通科技股份有限公司 | Public bus passenger flow volume detection system and method based on face identification and position positioning |
CN107330386A (en) * | 2017-06-21 | 2017-11-07 | 厦门中控智慧信息技术有限公司 | A kind of people flow rate statistical method and terminal device |
Non-Patent Citations (2)
Title |
---|
Real-time bus passenger flow statistics scheme based on light-sensitive wireless sensor network;Peiji Chen 等;《Proceedings of the 35th Chinese Control Conference》;20160729;第8490-8494页 * |
基于计算机视觉的公交车人流量统计系统;李衡宇 等;《四川大学学报(自然科学版)》;20070831;第44卷(第4期);第825-830页 * |
Also Published As
Publication number | Publication date |
---|---|
CN108875562A (en) | 2018-11-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108710865B (en) | Driver abnormal behavior detection method based on neural network | |
CN103824066B (en) | A kind of licence plate recognition method based on video flowing | |
CN102542289B (en) | Pedestrian volume statistical method based on plurality of Gaussian counting models | |
CN102968625B (en) | Ship distinguishing and tracking method based on trail | |
CN103745224B (en) | Image-based railway contact net bird-nest abnormal condition detection method | |
CN107330373A (en) | A kind of parking offense monitoring system based on video | |
CN104268596B (en) | A kind of Car license recognition device and its detection method of license plate and system | |
CN108090429A (en) | Face bayonet model recognizing method before a kind of classification | |
CN105740855A (en) | Front and rear license plate detection and recognition method based on deep learning | |
CN105809184A (en) | Vehicle real-time identification tracking and parking space occupancy determining method suitable for gas station | |
CN103034870B (en) | The boats and ships method for quickly identifying of feature based | |
Momin et al. | Vehicle detection and attribute based search of vehicles in video surveillance system | |
CN111062967B (en) | Electric power business hall passenger flow statistical method and system based on target dynamic tracking | |
CN108804987B (en) | Door opening and closing state detection method and device and people flow detection system | |
CN105551261A (en) | False-license-plate vehicle detection method and system | |
CN107578048B (en) | Vehicle type rough classification-based far-view scene vehicle detection method | |
CN104657724A (en) | Method for detecting pedestrians in traffic videos | |
CN108039044A (en) | The system and method that Vehicular intelligent based on multiple dimensioned convolutional neural networks is lined up | |
CN105095837B (en) | A kind of TV station symbol recognition method and system | |
Su et al. | A new local-main-gradient-orientation HOG and contour differences based algorithm for object classification | |
Chen et al. | Traffic congestion classification for nighttime surveillance videos | |
CN106127124A (en) | The automatic testing method of the abnormal image signal in region, taxi front row | |
CN103337175A (en) | Vehicle type recognition system based on real-time video steam | |
CN109711313B (en) | Real-time video monitoring method for identifying sewage poured into river channel | |
CN103778790B (en) | A kind of vehicle flowrate square wave statistic law based on video sequence |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20191011 |
|
CF01 | Termination of patent right due to non-payment of annual fee |