CN109308704A - Background elimination method, device, computer equipment and storage medium - Google Patents
Background elimination method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109308704A CN109308704A CN201810872355.8A CN201810872355A CN109308704A CN 109308704 A CN109308704 A CN 109308704A CN 201810872355 A CN201810872355 A CN 201810872355A CN 109308704 A CN109308704 A CN 109308704A
- Authority
- CN
- China
- Prior art keywords
- pixel
- value
- video
- numerical value
- whereabouts
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 230000008030 elimination Effects 0.000 title claims abstract description 35
- 238000003379 elimination reaction Methods 0.000 title claims abstract description 35
- 238000003860 storage Methods 0.000 title claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 97
- 238000009795 derivation Methods 0.000 claims abstract description 50
- 238000004590 computer program Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 3
- 238000000926 separation method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000001727 in vivo Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 208000001491 myopia Diseases 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/215—Motion-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of background elimination methods, device, computer equipment and storage medium, by carrying out first derivation to the corresponding pixel value of each pixel in the initial picture being previously obtained, obtain the first processing numerical value of the corresponding pixel value of each pixel, second order derivation is carried out to the first processing numerical value of the corresponding pixel value of each pixel, obtain the second processing numerical value of the corresponding pixel value of each pixel, if the first processing numerical value of the corresponding pixel value of the pixel meets preset first condition, if and the second processing numerical value of the corresponding pixel value of the pixel meets preset second condition, then determine that the pixel is the target person in the initial picture and the marginal point between target person background, according to edge line, the target person background in the initial picture is rejected, obtain target figure Piece.It is computationally intensive to solve the problems, such as.
Description
Technical field
The present invention relates to financial technology field more particularly to a kind of background elimination method, device, computer equipment and storages
Medium.
Background technique
Currently, the crime rate of society persistently rises, the arresting criminal of the task is more and more heavier.
In order to arrest offender, there is pedestrian's weight identification technology.In the traditional pedestrian's weight recognizer of use to video
In monitoring during the investigation of target person, traditional pedestrian's weight recognizer can be neural network algorithm or other calculations
Interception target person usually can all be extracted from picture, realize the separation of target person and its background by method etc..But
Conventional method is in the colour-difference for calculating target person and target person background away from the feelings such as larger or target person background light is stronger
It is computationally intensive when picture under condition.
Summary of the invention
Based on this, it is necessary to which in view of the above technical problems, providing one kind can be to avoid calculating target person and target person
When the colour-difference of background is away from picture of the larger or target person background light when relatively strong, the back of computationally intensive problem
Scape elimination method, device, computer equipment and storage medium.
A kind of background elimination method, comprising:
First derivation is carried out to the corresponding pixel value of each pixel in the initial picture being previously obtained respectively, is obtained each
First processing numerical value of the corresponding pixel value of a pixel, wherein the corresponding pixel value of pixel and the pixel pair
There are one-to-one relationships for first processing numerical value of the pixel value answered;
Second order derivation is carried out to the first processing numerical value of the corresponding pixel value of each pixel respectively, obtains each picture
The second processing numerical value of the corresponding pixel value of vegetarian refreshments, wherein the first processing numerical value of the corresponding pixel value of the pixel and institute
Stating the second processing numerical value of the corresponding pixel value of pixel, there are one-to-one relationships;
If the first processing numerical value of the corresponding pixel value of the pixel meets preset first condition, and the pixel
The second processing numerical value of corresponding pixel value meets preset second condition, it is determined that the pixel is the mesh in the picture
Mark the marginal point between personage and target person background;
The target person background in the initial picture is rejected according to edge line, obtains Target Photo,
In, the edge line by the target person in each initial picture and the marginal point between target person background connection and
At.
A kind of background device for eliminating, comprising:
First derivation module, for the corresponding pixel value of each pixel in the initial picture being previously obtained respectively into
Row first derivation obtains the first processing numerical value of the corresponding pixel value of each pixel, wherein the corresponding pixel of the pixel
There are one-to-one relationships for first processing numerical value of value pixel value corresponding with the pixel;
Second derivation module carries out two for the first processing numerical value to the corresponding pixel value of each pixel respectively
Rank derivation obtains the second processing numerical value of the corresponding pixel value of each pixel, wherein the corresponding pixel value of the pixel
There are one-to-one relationships for the second processing numerical value of first processing numerical value pixel value corresponding with the pixel;
Determining module, if the first processing numerical value for the corresponding pixel value of the pixel meets preset first
Part, and the second processing numerical value of the corresponding pixel value of the pixel meets preset second condition, it is determined that the pixel
For the marginal point between the target person in the picture and target person background;
Module is rejected to obtain for rejecting the target person background in the initial picture according to edge line
To Target Photo, wherein the edge line is by the target person in each initial picture and between target person background
Marginal point is formed by connecting.
A kind of computer equipment, including memory, processor and storage are in the memory and can be in the processing
The computer program run on device, the processor realize the step of above-mentioned background elimination method when executing the computer program
Suddenly.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter
The step of calculation machine program realizes above-mentioned background elimination method when being executed by processor.
Above-mentioned background elimination method, device, computer equipment and storage medium, by the initial picture being previously obtained
The corresponding pixel value of each pixel carry out first derivation, obtain the first processing number of the corresponding pixel value of each pixel
Value carries out second order derivation to the first processing numerical value of the corresponding pixel value of each pixel, obtains the corresponding picture of each pixel
The second processing numerical value of element value, if the first processing numerical value of the corresponding pixel value of the pixel meets preset first condition,
If the second processing numerical value of pixel value corresponding with the pixel meets preset second condition, it is determined that the pixel is
Marginal point between target person in the initial picture and target person background, according to edge line, by the initial picture
In the target person background rejected, obtain Target Photo.Since the Method of Seeking Derivative only by calculation amount very little just can
The marginal point between target person and target person background is determined, according to the edge line being formed by connecting by marginal point by target person
Object background is rejected from picture, realize target person and target person background separation, so as to avoid calculate target person and
It is computationally intensive when the colour-difference of target person background is away from picture of the larger or target person background light when relatively strong
Problem.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is an application environment schematic diagram of background elimination method in one embodiment of the invention;
Fig. 2 is a flow chart of background elimination method in one embodiment of the invention;
Fig. 3 is the flow chart for obtaining initial picture in one embodiment of the invention in background elimination method;
Fig. 4 is the flow chart for obtaining whereabouts video in one embodiment of the invention in background elimination method;
Fig. 5 is the flow chart for calculating pixel in one embodiment of the invention in background elimination method;
Fig. 6 is a schematic diagram of background device for eliminating in one embodiment of the invention;
Fig. 7 is a schematic diagram of computer equipment in one embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Background elimination method provided by the present application, can be applicable in the application environment such as Fig. 1, wherein computer equipment is logical
Network is crossed to be communicated with server-side.Firstly, server-side is to the corresponding picture of each pixel in the initial picture being previously obtained
Plain value carries out first derivation respectively, obtains the first processing numerical value of the corresponding pixel value of each pixel, then, server-side is to each
First processing numerical value of the corresponding pixel value of a pixel carries out second order derivation respectively, obtains the corresponding pixel value of each pixel
Second processing numerical value, if the corresponding pixel value of pixel first processing numerical value meet preset first condition, and pixel
The second processing numerical value of corresponding pixel value meets preset second condition, then server-side determines that pixel is in initial picture
Marginal point between target person and target person background;Server-side is according to edge line by the target person in initial picture
Background is rejected, and Target Photo is obtained.Wherein, computer equipment can be, but not limited to various personal computers, notebook electricity
Brain, smart phone, tablet computer and portable wearable device.Server-side can use the either multiple services of independent server
The server cluster of device composition is realized.
In one embodiment, as shown in Fig. 2, providing a kind of background elimination method, which is applied in finance
In industry, applies be illustrated for the server-side in Fig. 1 in this way, include the following steps:
S10: first derivation is carried out to the corresponding pixel value of each pixel in the initial picture being previously obtained respectively, is obtained
First to the corresponding pixel value of each pixel handles numerical value;
In the present embodiment, pixel is pixel, and the corresponding pixel value of pixel is the mould of the corresponding position vector of pixel
And color value.
Specifically, the first processing numerical value of the corresponding pixel value of each pixel in picture selected in order to obtain, needs
The corresponding pixel value of each pixel is first obtained from each pixel in the selected picture being previously obtained, then to obtaining
The corresponding pixel value of each pixel got carries out first derivation respectively, to obtain the corresponding pixel value of each pixel
First processing numerical value.Such as, at the first of the corresponding pixel value of each pixel for the picture " basketball .jpg " selected in order to obtain
Numerical value is managed, needs first to obtain each pixel pair from each pixel in the selected picture " basketball .jpg " being previously obtained
The pixel value answered, the corresponding pixel value of each pixel then got carry out first derivation respectively, it is preferable that the first pixel
The corresponding pixel value of point is xn, then to xnFirst derivation is carried out, the first processing number of the corresponding pixel value of the first pixel is obtained
Value is nxn-1, wherein n is more than or equal to the positive integer of number 1, and x is more than or equal to the positive integer of number 1.
It should be noted that the first processing numerical value of the corresponding pixel value of pixel pixel value corresponding with pixel exists
One-to-one relationship.
S20: second order derivation is carried out to the first processing numerical value of the corresponding pixel value of each pixel respectively, obtains each picture
The second processing numerical value of the corresponding pixel value of vegetarian refreshments;
In the present embodiment, the second processing number for the corresponding pixel value of each pixel in picture selected in order to obtain
It is worth, the first processing numerical value of the corresponding pixel value of each pixel in the selected picture for needing to obtain first derivation carries out
Second order derivation, to obtain the second processing numerical value of the desired corresponding pixel value of each pixel.Such as, it selectes in order to obtain
The second processing numerical value of the corresponding pixel value of each pixel of picture " bicycle .jpg ", needs the choosing obtained to first derivation
First processing numerical value of the corresponding pixel value of each pixel in fixed picture " bicycle .jpg " carries out second order derivation, preferably
The first processing numerical value on ground, the corresponding pixel value of the second pixel is nxn-1, then to nxn-1Second order derivation is carried out, obtains first
The second processing numerical value of the corresponding pixel value of pixel is n* (n-1) xn-2, wherein n is more than or equal to the positive integer of number 2, x
For the positive integer more than or equal to number 1.
It should be noted that the of the first corresponding with pixel pixel value of processing numerical value of the corresponding pixel value of pixel
There are one-to-one relationships for two processing numerical value.
S30: if the first processing numerical value of the corresponding pixel value of pixel meets preset first condition, and pixel is corresponding
The second processing numerical value of pixel value meet preset second condition, it is determined that pixel is target person and target in picture
Marginal point between personage's background;
In the present embodiment, when the first processing numerical value of the corresponding pixel value of obtained pixel meets preset first
When part, and when the second processing numerical value of the corresponding pixel value of obtained pixel meets preset second condition, it is determined that should
Pixel is the target person in picture and the marginal point between target person background, the marginal point namely target person and target
The tangent boundary point of personage's background.
It should be noted that preset first condition can be for greater than preset first threshold or equal to preset first threshold
Value, preset second condition can be for greater than preset second thresholds or equal to preset second threshold, preset first condition
With the particular content of preset second condition, can be set according to practical application, herein with no restrictions.
S40: the target person background in picture is rejected according to edge line, obtains Target Photo;
In the present embodiment, first by the target person in obtained each picture and the edge between target person background
Point connects, and obtains edge line, then along the edge line, target person is distinguished with target person background, next
The pixel value of the corresponding pixel value of each pixel in the target person background is substituted using preset third threshold value, therefore
The region for being apparent that the corresponding color of preset third threshold value is target person background, to avoid in picture
The interference of background when target person detects, and then play the effect for rejecting background.
It should be noted that the particular content that preset third threshold value can be, can be set according to practical application,
Herein with no restrictions.
In the corresponding embodiment of Fig. 2, by the corresponding pixel of each pixel in the initial picture being previously obtained
Value carries out first derivation, obtains the first processing numerical value of the corresponding pixel value of each pixel, picture corresponding to each pixel
First processing numerical value of element value carries out second order derivation, the second processing numerical value of the corresponding pixel value of each pixel is obtained, if institute
The the first processing numerical value for stating the corresponding pixel value of pixel meets preset first condition, and if the corresponding pixel of the pixel
The second processing numerical value of value meets preset second condition, it is determined that the pixel is the target person in the initial picture
The target person background in the initial picture is carried out according to edge line with the marginal point between target person background
It rejects, obtains Target Photo.Since the Method of Seeking Derivative only by calculation amount very little just can determine that out target person and target person
Marginal point between background rejects target person background according to the edge line being formed by connecting by marginal point from picture, realizes
The separation of target person and target person background, so as to avoid calculate the colour-difference of target person and target person background away from compared with
Greatly or target person background light it is relatively strong when picture when, computationally intensive problem.
Further, in one embodiment, which applies in financial industry, a kind of background elimination method
First processing numerical value of the corresponding pixel value of middle pixel meets preset first condition, and the of the corresponding pixel value of pixel
Two processing numerical value meet preset second condition are as follows:
Specifically, the first absolute value is greater than preset first threshold, and the second absolute value is greater than preset second threshold,
In, the one of the first derivation result for the pixel that the first absolute value the is default pixel laterally adjacent with the pixel
The absolute value of the difference of rank derivation result, the second order derivation result for the pixel that the second absolute value is and the pixel are lateral
The absolute value of the difference of the second order derivation result of default adjacent pixel.
It should be noted that the transverse direction can be left or right, or while including default of left and right pixel
Can be 3 or 5 pixels, e.g., 3, left side pixel or 3, the right pixel, or simultaneously include 3, left side pixel
With 3, the right pixel namely 6 pixels, the particular content of a pixel is preset, can be set according to practical application
It is fixed, herein with no restrictions.
In this embodiment, by by laterally adjacent pre- of the first derivation result of obtained pixel and the pixel
If the absolute value of the difference of the first derivation result of a pixel is compared with preset first threshold, and by obtained pixel
Compared with the absolute value of the difference of the second order derivation result of the second order derivation result default pixel laterally adjacent with the pixel,
It can determine that the obtained pixel is the target person in selected picture and the marginal point between target person background.Due to
The size of first derivation result represents the change rate size namely the more big picture of first derivation result of the color of the pixel
The variation of the color of vegetarian refreshments is bigger, and the variation of the color on the contrary for representing the pixel is with regard to smaller, while second order derivation result
Size represent the pixel color the speed of variation namely the bigger of second order derivation result represent the pixel
The variation of color is faster, otherwise the variation of the color that represents the pixel is slower, the variation of the color of two pixels greatly and
Variation represents two pixels as two different colors, so as to easily tell different color in picture slowly
Point, and then improve the convenience for distinguishing marginal point.
Further, in one embodiment, which applies in financial industry, and Fig. 2 as shown in Figure 3 is corresponding
Flow chart of the initial picture under an application scenarios is obtained in embodiment in a kind of background elimination method, the initial picture is logical
Cross following steps acquisition:
S101: according to the time difference in the whereabouts video being previously obtained between each video frame and the figure between each video frame
As the similarity of color character, the distance between each video frame is calculated;
S102: whereabouts video is split using hierarchical clustering method according to the distance between each video frame, is obtained initial
Picture.
For above-mentioned steps S101, it can be understood as, firstly, being determined as view to be split for the whereabouts video taken the photograph in advance is clapped
Frequently, then, according to the time difference between video frame each in video to be split and the image color feature between each video frame
Similarity calculates the distance between each video frame using video frame distance calculation formula.
For above-mentioned steps S102, it can be understood as, by taking video to be split includes M video frame as an example, according to be split
The image between time difference and the 1st video frame and the 2nd video frame in video between the 1st video frame and the 2nd video frame
The similarity of color character gets the distance between the 1st video frame and the 2nd video frame;It successively executes, until according to be split
The image between time difference and the 1st video frame and k-th of video frame in video between the 1st video frame and kth video frame
The similarity of color character gets the distance between the 1st video frame and kth video frame;According to the 2nd video in video to be split
The phase of the image color feature between time difference and the 2nd video frame and the 3rd video frame between frame and the 3rd video frame
Like degree, the distance between the 2nd video frame and the 3rd video frame are got;It successively executes, until according to the 2nd view in video to be split
The image color feature between time difference and the 2nd video frame and k-th of video frame between frequency frame and kth video frame
Similarity gets the distance between the 1st video frame and kth video frame;And so on, it is regarded according to q-1 in video to be split
The image color feature between time difference and q-1 video frame and q-th of video frame between frequency frame and q video frame
Similarity, get the distance between q-1 video frame and q video frame.By the above-mentioned means, each video can be obtained
The distance between frame is sent out using hierarchical clustering then according to the distance and is split the video to be split, wanted
Picture, selected picture is by selecting in the picture.
In the corresponding embodiment of Fig. 3, pass through step S101 and step S102, available desired picture, due to belonging to
Image in two video frames of the same Video Events, scene is identical with the color character similarity of target person, from
And the video frame belonged in the same video can be divided into the picture of the video according to time difference and similarity, to avoid
The video frame belonged in the same video is divided into the picture of different videos, and then ensure that the complete of video content segmentation
Whole property.
Further, in one embodiment, which applies in financial industry, a kind of background elimination method
The image color feature between time difference and each video frame in the whereabouts video that middle basis is previously obtained between each video frame
Similarity, calculate the distance between each video frame are as follows:
Specifically, it firstly, the chi-Square measure between each video frame is calculated according to each video frame, then, will clap in advance
The following calculating of chi-Square measure input between time difference, each video frame in the whereabouts video taken the photograph between each video frame is public
Formula obtains the distance between each video frame:
Wherein, whereabouts video includes q video frame, and q is greater than 1;kiFor i-th of video frame in whereabouts video, kjFor whereabouts
J-th of video frame in video, i are more than or equal to 1 and are less than or equal to q, and j is more than or equal to 1 and is less than or equal to q, s (ki,kj) it is i-th
The distance between a video frame and j-th of video frame, x2(ki, kj) color between i-th of video frame and j-th of video frame
The chi-Square measure of histogram, w1For preset chi-Square measure, r is preset positive integer, | i-j | indicate i-th of video frame and jth
Time difference between a video frame, the unit of the time difference be the second, point or hour, max (0, r- | i-j |) indicate 0 with
R- | i-j | between maximum value, r- | i-j | for natural number greater than 0.As r be 30, i 20, j 15, | i-j | indicate the 20th
Time difference between a video frame and the 15th video frame is 20 seconds, then r- | i-j | it be 10, max (0, r- | i-j |) is 10,
Or if r is 7, i 8, j 5, | i-j | indicate that the time difference between the 8th video frame and the 5th video frame is 1.6 points, then
R- | i-j | it be 5.4, max (0, r- | i-j |) is 5.4.
It should be noted that the bigger expression similarity of chi-Square measure is lower.
In this embodiment, by above-mentioned calculation formula, according to s (ki,kjEven if) can be by one containing some unrelated
The short-sighted frequency of video frame will not be further divided into shorter video, to ensure that the integrality of video.
Further, in one embodiment, which applies in financial industry, and Fig. 3 as shown in Figure 4 is corresponding
Flow chart of the whereabouts video under an application scenarios is obtained in embodiment in a kind of background elimination method, the whereabouts video is logical
Cross following steps acquisition:
S1011: the whereabouts video of continuous several times photographic subjects personage, and judge continuously during shooting whereabouts video
Whether the total time length for repeatedly shooting obtained whereabouts video reaches predetermined time period;
S1012: if the total time length for the whereabouts video that continuous several times are shot during shooting whereabouts video
Reach predetermined time period, then stop the shooting of whereabouts video, and by the whereabouts video-splicing of continuous several times photographic subjects personage
At a whereabouts video, spliced whereabouts video is obtained;
S1013: the whereabouts video that spliced whereabouts video is determined as.
For above-mentioned steps S1011, it can be understood as, target person can be pedestrian or other objects, target person
Particular content can be set, herein with no restrictions according to the actual situation.The whereabouts video of target person may exist in number
It according in library, also can store in optical disc or hard disk, the particular content of the storage location of the whereabouts video of target person, Ke Yigen
It is set according to actual conditions, herein with no restrictions.
Specifically, firstly, determining target person from multiple objects, that is, artificial selected target personage, then from
Target person is selected in multiple objects, next, the whereabouts video of continuous several times photographic subjects personage, and in shooting whereabouts every time
Judge whether the total time length for the whereabouts video that continuous several times are shot reaches predetermined time period during video,
In, judge whether the total time length of the whereabouts video of shooting reaches predetermined time period, it may be possible to neatly obtain and want
Time span whereabouts video, preset time span can be 30 minutes or 1 hour etc., the tool of preset time span
Hold in vivo, can be set according to the actual situation, herein with no restrictions.
It should be noted that the whereabouts video of target person can be acquired by picture pick-up device, it is then stored in institute
Storage location is stated, picture pick-up device can be video camera or digital camera etc., and the particular content of picture pick-up device can be according to practical feelings
Condition is set, herein with no restrictions.
For above-mentioned steps S1012, it can be understood as, if continuous several times shoot to obtain during shooting whereabouts video
The total time length of whereabouts video reach predetermined time period, then stop the shooting of whereabouts video, and show that continuous several times are clapped
The corresponding mark of whereabouts video of target object is taken the photograph, and corresponding according to whereabouts video of the user to continuous several times photographic subjects object
The result that is ranked up of mark the whereabouts video-splicing of continuous several times photographic subjects object is connected at a video, or record
Continue the whereabouts video corresponding time of multiple photographic subjects object, and according to user to the whereabouts of continuous several times photographic subjects object
The result that the video corresponding time is ranked up is by the whereabouts video-splicing of continuous several times photographic subjects object at a video.If
It is long that the total time length for the whereabouts video that continuous several times are shot during shooting whereabouts video is not up to preset time
Degree then subtracts total time length using predetermined time period, obtains remaining time length, then calculates remaining time length pre-
If the first accounting in time span, and second accounting of the total time length in predetermined time period is calculated, thus clear
Shooting time length progress in predetermined time period, and return step S1011.
For above-mentioned steps S1013, it can be understood as, the whereabouts video obtained after splicing is determined as the whereabouts got
Video, and be saved in video database.Wherein, video database can be the databases such as sql or Oracle, video database
Particular content, can be set according to practical application, herein with no restrictions.
In the corresponding embodiment of Fig. 4, S1011~step S1013, may be implemented multiple selectivity through the above steps
Shooting obtains a whereabouts video, while independently being screened to the content of the whereabouts video of shooting, improves and obtains video
Flexibility.
Further, in one embodiment, which applies in financial industry, Fig. 2 as shown in Figure 5 to figure
A kind of flow chart in 4 corresponding embodiments in background elimination method before step S10 under an application scenarios, the background are picked
Except method further include:
S50: the corresponding first object pixel value of each pixel of initial picture, each picture of target person background are obtained
Corresponding second target pixel value of vegetarian refreshments and pixel summation, wherein the pixel of picture include target person pixel and
The pixel of target person background, pixel summation are the total number of the pixel of picture or the pixel of target person background
Total number;
S60: preset calculating method is used according to first object pixel value, second target pixel value and pixel summation
The corresponding pixel value of each pixel of picture is calculated, the corresponding pixel value of the pixel after being calculated.
For above-mentioned steps S50, it can be understood as, selected picture includes target person and target person background.It is selected
Picture be to be composed of each pixel.Obtain the corresponding first object picture of each pixel in the picture of whole Zhang Xuanding
Element value, corresponding second target pixel value of each pixel of background and pixel summation, so as to corresponding to each pixel
Pixel value does the processing of inequality method.Wherein, pixel summation is the total number of the pixel of picture or the pixel of target person background
The total number of point.
It should be noted that the pixel value for the pixel that target person is overlapped with target person background, is carried on the back in target person
It is 0 in scape.
For above-mentioned steps S60, it can be understood as, according to first object pixel value, second target pixel value and picture
Vegetarian refreshments summation calculates the corresponding pixel value of each pixel of picture using preset calculating method, the picture after being calculated
The corresponding pixel value of vegetarian refreshments, that is, the corresponding first object pixel value of each pixel for the picture that first will acquire subtracts figure
Corresponding second target pixel value of each pixel of piece background, obtains each deviation, then by the summation of all deviations
Divided by the total number of pixel, equal difference is obtained, each deviation is subtracted into equal difference, the pixel after obtaining each inequality.
In the corresponding embodiment of Fig. 5, S50 and step S60, available each pixel are corresponding through the above steps
The average pixel value of pixel value, since average pixel value represents the mean pixel of all pixels point, to improve assessment pixel
The accuracy of the pixel of point.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
In one embodiment, a kind of background device for eliminating is provided, which picks with background in above-described embodiment
Except method corresponds.As shown in fig. 6, the background device for eliminating include the first derivation module 701, the second derivation module 702, really
Cover half block 703 and rejecting module 704.Detailed description are as follows for each functional module:
First derivation module 701, for the corresponding pixel value point of each pixel in the initial picture being previously obtained
Not carry out first derivation, obtain the corresponding pixel value of each pixel first processing numerical value, wherein the pixel value of pixel with
There are one-to-one relationships for first processing numerical value of the corresponding pixel value of pixel;
Second derivation module 702 carries out second order derivation for handling numerical value to each pixel corresponding first respectively, obtains
To the second processing numerical value of the corresponding pixel value of each pixel, wherein the first processing numerical value of the corresponding pixel value of pixel
There are one-to-one relationships for the second processing numerical value of pixel value corresponding with pixel;
First determining module 703, if the first processing numerical value for the corresponding pixel value of pixel meets preset first
Condition, and the second processing numerical value of the corresponding pixel value of pixel meets preset second condition, it is determined that pixel is picture
In target person and target person background between marginal point;
It rejects module 704 and obtains target figure for rejecting the target person background in picture according to edge line
Piece, wherein edge line is formed by connecting by the marginal point between the target person in each picture and target person background.
Further, it is determined that first processing numerical value of the corresponding pixel value of pixel meets preset first in module 703
Condition, and the second processing numerical value of the corresponding pixel value of pixel meets preset second condition are as follows: first judgment module is used for
First absolute value is greater than the first preset threshold, and the second absolute value is greater than the second preset threshold, wherein the first absolute value is pixel
The first of first processing numerical value of the corresponding pixel value of point pixel value corresponding with default laterally adjacent pixel of pixel
The absolute value of the difference of numerical value is handled, the second absolute value is the second processing numerical value and pixel cross of the corresponding pixel value of pixel
To the absolute value of the difference of the second processing numerical value of the default adjacent corresponding pixel value of pixel.
Further, initial picture with lower module by being obtained:
First computing module, for according between each video frame in the whereabouts video that is previously obtained time difference and each view
The similarity of image color feature between frequency frame, calculates the distance between each video frame;
Divide module, for being split using hierarchical clustering method to whereabouts video according to the distance between each video frame,
Obtain initial picture.
Further, according to the time difference in the whereabouts video being previously obtained between each video frame in the first computing module
The similarity of image color feature between each video frame, calculates the distance between each video frame specifically: second calculates mould
Block, the chi-Square measure between time difference, each video frame in the whereabouts video for will be previously obtained between each video frame are defeated
Enter following calculation formula, obtain the distance between each video frame:
Wherein, whereabouts video includes q video frame, and q is greater than 1;kiFor i-th of video frame in whereabouts video, kjFor whereabouts
J-th of video frame in video, i are more than or equal to 1 and are less than or equal to q, and j is more than or equal to 1 and is less than or equal to q, s (ki,kj) it is i-th
The distance between a video frame and j-th of video frame, x2(ki, kj) color between i-th of video frame and j-th of video frame
The chi-Square measure of histogram, w1For preset chi-Square measure, r is preset positive integer, | i-j | indicate i-th of video frame and jth
Time difference between a video frame, max (0, r- | i-j |) indicate 0 and r- | i-j | between maximum value, r- | i-j | for greater than
0 natural number.
Further, whereabouts video with lower module by being obtained:
Second judgment module, for the whereabouts video of continuous several times photographic subjects personage, and in the mistake of shooting whereabouts video
Judge whether the total time length for the whereabouts video that continuous several times are shot reaches predetermined time period in journey;
Splicing module, if for continuous several times are shot during shooting whereabouts video whereabouts video it is total when
Between length reach predetermined time period, then stop the shooting of whereabouts video, and the whereabouts of continuous several times photographic subjects personage are regarded
Frequency is spliced into a whereabouts video, obtains spliced whereabouts video;
Second determining module, the whereabouts video for spliced whereabouts video to be determined as.
Further, before the first derivation module 701, the background device for eliminating further include:
Module is obtained, the corresponding first object pixel value of each pixel, target person back for obtaining initial picture
Corresponding second target pixel value of each pixel of scape and pixel summation, wherein the pixel of picture includes target person
Pixel and target person background pixel, pixel summation be picture pixel total number or target person background
Pixel total number;
Third computing module, for being adopted according to first object pixel value, second target pixel value and pixel summation
It is calculated with each pixel corresponding pixel value of the preset calculating method to picture, the pixel after being calculated is corresponding
Pixel value.
Specific about background device for eliminating limits the restriction that may refer to above for background elimination method, herein not
It repeats again.Modules in above-mentioned background device for eliminating can be realized fully or partially through software, hardware and combinations thereof.On
Stating each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also store in a software form
In memory in computer equipment, the corresponding operation of the above modules is executed in order to which processor calls.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 7.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is for storing background elimination method related data.The network interface of the computer equipment is used for and outside
Terminal passes through network connection communication.To realize a kind of background elimination method when the computer program is executed by processor.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor realize above-described embodiment background rejecting side when executing computer program
The step of method, such as step S10 shown in Fig. 2 to step S40.Alternatively, processor realizes above-mentioned reality when executing computer program
Apply the function of each module/unit of background device for eliminating in example, for example, the first derivation module 701 shown in Fig. 6 to reject module 704
Function.To avoid repeating, which is not described herein again.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Background elimination method in above method embodiment is realized when machine program is executed by processor, alternatively, the computer program is processed
The function of each module/unit in background device for eliminating in above-mentioned apparatus embodiment is realized when device executes.To avoid repeating, here not
It repeats again.Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing
The all or part of function of description.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations;Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features;And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of background elimination method, which is characterized in that the background elimination method includes:
First derivation is carried out to the corresponding pixel value of each pixel in the initial picture being previously obtained respectively, obtains each picture
First processing numerical value of the corresponding pixel value of vegetarian refreshments, wherein the corresponding pixel value of the pixel is corresponding with the pixel
There are one-to-one relationships for first processing numerical value of pixel value;
Second order derivation is carried out to the first processing numerical value of the corresponding pixel value of each pixel respectively, obtains each pixel
The second processing numerical value of corresponding pixel value, wherein the first processing numerical value of the corresponding pixel value of the pixel and the picture
There are one-to-one relationships for the second processing numerical value of the corresponding pixel value of vegetarian refreshments;
If the first processing numerical value of the corresponding pixel value of the pixel meets preset first condition, and the pixel is corresponding
The second processing numerical value of pixel value meet preset second condition, it is determined that the pixel is the target person in the picture
Marginal point between object and target person background;
The target person background in the initial picture is rejected according to edge line, obtains Target Photo, wherein institute
Edge line is stated to be formed by connecting by the target person in each initial picture and the marginal point between target person background.
2. background elimination method as described in claim 1, which is characterized in that at the first of the corresponding pixel value of the pixel
Reason numerical value meets preset first condition, and the second processing numerical value of the corresponding pixel value of the pixel meets preset second
Condition are as follows: the first absolute value is greater than the first preset threshold, and the second absolute value is greater than the second preset threshold, wherein described first
Absolute value is the first processing numerical value of the corresponding pixel value of the pixel default pixel laterally adjacent with the pixel
The absolute value of the difference of first processing numerical value of the corresponding pixel value of point, second absolute value is the corresponding picture of the pixel
The second processing number of the second processing numerical value of element value pixel value corresponding with default laterally adjacent pixel of the pixel
The absolute value of the difference of value.
3. background elimination method as described in claim 1, which is characterized in that the initial picture is obtained by following steps:
It is special according to the time difference in the whereabouts video being previously obtained between each video frame and the image color between each video frame
The similarity of sign calculates the distance between each video frame;
The whereabouts video is split using hierarchical clustering method according to the distance between described each video frame, is obtained described first
Beginning picture.
4. background elimination method as claimed in claim 3, which is characterized in that each in the whereabouts video that the basis is previously obtained
The similarity of the image color feature between time difference and each video frame between video frame, calculate between each video frame away from
From are as follows: by the time difference in the whereabouts video being previously obtained between each video frame, the chi-Square measure between each video frame
Following calculation formula is inputted, the distance between each video frame is obtained:
Wherein, the whereabouts video includes q video frame, and the q is greater than 1;kiFor i-th of video frame in the whereabouts video,
kjFor j-th of video frame in the whereabouts video, the i is more than or equal to 1 and is less than or equal to the q, and the j is more than or equal to 1
And it is less than or equal to the q, s (ki,kj) it is the distance between i-th of video frame and j-th of video frame, the x2(ki,kj) for institute
State the chi-Square measure of the color histogram between i-th of video frame and j-th of video frame, the w1For preset chi-Square measure,
The r is preset positive integer, | i-j | the time difference between i-th of video frame of expression and j-th of video frame, max (0, r- |
I-j |) indicate 0 and r- | i-j | between maximum value, r- | i-j | for natural number greater than 0.
5. background elimination method as claimed in claim 3, which is characterized in that the whereabouts video is obtained by following steps:
The whereabouts video of continuous several times photographic subjects personage, and judge that continuous several times are shot during shooting whereabouts video
To the total time length of whereabouts video whether reach predetermined time period;
If the total time length for the whereabouts video that continuous several times are shot during shooting whereabouts video reaches default
Between length, then stop the shooting of whereabouts video, and by the whereabouts video-splicing of the continuous several times photographic subjects personage at one
Whereabouts video obtains spliced whereabouts video;
The whereabouts video that the spliced whereabouts video is determined as.
6. the background elimination method as described in any one of claims 1 to 5, which is characterized in that be previously obtained at described Dui
Before the corresponding pixel value of each pixel in initial picture carries out first derivation, the background elimination method further include:
Obtain the corresponding first object pixel value of each pixel of the initial picture, each picture of the target person background
Corresponding second target pixel value of vegetarian refreshments and pixel summation, wherein the pixel of the picture includes the target person
The pixel of pixel and the target person background, the total number of pixel of the pixel summation for the picture or institute
State the total number of the pixel of target person background;
Preset calculating method is used according to the first object pixel value, second target pixel value and the pixel summation
The corresponding pixel value of each pixel of the picture is calculated, the corresponding pixel value of the pixel after being calculated.
7. a kind of background device for eliminating, which is characterized in that the background device for eliminating includes:
First derivation module, for carrying out one respectively to the corresponding pixel value of each pixel in the initial picture being previously obtained
Rank derivation obtains the first processing numerical value of the corresponding pixel value of each pixel, wherein the corresponding pixel value of the pixel with
There are one-to-one relationships for first processing numerical value of the corresponding pixel value of the pixel;
Second derivation module carries out second order for the first processing numerical value to the corresponding pixel value of each pixel respectively and asks
It leads, obtains the second processing numerical value of the corresponding pixel value of each pixel, wherein the first of the corresponding pixel value of the pixel
There are one-to-one relationships for the second processing numerical value of processing numerical value pixel value corresponding with the pixel;
Determining module, if the first processing numerical value for the corresponding pixel value of the pixel meets preset first condition, and
The second processing numerical value of the corresponding pixel value of the pixel meets preset second condition, it is determined that the pixel is described
Marginal point between target person in picture and target person background;
It rejects module and obtains mesh for rejecting the target person background in the initial picture according to edge line
It marks on a map piece, wherein the edge line is by the target person in each initial picture and the edge between target person background
Point is formed by connecting.
8. background device for eliminating as claimed in claim 7, which is characterized in that at the first of the corresponding pixel value of the pixel
Reason numerical value meets preset first condition, and the second processing numerical value of the corresponding pixel value of the pixel meets preset second
Condition specifically: first judgment module is greater than the first preset threshold for the first absolute value, and the second absolute value is greater than second in advance
If threshold value, wherein the first absolute value is laterally adjacent pre- of the first processing numerical value and pixel of the corresponding pixel value of pixel
If the absolute value of the difference of the first processing numerical value of the corresponding pixel value of a pixel, the second absolute value is the corresponding picture of pixel
The second processing numerical value of the second processing numerical value of element value pixel value corresponding with default laterally adjacent pixel of pixel
The absolute value of difference.
9. a kind of computer equipment, including memory, processor and storage are in the memory and can be in the processor
The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to
Described in any one of 5 the step of background elimination method.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In the step of realization background elimination method as described in any one of claims 1 to 5 when the computer program is executed by processor
Suddenly.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810872355.8A CN109308704B (en) | 2018-08-02 | 2018-08-02 | Background eliminating method, device, computer equipment and storage medium |
PCT/CN2018/106379 WO2020024394A1 (en) | 2018-08-02 | 2018-09-19 | Background elimination method and device, computer device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810872355.8A CN109308704B (en) | 2018-08-02 | 2018-08-02 | Background eliminating method, device, computer equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109308704A true CN109308704A (en) | 2019-02-05 |
CN109308704B CN109308704B (en) | 2024-01-16 |
Family
ID=65226067
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810872355.8A Active CN109308704B (en) | 2018-08-02 | 2018-08-02 | Background eliminating method, device, computer equipment and storage medium |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN109308704B (en) |
WO (1) | WO2020024394A1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110502205A (en) * | 2019-08-29 | 2019-11-26 | 百度在线网络技术(北京)有限公司 | Picture showing edge processing method, device, electronic equipment and readable storage medium storing program for executing |
CN110765935A (en) * | 2019-10-22 | 2020-02-07 | 上海眼控科技股份有限公司 | Image processing method, image processing device, computer equipment and readable storage medium |
CN111860200A (en) * | 2020-06-28 | 2020-10-30 | 维沃移动通信有限公司 | Video frame processing method and device |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117197706B (en) * | 2023-04-23 | 2024-06-18 | 青岛尘元科技信息有限公司 | Method and system for dividing progressive lens, storage medium and electronic device |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR19980028903A (en) * | 1996-10-24 | 1998-07-15 | 배순훈 | Target tracking device and method of video telephone |
CN105719274A (en) * | 2014-12-10 | 2016-06-29 | 全视技术有限公司 | Edge Detection System And Methods |
WO2016131300A1 (en) * | 2015-07-22 | 2016-08-25 | 中兴通讯股份有限公司 | Adaptive cross-camera cross-target tracking method and system |
CN106056532A (en) * | 2016-05-20 | 2016-10-26 | 深圳市奥拓电子股份有限公司 | Method and device of removing background images |
CN106534951A (en) * | 2016-11-30 | 2017-03-22 | 北京小米移动软件有限公司 | Method and apparatus for video segmentation |
WO2017116808A1 (en) * | 2015-12-30 | 2017-07-06 | Ebay, Inc. | Background removal |
CN108038869A (en) * | 2017-11-20 | 2018-05-15 | 江苏省特种设备安全监督检验研究院 | Passenger falls down to the ground behavioral value method in a kind of lift car |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005100176A (en) * | 2003-09-25 | 2005-04-14 | Sony Corp | Image processor and its method |
-
2018
- 2018-08-02 CN CN201810872355.8A patent/CN109308704B/en active Active
- 2018-09-19 WO PCT/CN2018/106379 patent/WO2020024394A1/en active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR19980028903A (en) * | 1996-10-24 | 1998-07-15 | 배순훈 | Target tracking device and method of video telephone |
CN105719274A (en) * | 2014-12-10 | 2016-06-29 | 全视技术有限公司 | Edge Detection System And Methods |
WO2016131300A1 (en) * | 2015-07-22 | 2016-08-25 | 中兴通讯股份有限公司 | Adaptive cross-camera cross-target tracking method and system |
WO2017116808A1 (en) * | 2015-12-30 | 2017-07-06 | Ebay, Inc. | Background removal |
CN106056532A (en) * | 2016-05-20 | 2016-10-26 | 深圳市奥拓电子股份有限公司 | Method and device of removing background images |
CN106534951A (en) * | 2016-11-30 | 2017-03-22 | 北京小米移动软件有限公司 | Method and apparatus for video segmentation |
CN108038869A (en) * | 2017-11-20 | 2018-05-15 | 江苏省特种设备安全监督检验研究院 | Passenger falls down to the ground behavioral value method in a kind of lift car |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110502205A (en) * | 2019-08-29 | 2019-11-26 | 百度在线网络技术(北京)有限公司 | Picture showing edge processing method, device, electronic equipment and readable storage medium storing program for executing |
CN110502205B (en) * | 2019-08-29 | 2023-08-01 | 百度在线网络技术(北京)有限公司 | Picture display edge processing method and device, electronic equipment and readable storage medium |
CN110765935A (en) * | 2019-10-22 | 2020-02-07 | 上海眼控科技股份有限公司 | Image processing method, image processing device, computer equipment and readable storage medium |
CN111860200A (en) * | 2020-06-28 | 2020-10-30 | 维沃移动通信有限公司 | Video frame processing method and device |
CN111860200B (en) * | 2020-06-28 | 2024-04-19 | 维沃移动通信有限公司 | Video frame processing method and device |
Also Published As
Publication number | Publication date |
---|---|
WO2020024394A1 (en) | 2020-02-06 |
CN109308704B (en) | 2024-01-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109308704A (en) | Background elimination method, device, computer equipment and storage medium | |
CN110503076B (en) | Video classification method, device, equipment and medium based on artificial intelligence | |
CN110473137A (en) | Image processing method and device | |
CN108629791A (en) | Pedestrian tracting method and device and across camera pedestrian tracting method and device | |
CN108197532A (en) | The method, apparatus and computer installation of recognition of face | |
CN111062871A (en) | Image processing method and device, computer equipment and readable storage medium | |
CN109685100A (en) | Character identifying method, server and computer readable storage medium | |
CN106161980A (en) | Photographic method and system based on dual camera | |
CN106203333A (en) | Face identification method and system | |
CN104899579A (en) | Face recognition method and face recognition device | |
CN110443210A (en) | A kind of pedestrian tracting method, device and terminal | |
CN109299658B (en) | Face detection method, face image rendering device and storage medium | |
CN110176024B (en) | Method, device, equipment and storage medium for detecting target in video | |
US8634589B2 (en) | Information processing apparatus and method for registering moving objects as metadata of an image | |
CN108875492A (en) | Face datection and crucial independent positioning method, device, system and storage medium | |
CN108875481A (en) | Method, apparatus, system and storage medium for pedestrian detection | |
CN112101195B (en) | Crowd density estimation method, crowd density estimation device, computer equipment and storage medium | |
CN112364827A (en) | Face recognition method and device, computer equipment and storage medium | |
CN109815370A (en) | A kind of archiving method and device | |
US20200311962A1 (en) | Deep learning based tattoo detection system with optimized data labeling for offline and real-time processing | |
CN109800318A (en) | A kind of archiving method and device | |
CN108717520A (en) | A kind of pedestrian recognition methods and device again | |
CN109784220A (en) | A kind of method and device of determining passerby track | |
CN110766077A (en) | Method, device and equipment for screening sketch in evidence chain image | |
CN112597995B (en) | License plate detection model training method, device, equipment and 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |