CN109753842A - A kind of method and device that flow of the people counts - Google Patents
A kind of method and device that flow of the people counts Download PDFInfo
- Publication number
- CN109753842A CN109753842A CN201711058966.0A CN201711058966A CN109753842A CN 109753842 A CN109753842 A CN 109753842A CN 201711058966 A CN201711058966 A CN 201711058966A CN 109753842 A CN109753842 A CN 109753842A
- Authority
- CN
- China
- Prior art keywords
- macro block
- video image
- block number
- sub
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Image Analysis (AREA)
Abstract
The present invention is suitable for field of image processing, provides a kind of method and device that flow of the people counts, comprising: acquire the video image of target area in real time first；Then video image is divided into multiple rectangular areas to obtain multiple sub-video images；Multiple sub-video images are handled respectively to obtain multiple movement tendency characteristic values；Movement tendency characteristic value includes entry time characteristic value, time departure characteristic value, foreground macro block number maximum average value and invariant feature value；Judge whether multiple invariant feature values are all larger than stable threshold again；If it is not, the step of then executing the video image for acquiring target area in real time；If so, estimating number characteristic value according to the acquisition of multiple entry time characteristic values, multiple time departure characteristic values and multiple foreground macro block number maximum average values is multiple；Finally flow of the people is obtained according to multiple number characteristic values of estimating；So there is no need to be trained using a large amount of given datas, and operand is small, improves the timeliness and accuracy of people flow rate statistical.
Description
Technical field
The invention belongs to the method and devices that field of image processing more particularly to a kind of flow of the people count.
Background technique
Computer vision technique is to study how to allow the method for counting of machine " seeing ", passes through and utilizes camera and computer generation
Target is identified for human eye, tracked and is measured, and further does image procossing, it is become with computer disposal and is more suitable for
Eye-observation or the image for sending instrument detection to.
Traditional flow of the people method of counting based on computer vision passes through the detection and tracking of movement human, is moved
Motion profile of the human body in sequence image, realizes the purpose of people flow rate statistical.Usually in traditional flow of the people method of counting,
Need to input largely the training that the priori data for having calibration feature by hand carries out neural network for the detection of human motion, to reach
To the function of detection movement human.Tracking for movement human, it is complete usually using mixed Gauss model and Kalman filter
At motion tracking.Traditional needs to be trained using a large amount of given datas based on computer vision method of counting, and operation
Amount is big, has certain requirement for hardware in practical applications, and need early period data volume of the acquisition for training huge, because
This needs the new real-time height of one kind, the simple flow of the people method of counting in real time of training early period.
Summary of the invention
In view of this, the embodiment of the invention provides the method and devices that a kind of flow of the people counts, it is intended to solve existing skill
The problem that preoperative phase training is complicated and operand is big.
The first aspect of the embodiment of the present invention provides a kind of method that flow of the people counts, comprising:
The video image of acquisition target area in real time；
The video image is divided into multiple rectangular areas to obtain multiple sub-video images；
Multiple sub-video images are handled to obtain multiple movement tendency characteristic values respectively；The movement tendency
Characteristic value includes entry time characteristic value, time departure characteristic value, foreground macro block number maximum average value and invariant feature value；
Judge whether the multiple invariant feature value is all larger than stable threshold；
If the multiple invariant feature value is not all larger than stable threshold, the video figure for acquiring target area in real time is executed
The step of picture；
If the multiple invariant feature value is all larger than stable threshold, according to multiple entry time characteristic values, multiple leave
Temporal characteristics value and multiple foreground macro block number maximum average values, which obtain, multiple estimates number characteristic value.
Flow of the people is obtained according to multiple number characteristic values of estimating.
The second aspect of the embodiment of the present invention provides a kind of device that flow of the people counts, comprising:
Acquisition module, for acquiring the video image of target area in real time；
Sub-video image collection module, for the video image to be divided into multiple rectangular areas to obtain multiple sub- views
Frequency image；
Movement tendency characteristic value acquisition module is handled multiple sub-video images for respectively multiple to obtain
Movement tendency characteristic value；The movement tendency characteristic value includes entry time characteristic value, time departure characteristic value, foreground macro block number
Maximum average value and invariant feature value；
Stable threshold judgment module, for judging whether the multiple invariant feature value is all larger than stable threshold；
If stable threshold judgment module judges that the multiple invariant feature value is not all larger than stable threshold, triggering collection mould
Block；
Number characteristic value acquisition module is estimated, if judging that the multiple invariant feature value is equal for stable threshold judgment module
It is greater than stable threshold, then maximum according to multiple entry time characteristic values, multiple time departure characteristic values and multiple foreground macro block numbers
Average value acquisition is multiple to estimate number characteristic value；
Flow of the people obtains module, for obtaining flow of the people according to multiple number characteristic values of estimating.
The third aspect of the embodiment of the present invention provide the device that a kind of flow of the people counts, including memory, processor with
And the computer program that can be run in the memory and on the processor is stored, the processor executes the calculating
The step of method that flow of the people described above counts is realized when machine program.
The fourth aspect of the embodiment of the present invention provides a kind of computer readable storage medium, the computer-readable storage
Media storage has computer program, and the computer program realizes the method that flow of the people described above counts when being executed by processor
The step of.
Existing beneficial effect is the embodiment of the present invention compared with prior art: acquiring the video of target area in real time first
Image；Then the video image is divided into multiple rectangular areas to obtain multiple sub-video images；Respectively to multiple described
Sub-video image is handled to obtain multiple movement tendency characteristic values；The movement tendency characteristic value includes entry time feature
Value, time departure characteristic value, foreground macro block number maximum average value and invariant feature value；Judging the multiple invariant feature value again is
It is no to be all larger than stable threshold；If multiple invariant feature values are not all larger than stable threshold, execute and acquire target area in real time
The step of video image；If multiple invariant feature values are all larger than stable threshold, according to multiple entry time characteristic values, Duo Geli
ETAD expected time of arrival and departure characteristic value and multiple foreground macro block number maximum average values, which obtain, multiple estimates number characteristic value；Finally according to multiple described
It estimates number characteristic value and obtains flow of the people；Therefore it effectively prevents needing to be trained using a large amount of given datas, and operand
Big defect improves the timeliness and accuracy of people flow rate statistical.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is a kind of implementation process schematic diagram for the method that flow of the people provided in an embodiment of the present invention counts；
Fig. 2 is another implementation process schematic diagram for the method that flow of the people provided in an embodiment of the present invention counts；
Fig. 3 is a kind of schematic diagram for the device that flow of the people provided in an embodiment of the present invention counts；
Fig. 4 is another schematic diagram for the device that flow of the people provided in an embodiment of the present invention counts；
Fig. 5 is that the one kind for the device movement tendency characteristic value acquisition module that flow of the people provided in an embodiment of the present invention counts is shown
It is intended to；
Fig. 6 is the one of device the first movement tendency characteristic value acquisition module that flow of the people provided in an embodiment of the present invention counts
Kind schematic diagram；
Fig. 7 is the one of device the first movement tendency characteristic value calculating module that flow of the people provided in an embodiment of the present invention counts
Kind schematic diagram；
Fig. 8 is the one of device the second movement tendency characteristic value calculating module that flow of the people provided in an embodiment of the present invention counts
Kind schematic diagram；
Fig. 9 is a kind of schematic diagram that the device foreground macro block that flow of the people provided in an embodiment of the present invention counts obtains module；
Figure 10 is one kind that the device that flow of the people provided in an embodiment of the present invention counts estimates number characteristic value acquisition module
Schematic diagram；
Figure 11 is another schematic diagram for the device that flow of the people provided in an embodiment of the present invention counts.
Specific embodiment
In being described below, for illustration and not for limitation, the tool of such as particular system structure, technology etc is proposed
Body details, to understand thoroughly the embodiment of the present invention.However, it will be clear to one skilled in the art that there is no these specific
The present invention also may be implemented in the other embodiments of details.In other situations, it omits to well-known system, device, electricity
The detailed description of road and method, in case unnecessary details interferes description of the invention.
In order to illustrate technical solutions according to the invention, the following is a description of specific embodiments.
Fig. 1 shows the implementation process for the method that flow of the people provided in an embodiment of the present invention counts, for ease of description, only
Part related to the embodiment of the present invention is shown, details are as follows:
In a step 101, the video image of target area is acquired in real time.
Wherein, target area is the region for calculating flow of the people.
In a step 102, video image is divided into multiple rectangular areas to obtain multiple sub-video images.
In specific implementation, if the flow of the people of image acquisition region is horizontal direction, video image is divided into multiple shapes
Shape size is identical, horizontal direction side by side, the rectangle sub-video image that does not overlap；If the flow of the people of image acquisition region is vertical
Histogram to, then by video image be divided into multiple shape sizes are identical, vertical direction side by side, the rectangle sub-video that does not overlap
Image.
In step 103, multiple sub-video images are handled respectively to obtain multiple movement tendency characteristic values；Movement
Trend feature value includes entry time characteristic value, time departure characteristic value, foreground macro block number maximum average value and invariant feature value.
In specific implementation, step 103 may include step 103-1 to step 103-4.
In step 103-1, by the macro block of each macro block of sub-video image and former frame sub-video image same position
It compares to obtain foreground macro block；Macro block is the rectangle of presetted pixel width and presetted pixel length.
In specific implementation, step 103-1 includes step a and step b.
A. the similarity of each macro block of sub-video image and the macro block of former frame sub-video image same position is obtained.
In specific implementation, can with computing macro block current sub-video image and the sub-video image of former frame similarity,
All coordinate points pixel value difference at current sub-video image and former frame sub-video image same coordinate point is greater than i.e. in macro block
λ1Ratio, wherein λ1For pixel similarity threshold value, the calculation formula of similarity are as follows:
Wherein P (i, j) be macroblock coordinate be (i, j) macro block in current sub-video image and former frame sub-video image phase
It is greater than λ with pixel value difference at coordinate points1Pixel ratio, (x, y) is pixel coordinate, and A (i, j) is that belong to coordinate be (i, j)
Macro block pixel collection, D (x, y) is each pixel difference situation, wherein the calculating of each pixel difference situation is public
Formula are as follows:
Wherein, IprePixel value for the pixel in former frame sub-video image, IcurIt is the pixel in current sub-video figure
The pixel value of picture.
B. similarity is greater than the macro block of similarity threshold as foreground macro block.Similarity is greater than the macro block of similarity threshold
It can be determined that as stream of people position, pretend as foreground macro block.
In step 103-2, sub-video image is filled to obtain foreground area according to foreground macro block.
Step 103-2 can be with specifically: foreground area filling is carried out from horizontal direction according to foreground macro block, further according to prospect
Macro block carries out foreground area filling from vertical direction.
First with the sequence in sub-video image from top to bottom, every a line macro block is handled, every a line is respectively from most
Left side macro block begins look for the foreground macro block of the leftmost side, and from the rightmost side, macro block begins look for the foreground macro block of the rightmost side, if this journey
There are foreground macro blocks, and all macro blocks in this journey between leftmost side foreground macro block and rightmost side foreground macro block are all labeled as prospect
Macro block；
Sequence in sub-video image from left to right again handles each column macro block, and each column are respectively from the top
Macro block begins look for the foreground macro block of the top, and the foreground macro block of bottom is begun look for from bottom macro block, if this column exists
Foreground macro block, by this arrange in all macro blocks between the top foreground macro block and bottom foreground macro block to be all labeled as prospect macro
Block.
In step 103-3, the foreground macro block number for belonging to foreground area is obtained.It counts and is labeled as in each sub-video image
The number of the macro block of foreground macro block.
In step 103-4, before sub-video image before the foreground macro block number of current sub-video image, preset time
The foreground macro block number of sub-video image obtains movement tendency characteristic value before scape macroblock number and double preset time.
In specific implementation, step 103-4 may include step A to step E.
A. judge whether the foreground macro block number of current sub-video image is greater than foreground macro block number threshold value, and current sub-video figure
Whether the foreground macro block number of picture is not equal to the foreground macro block number of sub-video image before preset time；If before current sub-video image
Scape macroblock number is greater than foreground macro block number threshold value, and the foreground macro block number of current sub-video image is not equal to sub-video before preset time
The foreground macro block number of image, thens follow the steps B；If the foreground macro block number of current sub-video image is not more than foreground macro block number threshold
Value, and/or currently the foreground macro block number of sub-video image is equal to the foreground macro block number of sub-video image before preset time, then executes
Step C.
B. the movement tendency characteristic value under motion state is calculated.
Wherein, step B may include step B-1 to step B-4.
B-1. invariant feature value and foreground macro block number maximum average value are set as zero.
B-2a. when foreground macro block number of the foreground macro block number of current sub-video image greater than sub-video image before preset time
When, the sum of entry time characteristic value and 1 is set as new entry time characteristic value.
B-2b. the prospect of sub-video image is macro before the foreground macro block number of current sub-video image is less than or equal to preset time
When block number, the sum of time departure characteristic value and 1 is set as new time departure characteristic value.
B-3. calculate the foreground macro block number of sub-video image before the foreground macro block number of current sub-video image, preset time,
And before double preset time the foreground macro block number of sub-video image foreground macro block number average value.
B-4. when prospect macroblock number average value is greater than foreground macro block number maximum average value, foreground macro block number maximum is averaged
Value is set as foreground macro block number maximum average value.
C. judge whether invariant feature value is zero, and whether the foreground macro block number of current sub-video image is less than preset time
The foreground macro block number of preceding sub-video image；If invariant feature value is zero, and the foreground macro block number of current sub-video image is less than in advance
If the foreground macro block number of sub-video image, thens follow the steps D before the time；If invariant feature value is not zero and/or current sub-video
The foreground macro block number of image is not less than the foreground macro block number of sub-video image before preset time, thens follow the steps E.
D. the movement tendency characteristic value under stable state is calculated.Present application contemplates the foreground macro blocks of current sub-video image
Number is not more than foreground macro block number threshold value, and whether the foreground macro block number of current sub-video image is less than sub-video figure before preset time
Situation when the foreground macro block number of picture, above-mentioned condition are at the end of the stream of people is fast, and target area number is seldom and the stream of people leaves target
Situation when region, and calculate the movement tendency characteristic value under the situation.
Wherein, step D may include step D-1 to step D-4.
D-1. the sum of time departure characteristic value and 1 is set as new time departure characteristic value.
D-2. the sum of invariant feature value and 1 is set as new invariant feature value.
D-3. calculate the foreground macro block number of sub-video image before the foreground macro block number of current sub-video image, preset time,
And before double preset time the foreground macro block number of sub-video image foreground macro block number average value.
D-4. when prospect macroblock number average value is greater than foreground macro block number maximum average value, foreground macro block number maximum is averaged
Value is set as foreground macro block number average value.
E. the sum of invariant feature value and 1 is set as new invariant feature value.
At step 104, judge whether multiple invariant feature values are all larger than stable threshold, if multiple invariant feature values are big
In stable threshold, 105 are thened follow the steps, if multiple invariant feature values are not all larger than stable threshold, thens follow the steps 101.
In step 105, according to multiple entry time characteristic values, multiple time departure characteristic values and multiple foreground macro block numbers
Maximum average value acquisition is multiple to estimate number characteristic value.
In specific implementation, step 105 includes step 105-1 to step 105-4.
In step 105-1, before being judged in sub-video image according to entry time characteristic value and time departure characteristic value
Whether scene area is noise；If the foreground area in sub-video image is not noise, 105-2 is thened follow the steps.
Step 105-1 can be with specifically: for current sub-video image, when judgement is currently entering temporal characteristics value and leaves
Between the absolute value of characteristic value difference whether be less than noise difference threshold, and whether entry time characteristic value and time departure characteristic value same
When be less than noise independent threshold, if be both it is yes, foreground area is not noise, and otherwise, foreground area is noise.Because of lamp
Opening and closing will lead to the erroneous judgement of foreground area, therefore need to judge whether foreground area is noise.
In step 105-2, according to entry time characteristic value and time departure characteristic value acquisition time characteristic value.It will enter
The sum of temporal characteristics value and time departure characteristic value is divided by time threshold with acquisition time characteristic value.
In step 105-3, number characteristic value is obtained according to foreground macro block number maximum average value.Most by foreground macro block number
Big average value obtains number characteristic value divided by number threshold value.
In step 105-4, number characteristic value is estimated according to temporal characteristics value and the acquisition of number characteristic value.By temporal characteristics
The product of value and number characteristic value is used as and estimates number characteristic value.
In step 106, flow of the people is obtained according to multiple number characteristic values of estimating.Multiple sub-video images are corresponding pre-
The average value of number characteristic value is estimated as flow of the people.
As shown in Fig. 2, can also include step 100 before step 101.
In step 100, movement tendency characteristic value is initialized.
As shown in Fig. 2, can also include step 107 and step 108 after step 106.
In step 107, subregion is entered according to multiple invariant feature values acquisition stream of people and the stream of people leaves subregion.It will most
The corresponding sub-video image of big invariant feature value enters subregion as the stream of people, by the corresponding sub- view of the smallest invariant feature value
Frequency image leaves subregion as the stream of people.
In step 108, subregion is entered according to the stream of people and the stream of people leaves subregion and obtains stream of people direction.
Method in order to realize above-mentioned flow of the people counting, the embodiment of the invention also provides the dresses that a kind of flow of the people counts
It sets, as shown in figure 3, the device 30 that the people's flow counts includes acquisition module 310, sub-video image collection module 320, moves
Gesture characteristic value acquisition module 330, stable threshold judgment module 340 estimate number characteristic value acquisition module 350 and flow of the people acquisition
Module 360.
Acquisition module 310, for acquiring the video image of target area in real time；
Sub-video image collection module 320, for video image to be divided into multiple rectangular areas to obtain multiple sub- views
Frequency image；
Movement tendency characteristic value acquisition module 330 is handled multiple sub-video images for respectively multiple to obtain
Movement tendency characteristic value；Movement tendency characteristic value includes entry time characteristic value, time departure characteristic value, foreground macro block number maximum
Average value and invariant feature value；
Stable threshold judgment module 340, for judging whether multiple invariant feature values are all larger than stable threshold；
If stable threshold judgment module 340 judges that multiple invariant feature values are not all larger than stable threshold, triggering collection mould
Block 310；
Number characteristic value acquisition module 350 is estimated, if judging that multiple invariant feature values are equal for stable threshold judgment module
It is greater than stable threshold, then maximum according to multiple entry time characteristic values, multiple time departure characteristic values and multiple foreground macro block numbers
Average value acquisition is multiple to estimate number characteristic value；
Flow of the people obtains module 360, for obtaining flow of the people according to multiple number characteristic values of estimating.
Wherein, as shown in figure 4, the device 30 that the people's flow counts can also include that the stream of people leaves subregion acquisition module
370, stream of people direction obtains module 380 and movement tendency characteristic value initialization module 390.
The stream of people leave subregion obtain module 370, for according to multiple invariant feature values obtain the stream of people enter subregion and
The stream of people leaves subregion.
Stream of people direction obtains module 380, for entering subregion according to the stream of people and the stream of people leaves subregion and obtains stream of people side
To.
Movement tendency characteristic value initialization module 390, for being initialized to movement tendency characteristic value.
As shown in figure 5, movement tendency characteristic value acquisition module 330 includes that foreground macro block obtains module 331, foreground area obtains
Modulus block 332, foreground macro block number obtain module 333 and the first movement tendency characteristic value acquisition module 334.
Foreground macro block obtains module 331, for each macro block of sub-video image is identical as former frame sub-video image
The macro block of position is compared to obtain foreground macro block；Macro block is the rectangle of presetted pixel width and presetted pixel length.
Foreground area obtains module 332, for filling sub-video image according to foreground macro block to obtain foreground area.
Foreground macro block number obtains module 333, for obtaining the foreground macro block number for belonging to foreground area.
First movement tendency characteristic value acquisition module 334, for according to the foreground macro block number of current sub-video image, default
The foreground macro block number of sub-video image obtains movement before the foreground macro block number of sub-video image and double preset time before time
Trend feature value.
Wherein, as shown in fig. 6, the first movement tendency characteristic value acquisition module 334 include motion state judgment module 3341,
First movement tendency characteristic value calculating module 3342, rim condition judgment module 3343, the second movement tendency characteristic value calculate mould
Block 3344 and the first invariant feature value setting module 3345.
Motion state judgment module 3341, for judging it is macro whether the foreground macro block number of current sub-video image is greater than prospect
Block number threshold value, and whether the foreground macro block number of current sub-video image is not equal to the foreground macro block of sub-video image before preset time
Number；
First movement tendency characteristic value calculating module 3342, if judging current sub-video figure for motion state judgment module
The foreground macro block number of picture is greater than foreground macro block number threshold value, and the foreground macro block number of current sub-video image is not equal to before preset time
The foreground macro block number of sub-video image then calculates the movement tendency characteristic value under motion state；
Rim condition judgment module 3343, if judging that the prospect of current sub-video image is macro for motion state judgment module
Block number is not more than foreground macro block number threshold value, and/or the foreground macro block number of current sub-video image is equal to sub-video before preset time
The foreground macro block number of image then judges whether invariant feature value is zero, and whether the foreground macro block number of current sub-video image is small
The foreground macro block number of sub-video image before preset time；
Second movement tendency characteristic value calculating module 3344, if judging that invariant feature value is for rim condition judgment module
Zero, and the foreground macro block number of current sub-video image is less than the foreground macro block number of sub-video image before preset time, then calculates steady
Determine the movement tendency characteristic value under state；
First invariant feature value setting module 3345 is if judging invariant feature value not for rim condition judgment module
Zero, and/or currently the foreground macro block number of sub-video image then will not less than the foreground macro block number of sub-video image before preset time
Invariant feature value is set as new invariant feature value with 1 sum.
Wherein, as shown in fig. 7, the first movement tendency characteristic value calculating module 3342 includes invariant feature value initialization module
3342-1, entry time characteristic value setting module 3342-2, the first time departure characteristic value setting module 3342-3, the first prospect
Macroblock number mean value calculation module 3342-4 and the first foreground macro block number maximum average value setting module 3342-5.
Invariant feature value initialization module 3342-1, for invariant feature value and foreground macro block number maximum average value to be all provided with
It is zero；
Entry time characteristic value setting module 3342-2 is greater than default for the foreground macro block number when current sub-video image
Before time when the foreground macro block number of sub-video image, the sum of entry time characteristic value and 1 is set as new entry time characteristic value；
First time departure characteristic value setting module 3342-3 is less than for the foreground macro block number when current sub-video image
Equal to before preset time when the foreground macro block number of sub-video image, time departure characteristic value is set as new when leaving with 1 sum
Between characteristic value；
First foreground macro block number mean value calculation module 3342-4, for calculating the foreground macro block of current sub-video image
The foreground macro block number of sub-video image before several, sub-video image before preset time foreground macro block number and double preset time
Foreground macro block number average value；
First foreground macro block number maximum average value setting module 3342-5, for being greater than prospect when prospect macroblock number average value
When macroblock number maximum average value, foreground macro block number maximum average value is set as foreground macro block number maximum average value.
Wherein, as shown in figure 8, the second movement tendency characteristic value calculating module 3344 includes that the second time departure characteristic value is set
Cover half block 3344-1, the second invariant feature value setting module 3344-2, the second foreground macro block number mean value calculation module 3344-3
With the second foreground macro block number maximum average value setting module 3344-4.
Second time departure characteristic value setting module 3344-1, for the sum of time departure characteristic value and 1 to be set as new
Time departure characteristic value；
Second invariant feature value setting module 3344-2, for the sum of invariant feature value and 1 to be set as new invariant feature
Value；
Second foreground macro block number mean value calculation module 3344-3, for calculating the foreground macro block of current sub-video image
The foreground macro block number of sub-video image before several, sub-video image before preset time foreground macro block number and double preset time
Foreground macro block number average value；
Second foreground macro block number maximum average value setting module 3344-4, for being greater than prospect when prospect macroblock number average value
When macroblock number maximum average value, foreground macro block number maximum average value is set as foreground macro block number average value.
Wherein, as shown in figure 9, it includes that similarity acquisition module 3311 and foreground module are set that foreground macro block, which obtains module 331,
Cover half block 3312.
Similarity obtains module 3311, and each macro block for obtaining sub-video image is identical as former frame sub-video image
The similarity of the macro block of position；
Foreground module setting module 3312, for similarity to be greater than to the macro block of similarity threshold as foreground macro block.
Wherein, as shown in Figure 10, estimating number characteristic value acquisition module 350 includes noise judgment module 351, temporal characteristics
Value obtains module 352, number characteristic value acquisition module 353 and estimates number characteristic value acquisition module 354.
Noise judgment module 351, for judging sub-video image according to entry time characteristic value and time departure characteristic value
Whether interior foreground area is noise；
Temporal characteristics value obtains module 352, if judging the foreground area in sub-video image not for noise judgment module
For noise, then according to entry time characteristic value and time departure characteristic value acquisition time characteristic value；
Number characteristic value acquisition module 353, for obtaining number characteristic value according to foreground macro block number maximum average value；
Number characteristic value acquisition module 354 is estimated, for estimating number according to temporal characteristics value and the acquisition of number characteristic value
Characteristic value.
In conclusion the embodiment of the present invention acquires the video image of target area in real time first；Then by the video figure
As being divided into multiple rectangular areas to obtain multiple sub-video images；Multiple sub-video images are handled to obtain respectively
Take multiple movement tendency characteristic values；The movement tendency characteristic value includes entry time characteristic value, time departure characteristic value, prospect
Macroblock number maximum average value and invariant feature value；Judge whether the multiple invariant feature value is all larger than stable threshold again；If more
The step of a invariant feature value is not all larger than stable threshold, then executes the video image for acquiring target area in real time；If multiple
Invariant feature value is all larger than stable threshold, then according to multiple entry time characteristic values, multiple time departure characteristic values and it is multiple before
The acquisition of scape macroblock number maximum average value is multiple to estimate number characteristic value；Finally people is obtained according to multiple number characteristic values of estimating
Flow；Therefore effectively prevent needing to be trained using a large amount of given datas, and the defect that operand is big, improve flow of the people
The timeliness and accuracy of statistics.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
Figure 11 is the schematic diagram for the device that a kind of flow of the people that one embodiment of the invention provides counts.As shown in figure 11, should
The device 11 that a kind of flow of the people of embodiment counts includes: processor 110, memory 111 and is stored in memory 111 simultaneously
The computer program 112 that can be run on processor 110, such as the program that flow of the people counts.Processor 110 executes computer journey
Realize the step in embodiment of the method that above-mentioned each flow of the people counts when sequence 112, such as step 101 shown in FIG. 1 is to 106.
Alternatively, processor 110 realizes the function of each module/unit in above-mentioned each Installation practice when executing computer program 112, such as
The function of module 310 to 360 shown in Fig. 3.
Illustratively, computer program 112 can be divided into one or more module/units, one or more mould
Block/unit is stored in memory 111, and is executed by processor 110, to complete the present invention.One or more module/units
It can be the series of computation machine program instruction section that can complete specific function, the instruction segment is for describing computer program 112
Implementation procedure in the device 11 that flow of the people counts.For example, computer program 112 can be divided into acquisition module 310, son
Video image obtains module 320, movement tendency characteristic value acquisition module 330, stable threshold judgment module 340, estimates number spy
Value indicative obtains module 350 and flow of the people obtains module 360 (module in virtual bench), and each module concrete function is as follows:
Acquisition module 310, for acquiring the video image of target area in real time；
Sub-video image collection module 320, for video image to be divided into multiple rectangular areas to obtain multiple sub- views
Frequency image；
Movement tendency characteristic value acquisition module 330 is handled multiple sub-video images for respectively multiple to obtain
Movement tendency characteristic value；Movement tendency characteristic value includes entry time characteristic value, time departure characteristic value, foreground macro block number maximum
Average value and invariant feature value；
Stable threshold judgment module 340, for judging whether multiple invariant feature values are all larger than stable threshold；
If stable threshold judgment module 340 judges that multiple invariant feature values are not all larger than stable threshold, triggering collection mould
Block 310；
Number characteristic value acquisition module 350 is estimated, if judging that multiple invariant feature values are equal for stable threshold judgment module
It is greater than stable threshold, then maximum according to multiple entry time characteristic values, multiple time departure characteristic values and multiple foreground macro block numbers
Average value acquisition is multiple to estimate number characteristic value；
Flow of the people obtains module 360, for obtaining flow of the people according to multiple number characteristic values of estimating.
The device 11 that the flow of the people counts can be gateway or other access devices.The device that the flow of the people counts can
It include but are not limited to, processor 110, memory 111.It will be understood by those skilled in the art that Fig. 7 is only stream of people's meter
The example of several devices 11 does not constitute the restriction of the device 11 counted to flow of the people, may include more more or less than illustrating
Component, perhaps combine certain components or different components, such as the device that the flow of the people counts can also include input
Output equipment, network access equipment, bus etc..
Alleged processor 110 can be central processing unit (Central Processing Unit, CPU), can also be
Other general processors, digital signal processor (Digital Signal Processor, DSP), specific integrated circuit
(Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng.
The memory 111 can be the internal storage unit for the device 11 that the flow of the people counts, such as stream of people's meter
The hard disk or memory of several devices 11.The memory 111 is also possible to the external storage for the device 11 that the flow of the people counts
The plug-in type hard disk being equipped in equipment, such as the device 11 of flow of the people counting, intelligent memory card (Smart Media
Card, SMC), secure digital (Secure Digital, SD) card, flash card (Flash Card) etc..Further, described to deposit
Reservoir 111 can also both include the flow of the people count device 11 internal storage unit and also including External memory equipment.Institute
Memory 111 is stated for other program sum numbers needed for storing the device that the computer program and the flow of the people count
According to.The memory 111 can be also used for temporarily storing the data that has exported or will export.
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing
The all or part of function of description.Each functional unit in embodiment, module can integrate in one processing unit, can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units, it is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.In addition, each function list
Member, the specific name of module are also only for convenience of distinguishing each other, the protection scope being not intended to limit this application.Above system
The specific work process of middle unit, module, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, is not described in detail or remembers in some embodiment
The part of load may refer to the associated description of other embodiments.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
In embodiment provided by the present invention, it should be understood that disclosed device/terminal device and method, it can be with
It realizes by another way.For example, device described above/terminal device embodiment is only schematical, for example, institute
The division of module or unit is stated, only a kind of logical function partition, there may be another division manner in actual implementation, such as
Multiple units or components can be combined or can be integrated into another system, or some features can be ignored or not executed.Separately
A bit, shown or discussed mutual coupling or direct-coupling or communication connection can be through some interfaces, device
Or the INDIRECT COUPLING or communication connection of unit, it can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
'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, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated module/unit be realized in the form of SFU software functional unit and as independent product sale or
In use, can store in a computer readable storage medium.Based on this understanding, the present invention realizes above-mentioned implementation
All or part of the process in example method, can also instruct relevant hardware to complete, the meter by computer program
Calculation machine program can be stored in a computer readable storage medium, the computer program when being executed by processor, it can be achieved that on
The step of stating each embodiment of the method.Wherein, the computer program includes computer program code, the computer program
Code can be source code form, object identification code form, executable file or certain intermediate forms etc..Computer-readable Jie
Matter may include: can carry the computer program code any entity or device, recording medium, USB flash disk, mobile hard disk,
Magnetic disk, CD, computer storage, read-only memory (ROM, Read-Only Memory), random access memory (RAM,
Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium etc..It should be noted that described
The content that computer-readable medium includes can carry out increasing appropriate according to the requirement made laws in jurisdiction with patent practice
Subtract, such as does not include electric carrier signal and electricity according to legislation and patent practice, computer-readable medium in certain jurisdictions
Believe signal.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (20)
1. a kind of method that flow of the people counts, which is characterized in that the described method includes:
The video image of acquisition target area in real time；
The video image is divided into multiple rectangular areas to obtain multiple sub-video images；
Multiple sub-video images are handled to obtain multiple movement tendency characteristic values respectively；The movement tendency feature
Value includes entry time characteristic value, time departure characteristic value, foreground macro block number maximum average value and invariant feature value；
Judge whether the multiple invariant feature value is all larger than stable threshold；
If the multiple invariant feature value is not all larger than the stable threshold, the video figure for acquiring target area in real time is executed
The step of picture；
If the multiple invariant feature value is all larger than the stable threshold, according to multiple entry time characteristic values, multiple leave
Temporal characteristics value and multiple foreground macro block number maximum average values, which obtain, multiple estimates number characteristic value.
Flow of the people is obtained according to multiple number characteristic values of estimating.
2. the method that flow of the people as described in claim 1 counts, which is characterized in that described respectively to multiple sub-video figures
As being handled to obtain multiple movement tendency characteristic values and include:
By the macro block of sub-video image same position described in each macro block of the sub-video image and former frame compare with
Obtain foreground macro block；The macro block is the rectangle of presetted pixel width and presetted pixel length；
The sub-video image is filled according to the foreground macro block to obtain foreground area；
Obtain the foreground macro block number for belonging to the foreground area；
According to the foreground macro block number of sub-video image before the foreground macro block number of current sub-video image, preset time and double
The foreground macro block number of sub-video image obtains movement tendency characteristic value before the preset time.
3. the method that flow of the people as claimed in claim 2 counts, which is characterized in that before the current sub-video image of basis
Sub-video image before the foreground macro block number of sub-video image and the double preset time before scape macroblock number, preset time
Foreground macro block number obtains movement tendency characteristic value
Judge whether the foreground macro block number of current sub-video image is greater than foreground macro block number threshold value, and before current sub-video image
Whether scape macroblock number is not equal to the foreground macro block number of sub-video image before preset time；
If the foreground macro block number of current sub-video image is greater than the foreground macro block number threshold value, and the prospect of current sub-video image
Macroblock number is not equal to the foreground macro block number of sub-video image before preset time, then calculates the movement tendency feature under motion state
Value；
Before if the foreground macro block number of current sub-video image is not more than foreground macro block number threshold value, and/or current sub-video image
Scape macroblock number is equal to the foreground macro block number of sub-video image before preset time, then judges whether the invariant feature value is zero, and
Whether the foreground macro block number of current sub-video image is less than the foreground macro block number of sub-video image before preset time；
If judging, the invariant feature value is zero, and the foreground macro block number of current sub-video image is less than sub-video before preset time
The foreground macro block number of image then calculates the movement tendency characteristic value under stable state；
If judging, the invariant feature value is not zero, and/or the foreground macro block number of current sub-video image is not less than preset time
The sum of the invariant feature value and 1 is then set as the new invariant feature value by the foreground macro block number of preceding sub-video image.
4. the method that flow of the people as claimed in claim 3 counts, which is characterized in that the movement calculated under motion state becomes
Gesture characteristic value includes:
The invariant feature value and foreground macro block number maximum average value are set as zero；
Before the foreground macro block number of current sub-video image is greater than preset time when the foreground macro block number of sub-video image, it will enter
Temporal characteristics value is set as new entry time characteristic value with 1 sum；
It, will before the foreground macro block number of current sub-video image is less than or equal to preset time when the foreground macro block number of sub-video image
Time departure characteristic value is set as new time departure characteristic value with 1 sum；
Calculate before the foreground macro block number of current sub-video image, preset time the foreground macro block number of sub-video image and double
The foreground macro block number average value of the foreground macro block number of sub-video image before the preset time；
When prospect macroblock number average value is greater than the foreground macro block number maximum average value, the foreground macro block number maximum is averaged
Value is set as the foreground macro block number maximum average value.
5. the method that flow of the people as claimed in claim 3 counts, which is characterized in that the movement calculated under stable state becomes
Gesture characteristic value includes:
The sum of time departure characteristic value and 1 is set as new time departure characteristic value；
The sum of the invariant feature value and 1 is set as the new invariant feature value
Calculate before the foreground macro block number of current sub-video image, preset time the foreground macro block number of sub-video image and double
The foreground macro block number average value of the foreground macro block number of sub-video image before the preset time；
When prospect macroblock number average value is greater than the foreground macro block number maximum average value, the foreground macro block number maximum is averaged
Value is set as the foreground macro block number average value.
6. the method that flow of the people as described in claim 1 counts, which is characterized in that described by each of described sub-video image
Macro block and the macro block of sub-video image same position described in former frame are compared to obtain foreground macro block and include:
Obtain the similarity of each macro block and the macro block of sub-video image same position described in former frame of the sub-video image；
Similarity is greater than the macro block of similarity threshold as foreground macro block.
7. the method that flow of the people as described in claim 1 counts, which is characterized in that institute according to multiple entry time characteristic values,
Multiple time departure characteristic values and multiple foreground macro block number maximum average values obtain multiple number characteristic values of estimating
The prospect in the sub-video image is judged according to the entry time characteristic value and the time departure characteristic value
Whether region is noise；
If the foreground area in the sub-video image is not noise, according to the entry time characteristic value and it is described from
ETAD expected time of arrival and departure characteristic value acquisition time characteristic value；
Number characteristic value is obtained according to the foreground macro block number maximum average value；
Number characteristic value is estimated according to the temporal characteristics value and number characteristic value acquisition.
8. the method that flow of the people as described in claim 1 counts, which is characterized in that described described to estimate number spy according to multiple
Value indicative obtains after flow of the people further include:
Enter subregion according to multiple invariant feature values acquisition stream of people and the stream of people leaves subregion；
Enter subregion according to the stream of people and the stream of people leaves subregion and obtains stream of people direction.
9. the method that flow of the people as described in claim 1 counts, which is characterized in that the video of the real-time acquisition target area
Before image further include:
The movement tendency characteristic value is initialized.
10. the device that a kind of flow of the people counts characterized by comprising
Acquisition module, for acquiring the video image of target area in real time；
Sub-video image collection module, for the video image to be divided into multiple rectangular areas to obtain multiple sub-video figures
Picture；
Movement tendency characteristic value acquisition module, for respectively being handled multiple sub-video images to obtain multiple movements
Trend feature value；The movement tendency characteristic value includes entry time characteristic value, time departure characteristic value, foreground macro block number maximum
Average value and invariant feature value；
Stable threshold judgment module, for judging whether the multiple invariant feature value is all larger than stable threshold；
If stable threshold judgment module judges that the multiple invariant feature value is not all larger than stable threshold, triggering collection module；
Number characteristic value acquisition module is estimated, if judging that the multiple invariant feature value is all larger than for stable threshold judgment module
Stable threshold, then it is average according to multiple entry time characteristic values, multiple time departure characteristic values and multiple foreground macro block number maximums
Value acquisition is multiple to estimate number characteristic value；
Flow of the people obtains module, for obtaining flow of the people according to multiple number characteristic values of estimating.
11. the device that flow of the people according to claim 10 counts, which is characterized in that the movement tendency characteristic value obtains
Module includes:
Foreground macro block obtains module, for each macro block of the sub-video image is identical as sub-video image described in former frame
The macro block of position is compared to obtain foreground macro block；The macro block is the rectangle of presetted pixel width and presetted pixel length；
Foreground area obtains module, for filling the sub-video image according to the foreground macro block to obtain foreground area；
Foreground macro block number obtains module, for obtaining the foreground macro block number for belonging to the foreground area；
First movement tendency characteristic value acquisition module, before according to the foreground macro block number of current sub-video image, preset time
The foreground macro block number of sub-video image, which obtains, before the foreground macro block number of sub-video image and the double preset time moves
Gesture characteristic value.
12. the device that flow of the people according to claim 11 counts, which is characterized in that the first movement tendency characteristic value
Obtaining module includes:
Motion state judgment module, for judging whether the foreground macro block number of current sub-video image is greater than foreground macro block number threshold
Value, and whether the foreground macro block number of current sub-video image is not equal to the foreground macro block number of sub-video image before preset time；
First movement tendency characteristic value calculating module, if judging the prospect of current sub-video image for motion state judgment module
Macroblock number is greater than the foreground macro block number threshold value, and the foreground macro block number of current sub-video image is not equal to son view before preset time
The foreground macro block number of frequency image then calculates the movement tendency characteristic value under motion state；
Rim condition judgment module, if judging that the foreground macro block number of current sub-video image is little for motion state judgment module
In the foreground macro block number threshold value, and/or the foreground macro block number of current sub-video image is equal to sub-video image before preset time
Foreground macro block number, then judge whether the invariant feature value is zero, and whether the foreground macro block number of current sub-video image small
The foreground macro block number of sub-video image before preset time；
Second movement tendency characteristic value calculating module, if judging that the invariant feature value is zero for rim condition judgment module,
And the foreground macro block number of current sub-video image is less than the foreground macro block number of sub-video image before preset time, then calculates stable shape
Movement tendency characteristic value under state；
First invariant feature value setting module, if judge that the invariant feature value is not zero for rim condition judgment module, and/
Or the foreground macro block number of current sub-video image is not less than the foreground macro block number of sub-video image before preset time, then it will be described steady
The sum for determining characteristic value with 1 is set as the new invariant feature value.
13. the device that flow of the people according to claim 12 counts, which is characterized in that the first movement tendency characteristic value
Computing module includes:
Invariant feature value initialization module, for the invariant feature value and foreground macro block number maximum average value to be set as zero；
Entry time characteristic value setting module, for the son view before the foreground macro block number of current sub-video image is greater than preset time
When the foreground macro block number of frequency image, the sum of entry time characteristic value and 1 is set as new entry time characteristic value；
First time departure characteristic value setting module, for when the foreground macro block number of current sub-video image is less than or equal to default
Between preceding sub-video image foreground macro block number when, time departure characteristic value is set as new time departure characteristic value with 1 sum；
First foreground macro block number mean value calculation module, for calculating foreground macro block number, the preset time of current sub-video image
The prospect of the foreground macro block number of sub-video image is macro before the foreground macro block number of preceding sub-video image and the double preset time
Block number average value；
First foreground macro block number maximum average value setting module, for being greater than the foreground macro block number when prospect macroblock number average value
When maximum average value, the foreground macro block number maximum average value is set as the foreground macro block number maximum average value.
14. the device that flow of the people according to claim 12 counts, which is characterized in that the second movement tendency characteristic value
Computing module includes:
Second time departure characteristic value setting module, for the sum of time departure characteristic value and 1 to be set as new time departure spy
Value indicative；
Second invariant feature value setting module, for the sum of the invariant feature value and 1 to be set as the new invariant feature value；
Second foreground macro block number mean value calculation module, for calculating foreground macro block number, the preset time of current sub-video image
The prospect of the foreground macro block number of sub-video image is macro before the foreground macro block number of preceding sub-video image and the double preset time
Block number average value；
Second foreground macro block number maximum average value setting module, for being greater than the foreground macro block number when prospect macroblock number average value
When maximum average value, the foreground macro block number maximum average value is set as the foreground macro block number average value.
15. the device that flow of the people according to claim 11 counts, which is characterized in that the foreground macro block obtains module packet
It includes:
Similarity obtains module, and each macro block for obtaining the sub-video image is identical as sub-video image described in former frame
The similarity of the macro block of position；
Foreground module setting module, for similarity to be greater than to the macro block of similarity threshold as foreground macro block.
16. the device that flow of the people according to claim 10 counts, which is characterized in that described to estimate the acquisition of number characteristic value
Module includes:
Noise judgment module, for judging the sub-video according to the entry time characteristic value and the time departure characteristic value
Whether the foreground area in image is noise；
Temporal characteristics value obtains module, if judging the foreground area in the sub-video image not for noise judgment module
For noise, then according to the entry time characteristic value and the time departure characteristic value acquisition time characteristic value；
Number characteristic value acquisition module, for obtaining number characteristic value according to the foreground macro block number maximum average value；
Number characteristic value acquisition module is estimated, for estimating number according to the temporal characteristics value and number characteristic value acquisition
Characteristic value.
17. the device that flow of the people according to claim 10 counts, which is characterized in that further include:
The stream of people leaves subregion and obtains module, and for being obtained according to multiple invariant feature values, the stream of people enters subregion and the stream of people leaves
Subregion；
Stream of people direction obtains module, for entering subregion according to the stream of people and the stream of people leaves subregion and obtains stream of people side
To.
18. the device that flow of the people according to claim 10 counts, which is characterized in that further include:
Movement tendency characteristic value initialization module, for being initialized to the movement tendency characteristic value.
19. the device that a kind of flow of the people counts, including memory, processor and storage are in the memory and can be described
The computer program run on processor, which is characterized in that the processor realizes such as right when executing the computer program
It is required that the step of method of any one of 1 to the 9 flow of the people counting.
20. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In realizing the method that the flow of the people as described in any one of claim 1 to 9 counts when the computer program is executed by processor
Step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711058966.0A CN109753842B (en) | 2017-11-01 | 2017-11-01 | People flow counting method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711058966.0A CN109753842B (en) | 2017-11-01 | 2017-11-01 | People flow counting method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109753842A true CN109753842A (en) | 2019-05-14 |
| CN109753842B CN109753842B (en) | 2021-07-16 |
Family
ID=66397872
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711058966.0A Active CN109753842B (en) | 2017-11-01 | 2017-11-01 | People flow counting method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109753842B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110115920A1 (en) * | 2009-11-18 | 2011-05-19 | Industrial Technology Research Institute | Multi-state target tracking mehtod and system |
| CN102542289A (en) * | 2011-12-16 | 2012-07-04 | 重庆邮电大学 | Pedestrian volume statistical method based on plurality of Gaussian counting models |
| WO2013110780A1 (en) * | 2012-01-25 | 2013-08-01 | Bruno Delean | Systems, methods and computer program products for identifying objects in video data |
| CN104504394A (en) * | 2014-12-10 | 2015-04-08 | 哈尔滨工业大学深圳研究生院 | Dese population estimation method and system based on multi-feature fusion |
-
2017
- 2017-11-01 CN CN201711058966.0A patent/CN109753842B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110115920A1 (en) * | 2009-11-18 | 2011-05-19 | Industrial Technology Research Institute | Multi-state target tracking mehtod and system |
| CN102542289A (en) * | 2011-12-16 | 2012-07-04 | 重庆邮电大学 | Pedestrian volume statistical method based on plurality of Gaussian counting models |
| WO2013110780A1 (en) * | 2012-01-25 | 2013-08-01 | Bruno Delean | Systems, methods and computer program products for identifying objects in video data |
| CN104504394A (en) * | 2014-12-10 | 2015-04-08 | 哈尔滨工业大学深圳研究生院 | Dese population estimation method and system based on multi-feature fusion |
Non-Patent Citations (2)
| Title |
|---|
| BOGGAVARAPU R K ,PATERIYA P K: "Efficient method for counting and tracking of moving objects through region segmentation", 《2016 IEEE INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND COMPUTING RESEARCH (ICCIC)》 * |
| 程栗: "基于视频监控的人流量统计系统研究", 《中国优秀硕士论文全文数据库》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109753842B (en) | 2021-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106558053B (en) | Object segmentation methods and Object Segmentation device | |
| CN108921022A (en) | A kind of human body attribute recognition approach, device, equipment and medium | |
| CN105184812B (en) | A kind of pedestrian based on target following hovers detection method | |
| CN107274472A (en) | A kind of method and apparatus of raising VR play frame rate | |
| CN107403168A (en) | A kind of facial-recognition security systems | |
| US9406140B2 (en) | Method and apparatus for generating depth information | |
| CN107483953B (en) | Inter frame motion estimation method, apparatus and electronic equipment | |
| CN109840881A (en) | A kind of 3D special efficacy image generating method, device and equipment | |
| CN103400142B (en) | A kind of pedestrian counting method | |
| CN109753917A (en) | Face quality optimization method, system, computer readable storage medium and equipment | |
| CN109214324A (en) | Most face image output method and output system based on polyphaser array | |
| CN109711267A (en) | A kind of pedestrian identifies again, pedestrian movement's orbit generation method and device | |
| CN109461133A (en) | Bridge bolt dropping detection method and terminal device | |
| CN102054306A (en) | Method and system for detecting pedestrian flow by adopting deformable two-dimensional curves | |
| Asad et al. | Kinect depth stream pre-processing for hand gesture recognition | |
| CN103500454A (en) | Method for extracting moving target of shaking video | |
| CN106774846B (en) | Alternative projection method and device | |
| CN104142730A (en) | Method for mapping gesture tracking results to mouse events | |
| CN107705414B (en) | A kind of recognition methods of bank note, device, terminal device and storage medium | |
| CN109954854A (en) | Conticaster crystallizer bleed-out monitoring method, device, storage medium and electric terminal | |
| CN109753842A (en) | A kind of method and device that flow of the people counts | |
| CN109379384A (en) | Wireless network secure partition method, device and electronic equipment | |
| CN105631868A (en) | Depth information extraction method based on image classification | |
| CN104978558B (en) | The recognition methods of target and device | |
| CN105138979A (en) | Method for detecting the head of moving human body based on stereo visual sense |
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 |