CN108257082A - Method and apparatus based on fixed area removal image finger - Google Patents
Method and apparatus based on fixed area removal image finger Download PDFInfo
- Publication number
- CN108257082A CN108257082A CN201810100143.8A CN201810100143A CN108257082A CN 108257082 A CN108257082 A CN 108257082A CN 201810100143 A CN201810100143 A CN 201810100143A CN 108257082 A CN108257082 A CN 108257082A
- Authority
- CN
- China
- Prior art keywords
- rect
- regions
- subgraph
- lines
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000013507 mapping Methods 0.000 claims abstract description 64
- 230000035515 penetration Effects 0.000 claims abstract description 12
- 230000011218 segmentation Effects 0.000 claims description 14
- 238000005194 fractionation Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 6
- 230000008595 infiltration Effects 0.000 claims 1
- 238000001764 infiltration Methods 0.000 claims 1
- 238000009499 grossing Methods 0.000 abstract description 4
- 238000001514 detection method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G06T3/04—
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of method and apparatus based on fixed area removal image finger, this method includes：Acquire the Rect area images of presumptive area in image；It is upper Rect regions subgraph and lower Rect regions subgraph to split Rect area images；The image of more than Rect regions subgraph upper edge region is artwork, Rect regions subgraph in filling；The image of following Rect regions subgraph lower edge margin is artwork, fills lower Rect regions subgraph；The upper Rect regions subgraph of color penetration processing and lower Rect regions subgraph intersection.Frame choosing limitation is first carried out to the position of finger, then projection filling is carried out in the range selected frame, color penetration is carried out mutually in the stitching portion to projecting up and down, so as to fulfill stitching portion smoothing processing.Using the above scheme, no matter we acquire image, not in the pressing for being limited by fixed colour of skin finger under which kind of environment, may be implemented in the mapping filling in frame favored area, realize the versatility and accuracy of removal finger.
Description
Technical field
The present invention relates to a kind of minimizing technologies of foreign matter at computer picture, more particularly to high photographing instrument shooting file and picture
The method and device of middle fixed area removal finger.
Background technology
Made thing in image acquisition process is frequently necessary to the pressing of finger to ensure complete imaging, especially applies and is clapped in height
During instrument and other non-contacting scanning device scanned document, books.Finger in image needs to remove, and ensures that image is clear
It is clean.
The method used at present is the algorithm removal finger based on Face Detection, according to pre- in color space RGB, HSV
If threshold value, the color in image is judged.There is certain limitation using the above scheme, the method is for personal skin
Color can there are the phase advancing coloud nearsides of the colour of skin in different differences and image all cannot effectively distinguish.If imaging device is quick to light source
Sense, the imaging of different light sources may be multifarious, and in some cases, image quality is bad for current high photographing instrument, at this time hand
The Face Detection of finger can be judged by accident.The above method is a kind of colour of skin of acquiescence, and does not have colour of skin phase advancing coloud nearside, image image stabilization in image
Ideally, it is general in the case of other that finger Face Detection can all be caused to judge by accident, make finger in image that can not remove.
Invention content
It is current to solve the object of the present invention is to provide a kind of method and apparatus based on fixed area removal image finger
The easy detection error of finger in image, leads to the problem of finger can not remove in image.
In order to solve the above technical problems, as one aspect of the present invention, provide a kind of based on fixed area removal figure
As the method for finger, including：
Acquire the Rect area images of presumptive area in image；
The Rect area images are split as upper Rect regions subgraph and lower Rect regions subgraph；With the upper Rect
The image of region subgraph top edge presumptive area is upper artwork, and the upper artwork mapping is filled the upper Rect regions subgraph
Picture；Using the image of the lower Rect regions subgraph lower edge presumptive area as lower artwork, by the lower artwork mapping filling institute
State lower Rect regions subgraph；
Color penetration handles the upper Rect regions subgraph and the lower Rect regions subgraph intersection.
Further, it is described to split the Rect area images as upper Rect regions subgraph and lower Rect regions subgraph
Picture；The step of in, including：
By Rect area images by intermediate L lines be divided into upper and lower two contour size areas upper Rect regions subgraph and
Lower Rect regions subgraph；
The image using the upper Rect regions subgraph top edge presumptive area reflects the upper artwork as upper artwork
Penetrate the filling upper Rect regions subgraph；The step of in, including：
Determine the upper Rect regions subgraph top edge L1 lines, it is highly N1 regions to select on the top edge L1 lines
Image is upper artwork, and using the top edge L1 lines as mapping line of demarcation, the mapping of upper artwork is filled into upper Rect regions
In image；
If the distance between the top edge L1 lines and the intermediate L lines be less than or equal to N1, by upper artwork with institute
It states Rect regions subgraph and corresponds to the area maps of size and be filled into the upper Rect regions subgraph；
If the distance between the top edge L1 lines and the intermediate L lines be more than N1, by upper artwork according to it is described on
The region contiguous segmentation mapping that Rect regions subgraph corresponds to size is filled into the upper Rect regions subgraph.
Further, the image using the lower Rect regions subgraph lower edge presumptive area is lower artwork, by institute
It states lower artwork mapping and fills the lower Rect regions subgraph；The step of in, including：
Determine the lower Rect regions subgraph lower edge L2 lines, it is highly N1 regions to select under the lower edge L1 lines
Image is lower artwork, and using the lower edge L2 lines as mapping line of demarcation, the mapping of lower artwork is filled into lower Rect regions
In image；
If the distance between the lower edge L2 lines and the intermediate L lines be less than or equal to N1, by lower artwork with institute
It states lower Rect regions subgraph and corresponds to the area maps of size and be filled into the lower Rect regions subgraph；
If the distance between the lower edge L1 lines and the intermediate L lines be more than N1, by lower artwork according to it is described under
The region contiguous segmentation mapping that Rect regions subgraph corresponds to size is filled into the lower Rect regions subgraph.
Further, the color penetration handles the upper Rect regions subgraph and the lower Rect regions subgraph is handed over
At boundary, the step of include：
The original Fill Color value weights of Rect1 are set as weights Q1, Rect2 upwardly extends the color value power of intended pixel
Weight is weights Q2；The sum of weights Q1 and weights Q2 remain 1, and weights Q1 and weights Q2 are at the intermediate L lines
0.5, the transformation of weights Q1 and weights Q2 can be calculated by equation below：
Q1=0.5+F*i,
Q2=0.5-F*i,
Wherein, i is the distance away from the intermediate L lines, and as unit of pixel, F is the amplitude of weight variation, and F can pass through
Equation below calculates：
F=0.5/N,
Wherein N is extension number of pixels；
Image intersection transitional region thinks that color color value is calculated by formula below：
Value (i, j)=Q1*Value1 (i, j)+Q2*Value2 (i, j)；
It is color value new at pixel position (i, j) that wherein the sum of weights Q1 and weights Q2, which are equal to 1, Value (i, j),
Value1 is the value of upper Rect regions sub-image section projection filling, and Q1 is respective weights value；Value2 is lower Rect regions
The value of image section projection filling, Q2 respective weights values.
Further, in the acquisition image presumptive area Rect area images, in step,；
The presumptive area is the image comprising finger areas.
As another aspect of the present invention, the present invention also provides a kind of device based on fixed area removal image finger,
Including：
Collecting unit is configured to, and acquires the Rect area images of presumptive area in image；
Map unit is split, is configured to, splits the Rect area images as upper Rect regions subgraph and lower Rect
Region subgraph；Using the image of the upper Rect regions subgraph top edge presumptive area as upper artwork, the upper artwork is reflected
Penetrate the filling upper Rect regions subgraph；Using the image of the lower Rect regions subgraph lower edge presumptive area as lower original
The lower artwork mapping is filled the lower Rect regions subgraph by figure；
Permeation unit, color penetration handles the upper Rect regions subgraph and the lower Rect regions subgraph has a common boundary
Place.
Further, the fractionation map unit includes：
By Rect area images by intermediate L lines be divided into upper and lower two contour size areas upper Rect regions subgraph and
Lower Rect regions subgraph；
Determine the upper Rect regions subgraph top edge L1 lines, it is highly N1 regions to select on the top edge L1 lines
Image is upper artwork, and using the top edge L1 lines as mapping line of demarcation, the mapping of upper artwork is filled into upper Rect regions
In image；
If the distance between the top edge L1 lines and the intermediate L lines be less than or equal to N1, by upper artwork with institute
It states Rect regions subgraph and corresponds to the area maps of size and be filled into the upper Rect regions subgraph；
If the distance between the top edge L1 lines and the intermediate L lines be more than N1, by upper artwork according to it is described on
The region contiguous segmentation mapping that Rect regions subgraph corresponds to size is filled into the upper Rect regions subgraph.
Further, the fractionation map unit further includes：
Determine the lower Rect regions subgraph lower edge L2 lines, it is highly N1 regions to select under the lower edge L1 lines
Image is lower artwork, and using the lower edge L2 lines as mapping line of demarcation, the mapping of lower artwork is filled into lower Rect regions
In image；
If the distance between the lower edge L2 lines and the intermediate L lines be less than or equal to N1, by lower artwork with institute
It states lower Rect regions subgraph and corresponds to the area maps of size and be filled into the lower Rect regions subgraph；
If the distance between the lower edge L1 lines and the intermediate L lines be more than N1, by lower artwork according to it is described under
The region contiguous segmentation mapping that Rect regions subgraph corresponds to size is filled into the lower Rect regions subgraph.
Further, the permeation unit includes：
The original Fill Color value weights of Rect1 are set as weights Q1, Rect2 upwardly extends the color value power of intended pixel
Weight is weights Q2；The sum of weights Q1 and weights Q2 remain 1, and weights Q1 and weights Q2 are at the intermediate L lines
0.5, the transformation of weights Q1 and weights Q2 can be calculated by equation below：
Q1=0.5+F*i,
Q2=0.5-F*i,
Wherein, i is the distance away from the intermediate L lines, and as unit of pixel, F is the amplitude of weight variation, and F can pass through
Equation below calculates：
F=0.5/N,
Wherein N is extension number of pixels；
Image intersection transitional region thinks that color color value is calculated by formula below：
Value (i, j)=Q1*Value1 (i, j)+Q2*Value2 (i, j)；
It is color value new at pixel position (i, j) that wherein the sum of weights Q1 and weights Q2, which are equal to 1, Value (i, j),
Value1 is the value of upper Rect regions sub-image section projection filling, and Q1 is respective weights value；Value2 is lower Rect regions
The value of image section projection filling, Q2 respective weights values.
Further, the collecting unit includes：
The presumptive area is the image comprising finger areas.
Compared with prior art, beneficial aspects of the invention are：
Invention provides a kind of method based on fixed area removal image finger, including：Acquire presumptive area in image
Rect area images；The Rect area images are split as upper Rect regions subgraph and lower Rect regions subgraph；With described
The image of upper Rect regions subgraph upper edge region is artwork, fills the upper Rect regions subgraph；With the lower Rect
The image of region subgraph lower edge margin is artwork, fills the lower Rect regions subgraph；On color penetration processing is described
Rect regions subgraph and the lower Rect regions subgraph intersection.Frame choosing limitation, Ran Hou are first carried out to the position of finger
Projection filling is carried out to the range of frame choosing, color penetration is carried out mutually in the stitching portion to projecting up and down, so as to fulfill stitching portion
Smoothing processing.Using the above scheme, no matter we acquired under which kind of environment image, not be limited by fixed colour of skin finger by
Pressure may be implemented in the mapping filling in frame favored area, realize the versatility and accuracy of removal finger.
Description of the drawings
Fig. 1 diagrammatically illustrates a kind of method flow diagram based on fixed area removal image finger of the embodiment of the present invention；
The edge that Fig. 2 diagrammatically illustrates image in the embodiment of the present invention divides schematic diagram；
Fig. 3 diagrammatically illustrates mapping filling flow diagram of the embodiment of the present invention；
Fig. 4 diagrammatically illustrates a kind of apparatus structure signal based on fixed area removal image finger of the embodiment of the present invention
Figure：
Specific embodiment
The embodiment of the present invention is described in detail below, but what the present invention can be defined by the claims and cover
Multitude of different ways is implemented.
As shown in Figure 1, the embodiment of the present invention provides a kind of method based on fixed area removal image finger, it is specific to flow
Journey includes：
Step 10, the Rect area images of presumptive area in image are acquired；
Step 20, it is upper Rect regions subgraph and lower Rect regions subgraph to split Rect area images；More than Rect
The image of region subgraph top edge presumptive area is upper artwork, by the upper Rect regions subgraph of upper artwork mapping filling；Below
The image of Rect regions subgraph lower edge presumptive area is lower artwork, by the lower Rect regions subgraph of lower artwork mapping filling；
Step 30, the upper Rect regions subgraph of color penetration processing and lower Rect regions subgraph intersection.
Wherein, in step 20, horizontal mid-line is carried out to Rect area images and is divided into Rect regions subgraph on two parts
Picture and lower Rect regions subgraph.It is respectively adopted in the upper portion mapping filling of top edge L1 lines in the subgraph of Rect regions
Rect regions subgraph；The lower portion mapping of lower edge L2 lines is filled into lower Rect regions subgraph in lower Rect regions subgraph
As in.It is equivalent to Rect regions subgraph and lower Rect regions subgraph is spliced into Rect area images.
Frame choosing limitation is first carried out to the position of finger, then projection filling is carried out in the range selected frame, to throwing up and down
The stitching portion of shadow carries out mutually color penetration, so as to fulfill stitching portion smoothing processing.Using the above scheme, no matter we are at which kind of
Image is acquired under environment, not in the pressing for being limited by fixed colour of skin finger, may be implemented in the mapping filling in frame favored area,
Realize the versatility and accuracy of removal finger.
Wherein, in step 10, presumptive area is the image comprising finger areas.
As shown in Figures 2 and 3, step 20 specifically includes：
Step 21, Rect regions are divided into the upper Rect regions subgraph of upper and lower two contour size areas by intermediate L lines
Picture and lower Rect regions subgraph；
Rect area images are divided into two in the intermediate L lines of Rect area image vertical direction, respectively upper Rect areas
Domain subgraph and lower Rect regions subgraph.
Step 22, upper Rect regions subgraph top edge L1 lines are determined, it is highly N1 regions to select on top edge L1 lines
Image is upper artwork, and using top edge L1 lines as mapping line of demarcation, the mapping of upper artwork is filled into the subgraph of upper Rect regions；
Step 23, lower Rect regions subgraph lower edge L2 lines are determined, it is highly N1 regions to select under lower edge L1 lines
Image is lower artwork, and using lower edge L2 lines as mapping line of demarcation, the mapping of lower artwork is filled into the subgraph of lower Rect regions；
As shown in Fig. 2, determine upper Rect regions subgraph filling range, the filling of upper Rect regions subgraph (Rect1)
Between top edge L1 lines and L+N, lower Rect regions subgraph (Rect2) filling range, lower Rect regions subgraph are determined
Filling be between lower edge L2 lines and L-N, respectively upper Rect regions subgraph and lower Rect regions subgraph have height
Overlapping region for N.
In order to ensure finally to map the continuity of filling on the image, need setting that can map the range of filling, such as Fig. 2 institutes
Show, the range away from N1 can all map filling on the downside of the range and L2 on the upside of L1 away from N1.It is preferred that 100 pixel distances.
Step 24, if the distance between top edge L1 lines and centre L lines be less than or equal to N1, by upper artwork with it is upper
The area maps that Rect regions subgraph corresponds to size are filled into the subgraph of Rect regions；If top edge L1 lines and centre
The distance between L lines are more than N1, by upper artwork according to the region contiguous segmentation mapping of size corresponding with upper Rect regions subgraph
It is filled into the subgraph of Rect regions.
Step 25, if the distance between lower edge L2 lines and centre L lines be less than or equal to N1, by lower artwork under
The area maps that Rect regions subgraph corresponds to size are filled into the subgraph of lower Rect regions；If lower edge L1 lines and centre
The distance between L lines are more than N1, by lower artwork according to the region contiguous segmentation mapping of size corresponding with lower Rect regions subgraph
It is filled into the subgraph of lower Rect regions.
Contiguous segmentation is mapped to the step in the subgraph of Rect regions specifically, more than edge line is mapped to for symmetry axis
After in the subgraph of upper Rect regions, the following boundary line just mapped in the subgraph of continuation more than Rect regions continues down for symmetry axis
Symmetrical mapping, Rect regions subgraph intersection continuously smooth in guarantee, until will all be mapped in the subgraph of upper Rect regions
Terminate.
Lower Rect regions subgraph fill method is substantially consistent with upper Rect regions subgraph fill method, ensures filling
Image transition is smooth in region, and contiguous segmentation is mapped to the step in the subgraph of lower Rect regions specifically, following edge line is
After symmetry axis is mapped in the subgraph of lower Rect regions, it is pair to continue the upper border line just mapped in the subgraph of following Rect regions
Axis is claimed to continue up symmetrical mapping, ensures the intersection continuously smooth in lower Rect regions subgraph, until by lower Rect regions
All mapping terminates in subgraph.
As shown in Fig. 2, using top edge L1 lines as line of demarcation, using N1 position on the upside of top edge L1 lines fill L1~L it
Between image.
If the distance between L1~L be less than or equal to N1, can in artwork according to L1 symmetrically by the region between L1~L
Mapping filling.
If the distance between L1~L is more than N1, need to carry out subsection compression filling.D for filling position away from L2 away from
From.
D<In the range of=N1, by the way of symmetrical mapping filling.N1<D<In the range of 2*N1, the position that maps at this time
The extreme position that has reached and can map is put, if continuing mapping filling, needs the inverted sequence in the artwork of mapping, this ensure that
Image inverted sequence position seamlessly transits, but image final in this way has apparent linear marking, preferably N1=100 pixel,
Apparent strip lines can preferably be avoided.
Above-mentioned principle successively is filled lower Rect regions subgraph.
Step 30 specifically includes：
Rect area images mapping filling at this time is equivalent to the upper half using upper Rect regions subgraph top edge L1 lines
Part and the lower half portion of lower Rect regions subgraph lower edge L2 lines are spliced, under normal circumstances, can color in stitching portion
Have apparent jump.It is needed at this time to stitching portion smoothing processing, extended separately down in N number of pixel in intermediate L line positions
Subgraphs filling in Rect regions upwardly extends Rect regions subgraph under N number of pixel and fills, preferably N=20 pixel and
In the range of Rect area images.There is two-part mapped color value in the range of extension, be added to obtain using different weights
New color value.Specific steps are as follows：
By taking the overlay region of upward 20 pixels of intermediate L lines as an example, the original Fill Color value weights of Rect1 are set as weights
The color value weight that Q1, Rect2 upwardly extend intended pixel is weights Q2；The sum of weights Q1 and weights Q2 remain 1, power
Value Q1 and weights Q2 is 0.5 at intermediate L lines, and the transformation of weights Q1 and weights Q2 can be calculated by equation below：
Q1=0.5+F*i；
Q2=0.5-F*i；
Wherein, i is the distance away from intermediate L lines, and as unit of pixel, F is the amplitude of weight variation, and F can be by as follows
Formula calculates：
F=0.5/N；
Wherein N is extension number of pixels；
Image intersection transitional region thinks that color color value is calculated by formula below：
Value (i, j)=Q1*Value1 (i, j)+Q2*Value2 (i, j)；
It is color value new at pixel position (i, j) that wherein the sum of weights Q1 and weights Q2, which are equal to 1, Value (i, j),
Value1 is the value of upper Rect regions sub-image section projection filling, and Q1 is respective weights value；Value2 is lower Rect regions
The value of image section projection filling, Q2 respective weights values.
Another aspect according to embodiments of the present invention removes image finger the present invention also provides a kind of based on fixed area
Device, as shown in figure 4, including：
Collecting unit is configured to, and acquires the Rect area images of presumptive area in image；
Map unit is split, is configured to, it is upper Rect regions subgraph and lower Rect regions to split Rect area images
Subgraph；The image of more than Rect regions subgraph top edge presumptive area is upper artwork, by the upper Rect of upper artwork mapping filling
Region subgraph；The image of following Rect regions subgraph lower edge presumptive area is lower artwork, will be under the mapping filling of lower artwork
Rect regions subgraph；
Permeation unit, the upper Rect regions subgraph of color penetration processing and lower Rect regions subgraph intersection.
The fractionation map unit includes：
Rect regions are divided into the upper Rect regions subgraph of upper and lower two contour size areas under by intermediate L lines
Rect regions subgraph；
Upper Rect regions subgraph top edge L1 lines are determined, on the image on selected top edge L1 lines being highly N1 regions is
Using top edge L1 lines as mapping line of demarcation, the mapping of upper artwork is filled into the subgraph of upper Rect regions for artwork；
If the distance between top edge L1 lines and centre L lines be less than or equal to N1, by upper artwork with upper Rect regions
The area maps that subgraph corresponds to size are filled into the subgraph of Rect regions；
If the distance between top edge L1 lines and centre L lines be more than N1, by upper artwork according to upper Rect regions subgraph
As the region contiguous segmentation mapping of corresponding size is filled into the subgraph of Rect regions.
The fractionation map unit further includes：
Lower Rect regions subgraph lower edge L2 lines are determined, under the image under selected lower edge L1 lines being highly N1 regions is
Using lower edge L2 lines as mapping line of demarcation, the mapping of lower artwork is filled into the subgraph of lower Rect regions for artwork；
If the distance between lower edge L2 lines and centre L lines be less than or equal to N1, by lower artwork with lower Rect regions
The area maps that subgraph corresponds to size are filled into the subgraph of lower Rect regions；
If the distance between lower edge L1 lines and centre L lines be more than N1, by lower artwork according to lower Rect regions subgraph
As the region contiguous segmentation mapping of corresponding size is filled into the subgraph of lower Rect regions.
The permeation unit includes：
The original Fill Color value weights of Rect1 are set as weights Q1, Rect2 upwardly extends the color value power of intended pixel
Weight is weights Q2；The sum of weights Q1 and weights Q2 remain 1, and weights Q1 and weights Q2 are 0.5 at intermediate L lines, power
The transformation of value Q1 and weights Q2 can be calculated by equation below：
Q1=0.5+F*i,
Q2=0.5-F*i,
Wherein, i is the distance away from intermediate L lines, and as unit of pixel, F is the amplitude of weight variation, and F can be by as follows
Formula calculates：
F=0.5/N,
Wherein N is extension number of pixels；
Image intersection transitional region thinks that color color value is calculated by formula below：
Value (i, j)=Q1*Value1 (i, j)+Q2*Value2 (i, j)；
It is color value new at pixel position (i, j) that wherein the sum of weights Q1 and weights Q2, which are equal to 1, Value (i, j),
Value1 is the value of upper Rect regions sub-image section projection filling, and Q1 is respective weights value；Value2 is lower Rect regions
The value of image section projection filling, Q2 respective weights values.
The collecting unit includes：
Presumptive area is the image comprising finger areas.
It these are only the preferred embodiment of the present invention, be not intended to restrict the invention, for those skilled in the art
For ability person, the invention may be variously modified and varied.All within the spirits and principles of the present invention, that is made any repaiies
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (10)
- A kind of 1. method based on fixed area removal image finger, which is characterized in that including：Acquire the Rect area images of presumptive area in image；The Rect area images are split as upper Rect regions subgraph and lower Rect regions subgraph；With the upper Rect regions The image of subgraph top edge presumptive area is upper artwork, and the upper artwork mapping is filled the upper Rect regions subgraph； Using the image of the lower Rect regions subgraph lower edge presumptive area as lower artwork, by the lower artwork mapping filling it is described under Rect regions subgraph；Color penetration handles the upper Rect regions subgraph and the lower Rect regions subgraph intersection.
- 2. the method according to claim 1 based on fixed area removal image finger, which is characterized in that the fractionation institute Rect area images are stated as upper Rect regions subgraph and lower Rect regions subgraph；The step of in, including：Rect area images are divided into the upper Rect regions subgraph of upper and lower two contour size areas under by intermediate L lines Rect regions subgraph；The image using the upper Rect regions subgraph top edge presumptive area fills out the upper artwork mapping as upper artwork Fill the upper Rect regions subgraph；The step of in, including：Determine the upper Rect regions subgraph top edge L1 lines, it is highly the image in N1 regions to select on the top edge L1 lines For upper artwork, using the top edge L1 lines as mapping line of demarcation, the mapping of upper artwork is filled into the upper Rect regions subgraph It is interior；If the distance between the top edge L1 lines and the intermediate L lines be less than or equal to N1, by upper artwork with it is described on The area maps that Rect regions subgraph corresponds to size are filled into the upper Rect regions subgraph；If the distance between the top edge L1 lines and the intermediate L lines be more than N1, by upper artwork according to the upper Rect The region contiguous segmentation mapping that region subgraph corresponds to size is filled into the upper Rect regions subgraph.
- 3. the method based on fixed area removal image finger according to right wants 2, which is characterized in that described under described The image of Rect regions subgraph lower edge presumptive area is lower artwork, and the lower artwork mapping is filled the lower Rect regions Subgraph；The step of in, including：Determine the lower Rect regions subgraph lower edge L2 lines, it is highly the image in N1 regions to select under the lower edge L1 lines For lower artwork, using the lower edge L2 lines as mapping line of demarcation, the mapping of lower artwork is filled into the lower Rect regions subgraph It is interior；If the distance between the lower edge L2 lines and the intermediate L lines be less than or equal to N1, by lower artwork with it is described under The area maps that Rect regions subgraph corresponds to size are filled into the lower Rect regions subgraph；If the distance between the lower edge L1 lines and the intermediate L lines be more than N1, by lower artwork according to the lower Rect The region contiguous segmentation mapping that region subgraph corresponds to size is filled into the lower Rect regions subgraph.
- 4. the method according to claim 3 based on fixed area removal image finger, which is characterized in that the color is oozed Handle the upper Rect regions subgraph and the lower Rect regions subgraph intersection thoroughly, the step of include：The original Fill Color value weights of Rect1 are set as weights Q1, the color value weight that Rect2 upwardly extends intended pixel is Weights Q2；The sum of weights Q1 and weights Q2 remain 1, and weights Q1 and weights Q2 are 0.5 at the intermediate L lines, power The transformation of value Q1 and weights Q2 can be calculated by equation below：Q1=0.5+F*i,Q2=0.5-F*i,Wherein, i is the distance away from the intermediate L lines, and as unit of pixel, F is the amplitude of weight variation, and F can be by as follows Formula calculates：F=0.5/N,Wherein N is extension number of pixels；Image intersection transitional region thinks that color color value is calculated by formula below：Value (i, j)=Q1*Value1 (i, j)+Q2*Value2 (i, j)；It is color value new at pixel position (i, j) that wherein the sum of weights Q1 and weights Q2, which are equal to 1, Value (i, j), Value1 is the value of upper Rect regions sub-image section projection filling, and Q1 is respective weights value；Value2 is lower Rect regions The value of image section projection filling, Q2 respective weights values.
- 5. the method according to claim 4 based on fixed area removal image finger, which is characterized in that the acquisition figure The Rect area images of presumptive area as in, in step,；The presumptive area is the image comprising finger areas.
- 6. a kind of device based on fixed area removal image finger, which is characterized in that including：Collecting unit is configured to, and acquires the Rect area images of presumptive area in image；Map unit is split, is configured to, splits the Rect area images as upper Rect regions subgraph and lower Rect regions Subgraph；Using the image of the upper Rect regions subgraph top edge presumptive area as upper artwork, the upper artwork mapping is filled out Fill the upper Rect regions subgraph；It, will using the image of the lower Rect regions subgraph lower edge presumptive area as lower artwork The lower Rect regions subgraph is filled in the lower artwork mapping；Permeation unit, color penetration handle the upper Rect regions subgraph and the lower Rect regions subgraph intersection.
- 7. the device according to claim 6 based on fixed area removal image finger, which is characterized in that the fractionation is reflected Unit is penetrated to include：Rect area images are divided into the upper Rect regions subgraph of upper and lower two contour size areas under by intermediate L lines Rect regions subgraph；Determine the upper Rect regions subgraph top edge L1 lines, it is highly the image in N1 regions to select on the top edge L1 lines For upper artwork, using the top edge L1 lines as mapping line of demarcation, the mapping of upper artwork is filled into the upper Rect regions subgraph It is interior；If the distance between the top edge L1 lines and the intermediate L lines be less than or equal to N1, by upper artwork with it is described on The area maps that Rect regions subgraph corresponds to size are filled into the upper Rect regions subgraph；If the distance between the top edge L1 lines and the intermediate L lines be more than N1, by upper artwork according to the upper Rect The region contiguous segmentation mapping that region subgraph corresponds to size is filled into the upper Rect regions subgraph.
- 8. the device according to claim 7 based on fixed area removal image finger, which is characterized in that the fractionation is reflected Unit is penetrated to further include：Determine the lower Rect regions subgraph lower edge L2 lines, it is highly the image in N1 regions to select under the lower edge L1 lines For lower artwork, using the lower edge L2 lines as mapping line of demarcation, the mapping of lower artwork is filled into the lower Rect regions subgraph It is interior；If the distance between the lower edge L2 lines and the intermediate L lines be less than or equal to N1, by lower artwork with it is described under The area maps that Rect regions subgraph corresponds to size are filled into the lower Rect regions subgraph；If the distance between the lower edge L1 lines and the intermediate L lines be more than N1, by lower artwork according to the lower Rect The region contiguous segmentation mapping that region subgraph corresponds to size is filled into the lower Rect regions subgraph.
- 9. the device according to claim 8 based on fixed area removal image finger, which is characterized in that the infiltration is single Member includes：The original Fill Color value weights of Rect1 are set as weights Q1, the color value weight that Rect2 upwardly extends intended pixel is Weights Q2；The sum of weights Q1 and weights Q2 remain 1, and weights Q1 and weights Q2 are 0.5 at the intermediate L lines, power The transformation of value Q1 and weights Q2 can be calculated by equation below：Q1=0.5+F*i,Q2=0.5-F*i,Wherein, i is the distance away from the intermediate L lines, and as unit of pixel, F is the amplitude of weight variation, and F can be by as follows Formula calculates：F=0.5/N,Wherein N is extension number of pixels；Image intersection transitional region thinks that color color value is calculated by formula below：Value (i, j)=Q1*Value1 (i, j)+Q2*Value2 (i, j)；It is color value new at pixel position (i, j) that wherein the sum of weights Q1 and weights Q2, which are equal to 1, Value (i, j), Value1 is the value of upper Rect regions sub-image section projection filling, and Q1 is respective weights value；Value2 is lower Rect regions The value of image section projection filling, Q2 respective weights values.
- 10. the device according to claim 9 based on fixed area removal image finger, which is characterized in that the acquisition Unit includes：The presumptive area is the image comprising finger areas.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810100143.8A CN108257082B (en) | 2018-02-01 | 2018-02-01 | Method and device for removing image fingers based on fixed area |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810100143.8A CN108257082B (en) | 2018-02-01 | 2018-02-01 | Method and device for removing image fingers based on fixed area |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108257082A true CN108257082A (en) | 2018-07-06 |
| CN108257082B CN108257082B (en) | 2021-08-17 |
Family
ID=62743213
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810100143.8A Active CN108257082B (en) | 2018-02-01 | 2018-02-01 | Method and device for removing image fingers based on fixed area |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108257082B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021121302A1 (en) * | 2019-12-19 | 2021-06-24 | 华为技术有限公司 | Video collection control method, electronic device, and computer-readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103886561A (en) * | 2014-04-09 | 2014-06-25 | 武汉科技大学 | Criminisi image inpainting method based on mathematical morphology |
| US8952979B2 (en) * | 2012-09-19 | 2015-02-10 | Autodesk, Inc. | Wave fill |
| CN105898322A (en) * | 2015-07-24 | 2016-08-24 | 乐视云计算有限公司 | Video watermark removing method and device |
| CN106408952A (en) * | 2016-12-14 | 2017-02-15 | 浙江工业大学 | Motor vehicle violation shooting system and method |
-
2018
- 2018-02-01 CN CN201810100143.8A patent/CN108257082B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8952979B2 (en) * | 2012-09-19 | 2015-02-10 | Autodesk, Inc. | Wave fill |
| CN103886561A (en) * | 2014-04-09 | 2014-06-25 | 武汉科技大学 | Criminisi image inpainting method based on mathematical morphology |
| CN105898322A (en) * | 2015-07-24 | 2016-08-24 | 乐视云计算有限公司 | Video watermark removing method and device |
| CN106408952A (en) * | 2016-12-14 | 2017-02-15 | 浙江工业大学 | Motor vehicle violation shooting system and method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021121302A1 (en) * | 2019-12-19 | 2021-06-24 | 华为技术有限公司 | Video collection control method, electronic device, and computer-readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108257082B (en) | 2021-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR100981003B1 (en) | Moving body image extraction apparatus and computer readable storage medium storing program | |
| EP3104332A1 (en) | Digital image manipulation | |
| EP1895469A2 (en) | Image display apparatus, image display method, and image display program | |
| JP5810628B2 (en) | Image processing apparatus and image processing program | |
| CN102484724A (en) | Projection image area detecting device | |
| US8334880B2 (en) | Image processing apparatus | |
| US20140348428A1 (en) | Dynamic range-adjustment apparatuses and methods | |
| JP2014195148A (en) | Image processing apparatus, region determination method, and computer program | |
| KR20160110038A (en) | Image processing apparatus and image processing method | |
| US9990701B2 (en) | Image-processing apparatus, image-processing method, and computer program product | |
| CN104361335B (en) | A kind of processing method that black surround is automatically removed based on scan image | |
| TW201630408A (en) | Image data segmentation | |
| CN103852034A (en) | Elevator guide rail perpendicularity detection method | |
| JP5193107B2 (en) | Concrete image segmentation method | |
| CN110782470B (en) | Carpal bone region segmentation method based on shape information | |
| JP2009037596A (en) | Information processing device and program | |
| CN108257082A (en) | Method and apparatus based on fixed area removal image finger | |
| JP6045429B2 (en) | Imaging apparatus, image processing apparatus, and image processing method | |
| JP2021099749A (en) | Detection method of nasolabial folds | |
| JP6643301B2 (en) | Defect inspection device and defect inspection method | |
| JP3733686B2 (en) | Image processing method and apparatus | |
| JP2004219291A (en) | Line defect detection method and device for screen | |
| KR102634726B1 (en) | Motion estimation method, display device and method for driving display device using the same | |
| JP2014225753A (en) | Marker detection device, marker detection method, and program | |
| JP2022077874A (en) | Information processing apparatus, and information processing program |
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 |