CN103679745B - A kind of moving target detecting method and device - Google Patents
A kind of moving target detecting method and device Download PDFInfo
- Publication number
- CN103679745B CN103679745B CN201210344459.4A CN201210344459A CN103679745B CN 103679745 B CN103679745 B CN 103679745B CN 201210344459 A CN201210344459 A CN 201210344459A CN 103679745 B CN103679745 B CN 103679745B
- Authority
- CN
- China
- Prior art keywords
- image
- area
- frame image
- region
- current frame
- 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.)
- Active
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of moving target detecting method and device, in order to improve the detection efficiency to moving target and accuracy rate.Described moving target detecting method includes: extract the prospect of current frame image according to current background；Judge that whether the number of pixels of prospect is more than presetted pixel number, if, judge whether the area of the moving region of current frame image is more than the area in predetermined movement region, if it is, record meets the area number of image frames more than the area in predetermined movement region of moving region；Judge whether the area meeting moving region in Δ t is more than pre-set image frame number, if it is, determine and there is currently moving target more than the number of image frames of the area in predetermined movement region.
Description
Technical field
The present invention relates to technical field of image processing, particularly relate to a kind of moving target detecting method and device.
Background technology
Video monitoring is to be analyzed the video image in monitoring scene based on computer video, extracts
Image information in monitored scene, forms the warning of moving target, in public safety protection and traffic administration
Aspect plays an important role.
Video monitoring system is to use video image acquisition device that a certain scene is continuously shot collection
Video flowing, processes the video flowing collected, detects moving target therein, and carry out moving target
Follow the tracks of and subsequent treatment.
In some cases, video monitoring system is without counting the exact position of the moving target in scene
Calculate, it is only necessary to determine under this scene, whether have notable moving target.
The existing detection method to moving target comprises the following steps:
The first, the current frame video image obtained in video flowing is carried out background modeling, obtain video image
Background image.
The second, utilize the background image obtained that the foreground target in video image is extracted accurately, and
Carry out context update.
3rd, it is tracked processing, to judge whether it moves to the foreground target extracted.
One general character of close shot video is exactly that environment is the least, is affected the biggest by illumination variation.The back of the body of prior art
Scape modeling method is difficult in adapt to such illumination variation, it will usually cause modeling unsuccessfully, foreground extraction mistake.
Further, the exact position of the foreground target prospect that prior art is extracted, and be tracked prospect processing,
Data processing amount is relatively big, relatively low to the detection efficiency ratio of moving target, under the scene that illumination variation is bigger,
Extract prospect coordinate position accurately extremely difficult, it is difficult to obtain moving target or acquisition motion mesh accurately
Mark unsuccessfully.
Prior art, under the scene that illumination variation is bigger, the detection efficiency of moving target is relatively low, and is difficult to
Moving target accurately detected.
Summary of the invention
The embodiment of the present invention provides a kind of moving target detecting method and device, in order to improve moving target
Detection efficiency and accuracy rate.
The moving target detecting method that the embodiment of the present invention provides, comprises the following steps:
A, according to current background extract current frame image prospect；
B, add up the number of pixels of described prospect, it is judged that described number of pixels whether more than presetted pixel number,
If it is, perform step C, otherwise perform step D；
C, calculating current frame image, relative to the area of the moving region of previous frame image, perform step E；Its
In, described previous frame image meets: the number of pixels of prospect is more than presetted pixel number；
D, current frame image is updated to current background, next frame image is performed step as current frame image
Rapid A；
E, judge the area of moving region of current frame image whether more than the area in predetermined movement region, as
Fruit is, performs step F, otherwise performs step D；
F, judge to add up number of image frames first and start the time t that timing is experienced after resetting each time, if
More than prefixed time interval Δ t, if it does not, perform step G, if it is, accumulative is met moving region
Area more than predetermined movement region area number of image frames reset, perform step D；Wherein, described tired
Image in meter number of image frames meets the area area more than predetermined movement region of moving region；
G, the accumulative area number of image frames more than the area in predetermined movement region meeting moving region；
Whether the number of image frames that H, judgement add up is more than pre-set image frame number, if it is, perform step I, no
Then perform step D.
I, determine and there is currently moving target.
The moving object detection device that the embodiment of the present invention provides, including:
Foreground extraction unit, for extracting the prospect of current frame image according to current background；
Pixels statistics unit, for adding up the number of pixels of described prospect；
First judging unit is the biggest for judging the number of pixels of prospect that described pixels statistics unit adds up
In presetted pixel number；
Moving region computing unit, for calculating the current frame image face relative to the moving region of previous frame image
Long-pending, wherein, described previous frame image meets: the number of pixels of the prospect of previous frame image is preset more than described
Number of pixels；
Second judging unit, for judging the moving region of the calculated present frame of moving region computing unit
Whether area is more than predetermined movement region area；
3rd judging unit, for judging that adding up number of image frames first after resetting each time starts timing institute warp
The time t gone through, if more than prefixed time interval Δ t；Wherein, the image in described accumulative number of image frames is full
The area of foot moving region is more than the area in predetermined movement region；
Number of image frames resets unit, for when the 3rd judging unit judges t more than Δ t, is met by accumulative
The area of moving region resets more than the number of image frames of the area in predetermined movement region；
Record unit, for when the 3rd judging unit judges described t less than Δ t, adds up to meet moving region
Area more than the number of image frames of area in predetermined movement region；
4th judging unit, for judging that whether the number of image frames of recording unit records is more than pre-set image frame
Number；
Moving target determines unit, for judging that when the 4th judging unit the number of image frames of record is more than presetting figure
During as frame number, determine and there is currently moving target；
Context update unit, for being updated to current background by current frame image.
The embodiment of the present invention, extracts the prospect of current frame image according to current background；The pixel of judgement prospect
Whether number is more than presetted pixel number, if it is, judge that the area of the moving region of current frame image is the biggest
In the area in predetermined movement region, if it is, record meets the area of moving region more than predetermined movement region
The number of image frames of area；Judge the area meeting moving region in the Δ t area more than predetermined movement region
Number of image frames whether more than pre-set image frame number, if it is, determine and there is currently moving target, above-mentioned
All judged results be no in the case of update background, using next frame image as current frame image continue above-mentioned
Step, wherein, described Δ t is by starting the time that timing is experienced from adding up image first.The present invention is without right
Each two field picture carries out foreground extraction accurately, and without determining the area of the moving region of each two field picture
Save the expense of calculating, improve the detection efficiency to moving target, owing to the present invention need not determine accurately
Prospect coordinate, when light changes greatly, remain able to detect target image accurately, improve detection
Target image and accuracy rate.
Accompanying drawing explanation
The moving target detecting method overall procedure schematic diagram that Fig. 1 provides for the embodiment of the present invention；
The background modeling method schematic flow sheet that Fig. 2 provides for the embodiment of the present invention；
The foreground image acquisition methods schematic flow sheet that Fig. 3 provides for the embodiment of the present invention；
The moving target detecting method idiographic flow schematic diagram that Fig. 4 provides for the embodiment of the present invention；
The moving object detection apparatus structure schematic diagram that Fig. 5 provides for the embodiment of the present invention；
The moving object detection apparatus structure schematic diagram that Fig. 6 provides for the embodiment of the present invention.
Detailed description of the invention
The embodiment of the present invention provides a kind of moving target detecting method and device, in order to improve the inspection of moving target
Survey efficiency and accuracy rate.
Because foreground target generally is all bigger or more significant target, the embodiment of the present invention provides
Moving target detecting method, does not extract prospect accurately to each frame video image, only whether need to judge it
There is the prospect of large area, if there is bigger foreground area, then determine whether significant moving target,
Otherwise update background to continue to judge whether next frame video image has the prospect of large area.If had significantly
Moving target then exports result, otherwise updates background, continues to judge next frame video image whether You compare great district
The prospect in territory.
The present invention judges according only to the size of foreground area, is not tracked each two field picture, carries
The high detection efficiency of moving target, and the present invention need not extract prospect accurately, when in illumination variation
Time under bigger scene, moving target can be detected, improve the accuracy rate to moving object detection.
The technical scheme that the embodiment of the present invention provides is illustrated below by accompanying drawing.
See Fig. 1, the moving target detecting method that the embodiment of the present invention provides, comprise the following steps:
S10, according to current background extract current frame image prospect, perform step S11.
S11, add up the number of pixels of described prospect.
S12, judge the number of pixels of described prospect whether more than presetted pixel number, if it is, perform step
Rapid S13, otherwise performs step S14；
Presetted pixel number can determine according to image size, typically takes image size (i.e. figure image width * image
High) the number of pixels corresponding to 20%.
S13, calculating current frame image, relative to the area of the moving region of previous frame image, perform step 15；
Wherein, described previous frame image meets: the number of pixels of prospect is more than presetted pixel number；
Described current frame image relative to the calculating of the area of the moving region of previous frame image, particularly as follows:
Calculate current frame image and the optical flow field of previous frame image, obtain the face of the moving region of current frame image
Long-pending, perform step S15；Wherein, described previous frame image meets: the number of pixels of prospect is more than described pre-
If number of pixels.
S14, current frame image is updated to current background, using next frame image as present frame, performs step
Rapid S10.
S15, judge the area of moving region of current frame image whether more than the area in predetermined movement region,
If it is, perform step S16, otherwise perform step S14.
S16, judge to add up number of image frames first and start the time t that timing is experienced after resetting each time, be
No more than prefixed time interval Δ t, if it does not, perform step S17, if it is, perform step S20, its
In, the image in described accumulative number of image frames meets the area face more than predetermined movement region of moving region
Long-pending.
Described predetermined movement region determines according to image size, typically takes image size (i.e. figure image width * image
High) 10%.
S17, the accumulative area number of image frames more than the area in predetermined movement region meeting moving region.
Whether the number of image frames that S18, judgement add up is more than pre-set image frame number, if it is, perform step S20,
Otherwise perform step S14.
S19, determine and there is currently moving target.
After performing step S19, the method also comprises determining that described moving target leaves.
Specifically, perform step A and B successively, and meet number of pixels in step B judged result and be less than
After presetted pixel number, also include: judge after determining and there is currently moving target, meet current
The number of pixels of two field picture occurs, if it is, determine first less than the judged result of presetted pixel number
Described moving target leaves, and the accumulative area meeting moving region is more than the area in predetermined movement region
The frame number of picture frame resets, and continues executing with step D, starts to detect next moving target；Otherwise continue to depend on
Secondary execution step A and B.
It should be noted that before determining that described moving target leaves, it is not necessary to update background, it is only necessary to more
According to the number of pixels of prospect, new prospect, judges whether moving target leaves.
It is preferred that after determining and there is currently moving target or determine that moving target leaves, perform warning,
Will moving target output.
Time used by the process of each two field picture is not more than the update cycle of current frame image, it is ensured that
Process to each two field picture is all the process to current frame image, when next two field picture arrives, and former frame
Image has been disposed.It is therefore not necessary to next frame picture delay is processed.
Described pre-set image number can set according to practical situation, and amount of calculation is less and detection is transported to choose one
The accurate empirical value of moving-target.When pre-set image number is bigger, calculative to meet moving region big
More in the image of predeterminable area, the False Rate of the moving target obtained is relatively low, but amount of calculation is relatively big, because of
This selects a suitable empirical value relatively reasonable.
Because moving target is a continuous print motor process, when there is no moving target in Preset Time Δ t
Time, there is also some erroneous judgements, as there is the figure meeting moving region area more than the area in predetermined movement region
During picture, need the number of image frames of record to reset, not affect the accurate judgement to subsequent motion target detection.
It is preferred that described image is video image, and described video image can be moving object detection device
Obtain the real time data gathered from video capture device, it is also possible to be moving object detection device in advance
The video flowing preserved.
Before performing step S10, the method also includes: carry out background modeling according to the image of default frame number,
The background obtained is described current background.
Seeing Fig. 2, carry out background modeling according to the image of default frame number, the background obtained is the described current back of the body
The detailed process of scape comprises the following steps:
S21, the gray scale summation of two field picture some to continuous print are averaged, and obtain cumulative mean frame.
It is preferred that the span of the some two field pictures of described continuous print can be 5 ~ 50 frames.
Such as, it is preferred that 25 frames Y under YUV color space before in video image can be divided
Amount summation is averaged, and obtains cumulative mean frame (or also referred to as average frame).
S22, obtained cumulative mean frame is carried out histogram specification process obtain background image.
Described histogram specification is to remove the pending image of regulationization by a default rectangular histogram so that image
Histogram distribution close to preset rectangular histogram.
Seeing Fig. 3, in step S10, the extracting method of the prospect of current frame image comprises the steps:
S31, current frame image is carried out histogram specification process.
S32, by the background image obtained in Fig. 2 and through histogram specification process after current frame image enter
The absolute Difference Calculation of row, obtains foreground image.
Described absolute difference, the gray scale of i.e. corresponding to two width images pixel carries out subtraction and then takes definitely
Value.The present invention is by pixel corresponding with the current frame image after histogram specification processes for background image
Gray scale carries out subtraction and then takes absolute value.
The present invention is to process by the way of histogram specification to the modeling process of background, and algorithm is transported
Calculation amount is little, when the moving region judging current frame image is less than predetermined movement region, it is not necessary to present frame
Image carries out relatively time-consuming optical flow computation, and method is easily met the requirement of real-time.
The above-mentioned acquisition mode to foreground image is only the size in the region of foreground image, does not obtain prospect
The concrete coordinate figure of image.Eliminate substantial amounts of amount of calculation, improve the efficiency of detection moving target.
It is described below in Fig. 1 and to calculate the current frame image motion relative to previous frame image described in step S13
The process of the area in region.
Specifically, before determining that prior image frame is relative by calculating current frame image with the optical flow field of previous frame image
The area of the moving region of one two field picture.
First of all for reducing the noise of video image impact on subsequent optical Flow Field Calculation, to present frame and former frame
The image of optical flow field to be calculated carries out Gaussian smoothing.
Light stream reflects the image change caused in a certain time interval due to motion.Due in the short time
The interior viewed brightness of any object point is invariable, the formula of image overall light stream such as (1) formula,
Represent (x, y) pixel of position moved to (x+dx, y+dy) within the dt time:
F (x+dx, y+dy, t+dt)=f (x, y, t) (1)
If dynamic image to be expressed as the function of room and time, and carried out Taylor expansion and can be obtained
Arrive, wherein
Expression higher order term:
Dx in most instances, dy, dt are the least, and higher order term is negligible, then comprehensive (1) and (2)
Two formulas can obtain (3) formula, and wherein u, v are image light flow field to be asked, and illustrates the movement velocity of object:
The present invention uses the method (Two-Frame based on polygon extension that Gunnar Farneback proposes
Motion Estimation Based on Polynomial Expansion, in Proceedings of the 13th
Scandinavian Conference on Image Analysis,LNCS2749,(Gothenburg,Sweden),
Pp.363 370, June-July 2003), and the optimization design realizing having carried out floating number fixed point to it, make
Obtaining its computing and be not related to floating number, be more easy to be transplanted on embedded device, the speed of service is the most faster.
The bigger prospect that one big effect of calculating optical flow field can cause light sudden change, scene changes etc. exactly is by mistake
Inspection is got rid of, and increases the practicality of method.Transport owing to the present invention uses the method for optical flow field to be concerned with significantly
Moving-target, is separated by some frames between two two field pictures of acquired calculating optical flow field, is so easier to highlight fortune
The motion amplitude of moving-target, it is easier to moving target detected, and the amount of calculation of method is declined, more
Readily satisfy the requirement of real-time.
The operation time of the most various Processing Algorithm the most all sizes to handled image are directly proportional, in order to subtract
The little operand to image improves the detection efficiency to moving target, it is preferred that carrying on the back present image
Before scape modeling, also include: current frame image is carried out compression of images or intercepting is allowed to image size in advance
In given image magnitude range.
Described image magnitude range can use QCIF(Quarter Common Intermediate Format)
Form.So-called QCIF form is conventional standard image format.In H.323 protocol family, it is stipulated that
The standard acquisition resolution of video capture device, QCIF=176 × 144 pixel.
Technical scheme that the embodiment of the present invention provide is described more detail below.
See Fig. 4, the moving target detecting method that the embodiment of the present invention provides, specifically include following steps:
S101, some two field pictures of the default frame number of acquisition, such as front 25 frames.
S102, respectively according to pre-set image size described front 25 two field pictures are compressed or intercept.
S103, the Y-component (gray scale) under the YUV color space of described front 25 two field pictures is averaged
Value, obtains cumulative mean frame, obtained cumulative mean frame carries out histogram specification process and obtains background
Image, described cumulative mean frame namely average frame.
When the 26th two field picture arrives, the 26th two field picture is current frame image, is presented herein below the 26th
The processing procedure of two field picture.
S104, according to pre-set image size current frame image it is compressed or intercepts.
S105, extract compression according to described background image or intercept after the prospect of current frame image.
S106, add up the number of pixels of the prospect of described current frame image.
S107, judge the number of pixels of described prospect whether more than presetted pixel number, if it is, perform step
Rapid S108, otherwise, performs step S109.
S108, calculating previous frame image (the 25th two field picture) and current frame image (the 26th two field picture)
Optical flow field, obtains the current frame image moving region relative to previous frame image.
S109, judge that the number of pixels meeting current frame image is little after determining and there is currently moving target
Judged result in presetted pixel number occurs first, if it is, perform step S115, otherwise performs
Step S104.
S110, judge the area of moving region of current frame image whether more than predetermined movement region area, as
Fruit is carried out step S112, otherwise performs step S116.
S111, judge to add up number of image frames first and start the time t that timing is experienced after resetting each time,
Whether more than prefixed time interval Δ t, if NO, perform step S112, the most then perform step
S116。
S112, the accumulative area number of image frames more than the area in predetermined movement region meeting moving region.
Whether the number of image frames that S113, judgement add up is more than pre-set image frame number, if it is, perform step
S114, otherwise performs step S116.
S114, determine and there is currently moving target, perform step S104.
S115, determine that the moving target occurred last time leaves, perform step S117.
S116, current frame image is updated to current background, next frame image is performed as current frame image
Step S104.
S117, meet moving region area reset accumulative more than the picture frame of predetermined movement region area,
Perform step S116.
Moving object detection device that the embodiment of the present invention provide is described below, sees Fig. 5, including:
Foreground extraction unit 11, for extracting the prospect of current frame image according to current background；
Pixels statistics unit 12, for adding up the number of pixels of described prospect；
Whether first judging unit 13, for judging the number of pixels of prospect that pixels statistics unit 12 adds up
More than presetted pixel number；
Moving region computing unit 14, for calculating the current frame image moving region relative to previous frame image
Area, wherein, described previous frame image meets: the number of pixels of the prospect of previous frame image is more than described pre-
If number of pixels；
Second judging unit 15, for judging the motor region of the calculated present frame of moving region computing unit
Whether the area in territory is more than the area in predetermined movement region；
3rd judging unit 16, for judging that the area from adding up to meet moving region first is more than predetermined movement
The number of image frames of the area in region starts the time t that timing is experienced, if more than prefixed time interval Δ t；
Number of image frames resets unit 17, for when the 3rd judging unit judges t more than Δ t, being expired by accumulative
The area of foot moving region resets more than the number of image frames of the area in predetermined movement region；
Record unit 18, for when the 3rd judging unit judges described t less than Δ t, adds up to meet motor region
The area in territory is more than the number of image frames of the area in predetermined movement region；
4th judging unit 19, for judging that whether the number of image frames of recording unit records is more than pre-set image frame
Number；
Moving target determines unit 20, for judging that when the 4th judging unit the number of image frames of record is more than presetting
During number of image frames, determine and there is currently moving target；
Context update unit 21, for being updated to current background by current frame image.
It is preferred that see Fig. 6, described device also includes:
5th judging unit 22, for judging after determining and there is currently moving target, meets present frame figure
The number of pixels of picture occurs first less than the judged result of presetted pixel number；
Moving target leaves and determines unit 23, for judging the picture of current frame image at the 5th judging unit 22
When element number occurs first less than the judged result of presetted pixel number, it is determined that described moving target leaves；
Number of image frames resets unit 17 and is additionally operable to, and after determining that moving target leaves, is met by accumulative
The area of moving region resets more than the frame number of the picture frame of the area in predetermined movement region.
It is preferred that foreground extraction unit 11 specifically for: current frame image is carried out at histogram specification
Reason；Current frame image after histogram specification processes and described current background are carried out absolute difference meter
Calculate, obtain the prospect of current frame image.
Seeing Fig. 6, this device also includes:
Background modeling unit 24, for carrying out background modeling according to the image presetting frame number, the background obtained is
Described current background；Specifically for, the gray scale summation to the most front some two field pictures is averaged, and is put down
All frame, carries out histogram specification process to described average frame, obtains current background.
Seeing Fig. 6, this device also includes:
Image editing unit 25, for according to the size of pre-set image current frame image carried out compression of images or
Intercept.
In sum, the embodiment of the present invention extracts the prospect of current frame image according to current background；Judgement prospect
Number of pixels whether more than presetted pixel number, if it is, judge the face of the moving region of current frame image
The long-pending area whether being more than predetermined movement region, if it is, record meets the area of moving region more than presetting
The number of image frames of the area of moving region；Judge that the area meeting moving region in Δ t is more than predetermined movement district
Whether the number of image frames of the area in territory is more than pre-set image frame number, if it is, determine and there is currently moving target,
Above-mentioned renewal background in the case of all judged results are no, continues next frame image as current frame image
Continuous above-mentioned steps, wherein, described Δ t is by starting, from first record image, the time that timing is experienced.The present invention
Without each two field picture being carried out foreground extraction accurately, and without determining the moving region of each two field picture
Area save the expense of calculating, improve the detection efficiency to moving target, owing to the present invention need not really
Fixed prospect coordinate accurately, when light changes greatly, remains able to detect target image accurately, carries
High detection target image and accuracy rate.
Those skilled in the art it should be appreciated that embodiments of the invention can be provided as method, system or
Computer program.Therefore, the present invention can use complete hardware embodiment, complete software implementation or
Person combines the form of the embodiment in terms of software and hardware.And, the present invention can use at one or more
The computer-usable storage medium wherein including computer usable program code (includes but not limited to that disk is deposited
Reservoir and optical memory etc.) form of the upper computer program implemented.
The present invention is with reference to method, equipment (system) and computer program product according to embodiments of the present invention
The flow chart of product and/or block diagram describe.It should be understood that can be by computer program instructions flowchart
And/or in each flow process in block diagram and/or square frame and flow chart and/or block diagram
Flow process and/or the combination of square frame.These computer program instructions can be provided to general purpose computer, special meter
The processor of calculation machine, Embedded Processor or other programmable data processing device to produce a machine,
Make the instruction performed by the processor of computer or other programmable data processing device produce for
Realize at one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame
The device of the middle function specified.
These computer program instructions may be alternatively stored in and computer or other programmable datas can be guided to process
In the computer-readable memory that equipment works in a specific way so that be stored in this computer-readable memory
In instruction produce and include the manufacture of command device, this command device realize in one flow process of flow chart or
The function specified in multiple flow processs and/or one square frame of block diagram or multiple square frame.
These computer program instructions also can be loaded in computer or other programmable data processing device,
Make on computer or other programmable devices, perform sequence of operations step to realize to produce computer
Process, thus on computer or other programmable devices perform instruction provide for realize in flow process
The merit specified in one flow process of figure or multiple flow process and/or one square frame of block diagram or multiple square frame
The step of energy.
Obviously, those skilled in the art can carry out various change and modification without deviating from this to the present invention
Bright spirit and scope.So, if the present invention these amendment and modification belong to the claims in the present invention and
Within the scope of its equivalent technologies, then the present invention is also intended to comprise these change and modification.
Claims (10)
1. a moving target detecting method, it is characterised in that the method comprises the following steps:
A, according to current background extract current frame image prospect；
B, add up the number of pixels of described prospect, it is judged that described number of pixels whether more than presetted pixel number,
If it is, perform step C, otherwise perform step D；
C, calculating current frame image, relative to the area of the moving region of previous frame image, perform step E；Its
In, described previous frame image meets: the number of pixels of prospect is more than presetted pixel number；
D, current frame image is updated to current background, next frame image is performed step as current frame image
Rapid A；
E, judge the area of moving region of current frame image whether more than the area in predetermined movement region, as
Fruit is, performs step F, otherwise performs step D；
F, judge to add up first after number of image frames each time resets time number of image frames starts that timing experienced
Between t, if more than prefixed time interval Δ t, if it does not, perform step G, if it is, accumulative is expired
The area of foot moving region resets more than the number of image frames of the area in predetermined movement region, performs step D；Its
In, the image in described accumulative number of image frames meets the area face more than predetermined movement region of moving region
Long-pending；
G, the accumulative area number of image frames more than the area in predetermined movement region meeting moving region；
Whether the number of image frames that H, judgement add up is more than pre-set image frame number, if it is, perform step I, no
Then perform step D；
I, determine and there is currently moving target.
Method the most according to claim 1, it is characterised in that after performing step I, the party
Method also includes:
Perform step A and B successively, and meet number of pixels less than presetted pixel in step B judged result
After number, also include: judge, after determining and there is currently moving target, to meet the picture of current frame image
Element number occurs, if it is, determine described motion mesh first less than the judged result of presetted pixel number
Mark leaves, and the accumulative area meeting moving region is more than the frame of the picture frame of the area in predetermined movement region
Number resets, and continues executing with step D；Otherwise continue to perform step A and B successively.
Method the most according to claim 1, it is characterised in that step A is extracted according to current background
The prospect of current frame image, particularly as follows:
Current frame image is carried out histogram specification process；
Current frame image after histogram specification processes and described current background are carried out absolute difference meter
Calculate, obtain the prospect of current frame image.
Method the most according to claim 1, it is characterised in that before performing step A, the party
Method also includes: carry out background modeling according to the image of default frame number, and the background obtained is described current background；
Particularly as follows: the gray scale summation to front some two field pictures is averaged, obtain average frame, described average frame is entered
Column hisgram specification processing, obtains current background.
Method the most according to claim 1, it is characterised in that in step, extracts present frame
Before the prospect of image, also include: according to the size of pre-set image current frame image carried out compression of images or
Intercept.
6. a moving object detection device, it is characterised in that this device includes:
Foreground extraction unit, for extracting the prospect of current frame image according to current background；
Pixels statistics unit, for adding up the number of pixels of described prospect；
First judging unit is the biggest for judging the number of pixels of prospect that described pixels statistics unit adds up
In presetted pixel number；
Moving region computing unit, for calculating the current frame image face relative to the moving region of previous frame image
Long-pending, wherein, described previous frame image meets: the number of pixels of the prospect of previous frame image is preset more than described
Number of pixels；
Second judging unit, for judging the moving region of the calculated present frame of moving region computing unit
Whether area is more than predetermined movement region area；
3rd judging unit, for judging that the area from adding up to meet moving region first is more than predetermined movement district
The number of image frames of the area in territory starts the time t that timing is experienced, if more than prefixed time interval Δ t；
Number of image frames resets unit, for when the 3rd judging unit judges t more than Δ t, is met by accumulative
The area of moving region resets more than the number of image frames of the area in predetermined movement region；
Record unit, for when the 3rd judging unit judges described t less than Δ t, adds up to meet moving region
Area more than the number of image frames of area in predetermined movement region；
4th judging unit, for judging that whether the number of image frames of recording unit records is more than pre-set image frame
Number；
Moving target determines unit, for judging that when the 4th judging unit the number of image frames of record is more than presetting figure
During as frame number, determine and there is currently moving target；
Context update unit, for being updated to current background by current frame image.
Device the most according to claim 6, it is characterised in that also include:
5th judging unit, for judging after determining and there is currently moving target, meets current frame image
Number of pixels occur first less than the judged result of presetted pixel number；
Moving target leaves and determines unit, for judging the number of pixels of current frame image at the 5th judging unit
When occurring first less than the judged result of presetted pixel number, it is determined that described moving target leaves；
Described number of image frames resets unit, is additionally operable to, after determining that described moving target leaves, be expired by accumulative
The area of foot moving region resets more than the frame number of the picture frame of the area in predetermined movement region.
Device the most according to claim 6, it is characterised in that described foreground extraction unit is specifically used
In: current frame image is carried out histogram specification process；To the present frame after histogram specification processes
Image and described current background carry out absolute Difference Calculation, obtain the prospect of current frame image.
Device the most according to claim 6, it is characterised in that this device also includes:
Background modeling unit, for carrying out background modeling according to the image presetting frame number, the background obtained is institute
State current background；Specifically for, the gray scale summation to the most front some two field pictures is averaged, and obtains average
Frame, carries out histogram specification process to described average frame, obtains current background.
Device the most according to claim 6, it is characterised in that this device also includes:
Image editing unit, is used for the size according to pre-set image and current frame image carries out compression of images or cuts
Take.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210344459.4A CN103679745B (en) | 2012-09-17 | 2012-09-17 | A kind of moving target detecting method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210344459.4A CN103679745B (en) | 2012-09-17 | 2012-09-17 | A kind of moving target detecting method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103679745A CN103679745A (en) | 2014-03-26 |
| CN103679745B true CN103679745B (en) | 2016-08-17 |
Family
ID=50317195
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210344459.4A Active CN103679745B (en) | 2012-09-17 | 2012-09-17 | A kind of moving target detecting method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103679745B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106710348A (en) * | 2016-12-20 | 2017-05-24 | 江苏前景信息科技有限公司 | Civil air defense interactive experience method and system |
| CN109255360B (en) * | 2017-07-12 | 2021-04-02 | 杭州海康威视数字技术股份有限公司 | Target classification method, device and system |
| CN107566723B (en) * | 2017-09-13 | 2019-11-19 | 维沃移动通信有限公司 | A kind of image pickup method, mobile terminal and computer readable storage medium |
| CN109379594B (en) * | 2018-10-31 | 2022-07-19 | 北京佳讯飞鸿电气股份有限公司 | Video coding compression method, device, equipment and medium |
| CN111832357A (en) * | 2019-04-19 | 2020-10-27 | 苏州涟漪信息科技有限公司 | Mobile event detection method and device |
| CN111191556A (en) * | 2019-12-25 | 2020-05-22 | 杭州宇泛智能科技有限公司 | Face recognition method and device and electronic equipment |
| CN111308969B (en) * | 2020-01-16 | 2021-04-06 | 浙江大学 | Carrier motion mode discrimination method based on time domain differential characteristics |
| CN111310733B (en) * | 2020-03-19 | 2023-08-22 | 成都云盯科技有限公司 | Personnel access detection method, device and equipment based on monitoring video |
| CN112037266B (en) * | 2020-11-05 | 2021-02-05 | 北京软通智慧城市科技有限公司 | Falling object identification method and device, terminal equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1449186A (en) * | 2003-04-03 | 2003-10-15 | 上海交通大学 | Abnormal object automatic finding and tracking video camera system |
| EP1667080A1 (en) * | 2001-10-17 | 2006-06-07 | Biodentity Systems Corporation | Face imaging system for recordal and automated identity confirmation |
| CN101261681A (en) * | 2008-03-31 | 2008-09-10 | 北京中星微电子有限公司 | Road image extraction method and device in intelligent video monitoring |
| CN102222349A (en) * | 2011-07-04 | 2011-10-19 | 江苏大学 | Prospect frame detecting method based on edge model |
| WO2012019417A1 (en) * | 2010-08-10 | 2012-02-16 | 中国科学院自动化研究所 | Device, system and method for online video condensation |
-
2012
- 2012-09-17 CN CN201210344459.4A patent/CN103679745B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1667080A1 (en) * | 2001-10-17 | 2006-06-07 | Biodentity Systems Corporation | Face imaging system for recordal and automated identity confirmation |
| CN1449186A (en) * | 2003-04-03 | 2003-10-15 | 上海交通大学 | Abnormal object automatic finding and tracking video camera system |
| CN101261681A (en) * | 2008-03-31 | 2008-09-10 | 北京中星微电子有限公司 | Road image extraction method and device in intelligent video monitoring |
| WO2012019417A1 (en) * | 2010-08-10 | 2012-02-16 | 中国科学院自动化研究所 | Device, system and method for online video condensation |
| CN102222349A (en) * | 2011-07-04 | 2011-10-19 | 江苏大学 | Prospect frame detecting method based on edge model |
Non-Patent Citations (4)
| Title |
|---|
| Data fusion for ground moving target tracking;Jost Koller，Martin Ulmke;《Aerospace Science and Technology》;20070223;第261-270页 * |
| 一种应用于嵌入式图像监控系统的运动检测方法;张宇 等;《2009年中国高校通信类院系学术研讨会论文集》;20090918;第295页2.2节第1段、296页第6段、第297页3.2节第1段 * |
| 基于自适应背景的实时运动物体检测;潘石柱 等;《计算机应用》;20041031;第24卷(第10期);第94-96页 * |
| 运动目标检测与跟踪系统设计;王笑雨;《中国优秀硕士学位论文全文数据库 信息科技辑》;20100615;正文第12页第1段、第13页2.2节第1段、第14页第2段 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103679745A (en) | 2014-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103679745B (en) | A kind of moving target detecting method and device | |
| CN106096577B (en) | A kind of target tracking method in camera distribution map | |
| CN108257146B (en) | Motion trail display method and device | |
| CN102542289B (en) | Pedestrian volume statistical method based on plurality of Gaussian counting models | |
| KR102153591B1 (en) | Method and apparatus for detecting garbage dumping action in real time on video surveillance system | |
| EP2858008B1 (en) | Target detecting method and system | |
| CN108665476B (en) | Pedestrian tracking method and electronic equipment | |
| CN103604371B (en) | Mobile terminal and object measurement method thereof | |
| WO2016149938A1 (en) | Video monitoring method, video monitoring system and computer program product | |
| CN111951313B (en) | Image registration method, device, equipment and medium | |
| JP2015062121A5 (en) | ||
| CN103678299A (en) | Method and device for monitoring video abstract | |
| CN102982537B (en) | A kind of method and system detecting scene change | |
| CN105678288A (en) | Target tracking method and device | |
| KR101348680B1 (en) | Target acquisition method for video tracker, and target acquisition apparatus using the same | |
| CN103745483A (en) | Mobile-target position automatic detection method based on stadium match video images | |
| CN111723656B (en) | Smog detection method and device based on YOLO v3 and self-optimization | |
| CN103996203A (en) | Method and device for detecting whether face in image is sheltered | |
| CN102222349B (en) | Prospect frame detecting method based on edge model | |
| CN111291587A (en) | Pedestrian detection method based on dense crowd, storage medium and processor | |
| CN111353496B (en) | Real-time detection method for infrared dim targets | |
| CN105141923A (en) | Method and device for video concentration | |
| CN113256683B (en) | Target tracking method and related equipment | |
| CN101916380B (en) | Video-based device and method for detecting smog | |
| CN101877135A (en) | Moving target detecting method based on background reconstruction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |