CN104504668B - Image containing face adds sharp method and device - Google Patents
Image containing face adds sharp method and device Download PDFInfo
- Publication number
- CN104504668B CN104504668B CN201410848840.3A CN201410848840A CN104504668B CN 104504668 B CN104504668 B CN 104504668B CN 201410848840 A CN201410848840 A CN 201410848840A CN 104504668 B CN104504668 B CN 104504668B
- Authority
- CN
- China
- Prior art keywords
- image
- sharp
- pending
- plus
- replaced
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The invention discloses a kind of images containing face to add sharp method, includes the following steps：Obtain pending image；When detecting facial image in the pending image, determine the human face region in the pending image as first area；The facial image is extracted from the first area as the first image, and the pending image is carried out plus sharp processing, generation first add sharp image, determines in described first plus sharp image and corresponds to the image of the first area as the second image；Second image in described first plus sharp image is replaced with into described first image, using replaced first plus sharp image as target image.The invention also discloses a kind of images containing face to add sharp device.The present invention improves plus sharp effect the image fidelity of face part, improves the Experience Degree of user.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to the image containing face adds sharp method and device.
Background technology
Acutance is the index for the sharp keen degree of clarity and image border for reflecting the plane of delineation.In existing image procossing
In algorithm, in order to improve the whole clarity of image, generally whole plus sharp processing method is carried out using to image so that image
Detail contrast higher in plane, it appears that become apparent from.But when containing facial image in pending image, add sharp
Processing operation can make face line-dividing apparent, and face contour is stiff unnatural so that the image fault of face part is serious,
Cause plus sharp effect is poor, so that the Experience Degree of user is poor.
The above is only used to facilitate the understanding of the technical scheme, and is not represented and is recognized that the above is existing skill
Art.
Invention content
The main purpose of the present invention is to provide a kind of images containing face to add sharp method and device, it is intended to solve to work as and wait for
When containing facial image in the image of processing, add sharp processing operation that face line-dividing can be made apparent, face contour is stiff not
It is natural so that the image fault of face point is serious, causes plus sharp effect is poor, so that the problem of the Experience Degree difference of user.
To achieve the above object, the present invention provides a kind of image containing face and adds sharp method, including step：
Obtain pending image；
When detecting facial image in the pending image, the human face region conduct in the pending image is determined
First area；
The facial image is extracted from the first area as the first image,
The pending image is carried out plus sharp processing, generation first add sharp image, is determined in described first plus sharp image
The image of the corresponding first area is as the second image；
Second image in described first plus sharp image is replaced with into described first image, replaced first is added
Sharp image is as target image.
Preferably, it after the step of human face region in the determination pending image is as first area, also wraps
It includes：
The region in the pending image in addition to the human face region is determined as second area, and extracts described
The corresponding image in two regions is as third image；
The third image is carried out plus sharp processing, generation second add sharp image；
The third image in the pending image is replaced with into described second plus sharp image, waits locating by replaced
Image is managed as target image.
Preferably, second image by described first plus sharp image replaces with described first image, will replace
First plus sharp image after changing include as the step of target image：
Obtain the pixel point coordinates and the corresponding color of the pixel of the second image described in described first plus sharp image
Value；
The corresponding color value of pixel acquired in described first plus sharp image is replaced with described first image to correspond to
The color value of pixel, using replaced first plus sharp image as target image.
Preferably, described using replaced first plus sharp image as after the step of target image, further include：
The fringe region for determining the first image described in the target image, to the color of pixel in the fringe region
Value carries out interpolation processing；
Denoising is carried out to the target image after interpolation processing and obtains the target image after denoising；
Preserve the target image after the denoising.
Preferably, after the step of acquisition pending image, further include：
When facial image is not detected in the pending image, the pending image is carried out to add sharp processing and life
At target image.
In addition, to achieve the above object, the present invention also provides a kind of images containing face to add sharp device, including：
Acquisition module, for obtaining pending image；
Determining module when for detecting facial image in the pending image, determines in the pending image
Human face region as first area；
Extraction module, for extracting the facial image from the first area as the first image；
Add sharp module, adds sharp processing for being carried out to the pending image, generate first plus sharp image；
Replacement module corresponds to the image of the first area as the second figure for determining in described first plus sharp image
Second image in described first plus sharp image is replaced with described first image by picture, by replaced first plus sharp figure
As being used as target image.
Preferably, the extraction module is additionally operable to determine the area in the pending image in addition to the human face region
The corresponding image of the second area is extracted as third image as second area in domain；
Described plus sharp module is additionally operable to carry out the third image plus sharp processing, generation second adds sharp image；
The replacement module is additionally operable to the third image in the pending image replacing with described second plus sharp figure
Picture, using replaced pending image as target image.
Preferably, the replacement module includes acquiring unit and replacement unit,
The acquiring unit, for obtaining the pixel point coordinates of the second image described in described first plus sharp image and described
The corresponding color value of pixel；
The replacement unit, for the corresponding color value of pixel acquired in described first plus sharp image to be replaced with institute
The color value for stating the first image corresponding pixel points, using replaced first plus sharp image as target image.
Preferably, the image processing apparatus containing face further includes：
Processing module, the fringe region for determining the first image described in the target image, to the fringe region
The color value of pixel carries out interpolation processing；It is additionally operable to obtain at denoising the target image progress denoising after interpolation processing
Target image after reason；
Memory module, for preserving the target image after the denoising.
Preferably, described plus sharp module, when being additionally operable to that facial image is not detected in the pending image, to described
Pending image carries out adding sharp processing and generates target image.
The present invention is by adding sharp processing procedure to the image containing face, extracting the face figure in pending image
Picture, with extract without add the facial image of sharp processing replace plus be sharp after pending image in facial image, protect
The image sharpness of face part is stayed.It effectively avoids, when containing facial image in pending image, adding sharp processing operation meeting
So that face line-dividing is apparent, face contour is stiff unnatural so that the image fault of face part is serious, causes to add sharp effect
Fruit is poor, so that the problem of the Experience Degree difference of user.It to the image fidelity of face part, and then improves and adds sharp effect, carry
The high Experience Degree of user.
Description of the drawings
Fig. 1 is the flow diagram for the first embodiment that the image of the invention containing face adds sharp method；
Fig. 2 is the refinement flow diagram of mono- embodiments of step S50 in Fig. 1；
Fig. 3 is the flow diagram for the second embodiment that the image of the invention containing face adds sharp method；
Fig. 4 is the flow diagram for the 3rd embodiment that the image of the invention containing face adds sharp method；
Fig. 5 is the high-level schematic functional block diagram for the first embodiment that the image of the invention containing face adds sharp device；
Fig. 6 is the refinement high-level schematic functional block diagram of one embodiment of replacement module in Fig. 5；
Fig. 7 is the high-level schematic functional block diagram for the second embodiment that the image of the invention containing face adds sharp device.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific implementation mode
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The primary solutions of the embodiment of the present invention are：Obtain pending image；It is detected in the pending image
When facial image, determine the human face region in the pending image where facial image as first area；From described first
The facial image is extracted in region as the first image, and the pending image is carried out plus sharp processing, generation first add sharp
Image determines in described first plus sharp image and corresponds to the image of the first area as the second image；Described first is added sharp
Second image in image replaces with described first image, using replaced first plus sharp image as target image.It is logical
It crosses in adding sharp processing procedure to the image containing face, extracts the facial image in pending image, with extracting not
The facial image in pending image after adding the facial image of sharp processing to replace plus is sharp, remains the image of face part
Acutance.It effectively avoids, when containing facial image in pending image, adding sharp processing operation that face line-dividing can be made bright
Aobvious, face contour is stiff unnatural so that the image fault of face part is serious, causes plus sharp effect is poor, so that user
Experience Degree difference problem.It to the image fidelity of face part, and then improves and adds sharp effect, improve the Experience Degree of user.
Due to, when containing facial image in pending image, adding sharp processing operation meeting under existing plus sharp processing mode
So that face line-dividing is apparent, face contour is stiff unnatural so that the image fault of face part is serious, causes to add sharp effect
Fruit is poor, so that the Experience Degree of user is poor.
Based on the above issues, the present invention provides a kind of image containing face and adds sharp method.
Referring to Fig.1, Fig. 1 is the flow diagram for the first embodiment that the image of the invention containing face adds sharp method.
In one embodiment, the image containing face adds the sharp method to include：
Step S10 obtains pending image；
The pending image can be carried out in image processing process plus the intermediate image of sharp processing, also can be with
It is the image of the camera induction of picture pick-up device by carrying out in front-end image processing procedure plus the middle graph of sharp processing
Picture.
Step S20 when detecting facial image in the pending image, determines the face in the pending image
Region is as first area；
Detection to facial image can be identified by the colour contrast of background environment and face complexion, determine institute
It states in pending image and whether contains facial image, the people in the pending image where facial image is returned if containing
The coordinate value in face region；It can also be judged by calculating pending image and the matching degree of preset standard faces template
Whether in the pending image contain facial image, face in the pending image is returned if containing facial image
The coordinate value of human face region where image；Alternatively, the character modules by comparing pending image and preset face and face
Type is matched, and judges whether contain facial image in the pending image, is waited for described in return if containing facial image
Handle the coordinate value of the human face region in image where facial image.
Step S30 extracts the facial image from the human face region；
According to the coordinate value of determining human face region, the pixel point coordinates and corresponding pixel points of the facial image are extracted
Color value, and the pixel point coordinates of the facial image and the corresponding color value data of the pixel are stored in interim storage
Area.
Step S40, carries out the pending image plus sharp processing, generation first add sharp image, determines that described first adds
The image of the first area is corresponded in sharp image as the second image；
Sharp processing is added to the pending image, high-pass filtering processing can be utilized to enhance the sharp of the pending image
Degree retains the radio-frequency component of the pending image so that described to wait for by the low-frequency component in the pending image of decaying
The acutance for handling image improves, and generates first plus sharp image；Unsharp masking processing can also be utilized to enhance the pending figure
The acutance of picture by the way that the pending image to be subtracted to the blurry versions of the pending image, and then obtains described pending
Image add it is sharp after first plus sharp image, specifically, the unsharp masking can be used following formula to indicate：F (x, y) is pending image,For the fuzzy version of the pending image
This, fs(x, y) is by unsharp masking treated image；Or the enhancing pending figure is filtered by height enhancing
The acutance of picture generates first and just adds sharp image, specifically, the high enhancing is filtered and is carried out by following formula：It is the pending image that A, which is more than or equal to 1, f (x, y),For
The blurry versions of the pending image, fhbFirst after (x, y) is filtered for high enhancing adds sharp image, when A is more than 1 and gets over
Greatly, the degree of the pending image sharpening is smaller.
Determine that the image that the first area is corresponded in described first plus sharp image adds as the second image, i.e., described first
Added sharp facial image in sharp image.
Second image in described first plus sharp image is replaced with described first image, after replacement by step S50
First plus sharp image as target image.
Specifically, with reference to figure 2, described first adds second image in sharp image to replace with described first image, will
Replaced first plus sharp image may include as the process of target image：
Step S51 obtains the pixel point coordinates of the second image and the pixel pair described in described first plus sharp image
The color value answered；
The corresponding color value of pixel acquired in described first plus sharp image is replaced with described first by step S52
The color value of image corresponding pixel points, using replaced first plus sharp image as target image.
Obtain the coordinate and the corresponding color value of the coordinate of the second image described in described first plus sharp image；And it determines
The color value of described first image respective coordinates, the color value of the first image in the pending image replace described first plus
Second image described in sharp image corresponds to the color value of the coordinate, by replaced image in described first plus sharp image and not
Replaced first plus sharp image are generated target image by the image construction target image of replacement.
The present embodiment is by adding sharp processing procedure to the image containing face, extracting the face figure in pending image
Picture, with extract without add the facial image of sharp processing replace plus be sharp after pending image in facial image, protect
The image sharpness of face part is stayed.It effectively avoids, when containing facial image in pending image, adding sharp processing operation meeting
So that face line-dividing is apparent, face contour is stiff unnatural so that the image fault of face part is serious, causes to add sharp effect
Fruit is poor, so that the problem of the Experience Degree difference of user.It to the image fidelity of face part, and then improves and adds sharp effect, carry
The high Experience Degree of user.
With reference to Fig. 3, Fig. 3 is the flow diagram for the second embodiment that the image of the invention containing face adds sharp method.Base
Add the first embodiment of sharp method in the above-mentioned image containing face, after the step S50, further includes：
Step S60 determines the fringe region of the first image described in the target image, to pixel in the fringe region
The color value of point carries out interpolation processing；
Step S70 carries out denoising to the target image after interpolation processing and obtains the target image after denoising；
Step S80 preserves the target image after the denoising.
The fringe region of first image described in the target image includes the region of the edge both sides of facial image, according to
Preset width value determines the fringe region, carries out interpolation processing to the color value of the pixel in the fringe region, makes
The image for obtaining the fringe region is smoothened naturally, avoiding the fringe region because replacing the problem for seeming stiff.To institute
It states target image and carries out denoising, preserve the target image after the denoising so that target image is more clear nature.
Further, when facial image is not detected in the pending image, the pending image is added
Sharp processing simultaneously generates target image.
When facial image is not detected in the pending image, to the pending image plus sharp processing, can be with
The acutance for enhancing the pending image using high-pass filtering processing, by the low-frequency component in the pending image of decaying,
Retain the radio-frequency component of the pending image so that the acutance of the pending image improves, and adds sharp image；It can also utilize
Unsharp masking processing enhances the acutance of the pending image, by the way that the pending image is subtracted the pending image
Blurry versions, and then obtain the pending image add it is sharp after first plus sharp image, specifically, the unsharp masking can
It is indicated with following formula：F (x, y) is pending image,It is waited for be described
Handle the blurry versions of image, fs(x, y) is by unsharp masking treated image；Or it is filtered by height enhancing
Enhance the acutance of the pending image, generate first and just add sharp image, specifically, the high enhancing is filtered by as follows
Formula carries out：It is the pending image that A, which is more than or equal to 1, f (x, y),For the blurry versions of the pending image, fhb(x, y) is first plus sharp image after high enhancing is filtered, and works as A
More than 1 and bigger, the degree of the pending image sharpening is smaller.
With reference to Fig. 4, Fig. 4 is the flow diagram for the 3rd embodiment that the image of the invention containing face adds sharp method.Base
Add the first embodiment of sharp method in the above-mentioned image containing face, after the step S20, further includes：
Step S90 determines the region in the pending image in addition to the human face region as second area, and carries
Take the corresponding image of the second area as third image；
The detection of facial image can be identified by the colour contrast of background environment and face complexion, described in determination
Whether contain facial image in pending image, the face in the pending image where facial image is returned if containing
Region, and determine that the region in the pending image in addition to the human face region as second area, returns to described second
The coordinate value in region, and the corresponding image of the second area is extracted as third image.
Step S100, carries out the third image plus sharp processing, generation second add sharp image；
The third image is carried out to add sharp processing, it is described to add in sharp process and above-mentioned first embodiment to the first image
Add sharp process identical, details are not described herein.The third image is being carried out to add sharp processing, is generating second plus sharp image.
The third image in the pending image is replaced with described second plus sharp image, will replaced by step S110
Pending image after changing is as target image.
Determine the pixel point coordinates of third image described in the pending image and the corresponding color value of the pixel,
The pixel color value that the respective coordinates of the third image are replaced with the color value of described second plus sharp image, will be replaced
Pending image is as target image as target image.It preferably, can in order to improve the smoothness at described first image edge
To be to determine described in the target image fringe region of second plus sharp image, interpolation processing is carried out to the fringe region,
So that the profile of target image because of replacement more naturally, will not cause the fringe region transition stiff.Further, to institute
It states target image and carries out denoising and smoothing processing, preserve the target image after the denoising.Make the target figure
As fidelity higher.
The present embodiment during the image procossing containing face, by pending image in addition to human face region
The image in region carries out plus sharp processing generates second plus sharp image, by second plus the sharp image replacement pending image with institute
State second plus the corresponding image of sharp image.It to the image fidelity of face part, and then improves and adds sharp effect, improve user's
Experience Degree.
The present invention further provides a kind of images containing face to add sharp device.
With reference to Fig. 5, Fig. 5 is the function module signal for the first embodiment that the image of the invention containing face adds sharp device
Figure.
In one embodiment, the image processing apparatus containing face includes：Acquisition module 10, carries determining module 20
Modulus block 30 plus sharp module 40, replacement module 50,
The acquisition module 10, for obtaining pending image；
The pending image can be carried out in image processing process plus the intermediate image of sharp processing, also can be with
It is the image of the camera induction of picture pick-up device by carrying out in front-end image processing procedure plus the middle graph of sharp processing
Picture.
The determining module 20 when for detecting facial image in the pending image, determines described pending
Human face region in image is as first area；
Detection to facial image can be identified by the colour contrast of background environment and face complexion, determine institute
It states in pending image and whether contains facial image, the people in the pending image where facial image is returned if containing
The coordinate value in face region；It can also be judged by calculating pending image and the matching degree of preset standard faces template
Whether in the pending image contain facial image, face in the pending image is returned if containing facial image
The coordinate value of human face region where image；Alternatively, the character modules by comparing pending image and preset face and face
Type is matched, and judges whether contain facial image in the pending image, is waited for described in return if containing facial image
Handle the coordinate value of the human face region in image where facial image.
The extraction module 30, for extracting the facial image from the human face region；
According to the coordinate value of determining human face region, the pixel point coordinates and corresponding pixel points of the facial image are extracted
Color value, and the pixel point coordinates of the facial image and the corresponding color value data of the pixel are stored in interim storage
Area.
Described plus sharp module 40 adds sharp processing for being carried out to the pending image, generates first plus sharp image；
Replacement module 50 corresponds to the image of the first area as the second figure for determining in described first plus sharp image
Picture；
Sharp processing is added to the pending image, high-pass filtering processing can be utilized to enhance the sharp of the pending image
Degree retains the radio-frequency component of the pending image so that described to wait for by the low-frequency component in the pending image of decaying
The acutance for handling image improves, and generates first plus sharp image；Unsharp masking processing can also be utilized to enhance the pending figure
The acutance of picture by the way that the pending image to be subtracted to the blurry versions of the pending image, and then obtains described pending
Image add it is sharp after first plus sharp image, specifically, the unsharp masking can be used following formula to indicate：F (x, y) is pending image,For the fuzzy version of the pending image
This, fs(x, y) is by unsharp masking treated image；Or the enhancing pending figure is filtered by height enhancing
The acutance of picture generates first and just adds sharp image, specifically, the high enhancing is filtered and is carried out by following formula：It is the pending image that A, which is more than or equal to 1, f (x, y),For
The blurry versions of the pending image, fhbFirst after (x, y) is filtered for high enhancing adds sharp image, when A is more than 1 and gets over
Greatly, the degree of the pending image sharpening is smaller.
Determine that the image that the first area is corresponded in described first plus sharp image adds as the second image, i.e., described first
Added sharp facial image in sharp image.
The replacement module 50, for second image in described first plus sharp image to be replaced with first figure
Picture, using replaced first plus sharp image as target image.
Specifically, with reference to figure 6, the replacement module 50 includes acquiring unit 51 and replacement unit 52,
The acquiring unit 51, for obtaining the pixel point coordinates of the second image and institute described in described first plus sharp image
State the corresponding color value of pixel；
The replacement unit 52, for replacing the corresponding color value of pixel acquired in described first plus sharp image
For the color value of described first image corresponding pixel points, add sharp image as target image using replaced first.
Obtain the coordinate and the corresponding color value of the coordinate of the second image described in described first plus sharp image；And it determines
The color value of described first image respective coordinates, the color value of the first image in the pending image replace described first plus
Second image described in sharp image corresponds to the color value of the coordinate, by replaced image in described first plus sharp image and not
Replaced first plus sharp image are generated target image by the image construction target image of replacement.
The present embodiment is by adding sharp processing procedure to the image containing face, extracting the face figure in pending image
Picture, with extract without add the facial image of sharp processing replace plus be sharp after pending image in facial image, protect
The image sharpness of face part is stayed.It effectively avoids, when containing facial image in pending image, adding sharp processing operation meeting
So that face line-dividing is apparent, face contour is stiff unnatural so that the image fault of face part is serious, causes to add sharp effect
Fruit is poor, so that the problem of the Experience Degree difference of user.It to the image fidelity of face part, and then improves and adds sharp effect, carry
The high Experience Degree of user.
With reference to Fig. 7, Fig. 7 is the function module signal for the second embodiment that the image of the invention containing face adds sharp device
Figure.The image containing face adds the sharp device to further include：Processing module 60 and memory module 70,
The processing module 60, the fringe region for determining the first image described in the target image, to the side
The color value of pixel carries out interpolation processing in edge region；It is additionally operable to obtain the target image progress denoising after interpolation processing
Target image after to denoising；
The memory module 70, for preserving the target image after the denoising.
The fringe region of first image described in the target image includes the region of the edge both sides of facial image, according to
Preset width value determines the fringe region, carries out interpolation processing to the color value of the pixel in the fringe region, makes
The image for obtaining the fringe region is smoothened naturally, avoiding the fringe region because replacing the problem for seeming stiff.To institute
It states target image and carries out denoising, preserve the target image after the denoising so that target image is more clear nature.
Further, described plus sharp module 40 is right when being additionally operable to that facial image is not detected in the pending image
The pending image carries out adding sharp processing and generates target image.
When facial image is not detected in the pending image, to the pending image plus sharp processing, can be with
The acutance for enhancing the pending image using high-pass filtering processing, by the low-frequency component in the pending image of decaying,
Retain the radio-frequency component of the pending image so that the acutance of the pending image improves, and adds sharp image；It can also utilize
Unsharp masking processing enhances the acutance of the pending image, by the way that the pending image is subtracted the pending image
Blurry versions, and then obtain the pending image add it is sharp after first plus sharp image, specifically, the unsharp masking can
It is indicated with following formula：F (x, y) is pending image,It is waited for be described
Handle the blurry versions of image, fs(x, y) is by unsharp masking treated image；Or it is filtered by height enhancing
Enhance the acutance of the pending image, generate first and just add sharp image, specifically, the high enhancing is filtered by as follows
Formula carries out：It is the pending image that A, which is more than or equal to 1, f (x, y),For the blurry versions of the pending image, fhb(x, y) is first plus sharp image after high enhancing is filtered, and works as A
More than 1 and bigger, the degree of the pending image sharpening is smaller.
Further, the extraction module 30 is additionally operable to determine in the pending image in addition to the human face region
Region as second area, and from the corresponding image of the second area is extracted as third image；
The detection of facial image can be identified by the colour contrast of background environment and face complexion, described in determination
Whether contain facial image in pending image, the face in the pending image where facial image is returned if containing
Region, and determine that the region in the pending image in addition to the human face region as second area, returns to described second
The coordinate value in region, and the corresponding image of the second area is extracted as third image, for example, extracting the second area pair
The color value on the image slices vegetarian refreshments left side and respective coordinates answered.
Described plus sharp module 40 is additionally operable to carry out the third image plus sharp processing, generation second adds sharp image；
The third image is carried out to add sharp processing, it is described to add in sharp process and above-mentioned first embodiment to the first image
Add sharp process identical, details are not described herein.The third image is being carried out to add sharp processing, is generating second plus sharp image.
The replacement module 50, be additionally operable to by the third image in the pending image replace with described second plus
Sharp image, using replaced pending image as target image.
Determine the pixel point coordinates of third image described in the pending image and the corresponding color value of the pixel,
The pixel color value that the respective coordinates of the third image are replaced with the color value of described second plus sharp image, will be replaced
Pending image is as target image as target image.It preferably, can in order to improve the smoothness at described first image edge
To be to determine described in the target image fringe region of second plus sharp image, interpolation processing is carried out to the fringe region,
So that the profile of target image because of replacement more naturally, will not cause the fringe region transition stiff.Further, to institute
It states target image and carries out denoising and smoothing processing, preserve the target image after the denoising.Make the target figure
As fidelity higher.
The present embodiment during the image procossing containing face, by pending image in addition to human face region
The image in region carries out plus sharp processing generates second plus sharp image, by second plus the sharp image replacement pending image with institute
State second plus the corresponding image of sharp image.It to the image fidelity of face part, and then improves and adds sharp effect, improve user's
Experience Degree.
The embodiments of the present invention are for illustration only, can not represent the quality of embodiment.Pass through above embodiment party
The description of formula, it is required general that those skilled in the art can be understood that above-described embodiment method can add by software
The mode of hardware platform is realized, naturally it is also possible to which by hardware, but the former is more preferably embodiment in many cases.It is based on
Such understanding, substantially the part that contributes to existing technology can be with software product in other words for technical scheme of the present invention
Form embody, which is stored in a storage medium (such as ROM/RAM, magnetic disc, CD), including
Some instructions are used so that a station terminal equipment (can be mobile phone, computer, server or the network equipment etc.) executes this hair
Method described in bright each embodiment.
It these are only the preferred embodiment of the present invention, be not intended to limit the scope of the invention, it is every to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of image containing face adds sharp method, which is characterized in that including step：
Obtain pending image；
When detecting facial image in the pending image, determine the human face region in the pending image as first
Region；
The facial image is extracted from the first area as the first image,
The pending image is carried out plus sharp processing, generation first add sharp image, is determined corresponding in described first plus sharp image
The image of the first area is as the second image；
Second image in described first plus sharp image is replaced with into described first image, by replaced first plus sharp figure
As being used as target image；
The fringe region for determining the first image described in the target image, to the color value of pixel in the fringe region into
Row interpolation processing；
Denoising is carried out to the target image after interpolation processing and obtains the target image after denoising.
2. the image containing face adds sharp method as described in claim 1, which is characterized in that the determination pending figure
As in human face region as first area the step of after, further include：
The region in the pending image in addition to the human face region is determined as second area, and extracts secondth area
The corresponding image in domain is as third image；
The third image is carried out plus sharp processing, generation second add sharp image；
The third image in the pending image is replaced with into described second plus sharp image, by replaced pending figure
As being used as target image.
3. the image containing face adds sharp method as described in claim 1, which is characterized in that described to add sharp figure by described first
Second image as in replaces with described first image, using replaced first plus sharp image as the step of target image
Including：
Obtain the pixel point coordinates and the corresponding color value of the pixel of the second image described in described first plus sharp image；
The corresponding color value of pixel acquired in described first plus sharp image is replaced with into described first image respective pixel
The color value of point, using replaced first plus sharp image as target image.
4. the image containing face adds sharp method as described in claim 1, which is characterized in that described to add replaced first
After the step of sharp image is as target image, further include：
Preserve the target image after the denoising.
5. the image for containing face such as Claims 1-4 any one of them adds sharp method, which is characterized in that the acquisition waits for
After the step of handling image, further include：
When facial image is not detected in the pending image, the pending image is carried out adding sharp processing and generates mesh
Logo image.
6. a kind of image containing face adds sharp device, which is characterized in that including：
Acquisition module, for obtaining pending image；
Determining module when for detecting facial image in the pending image, determines the people in the pending image
Face region is as first area；
Extraction module, for extracting the facial image from the first area as the first image；
Add sharp module, adds sharp processing for being carried out to the pending image, generate first plus sharp image；
Replacement module, for determining that the image for corresponding to the first area in described first plus sharp image, will as the second image
Described first plus sharp image in second image replace with described first image, will replaced first add sharp image as
Target image；
Processing module, the fringe region for determining the first image described in the target image, to the border area pixels
The color value of point carries out interpolation processing；It is additionally operable to after obtaining denoising to the target image progress denoising after interpolation processing
Target image.
7. the image containing face adds sharp device as claimed in claim 6, which is characterized in that the extraction module is additionally operable to
The region in the pending image in addition to the human face region is determined as second area, and extracts the second area pair
The image answered is as third image；
Described plus sharp module is additionally operable to carry out the third image plus sharp processing, generation second adds sharp image；
The replacement module is additionally operable to the third image in the pending image replacing with described second plus sharp image, will
Replaced pending image is as target image.
8. the image containing face adds sharp device as claimed in claim 6, which is characterized in that the replacement module includes obtaining
Unit and replacement unit,
The acquiring unit, for obtaining the pixel point coordinates of the second image and the pixel described in described first plus sharp image
The corresponding color value of point；
The replacement unit, for the corresponding color value of pixel acquired in described first plus sharp image to be replaced with described the
The color value of one image corresponding pixel points, using replaced first plus sharp image as target image.
9. the image containing face adds sharp device as claimed in claim 6, which is characterized in that at the image containing face
Managing device further includes：
Memory module, for preserving the target image after the denoising.
10. the image for containing face such as claim 6 to 9 any one of them adds sharp device, which is characterized in that described plus sharp mould
Block when being additionally operable to that facial image is not detected in the pending image, carries out the pending image to add sharp processing simultaneously
Generate target image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410848840.3A CN104504668B (en) | 2014-12-30 | 2014-12-30 | Image containing face adds sharp method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410848840.3A CN104504668B (en) | 2014-12-30 | 2014-12-30 | Image containing face adds sharp method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104504668A CN104504668A (en) | 2015-04-08 |
| CN104504668B true CN104504668B (en) | 2018-09-14 |
Family
ID=52946062
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410848840.3A Active CN104504668B (en) | 2014-12-30 | 2014-12-30 | Image containing face adds sharp method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104504668B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110988888B (en) * | 2019-11-08 | 2021-10-29 | 中科长城海洋信息系统有限公司 | Method and device for acquiring seabed information |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7844076B2 (en) * | 2003-06-26 | 2010-11-30 | Fotonation Vision Limited | Digital image processing using face detection and skin tone information |
| CN102025960B (en) * | 2010-12-07 | 2012-10-03 | 浙江大学 | Motion compensation de-interlacing method based on adaptive interpolation |
| US8823745B2 (en) * | 2011-06-02 | 2014-09-02 | Yoostar Entertainment Group, Inc. | Image processing based on depth information and color data of a scene |
| CN103095979A (en) * | 2011-11-07 | 2013-05-08 | 华晶科技股份有限公司 | Image processing method and imaging capture device of face overexposure |
| CN103049894A (en) * | 2012-09-14 | 2013-04-17 | 深圳市万兴软件有限公司 | Image processing method and device |
| CN104182721A (en) * | 2013-05-22 | 2014-12-03 | 华硕电脑股份有限公司 | Image processing system and image processing method capable of improving face identification rate |
| CN103686095B (en) * | 2014-01-02 | 2017-05-17 | 中安消技术有限公司 | Video concentration method and system |
| CN103927719B (en) * | 2014-04-04 | 2017-05-17 | 北京猎豹网络科技有限公司 | Picture processing method and device |
-
2014
- 2014-12-30 CN CN201410848840.3A patent/CN104504668B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104504668A (en) | 2015-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9569827B2 (en) | Image processing apparatus and method, and program | |
| JP6100744B2 (en) | Color document image segmentation and binarization using automatic restoration | |
| CN105243371A (en) | Human face beauty degree detection method and system and shooting terminal | |
| CN105096347B (en) | Image processing apparatus and method | |
| CN109325988A (en) | A kind of facial expression synthetic method, device and electronic equipment | |
| JP2016505186A (en) | Image processor with edge preservation and noise suppression functions | |
| CN107194869B (en) | Image processing method and terminal, computer storage medium and computer equipment | |
| CN107172354B (en) | Video processing method and device, electronic equipment and storage medium | |
| JP2007504719A (en) | System and method for detecting and correcting red eyes in an image suitable for embedded applications | |
| CN109214999A (en) | A kind of removing method and device of video caption | |
| CN105096330A (en) | Image processing method capable of automatically recognizing pure-color borders, system and a photographing terminal | |
| Ancuti et al. | Image and video decolorization by fusion | |
| CN112785572B (en) | Image quality evaluation method, apparatus and computer readable storage medium | |
| CN110390643A (en) | A kind of license plate Enhancement Method, device and electronic equipment | |
| CN104504703A (en) | Welding spot color image segmentation method based on chip element SMT (surface mounting technology) | |
| CN112258440A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
| CN111047619B (en) | Face image processing method and device and readable storage medium | |
| CN104504668B (en) | Image containing face adds sharp method and device | |
| CN114862729A (en) | Image processing method, image processing device, computer equipment and storage medium | |
| KR101921608B1 (en) | Apparatus and method for generating depth information | |
| WO2022016326A1 (en) | Image processing method, electronic device, and computer-readable medium | |
| Dang et al. | Inpainted image quality assessment | |
| CN110140150B (en) | Image processing method and device and terminal equipment | |
| KR101881795B1 (en) | Method for Detecting Edges on Color Image Based on Fuzzy Theory | |
| Liu et al. | Infrared ship target image smoothing based on adaptive mean shift |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |