CN109478328A - Method for tracking target, device and image processing equipment - Google Patents

Method for tracking target, device and image processing equipment Download PDF

Info

Publication number
CN109478328A
CN109478328A CN201680087596.0A CN201680087596A CN109478328A CN 109478328 A CN109478328 A CN 109478328A CN 201680087596 A CN201680087596 A CN 201680087596A CN 109478328 A CN109478328 A CN 109478328A
Authority
CN
China
Prior art keywords
candidate blocks
target
reference block
feature vector
tracking
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201680087596.0A
Other languages
Chinese (zh)
Inventor
白向晖
伍健荣
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fujitsu Ltd
Original Assignee
Fujitsu Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Fujitsu Ltd filed Critical Fujitsu Ltd
Publication of CN109478328A publication Critical patent/CN109478328A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)

Abstract

A kind of method for tracking target, device and image processing equipment.The method for tracking target includes: multiple and different sizes that the detection target is determined according to scheduled zoom factor；For each current size, the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of the detection target are calculated, and determines the matching value and candidate blocks of the current size according to described eigenvector distance；According to the multiple various sizes of matching value and candidate blocks, tracking result of the detection target in the present frame is determined.Even if detection target significantly zooms in or out in picture frame as a result, the precision of target following will not be reduced.

Description

Method for tracking target, device and image processing equipment Technical field
The present embodiments relate to graphics technology field, in particular to a kind of method for tracking target, device and image processing equipment.
Background technique
In field of video monitoring, generally requires and detect interested target.Such as in the vehicle detection in parking lot, need to carry out real-time monitoring to the vehicle occurred in video.The target in video can be tracked at present；Wherein due to movement, the size in picture frame may change for moving object.
It should be noted that the above description of the technical background is intended merely to conveniently carry out clear, complete explanation to technical solution of the present invention, and facilitates the understanding of those skilled in the art and illustrate.It cannot be merely because these schemes be expounded in background technology part of the invention and think that above-mentioned technical proposal is known to those skilled in the art.
Summary of the invention
But, inventor's discovery: size the case where changing of detection target is not accounted in the scheme of target following at present, if detection target significantly zooms in or out in picture frame (such as size of the vehicle to come on highway towards monitoring camera in picture frame can become larger), the precision of target following can be reduced.
The embodiment of the present invention provides a kind of method for tracking target, device and image processing equipment, even if expecting that detection target significantly zooms in or out in picture frame, will not reduce the precision of target following.
First aspect according to an embodiment of the present invention provides a kind of method for tracking target, tracks to the detection target in video, and the method for tracking target includes:
Multiple and different sizes of the detection target are determined according to scheduled zoom factor；
For each current size, the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of the detection target are calculated, and determines the matching value and candidate blocks of the current size according to described eigenvector distance；
According to the multiple various sizes of matching value and candidate blocks, tracking result of the detection target in the present frame is determined.
The second aspect according to an embodiment of the present invention provides a kind of target tracker, to the detection target in video It is tracked, the target tracker includes:
Size determination unit determines multiple and different sizes of the detection target according to scheduled zoom factor；
Candidate determination unit, it is for each current size, the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of the detection target are calculated, and determines the matching value and candidate blocks of the current size according to described eigenvector distance；
As a result determination unit determines tracking result of the detection target in the present frame according to the multiple various sizes of matching value and candidate blocks.
In terms of third according to an embodiment of the present invention, a kind of image processing equipment is provided, wherein described image processing equipment includes foregoing target tracker.
Another aspect according to an embodiment of the present invention, a kind of computer-readable program is provided, wherein when executing described program in target tracker perhaps image processing equipment, described program makes the target tracker or image processing equipment execute method for tracking target as described above.
Another aspect according to an embodiment of the present invention provides a kind of storage medium for being stored with computer-readable program, wherein the computer-readable program makes target tracker or image processing equipment execute method for tracking target as described above.
The beneficial effect of the embodiment of the present invention is: multiple and different sizes being calculated separately with the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of detection target, and determines the matching value and candidate blocks of current size according to described eigenvector distance；According to the multiple various sizes of matching value and candidate blocks, the tracking result of the detection target in the current frame is determined.Even if detection target significantly zooms in or out in picture frame as a result, the precision of target following will not be reduced.
Referring to following description and accompanying drawings, only certain exemplary embodiments of this invention is disclosed in detail, specifying the principle of the present invention can be in a manner of adopted.It should be understood that embodiments of the present invention are not so limited in range.In the range of the spirit and terms of appended claims, embodiments of the present invention include many changes, modifications and are equal.
The feature for describing and/or showing for a kind of embodiment can be used in one or more other embodiments in a manner of same or similar, be combined with the feature in other embodiment, or the feature in substitution other embodiment.
It should be emphasized that term "comprises/comprising" refers to the presence of feature, one integral piece, step or component when using herein, but the presence or additional of one or more other features, one integral piece, step or component is not precluded.
Detailed description of the invention
Many aspects of the invention may be better understood referring to attached drawing below.Component in attached drawing is not proportional drafting, and is intended merely to show the principle of the present invention.For the ease of showing and describing some parts of the invention, corresponding part may be exaggerated or minimized in attached drawing.
It can be combined with elements and features shown in one or more other attached drawings or embodiment in elements and features described in one drawing or one embodiment of the invention.In addition, in the accompanying drawings, similar label indicates corresponding component in several attached drawings, and may be used to indicate corresponding component used in more than one embodiment.
Fig. 1 is a schematic diagram of the method for tracking target of the embodiment of the present invention 1；
Fig. 2 is the schematic diagram that matching value and candidate blocks are calculated for a certain size of the embodiment of the present invention 1；
Fig. 3 is a schematic diagram of the search window of the embodiment of the present invention 1；
Fig. 4 is a schematic diagram of the extraction candidate blocks of the embodiment of the present invention 1；
Fig. 5 is a schematic diagram of the target tracker of the embodiment of the present invention 2；
Fig. 6 is another schematic diagram of the target tracker of the embodiment of the present invention 2；
Fig. 7 is a schematic diagram of the candidate determination unit of the embodiment of the present invention 2；
Fig. 8 is a schematic diagram of the vector generation unit of the embodiment of the present invention 2；
Fig. 9 is a schematic diagram of the vector structural unit of the embodiment of the present invention 2；
Figure 10 is a schematic diagram of the result determination unit of the embodiment of the present invention 2；
Figure 11 is a schematic diagram of the image processing equipment of the embodiment of the present invention 3.
Specific embodiment
Referring to attached drawing, by following specification, aforementioned and other feature of the invention be will be apparent.In the specification and illustrated in the drawings, specifically disclose only certain exemplary embodiments of this invention, which show some embodiments that can wherein use principle of the invention, it will be appreciated that, the present invention is not limited to described embodiments, on the contrary, the present invention includes whole modifications, modification and the equivalent fallen within the scope of the appended claims.
Embodiment 1
The embodiment of the present invention provides a kind of method for tracking target, tracks to the detection target in video.
Fig. 1 is a schematic diagram of the method for tracking target of the embodiment of the present invention, as shown in Figure 1, the method for tracking target includes:
Step 101, multiple and different sizes of the detection target are determined according to scheduled zoom factor；
Step 102, for each current size, the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of the detection target are calculated, and determines the matching value and candidate blocks of the current size according to described eigenvector distance；
Step 103, according to the multiple various sizes of matching value and candidate blocks, tracking result of the detection target in the present frame is determined.
In the present embodiment, multiple and different sizes may include: normal size, minification and expansion size.For example, by normal size multiplied by the available minification of the scheduled diminution factor (such as 0.95), by normal size multiplied by scheduled amplification factor (such as 1.05) available up-sizing.However, the present invention is not limited thereto, can zoom factor determine according to actual needs, can additionally have multiple and different sizes.
Only by normal size, minification and expand size below and be illustrated for these three.
In the present embodiment, matching value and candidate blocks can be calculated for every kind of size, specifically how to calculate can be as described later.It then can matching value under more multiple and different sizes；And candidate blocks corresponding to smallest match value are determined as the tracking block of the detection target in the current frame.
The case where size that the embodiment of the present invention considers detection target as a result, changes, even if detection target significantly zooms in or out in picture frame, will not reduce the precision of target following.
Following elder generation is by taking normal size as an example, to how calculating matching value and candidate blocks schematically illustrate.
Fig. 2 is the schematic diagram that matching value and candidate blocks are calculated for a certain size of the embodiment of the present invention, and as described in Figure 2, which may include:
Step 201, the reference block of detection target is determined for current size；
In the present embodiment, the initial position (x, y, w, h) of target to be tracked can be obtained in the frame (such as former frame of present frame) of video；Wherein, x is the horizontal coordinate of the top left corner pixel of the frame comprising target, and y is the vertical coordinate of the top left corner pixel of the frame comprising target, and w is the width of the frame comprising target, and h is the height of the frame comprising target.In the frame of initial video, the image zooming-out where the frame comprising target can be come out and be used as reference block.
Step 202, level sampling interval and Vertical Sampling interval are determined according to the size of reference block.
In the present embodiment, a parameter min_dim can be set first, then obtain the short side SE=min (w, h) of tracking target (corresponding to reference block), and obtain the long side LE=max (w, h) of tracking target.If short side SE is less than min_dim, the sampling interval SDS=1 of short side is set, otherwise SDS=round (SE/min_dim) is set, wherein round () indicates the operation that rounds up.In addition, the sampling interval SDL=round (SDS*LE/SE) of setting long side.
It is then possible to which level sampling interval SDH=w < h of tracking target is arranged SDS:SDL, setting tracking mesh Target Vertical Sampling interval SDV=w > h SDS:SDL.
It is worth noting that, only diagrammatically illustrating level sampling interval SDH and Vertical Sampling interval SDV how is arranged above, however, the present invention is not limited thereto can also be configured using other modes.
Step 203, according to the position of the reference block, the level sampling interval and the Vertical Sampling interval, the search window in the present frame is set；
In the present embodiment, for present frame, search window can be set around the position where the tracking target (such as corresponding to the reference block) that former frame obtains.Such as, one parameter search_range can be set first, the horizontal extent of search window is from-SDH*search_range+x to SDH*search_range+x+w, and the vertical range of search window is from-SDV*search_range+y to SDV*search_range+y+h.
Fig. 3 is a schematic diagram of the search window of the embodiment of the present invention, as shown in figure 3, search window in the current frame can be arranged around reference block.
It is worth noting that, only diagrammatically illustrating search window how is arranged above, however, the present invention is not limited thereto can also be configured using other modes.By the way that search window is arranged, calculation amount can be reduced in the case where not reducing the accuracy of search.
Step 204, according to the size of the reference block, the level sampling interval and the Vertical Sampling interval, one or more candidate blocks are extracted from described search window；
In the present embodiment, it such as can successively be extracted in search window with the block of reference block same size as candidate blocks, level is divided into SDH between extracting, and is divided into SDV between vertically extracting.
Fig. 4 is a schematic diagram of the extraction candidate blocks of the embodiment of the present invention, as shown in figure 4, one or more candidate blocks can be extracted in the search window.It is worth noting that, only diagrammatically illustrating how to extract candidate blocks above, however, the present invention is not limited thereto can also be configured using other modes.By sampling to candidate blocks or reference block, calculation amount can be reduced in the case where not reducing the accuracy of search.
Step 205, feature vector is generated for each candidate blocks and the reference block；
In the present embodiment, specifically for some candidate blocks or reference block, the pixel in described piece can be sampled according to the level sampling interval and the Vertical Sampling interval；And the characteristic value of the multiple pixels obtained using sampling constructs described piece of feature vector.
Wherein, feature vector may include as follows: gray feature vector sum histograms of oriented gradients (HoG, Histogram of Oriented Gradient) feature vector, however, the present invention is not limited thereto, such as can also be using other feature vectors.
In the present embodiment, the gray value for multiple pixels that sampling obtains can be configured to gray feature vector；The HoG value for multiple pixels that sampling obtains is configured to HoG feature vector；Then HoG feature vector described in the gray feature vector sum is merged, obtains the feature vector of the candidate blocks or the reference block.
For example, can be sampled in the candidate blocks for a certain candidate blocks, level sampling interval is, for example, 2*SDH, and Vertical Sampling interval is, for example, 2*SDV.The gray value for multiple pixels that sampling obtains is arranged successively and constitutes gray feature vector.Then HoG feature is calculated to these pixels that sampling obtains, merges obtained HoG feature vector to obtain final block eigenvector with gray feature vector.
Further, it is also possible to assign weight coefficient respectively for gray feature vector sum HoG feature vector；And the gray feature vector sum HoG feature vector after imparting weight coefficient is merged.For example, gray feature vector obtains the weight of Wy in final feature vector, HoG feature vector obtains the weight of Wh.
For reference block, identical processing can be carried out.Thus a feature vector can be obtained for reference block, a feature vector can also be obtained for each candidate blocks.
Step 206, the vector distance between the feature vector of reference block and the feature vector of each candidate blocks is calculated.
In the present embodiment, how vector distance is calculated about between two feature vectors, the relevant technologies can be referred to.
Step 207, the smallest candidate blocks of vector distance between the feature vector of reference block are determined as to the tracking block of current size；And
Step 208, using the smallest vector distance as the matching value of current size.
For example, can be determined that the tracking result under current size with the position where that the smallest candidate blocks of the vector distance of reference block, the matching value under current size can be determined that with that the smallest vector distance of the vector distance of reference block.
It is worth noting that, attached drawing 2 is only symbolically illustrated the embodiment of the present invention, however, the present invention is not limited thereto.Such as the sequence that executes between each step can be suitably adjusted, other some steps can be additionally increased or reduces certain steps therein.Those skilled in the art can carry out suitably modification according to above content, be not limited solely to the record of above-mentioned attached drawing.
Thus, it is possible to calculate the matching value and candidate blocks under such as normal size.Above-mentioned steps 201 be may then pass through to step 208, calculate separately matching value and candidate blocks under minification and up-sizing.
Such as it can be by tracking target scale to a lesser size (such as 0.95 of normal size), the center of small size target is consistent with the center of normal size target, length and width are that the length and width of normal size target are multiplied by a coefficient (such as 0.95) less than 1 respectively, and reference block is also zoomed in same size.
Then step 201 is repeated to step 208, calculates matching value and tracking result of the target under small size.If matching value is less than the matching value under normal size, otherwise retain the matching value and tracking result under normal size with the matching value and tracking result obtained under matching value and tracking result substitution normal size under small size.
Again for example, target scale will be tracked to a biggish size (such as 1.05 of normal size), the center of large scale target is consistent with the center of normal size target, length and width are that the length and width of normal size target are multiplied by a coefficient (such as 1.05) greater than 1 respectively, and reference block is also zoomed in same size.
Then step 201 is repeated to step 208, calculates matching value and tracking result of the target under large scale.If matching value be less than before retain matching value, with obtained under large scale matching value and tracking result instead before matching value and tracking result, otherwise, matching value and tracking result before retaining.
It is worth noting that, above only by normal size, minification and expand size for these three and be illustrated.However, the present invention is not limited thereto can also be three kinds or more of size, etc.；It can according to need determining specific embodiment.
In the present embodiment, the reference block can also be updated using the information of the tracking block.
For example, the block where obtained tracking result to be zoomed to the size of reference block, then reference block is updated according to the information of reference block (such as gray value) and the information (such as gray value) of block where tracking result.It is possible to further which the result after being weighted and averaged is updated reference block.
For example, Ref=learing_rate*Ref+ (1-learing_rate) * Trk.Wherein, Ref indicates the information of reference block, and learing_rate indicates that weight coefficient, Trk indicate the information of tracking result place block.
As can be seen from the above embodiments, multiple and different sizes are calculated separately with the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of detection target, and determines the matching value and candidate blocks of current size according to described eigenvector distance；According to the multiple various sizes of matching value and candidate blocks, the tracking result of the detection target in the current frame is determined.Even if detection target significantly zooms in or out in picture frame as a result, the precision of target following will not be reduced.
Embodiment 2
The embodiment of the present invention provides a kind of target tracker, tracks to the detection target in video.The present embodiment 2 corresponds to the method for tracking target of embodiment 1, and identical content repeats no more.
Fig. 5 is a schematic diagram of the target tracker of the embodiment of the present invention, as shown in figure 5, target tracker 500 includes:
Size determination unit 501 determines multiple and different sizes of the detection target according to scheduled zoom factor；
Candidate determination unit 502, it is for each current size, the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of the detection target are calculated, and determines the matching value and candidate blocks of the current size according to described eigenvector distance；
As a result determination unit 503 determine tracking result of the detection target in the present frame according to the multiple various sizes of matching value and candidate blocks.
In the present embodiment, the multiple different sizes include: normal size, minification and expansion size；However, the present invention is not limited thereto.
Fig. 6 is another schematic diagram of the target tracker of the embodiment of the present invention, as shown in fig. 6, target tracker 600 includes: size determination unit 501, candidate determination unit 502 and result determination unit 503, and as described above.
As shown in fig. 6, target tracker 600 can also include:
Reference block determination unit 601 determines the current size reference block of the detection target；
It is spaced determination unit 602, level sampling interval and Vertical Sampling interval are determined according to the size of the reference block.
As shown in fig. 6, target tracker 600 can also include:
Reference block updating unit 603 is updated the reference block using the information of tracking block.
Fig. 7 is a schematic diagram of the candidate determination unit 502 of the embodiment of the present invention, as shown in fig. 7, candidate determination unit 502 may include:
The search window in the present frame is arranged according to the position of the reference block, the level sampling interval and the Vertical Sampling interval in window setting unit 701；
Candidate blocks extracting unit 702 extracts one or more candidate blocks according to the size of the reference block, the level sampling interval and the Vertical Sampling interval from described search window；
Vector generation unit 703 is that each candidate blocks and the reference block generate feature vector；And
Metrics calculation unit 704 calculates the vector distance between the feature vector of the reference block and the feature vector of each candidate blocks.
As shown in fig. 7, candidate determination unit 502 can also include:
The smallest candidate blocks of vector distance between the feature vector of the reference block are determined as the tracking block of the current size by candidate blocks determination unit 705；And
Matching value determination unit 706, using the smallest vector distance as the matching value of the current size.
Fig. 8 is a schematic diagram of the vector generation unit 703 of the embodiment of the present invention, as shown in figure 8, vector generates list First 703 may include:
Pixel sampling unit 801 samples the pixel in the candidate blocks or the reference block according to the level sampling interval and the Vertical Sampling interval；
The characteristic value of vector structural unit 802, the multiple pixels obtained using sampling constructs the feature vector of the candidate blocks or the reference block.
Fig. 9 is a schematic diagram of the vector structural unit 802 of the embodiment of the present invention, as shown in figure 9, vector structural unit 802 may include:
The gray value for multiple pixels that sampling obtains is configured to gray feature vector by primary vector structural unit 901；
The histograms of oriented gradients value for multiple pixels that sampling obtains is configured to histograms of oriented gradients feature vector by secondary vector structural unit 902；And
Vector combining unit 903 merges histograms of oriented gradients feature vector described in the gray feature vector sum, obtains the feature vector of the candidate blocks or the reference block.
As shown in figure 9, vector structural unit 802 can also include:
Weight given unit 904 assigns weight coefficient for histograms of oriented gradients feature vector described in the gray feature vector sum respectively；
And vector combining unit 903 is also used to: histograms of oriented gradients feature vector described in the gray feature vector sum after imparting weight coefficient is merged.
Figure 10 is a schematic diagram of the result determination unit 503 of the embodiment of the present invention, and as shown in Figure 10, as a result determination unit 503 may include:
Matching value comparing unit 1001, it is more the multiple difference sizes under the matching value；And
Block determination unit 1002 is tracked, candidate blocks corresponding to smallest match value are determined as tracking block of the detection target in the present frame.
It is worth noting that, only each component related to the present invention is illustrated above, however, the present invention is not limited thereto.Target tracker can also include that perhaps module can refer to the prior art about these components or the particular content of module to other component.
As can be seen from the above embodiments, multiple and different sizes are calculated separately with the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of detection target, and determines the matching value and candidate blocks of current size according to described eigenvector distance；According to the multiple various sizes of matching value and candidate blocks, the tracking result of the detection target in the current frame is determined.Even if as a result, detection target in picture frame significantly amplification or It reduces, the precision of target following will not be reduced.
Embodiment 3
The embodiment of the present invention provides a kind of image processing equipment, which includes target tracker as described in Example 2.
Figure 11 is a schematic diagram of the image processing equipment of the embodiment of the present invention.As shown in figure 11, image processing equipment 1100 may include: central processing unit (CPU) 100 and memory 110；Memory 110 is coupled to central processing unit 100.Wherein the memory 110 can store various data；The program of information processing is additionally stored, and executes the program under the control of central processing unit 100.
In one embodiment, the function of target tracker 500 or 600 can be integrated into central processing unit 100.Wherein, central processing unit 100, which can be configured as, realizes method for tracking target as described in Example 1.
In another embodiment, target tracker 500 or 600 can be with 100 separate configuration of central processing unit, such as target tracker 500 or 600 can be configured to the chip connecting with central processing unit 100, the function of target tracker 500 or 600 is realized by the control of central processing unit 100.
In the present embodiment, central processing unit 100, which can be configured as, carries out following control: multiple and different sizes of the detection target are determined according to scheduled zoom factor；For each current size, the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of the detection target are calculated, and determines the matching value and candidate blocks of the current size according to described eigenvector distance；According to the multiple various sizes of matching value and candidate blocks, tracking result of the detection target in the present frame is determined.
In addition, as shown in figure 11, image processing equipment 1100 can also include: input and output (I/O) equipment 120 and display 130 etc.；Wherein, similarly to the prior art, details are not described herein again for the function of above-mentioned component.It is worth noting that, image processing equipment 1100 is also not necessary to include all components shown in Figure 11；In addition, image processing equipment 1100 can also include the component being not shown in Figure 11, the prior art can be referred to.
The embodiment of the present invention provides a kind of computer-readable program, wherein described program makes the target tracker or image processing equipment execute method for tracking target as described in Example 1 when executing described program in target tracker or image processing equipment.
The embodiment of the present invention provides a kind of storage medium for being stored with computer-readable program, wherein the computer-readable program makes target tracker or image processing equipment execute method for tracking target as described in Example 1.
The device and method more than present invention can be by hardware realization, can also be by combination of hardware software realization.The present invention It is related to such computer-readable program, when the program is performed by logical block, the logical block can be made to realize devices described above or component parts, or the logical block is made to realize various method or steps described above.The invention further relates to the storage mediums for storing procedure above, such as hard disk, disk, CD, DVD, flash memory.
Hardware, the software module executed by processor or both combination can be embodied directly in conjunction with the method, device that the embodiment of the present invention describes.Such as, one or more of functional block diagram and/or functional block diagram shown in Fig. 5 one or more combinations (such as, size determination unit, candidate determination unit, result determination unit etc.), both it can correspond to each software module of computer program process, each hardware module can also be corresponded to.These software modules can correspond respectively to each step shown in FIG. 1.These software modules are for example solidified using field programmable gate array (FPGA) and are realized by these hardware modules.
Software module can be located at the storage medium of RAM memory, flash memory, ROM memory, eprom memory, eeprom memory, register, hard disk, mobile disk, CD-ROM or any other form known in the art.A kind of storage medium can be coupled to processor, to enable a processor to from the read information, and information can be written to the storage medium；Or the storage medium can be the component part of processor.Pocessor and storage media can be located in ASIC.The software module can store in a memory in the mobile terminal, also can store in the storage card that can be inserted into mobile terminal.For example, the software module is storable in the flash memory device of the MEGA-SIM card or large capacity if equipment (such as mobile terminal) is using the MEGA-SIM card of larger capacity or the flash memory device of large capacity.
For one or more combinations of one or more of function box described in attached drawing and/or function box, it can be implemented as general processor for executing function described herein, digital signal processor (DSP), specific integrated circuit (ASIC), field programmable gate array (FPGA) either other programmable logic device, discrete gate or transistor logic, discrete hardware components or it is any appropriately combined.One or more combinations of one or more of function box for attached drawing description and/or function box, it is also implemented as calculating the combination of equipment, for example, the combination of DSP and microprocessor, multi-microprocessor, the one or more microprocessors or any other this configuration combined with DSP communication.
Combining specific embodiment above, invention has been described, it will be appreciated by those skilled in the art that these descriptions are all exemplary, it is not limiting the scope of the invention.Those skilled in the art can make various variants and modifications to the present invention with spirit according to the present invention and principle, these variants and modifications are also within the scope of the invention.

Claims (20)

  1. A kind of method for tracking target, tracks the detection target in video, and the method for tracking target includes:
    Multiple and different sizes of the detection target are determined according to scheduled zoom factor；
    For each current size, the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of the detection target are calculated, and determines the matching value and candidate blocks of the current size according to described eigenvector distance；
    According to the multiple various sizes of matching value and candidate blocks, tracking result of the detection target in the present frame is determined.
  2. Method for tracking target according to claim 1, wherein the multiple difference size includes: normal size, minification and expansion size.
  3. Method for tracking target according to claim 1, wherein the method for tracking target further include:
    The reference block of the detection target is determined for the current size；
    Level sampling interval and Vertical Sampling interval are determined according to the size of the reference block.
  4. Method for tracking target according to claim 3, wherein calculate the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of the detection target, comprising:
    According to the position of the reference block, the level sampling interval and the Vertical Sampling interval, the search window in the present frame is set；
    According to the size of the reference block, the level sampling interval and the Vertical Sampling interval, one or more candidate blocks are extracted from described search window；
    Feature vector is generated for each candidate blocks and the reference block；And
    Calculate the vector distance between the feature vector of the reference block and the feature vector of each candidate blocks.
  5. Method for tracking target according to claim 4, wherein the matching value and candidate blocks of the current size are determined according to described eigenvector distance, comprising:
    The smallest candidate blocks of vector distance between the feature vector of the reference block are determined as to the tracking block of the current size；And
    Using the smallest vector distance as the matching value of the current size.
  6. Method for tracking target according to claim 4, wherein generate feature vector for each candidate blocks and the reference block, comprising:
    According to the level sampling interval and the Vertical Sampling interval, the pixel in the candidate blocks or the reference block is sampled；
    The characteristic value of the multiple pixels obtained using sampling constructs the feature vector of the candidate blocks or the reference block.
  7. Method for tracking target according to claim 6, wherein the characteristic value of the multiple pixels obtained using sampling constructs the feature vector of the candidate blocks or the reference block, comprising:
    The gray value for multiple pixels that sampling obtains is configured to gray feature vector；
    The histograms of oriented gradients value for multiple pixels that sampling obtains is configured to histograms of oriented gradients feature vector；
    Histograms of oriented gradients feature vector described in the gray feature vector sum is merged, the feature vector of the candidate blocks or the reference block is obtained.
  8. Method for tracking target according to claim 7, wherein the method for tracking target further include:
    Weight coefficient is assigned respectively for histograms of oriented gradients feature vector described in the gray feature vector sum；
    And histograms of oriented gradients feature vector described in the gray feature vector sum after imparting weight coefficient is merged.
  9. Method for tracking target according to claim 1, wherein tracking result of the detection target in the present frame is determined according to the multiple various sizes of matching value and candidate blocks, comprising:
    The matching value under more the multiple difference size；And
    Candidate blocks corresponding to smallest match value are determined as tracking block of the detection target in the present frame.
  10. Method for tracking target according to claim 9, wherein the method for tracking target further include:
    The reference block is updated using the information of the tracking block.
  11. A kind of target tracker, tracks the detection target in video, and the target tracker includes:
    Size determination unit determines multiple and different sizes of the detection target according to scheduled zoom factor；
    Candidate determination unit, it is for each current size, the reference block and the detection target feature vector distance between one or more candidate blocks in the current frame of the detection target are calculated, and determines the matching value and candidate blocks of the current size according to described eigenvector distance；
    As a result determination unit determines tracking result of the detection target in the present frame according to the multiple various sizes of matching value and candidate blocks.
  12. Target tracker according to claim 11, wherein the multiple difference size includes: normal size, minification and expansion size.
  13. Target tracker according to claim 11, wherein the target tracker further include:
    Reference block determination unit determines the current size reference block of the detection target；
    It is spaced determination unit, level sampling interval and Vertical Sampling interval are determined according to the size of the reference block.
  14. Target tracker according to claim 13, wherein it is described candidate determination unit include:
    The search window in the present frame is arranged according to the position of the reference block, the level sampling interval and the Vertical Sampling interval in window setting unit；
    Candidate blocks extracting unit extracts one or more candidate blocks according to the size of the reference block, the level sampling interval and the Vertical Sampling interval from described search window；
    Vector generation unit is that each candidate blocks and the reference block generate feature vector；And
    Metrics calculation unit calculates the vector distance between the feature vector of the reference block and the feature vector of each candidate blocks.
  15. Target tracker according to claim 14, wherein candidate's determination unit further include:
    The smallest candidate blocks of vector distance between the feature vector of the reference block are determined as the tracking block of the current size by candidate blocks determination unit；And
    Matching value determination unit, using the smallest vector distance as the matching value of the current size.
  16. Target tracker according to claim 14, wherein the vector generation unit includes:
    Pixel sampling unit samples the pixel in the candidate blocks or the reference block according to the level sampling interval and the Vertical Sampling interval；
    The characteristic value of vector structural unit, the multiple pixels obtained using sampling constructs the feature vector of the candidate blocks or the reference block.
  17. Target tracker according to claim 16, wherein the vector structural unit includes:
    The gray value for multiple pixels that sampling obtains is configured to gray feature vector by primary vector structural unit；
    The histograms of oriented gradients value for multiple pixels that sampling obtains is configured to histograms of oriented gradients feature vector by secondary vector structural unit；And
    Vector combining unit merges histograms of oriented gradients feature vector described in the gray feature vector sum, obtains the feature vector of the candidate blocks or the reference block.
  18. Target tracker according to claim 11, wherein the result determination unit includes:
    Matching value comparing unit, it is more the multiple difference sizes under the matching value；And
    Block determination unit is tracked, candidate blocks corresponding to smallest match value are determined as tracking block of the detection target in the present frame.
  19. Target tracker according to claim 18, wherein the target tracker further include:
    Reference block updating unit is updated the reference block using the information of the tracking block.
  20. A kind of image processing equipment, wherein described image processing equipment includes target tracker as claimed in claim 11.
CN201680087596.0A 2016-09-30 2016-09-30 Method for tracking target, device and image processing equipment Pending CN109478328A (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/101094 WO2018058531A1 (en) 2016-09-30 2016-09-30 Target tracking method and device, and image processing apparatus

Publications (1)

Publication Number Publication Date
CN109478328A true CN109478328A (en) 2019-03-15

Family

ID=61762478

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201680087596.0A Pending CN109478328A (en) 2016-09-30 2016-09-30 Method for tracking target, device and image processing equipment

Country Status (2)

Country Link
CN (1) CN109478328A (en)
WO (1) WO2018058531A1 (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20000054329A (en) * 2000-06-01 2000-09-05 이성환 Object tracking method in the moving picture data
JP2005346425A (en) * 2004-06-03 2005-12-15 Matsushita Electric Ind Co Ltd Automatic tracking system and automatic tracking method
CN1921628A (en) * 2005-08-23 2007-02-28 松下电器产业株式会社 Motion vector detection apparatus and motion vector detection method
CN101170683A (en) * 2006-10-27 2008-04-30 松下电工株式会社 Target moving object tracking device
CN102456225A (en) * 2010-10-22 2012-05-16 深圳中兴力维技术有限公司 Video monitoring system and moving target detecting and tracking method thereof
CN103049749A (en) * 2012-12-30 2013-04-17 信帧电子技术（北京）有限公司 Method for re-recognizing human body under grid shielding
CN103578116A (en) * 2012-07-23 2014-02-12 三星泰科威株式会社 Apparatus and method for tracking object
CN103996208A (en) * 2014-05-21 2014-08-20 国家电网公司 Method for conducting automatic tracking of PTZ single target in video image
CN104424638A (en) * 2013-08-27 2015-03-18 深圳市安芯数字发展有限公司 Target tracking method based on shielding situation

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101888479B (en) * 2009-05-14 2012-05-02 汉王科技股份有限公司 Method and device for detecting and tracking target image
CN101739692B (en) * 2009-12-29 2012-05-30 天津市亚安科技股份有限公司 Fast correlation tracking method for real-time video target
US8582821B1 (en) * 2011-05-23 2013-11-12 A9.Com, Inc. Tracking objects between images
JP5746937B2 (en) * 2011-09-01 2015-07-08 ルネサスエレクトロニクス株式会社 Object tracking device
CN103632382B (en) * 2013-12-19 2016-06-22 中国矿业大学(北京) A kind of real-time multiscale target tracking based on compressed sensing
CN105184779B (en) * 2015-08-26 2018-04-06 电子科技大学 One kind is based on the pyramidal vehicle multiscale tracing method of swift nature
CN105787964A (en) * 2016-02-29 2016-07-20 深圳电科技有限公司 Target tracking method and device
CN105913453A (en) * 2016-04-01 2016-08-31 海信集团有限公司 Target tracking method and target tracking device

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20000054329A (en) * 2000-06-01 2000-09-05 이성환 Object tracking method in the moving picture data
JP2005346425A (en) * 2004-06-03 2005-12-15 Matsushita Electric Ind Co Ltd Automatic tracking system and automatic tracking method
CN1921628A (en) * 2005-08-23 2007-02-28 松下电器产业株式会社 Motion vector detection apparatus and motion vector detection method
CN101170683A (en) * 2006-10-27 2008-04-30 松下电工株式会社 Target moving object tracking device
CN102456225A (en) * 2010-10-22 2012-05-16 深圳中兴力维技术有限公司 Video monitoring system and moving target detecting and tracking method thereof
CN103578116A (en) * 2012-07-23 2014-02-12 三星泰科威株式会社 Apparatus and method for tracking object
CN103049749A (en) * 2012-12-30 2013-04-17 信帧电子技术（北京）有限公司 Method for re-recognizing human body under grid shielding
CN104424638A (en) * 2013-08-27 2015-03-18 深圳市安芯数字发展有限公司 Target tracking method based on shielding situation
CN103996208A (en) * 2014-05-21 2014-08-20 国家电网公司 Method for conducting automatic tracking of PTZ single target in video image

Also Published As

Publication number Publication date
WO2018058531A1 (en) 2018-04-05

Similar Documents

Publication Publication Date Title
US11302315B2 (en) Digital video fingerprinting using motion segmentation
CN109188457B (en) Object detection frame generation method, device, equipment, storage medium and vehicle
KR20180056685A (en) System and method for non-obstacle area detection
KR20180105876A (en) Method for tracking image in real time considering both color and shape at the same time and apparatus therefor
JP2016507834A (en) System and method for tracking and detecting a target object
US20180181816A1 (en) Handling Perspective Magnification in Optical Flow Proessing
CN110853353B (en) Vision-based density traffic vehicle counting and traffic flow calculating method and system
JP2019016116A (en) Outside recognition device for vehicle
US9323989B2 (en) Tracking device
CN111079669A (en) Image processing method, device and storage medium
CN104680504A (en) Scene change detection method and device thereof
US11657485B2 (en) Method for expanding image depth and electronic device
TW201539378A (en) Object detection system
CN109478333A (en) Object detection method, device and image processing equipment
JP6116765B1 (en) Object detection apparatus and object detection method
CN111294524A (en) Video editing method and device, electronic equipment and storage medium
US9349069B2 (en) Dynamic line-detection system for processors having limited internal memory
US11948312B2 (en) Object detection/tracking device, method, and program recording medium
CN113256683B (en) Target tracking method and related equipment
US10708600B2 (en) Region of interest determination in video
KR101806453B1 (en) Moving object detecting apparatus for unmanned aerial vehicle collision avoidance and method thereof
CN113542868A (en) Video key frame selection method and device, electronic equipment and storage medium
CN116977671A (en) Target tracking method, device, equipment and storage medium based on image space positioning
CN109478328A (en) Method for tracking target, device and image processing equipment
CN112561956B (en) Video target tracking method and device, electronic equipment and storage medium

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
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20190315

WD01 Invention patent application deemed withdrawn after publication