CN108564541A - A kind of image processing method and device - Google Patents
A kind of image processing method and device Download PDFInfo
- Publication number
- CN108564541A CN108564541A CN201810266817.1A CN201810266817A CN108564541A CN 108564541 A CN108564541 A CN 108564541A CN 201810266817 A CN201810266817 A CN 201810266817A CN 108564541 A CN108564541 A CN 108564541A
- Authority
- CN
- China
- Prior art keywords
- image
- fuzzy
- target
- rgb value
- processing
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 209
- 238000000034 method Methods 0.000 claims abstract description 40
- 238000005192 partition Methods 0.000 claims abstract description 22
- 238000009499 grossing Methods 0.000 claims description 25
- 239000011159 matrix material Substances 0.000 claims description 23
- 238000013459 approach Methods 0.000 claims description 11
- 239000000284 extract Substances 0.000 claims description 9
- 238000010586 diagram Methods 0.000 description 14
- 230000000694 effects Effects 0.000 description 11
- 238000012549 training Methods 0.000 description 11
- 238000003860 storage Methods 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
Classifications
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Abstract
The embodiment of the present application provides a kind of image processing method and device, wherein this method includes：After receiving the request of image Fuzzy Processing, the target image for waiting for image Fuzzy Processing is obtained；Region division processing is carried out to the target image, obtains multiple image subblocks for waiting for image Fuzzy Processing；The corresponding blur radius of each image subblock for dividing and obtaining is determined respectively；For each image subblock, according to the blur radius of the image subblock, image Fuzzy Processing is carried out to the image subblock respectively, obtains the target image after Fuzzy Processing.In the embodiment of the present application, by the way of carrying out fuzzy partition processing to target image, simultaneously, Fuzzy Processing is carried out using corresponding blur radius for different image subblocks, in this way so that image is fuzzy more targetedly, the stereovision for improving the target image after obscuring, meets the variety of images Fuzzy Processing demand of user, improves the user experience.
Description
Technical field
This application involves field of information processing more particularly to a kind of image processing methods and device.
Background technology
Currently, with the continuous development of image processing techniques, for different application scenarios, by being carried out to target image
Corresponding image procossing, for example, in order to improve the aesthetics of image, is needed to target image with obtaining required target image
U.S. face processing is carried out, for another example, in order to protect the privacy content for including in image, needs to carry out Fuzzy Processing to target image, with
It user to be continuously increased the emphasis degree of itself anonymity, the demand for carrying out image Fuzzy Processing is more and more.
Currently, a kind of image fuzzy processing method is provided in the prior art, and this method is mainly：Acquisition waits for that image is fuzzy
The target image of processing is determined and carries out used blur radius when image obscures to the target image, calculated using default obscure
Method is based on the blur radius and is obscured to target image progress entirety, specifically, in practical application there are many image fuzzy algorithmic approaches,
For example, Gaussian Blur algorithm and mean value fuzzy algorithmic approach etc..
It follows that image fuzzy processing method in the prior art directly carries out fuzzy place to the entirety of target image
Reason, there are image Fuzzy Processing mode is relatively simple, the problem of flexibility difference, the variety of images that cannot be satisfied user fuzzy is located
Reason needs.
Invention content
The purpose of the embodiment of the present application is to provide a kind of image processing method and device, and subregion is carried out using to target image
The mode of Fuzzy Processing, meanwhile, Fuzzy Processing is carried out using corresponding blur radius for different image subblocks, so that
Image is fuzzy more targetedly, improves the stereovision of the target image after obscuring, and the variety of images for meeting user is fuzzy
Process demand improves the user experience.
In order to solve the above technical problems, what the embodiment of the present application was realized in：
The embodiment of the present application provides a kind of image processing method, including：
After receiving the request of image Fuzzy Processing, the target image for waiting for image Fuzzy Processing is obtained；
Region division processing is carried out to the target image, obtains multiple image subblocks for waiting for image Fuzzy Processing；
Each corresponding blur radius of described image sub-block for dividing and obtaining is determined respectively, wherein the blur radius
It is determined according to the relevant information of information-setting by user and/or image subblock, for indicating the corresponding fuzzy journey of image subblock
Degree；
For each described image sub-block, according to the blur radius of the image subblock, respectively to the image subblock into
Row image Fuzzy Processing, obtains the target image after Fuzzy Processing.
The embodiment of the present application provides a kind of image processing apparatus, including：
Image collection module, for after receiving the request of image Fuzzy Processing, obtaining the target for waiting for image Fuzzy Processing
Image；
Image division module obtains multiple waiting for the fuzzy place of image for carrying out region division processing to the target image
The image subblock of reason；
Blur radius determining module, for determining each described image sub-block corresponding fuzzy half for dividing and obtaining respectively
Diameter, wherein the blur radius is determined according to the relevant information of information-setting by user and/or image subblock, for indicating image
The corresponding fog-level of sub-block；
Image obfuscation module, according to the blur radius of the image subblock, is divided for being directed to each described image sub-block
It is other that image Fuzzy Processing is carried out to the image subblock, obtain the target image after Fuzzy Processing.
The embodiment of the present application provides a kind of image processing equipment, including：Processor；And
It is arranged to the memory of storage computer executable instructions, the computer executable instructions make when executed
The processor realizes following below scheme：
After receiving the request of image Fuzzy Processing, the target image for waiting for image Fuzzy Processing is obtained；
Region division processing is carried out to the target image, obtains multiple image subblocks for waiting for image Fuzzy Processing；
Each corresponding blur radius of described image sub-block for dividing and obtaining is determined respectively, wherein the blur radius
It is determined according to the relevant information of information-setting by user and/or image subblock, for indicating the corresponding fuzzy journey of image subblock
Degree；
For each described image sub-block, according to the blur radius of the image subblock, respectively to the image subblock into
Row image Fuzzy Processing, obtains the target image after Fuzzy Processing.
The embodiment of the present application provides a kind of storage medium, and for storing computer executable instructions, the computer can
It executes instruction and realizes following below scheme when executed：
After receiving the request of image Fuzzy Processing, the target image for waiting for image Fuzzy Processing is obtained；
Region division processing is carried out to the target image, obtains multiple image subblocks for waiting for image Fuzzy Processing；
Each corresponding blur radius of described image sub-block for dividing and obtaining is determined respectively, wherein the blur radius
It is determined according to the relevant information of information-setting by user and/or image subblock, for indicating the corresponding fuzzy journey of image subblock
Degree；
For each described image sub-block, according to the blur radius of the image subblock, respectively to the image subblock into
Row image Fuzzy Processing, obtains the target image after Fuzzy Processing.
Image processing method in the embodiment of the present application and device, after receiving the request of image Fuzzy Processing, acquisition waits for
The target image of image Fuzzy Processing；Region division processing is carried out to the target image, obtains multiple image Fuzzy Processings of waiting for
Image subblock；The corresponding blur radius of each image subblock for dividing and obtaining is determined respectively；For each image subblock, according to
The blur radius of the image subblock carries out image Fuzzy Processing to the image subblock respectively, obtains the target figure after Fuzzy Processing
Picture.In the embodiment of the present application, by the way of carrying out fuzzy partition processing to target image, meanwhile, for different image
Block carries out Fuzzy Processing using corresponding blur radius, so that image is fuzzy more targetedly, improves after obscuring
The stereovision of target image meets the variety of images Fuzzy Processing demand of user, improves the user experience.
Description of the drawings
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments described in application, for those of ordinary skill in the art, in the premise of not making the creative labor property
Under, other drawings may also be obtained based on these drawings.
Fig. 1 is the first flow diagram of image processing method provided by the embodiments of the present application；
Fig. 2 is the realization principle schematic diagram provided by the embodiments of the present application that Fuzzy Processing is carried out to target image；
Fig. 3 is the realization principle schematic diagram of the coordinate system provided by the embodiments of the present application for establishing target image；
Fig. 4 is second of flow diagram of image processing method provided by the embodiments of the present application；
Fig. 5 is the third flow diagram of image processing method provided by the embodiments of the present application；
Fig. 6 is the realization principle for carrying out region division in image processing method provided by the embodiments of the present application to target image
Schematic diagram；
Fig. 7 is the 4th kind of flow diagram of image processing method provided by the embodiments of the present application；
Fig. 8 is the module composition schematic diagram of image processing apparatus provided by the embodiments of the present application；
Fig. 9 is the structural schematic diagram of image processing equipment provided by the embodiments of the present application.
Specific implementation mode
In order to make those skilled in the art better understand the technical solutions in the application, below in conjunction with the application reality
The attached drawing in example is applied, technical solutions in the embodiments of the present application is clearly and completely described, it is clear that described implementation
Example is merely a part but not all of the embodiments of the present application.Based on the embodiment in the application, this field is common
The every other embodiment that technical staff is obtained without creative efforts should all belong to the application protection
Range.
The embodiment of the present application provides a kind of image processing method and device, is carried out at fuzzy partition using to target image
The mode of reason, meanwhile, Fuzzy Processing is carried out using corresponding blur radius for different image subblocks, so that image mould
Paste is more targeted, improves the stereovision of the target image after obscuring, meets the variety of images Fuzzy Processing need of user
It asks, improves the user experience.
Fig. 1 is the first flow diagram of image processing method provided by the embodiments of the present application, and the method in Fig. 1 is held
Row main body can also be mobile terminal device with computer equipment, as shown in Figure 1, this method at least includes the following steps：
S101 obtains the target image for waiting for image Fuzzy Processing after receiving the request of image Fuzzy Processing；Specifically,
Image Fuzzy Processing request generating mode can be based on user's selection target image and click image obscure control generate,
It can also be and generated based on voice messaging input by user.
S102 carries out region division processing to the target image got, obtains multiple images for waiting for image Fuzzy Processing
Sub-block；Specifically, during carrying out region division to target image, it can be determined and be drawn based on the manual region division operation of user
The mode of dividing, the image information for being also based on target image automatically determine dividing mode, system can also be automatically determined and be used
Family manually selects two ways and is combined determining dividing mode, i.e., providing multiple subregions according to the image information of target image selects
, and dividing mode is determined for the specified option of subregion option selection according to user.
S103 determines the corresponding blur radius of each image subblock for dividing and obtaining, wherein the blur radius root respectively
It is determined according to the relevant information of information-setting by user and/or image subblock, for indicating the corresponding fog-level of image subblock；
Specifically, after to target image multidomain treat-ment, need to determine matched blur radius for each image subblock, wherein true
During the blur radius of fixed each image subblock, the numerical value that can be selected based on user determines the fuzzy of corresponding image subblock
Radius, the relevant information for being also based on image subblock automatically determine the blur radius of the image subblock, can also be by system certainly
Dynamic determination and user manually select the blur radius that two ways is combined determining image subblock, i.e., according to the phase of each image subblock
It closes information and provides multiple blur radius options, and each image is determined for the specified option of blur radius option selection according to user
The blur radius of sub-block.
S104, for each image subblock, according to the corresponding blur radius of the image subblock, respectively to the image subblock into
Row image Fuzzy Processing, obtains the target image after Fuzzy Processing；Specifically, being determined for each image subblock respective fuzzy
After radius, Fuzzy Processing is carried out to corresponding image subblock using the blur radius, generates required blurred picture.
In the embodiment of the present application, by the way of carrying out fuzzy partition processing to target image, meanwhile, for different figures
As sub-block carries out Fuzzy Processing using corresponding blur radius, so that image is fuzzy more targetedly, improves fuzzy
The stereovision of target image afterwards achievees the effect that the depth of field is adjusted, meets the variety of images Fuzzy Processing demand of user, improves
User experience.
Fig. 2 is the realization principle schematic diagram provided by the embodiments of the present application that Fuzzy Processing is carried out to target image, specifically,
As shown in Fig. 2, first, being asked based on image Fuzzy Processing, the target image for waiting for image Fuzzy Processing is obtained；
Then, it is determined that for the region division mode of target image, target image is carried out based on the region division mode
Region division obtains multiple images sub-block, wherein for target image to be divided into 4 image subblocks, is being embodied
When, it is divided according to actual demand, the graphics shape of each image subblock can be regular shape, can also be irregular shape
Shape；
Again, the corresponding blur radius of each image subblock for dividing and obtaining is determined respectively, i.e., determines image respectively
Block 1 is to 4 corresponding blur radius R of image subblock1To R4；
Finally, it is based on the respective blur radius of image subblock successively, Fuzzy Processing is carried out to each image subblock, obtains mould
Paste treated target image.
As shown in figure 3, be marked for the ease of the position to image subblock, and for the ease of being carried out to each pixel
Label is that target image establishes coordinate system according to preset rules, determines the coordinate (x, y) of each pixel in target image, each
Pixel corresponds to a coordinate, and before image is fuzzy, each pixel corresponds to a fuzzy preceding rgb value, after image is fuzzy, often
Therefore a pixel, which corresponds to a fuzzy rear rgb value, has certain correspondence between coordinate and rgb value；
Wherein it is possible to the coordinate of the central pixel point of the center of target image is defined as (0,0), at this point, in this
The coordinate of a upper pixel for imago vegetarian refreshments is (0,1), and the coordinate of next pixel of the central pixel point is (0, -1),
The coordinate of one pixel in the right side of the central pixel point is (1,0), the coordinate of the first from left pixel of the central pixel point for (-
1,0).
Further, it is contemplated that when carrying out Fuzzy Processing to target image, due to carrying out subregion using to target image
The mode of Fuzzy Processing, meanwhile, Fuzzy Processing, two such are carried out using different blur radius for different image subblocks
Image subblock intersection greatly reduces fuzzy at this time there may be the larger situation of the fog-level diversity ratio of neighbor pixel
The fuzzy naturalness of target image afterwards affects user's visual experience effect, this is based on, as shown in figure 4, in above-mentioned S104 needles
To each image subblock, according to the corresponding blur radius of the image subblock, image Fuzzy Processing is carried out to the image subblock respectively,
After obtaining the target image after Fuzzy Processing, further include：
S105 determines the location information of sub-block intersection in target image according to the location information of each image subblock；Specifically
, the coordinate system of the target image based on Fig. 3, when carrying out region division to target image, record has each figure that division obtains
As the location information of sub-block in a coordinate system, therefore, in conjunction with the location information of each image subblock, image in target image can be obtained
The location information of sub-block intersection.
S106, according to the location information for the sub-block intersection determined, determine the target pixel points for waiting for smoothing processing and
The fuzzy rear rgb value of the target pixel points；Specifically, the coordinate system of the target image based on Fig. 3, in determining target image
After the location information of sub-block intersection, the pixel in the preset range of adjacent image sub-block intersecting lens two-by-two is determined as waiting putting down
The target pixel points of sliding processing determine the coordinate of each target pixel points, and according to corresponding between coordinate and fuzzy rear rgb value
Relationship determines the fuzzy rear rgb value of each target pixel points.
S107 determines the smooth rear rgb value of the target pixel points respectively according to the fuzzy rear rgb value of each target pixel points；
Specifically, the fuzzy rear rgb value based on the target pixel points of the specified quantity of adjacent image sub-block intersecting lens both sides two-by-two, determines
The smoothing factor of each target pixel points determines the smooth rear rgb value of each target pixel points in conjunction with the smoothing factor, reaches to two-by-two
The fog-level of the target pixel points of the specified quantity of adjacent image sub-block intersecting lens both sides carries out smooth effect.
S108, rgb value after the rgb value of each target pixel points is adjusted to smooth by fuzzy rear rgb value, is obscured and is handed over
Target image after boundary is smooth；Specifically, determine respectively after the target pixel points of smoothing processing it is smooth after after rgb value, base
In each target pixel points it is smooth after rgb value adjust corresponding pixel points rgb value, to obscure and had a common boundary smoothly after mesh
Logo image.
Specifically, above-mentioned S107 determines the flat of the target pixel points respectively according to the fuzzy rear rgb value of each target pixel points
Rgb value after cunning, specifically includes：
Target pixel points are divided into multiple smoothing processings that wait for according to the location information of each target pixel points by step 1
Object pixel point set；Specifically, on the basis of adjacent image sub-block intersecting lens two-by-two, believed according to the position of each target pixel points
Target pixel points on the line segment vertical with intersecting lens are included in same object pixel point set by breath, for example, target image is drawn
It is divided into two image subblocks up and down, at this point, the abscissa for belonging to the pixel of same object pixel point set is identical, each target
Comprising the preset quantity nearest from point of interface positioned at the pixel and most from point of interface of the first image subblock in pixel collection
The pixel positioned at the second image subblock of close preset quantity.
Step 2, for each object pixel point set, using linear interpolation method according to each in the object pixel point set
The fuzzy rear rgb value of target pixel points, determines the smoothing factor of each target pixel points, wherein the smoothing factor plays tune respectively
Save the effect of the rgb value of adjacent target pixel two-by-two so that the rgb value difference between adjacent target pixel is default two-by-two
Within the scope of；
Step 3 determines respectively according to the smoothing factor of the fuzzy rear rgb value and the target pixel points of each target pixel points
The smooth rear rgb value of each target pixel points, specifically, each target pixel points are directed to, by the fuzzy rear RGB of the target pixel points
It is worth corresponding smoothing factor to be multiplied, you can obtain the smooth rear rgb value of the target pixel points.
In the embodiment of the present application, due to consideration that by the way of carrying out fuzzy partition processing to target image, meanwhile,
Fuzzy Processing is carried out using different blur radius for different image subblocks, two such image subblock intersection may be deposited
In the situation that the fog-level diversity ratio of neighbor pixel is larger, greatly reduce at this time it is fuzzy after target image it is fuzzy from
So degree, affects user's visual experience effect, therefore, increases and is put down to the image subblock intersection of the target image after obscuring
The process of sliding processing, improves the fuzzy naturalness of the image subblock intersection of the target image after obscuring, and promotes user's vision body
Test effect.
Wherein, for target image subregion process, as shown in figure 5, above-mentioned S102 carries out area to the target image got
Domain division processing, obtains multiple image subblocks for waiting for image Fuzzy Processing, specifically includes：
S1021 carries out image recognition processing to target image, the image information in the target image is extracted, specifically, base
In image recognition algorithm, the image information in target image is extracted；
S1022, according to for target image partition information and the image information that extracts, determine to be divided more
The location information of a image-region, specifically, the partition information includes the image district that image detaches information and do not isolate
The dividing mode in domain, for example, the separation of people's scape or object view separation etc., after determining the partition information for target image,
Positioning needs the image-region isolated, the figure for determining the location information at least one image-region isolated and not isolating
As again region is divided into the location information in multiple images region；
S1023 carries out region division processing to target image, obtains multiple according to the location information in multiple images region
Wait for the image subblock of image Fuzzy Processing.
Wherein, it is determined as follows for the partition information of target image, specially：
Mode one is asked according to region division input by user, determines the partition information for target image；
Mode two determines the partition information for target image according to division rule corresponding with target image；
Mode three is directed to the selection operation of the division option automatically determined according to user, determines the area for target image
Domain division information；Specifically, according to division rule corresponding with target image, determines and show multiple division options, so both
It disclosure satisfy that the demand that user independently selects, and user enabled to quickly to select required option.
Specifically, being directed to target image behaviour scape image or object view image, need to the region except portrait area or image
Except background image carry out Fuzzy Processing, due to consideration that usually background area range is bigger, in order to improve regarding for user
Feel effect, avoid the loftier of people's scape image or object view image peripheral blur, region division can be carried out to background area, with
The corresponding division rule of target image is determined as follows, specially：
If there is the image-region for not needing Fuzzy Processing in target image, not need the image-region of Fuzzy Processing
On the basis of, according to the graphics shape of the image-region for not needing Fuzzy Processing, determine division rule corresponding with target image,
Wherein, which is that will not need the image-region of Fuzzy Processing as an image subblock, and will need fuzzy place
The image-region of reason is divided into the image subblock of specified quantity.
For example, as shown in fig. 6, target image is to include the photo of a football, and require football without Fuzzy Processing,
Fuzzy Processing is carried out to the region in addition to football, at this time, it may be necessary to the location information of positioned football in the target image, with football
For the center of circle, multiple concentric circles are divided outward, and the multiple images sub-block that such target image divides has：Include the image of football
Image subblock 2 other than sub-block 1 and football is to image subblock 5.
Wherein, for the blur radius determination process of image subblock, above-mentioned S103 determines each image for dividing and obtaining respectively
The corresponding blur radius of sub-block, specifically includes：
Mode one receives blur radius setting request of the user for each image subblock, and asks to determine according to the setting
The blur radius of the image subblock；
Mode two determines the blur radius of the image subblock according to the relevant information of each image subblock, wherein the correlation
Information includes：At least one of image content information, location information, fuzzy preceding rgb value distribution；
Mode three receives the selection operation that user is directed to the multiple blur radius options automatically determined, and according to the selection
Operation determines the blur radius of each image subblock.
Specifically, after being divided to target image according to the graphics shape for the image-region for not needing Fuzzy Processing,
If relevant information is location information determines the blur radius of the image subblock, specifically according to the relevant information of each image subblock
For：
Spacing distance between the image subblock and the image subblock for not needing Fuzzy Processing of Fuzzy Processing as needed, really
Surely the blur radius of the image subblock of Fuzzy Processing is needed, wherein blur radius and spacing distance positive correlation, i.e. spacing distance are got over
Greatly, the blur radius of corresponding image subblock is bigger.For example, with the image-region division rule in Fig. 7 to target image into
After row divides, blur radius remoter from football is bigger, and fog-level gradually increases so that and the image linking after fuzzy is more natural,
More have a sense of hierarchy.
Specifically, in view of during automatically determining blur radius, the reference factor of introducing is more, the mould determined
It is more accurate to paste radius, in order to improve the accuracy for the blur radius determined, improving user expires final image blur effect
Meaning degree in conjunction with the multiple correlative factors for influencing blur effect, and combines the image of user fuzzy inclined in the embodiment of the present application
Good, the comprehensive blur radius for determining each image subblock is based on this, the type for the relevant information of the image subblock of introducing is more
The case where item, the above-mentioned relevant information according to each image subblock determines that the blur radius of the image subblock, detailed process are：
It obtains blur radius corresponding with the request target account of image Fuzzy Processing and determines model, wherein this fuzzy half
Diameter determines that model is obtained using deep learning method based on model training sample training corresponding with target account；
For each image subblock, the relevant information of the image subblock is input to the blur radius got and determines mould
Type；
Blur radius is determined that the numerical value of model output is determined as the blur radius of the image subblock.
Wherein, it needs to be that each target account trains to obtain corresponding blur radius and determines model in advance, specifically, passing through
Model is determined as under type trains to obtain the corresponding blur radius of each target account, specially：
Judge whether the fuzzy record of the history image under target account meets preset condition, wherein the history image is fuzzy
Record includes the correspondence between photographed image-related information and blur radius in historical time section, and preset condition is historical record
Item number is more than predetermined threshold value.
If satisfied, then by model training sample of the fuzzy record of history image as the target account under target account；
Using model training sample of the deep learning method based on target account, training obtains the corresponding mould of the target account
Paste radius determines model.
Specifically, when the fuzzy record of the history image under target account meets preset condition, illustrate that history image is fuzzy
The information recorded in record is enough, and the information recorded in the fuzzy record of history image is more, can more reflect the figure of user
As ambiguous preference, using the record as model training sample, and training obtains the corresponding blur radius of the target account and determines mould
Type, such later use blur radius determine that the blur radius accuracy that model is determined is higher.
If not satisfied, then by the under the reference account for marking the fuzzy record of the first history image under account and target account
Model training sample of the fuzzy record of two history images as the target account, wherein the reference account and target of target account
Account attribute information having the same, wherein the attribute information includes：The age of user, occupation, hobby, gender, identity
At least one of；
Using model training sample of the deep learning method based on target account, training obtains the corresponding mould of the target account
Paste radius determines model.
Specifically, when the fuzzy record of the history image under target account is unsatisfactory for preset condition, illustrate history image mould
The information that records is fewer in paste record, at this point, the image ambiguous preference of user can not be determined comprehensively, therefore, by with mesh
Marking account, there is the fuzzy record of the history image of the reference account of same alike result information, combined training to obtain target account correspondence
Blur radius determine model, the standard for the blur radius that the later use blur radius determines that model is determined can be improved in this way
Exactness.
Wherein, Fuzzy Processing process is carried out for each image subblock, as shown in fig. 7, above-mentioned S104 is directed to each image
Sub-block carries out image Fuzzy Processing to the image subblock respectively according to the corresponding blur radius of the image subblock, obtains fuzzy place
Target image after reason, specifically includes：
S1041 chooses in the multiple pixels for including in the image subblock at present Fuzzy for each image subblock
The target pixel points of reason, specifically, choosing one from the multiple pixels for including in image subblock successively according to preset rules
Pixel is as target pixel points；
S1042 determines the target pixel points according to the corresponding blur radius of image subblock where the target pixel points of selection
Multiple neighbor pixels, specifically, the quantity of the neighbor pixel of target pixel points is R × R-1, wherein R is fuzzy half
Diameter, for example, blur radius R=3, at this point, the quantity of the neighbor pixel of target pixel points is 8, using target pixel points as target
For central pixel point (0,0) in image, the neighbor pixel of the target pixel points is：(0,1), (0, -1), (1,0), (-
1,0), (1,1), (- 1, -1), (- 1,1), (1, -1)；
Wherein, by taking blur radius R=3 as an example, the quantity of the neighbor pixel of target pixel points is 8, central pixel point (0,
0) pixel coordinates matrix is：
S1043, according to target pixel points it is fuzzy before rgb value and each neighbor pixel of the target pixel points it is fuzzy before
Rgb value determines the fuzzy rear rgb value of the target pixel points, specifically, in the seat for the neighbor pixel for determining target pixel points
After mark, according to the correspondence between coordinate and fuzzy preceding rgb value, the fuzzy preceding rgb value of each neighbor pixel can be obtained, is based on
Rgb value obtains the fuzzy rear rgb value of target pixel points before fuzzy；
S1044 judges whether to determine the fuzzy rear rgb value of each pixel in image subblock；
If it is not, then continuing to execute step S1041；
If so, S1045 is executed, by the rgb value of each pixel in target image by the fuzzy preceding rgb value tune of the pixel
Whole is fuzzy rear rgb value, obtains the target image after Fuzzy Processing.
Specifically, each adjacent pixels of the above-mentioned S1043 according to the fuzzy preceding rgb value and the target pixel points of target pixel points
The fuzzy preceding rgb value of point determines the fuzzy rear rgb value of the target pixel points, specifically includes：
Step 1, according to the fuzzy preceding rgb value of target pixel points and obscuring for each neighbor pixel of the target pixel points
Preceding rgb value determines the fuzzy preceding rgb value matrix of the target pixel points；
Wherein, after the coordinate for the neighbor pixel for determining target pixel points, according between coordinate and fuzzy preceding rgb value
Correspondence, can obtain each neighbor pixel it is fuzzy before rgb value, obtain target pixel points it is fuzzy before rgb value matrix；
Still by taking blur radius R=3 as an example, the quantity of the neighbor pixel of target pixel points is 8, and target pixel points obscure
Preceding rgb value matrix is：
Wherein, a22For the fuzzy preceding rgb value of target pixel points, remaining is the mould of 8 neighbor pixels of target pixel points
Rgb value before paste.
Step 2 determines target using specified fuzzy algorithmic approach according to the blur radius of image subblock where target pixel points
The weight coefficient of the weight coefficient of pixel and each neighbor pixel of the target pixel points, obtains the fuzzy weight of target pixel points
Weight coefficient matrix；
Wherein, specified fuzzy algorithmic approach includes：Gaussian Blur algorithm, mean value fuzzy algorithmic approach, radial blur, surface blur, shape
Shape is fuzzy etc., can independently select by user that, using which kind of fuzzy algorithmic approach, image can also be prestored according to actual demand
Relevant information and fuzzy algorithmic approach between correspondence, therefore, which is to be based on the correspondence and image
What the relevant information of sub-block determined.
By taking specified fuzzy algorithmic approach is Gaussian Blur algorithm as an example, the fuzzy weighted values coefficient matrix of target pixel points is determined, have
Body is：
According to formulaCalculate the weight coefficient and the target pixel points of target pixel points
Each neighbor pixel weight coefficient, wherein γ (x, y) denotation coordination be (x, y) pixel weight coefficient, R is mould
Paste radius；
Still by taking blur radius R=3 as an example, the quantity of the neighbor pixel of target pixel points is 8, and target pixel points obscure
Weight coefficient matrix is：
Wherein, γ22For the weight coefficient of target pixel points, remaining is the weight of 8 neighbor pixels of target pixel points
Coefficient,
Step 3 determines the target picture according to the fuzzy preceding rgb value matrix and fuzzy weighted values coefficient matrix of target pixel points
The fuzzy rear rgb value of vegetarian refreshments, specifically, in the fuzzy preceding rgb value matrix and fuzzy weighted values coefficient square of determining target pixel points
After battle array, the sum of products of each fuzzy preceding RGB and corresponding fuzzy weighted values coefficient are calculated, you can obtain the mould of the target pixel points
Rgb value after paste；
Still by taking blur radius R=3 as an example, the quantity of the neighbor pixel of target pixel points is 8, and target pixel points obscure
Rgb value is afterwards：
Wherein, A is the fuzzy rear rgb value of target pixel points, and i is the row subscript of matrix, and j is matrix column subscript.
Image processing method in the embodiment of the present application, after receiving the request of image Fuzzy Processing, acquisition waits for image mould
Paste the target image of processing；Region division processing is carried out to the target image, obtains multiple image for waiting for image Fuzzy Processing
Block；The corresponding blur radius of each image subblock for dividing and obtaining is determined respectively；For each image subblock, according to the image
The blur radius of sub-block carries out image Fuzzy Processing to the image subblock respectively, obtains the target image after Fuzzy Processing.This Shen
Please in embodiment, by the way of carrying out fuzzy partition processing to target image, meanwhile, for different image subblocks using pair
The blur radius answered carries out Fuzzy Processing, so that image is fuzzy more targetedly, improves the target image after obscuring
Stereovision, meet the variety of images Fuzzy Processing demand of user, improve the user experience.
The image processing method that corresponding above-mentioned Fig. 1 to Fig. 7 is described, based on the same technical idea, the embodiment of the present application is also
A kind of image processing apparatus is provided, Fig. 8 is the module composition schematic diagram of image processing apparatus provided by the embodiments of the present application, should
Device is for executing the image processing method that Fig. 1 to Fig. 7 is described, as shown in figure 8, the device includes：
Image collection module 801, for after receiving the request of image Fuzzy Processing, obtaining the mesh for waiting for image Fuzzy Processing
Logo image；
Image division module 802 obtains multiple waiting for that image is fuzzy for carrying out region division processing to the target image
The image subblock of processing；
Blur radius determining module 803, for determining each corresponding mould of described image sub-block for dividing and obtaining respectively
Paste radius, wherein the blur radius is determined according to the relevant information of information-setting by user and/or image subblock, for indicating
The corresponding fog-level of image subblock；
Image obfuscation module 804, for being directed to each described image sub-block, according to described fuzzy the half of the image subblock
Diameter carries out image Fuzzy Processing to the image subblock respectively, obtains the target image after Fuzzy Processing.
Optionally, above-mentioned apparatus further includes：Sub-block boundary Leveling Block 805, wherein sub-block boundary Leveling Block
805, it is used for：
According to the location information of each described image sub-block, the location information of sub-block intersection in the target image is determined；
According to the location information of the sub-block intersection, the target pixel points and the object pixel for waiting for smoothing processing are determined
The fuzzy rear rgb value of point；
According to the fuzzy rear rgb value of each target pixel points, the smooth rear RGB of the target pixel points is determined respectively
Value；
Rgb value after the rgb value of each target pixel points is adjusted to smooth by fuzzy rear rgb value, is obscured and is handed over
Target image after boundary is smooth.
Optionally, sub-block boundary Leveling Block 805, is specifically used for：
According to the location information of each target pixel points, target pixel points are divided into multiple targets for waiting for smoothing processing
Pixel collection；
For each object pixel point set, using linear interpolation method according to each target in the object pixel point set
The fuzzy rear rgb value of pixel, determines the smoothing factor of each target pixel points respectively；
It is true respectively according to the smoothing factor of the fuzzy rear rgb value and the target pixel points of each target pixel points
The smooth rear rgb value of fixed each target pixel points.
Optionally, described image division module 802, is specifically used for：
Image recognition processing is carried out to the target image, extracts the image information in the target image；
According to for the target image partition information and the described image information that extracts, determine to be divided
The location information in multiple images region, wherein the partition information is to ask to determine according to region division input by user
, it is being determined according to division rule corresponding with the target image or the division option that automatically determines is directed to according to user
Selection operation determine；
According to the location information in multiple images region, region division processing is carried out to the target image, is obtained more
A image subblock for waiting for image Fuzzy Processing.
Optionally, the blur radius determining module 803, is specifically used for：
Blur radius setting request of the user for each image subblock is received, and the image is determined according to setting request
The blur radius of sub-block；
Alternatively,
According to the relevant information of each described image sub-block, the blur radius of described image sub-block is determined, wherein the correlation
Information includes：At least one of image content information, location information, fuzzy preceding rgb value distribution；
Alternatively,
The selection operation that user is directed to the multiple blur radius options automatically determined is received, and true according to the selection operation
The blur radius of fixed each image subblock.
Optionally, described image obfuscation module 804, is specifically used for：
Present Fuzzy processing is chosen for each described image sub-block, in the multiple pixels for including in the image subblock
Target pixel points；
According to the corresponding blur radius of image subblock where the target pixel points, the multiple of the target pixel points are determined
Neighbor pixel；
According to the fuzzy preceding rgb value of the fuzzy preceding rgb value of the target pixel points and each neighbor pixel, institute is determined
The fuzzy rear rgb value for stating target pixel points, the fuzzy rear rgb value until determining each pixel in described image sub-block；
RGB after the rgb value of each pixel in the target image is adjusted to fuzzy by the fuzzy preceding rgb value of the pixel
Value, obtains the target image after Fuzzy Processing.
Optionally, described image obfuscation module 804, is further specifically used for：
According to the fuzzy preceding rgb value of the fuzzy preceding rgb value of the target pixel points and each neighbor pixel, institute is determined
State the fuzzy preceding rgb value matrix of target pixel points；
Using specified fuzzy algorithmic approach according to the blur radius of image subblock where the target pixel points, the target is determined
The weight coefficient of the weight coefficient of pixel and each neighbor pixel obtains the fuzzy weighted values coefficient of the target pixel points
Matrix；
According to the fuzzy preceding rgb value matrix and the fuzzy weighted values coefficient matrix, the mould of the target pixel points is determined
Rgb value after paste.
Image processing apparatus in the embodiment of the present application, after receiving the request of image Fuzzy Processing, acquisition waits for image mould
Paste the target image of processing；Region division processing is carried out to the target image, obtains multiple image for waiting for image Fuzzy Processing
Block；The corresponding blur radius of each image subblock for dividing and obtaining is determined respectively；For each image subblock, according to the image
The blur radius of sub-block carries out image Fuzzy Processing to the image subblock respectively, obtains the target image after Fuzzy Processing.This Shen
Please in embodiment, by the way of carrying out fuzzy partition processing to target image, meanwhile, for different image subblocks using pair
The blur radius answered carries out Fuzzy Processing, so that image is fuzzy more targetedly, improves the target image after obscuring
Stereovision, meet the variety of images Fuzzy Processing demand of user, improve the user experience.
It should be noted that at image processing apparatus provided by the embodiments of the present application and image provided by the embodiments of the present application
Reason method is based on same inventive concept, therefore the specific implementation of the embodiment may refer to the implementation of aforementioned image processing method,
Overlaps will not be repeated.
Further, corresponding above-mentioned Fig. 1 is to method shown in Fig. 7, and based on the same technical idea, the embodiment of the present application is also
A kind of image processing equipment is provided, for executing above-mentioned image processing method, Fig. 9 is provided the equipment for the embodiment of the present application
Image processing equipment structural schematic diagram..
As shown in figure 9, image processing equipment can generate bigger difference because configuration or performance are different, may include one
A or more than one processor 901 and memory 902 can be stored with one or more storages in memory 902 and answer
With program or data.Wherein, memory 902 can be of short duration storage or persistent storage.It is stored in the application program of memory 902
May include one or more modules (diagram is not shown), each module may include to the system in image processing equipment
Column count machine executable instruction.Further, processor 901 could be provided as communicating with memory 902, be set in image procossing
Series of computation machine executable instruction in standby upper execution memory 902.Image processing equipment can also include one or one
The above power supply 903, one or more wired or wireless network interfaces 904, one or more input/output interfaces
905, one or more keyboards 906 etc..
In a specific embodiment, image processing equipment includes memory and one or more journey
Sequence, either more than one program is stored in memory and one or more than one program may include one for one of them
Or more than one module, and each module may include to the series of computation machine executable instruction in image processing equipment, and
Be configured to by one either more than one processor execute this or more than one program include by carry out it is following based on
Calculation machine executable instruction：
After receiving the request of image Fuzzy Processing, the target image for waiting for image Fuzzy Processing is obtained；
Region division processing is carried out to the target image, obtains multiple image subblocks for waiting for image Fuzzy Processing；
Each corresponding blur radius of described image sub-block for dividing and obtaining is determined respectively, wherein the blur radius
It is determined according to the relevant information of information-setting by user and/or image subblock, for indicating the corresponding fuzzy journey of image subblock
Degree；
For each described image sub-block, according to the blur radius of the image subblock, respectively to the image subblock into
Row image Fuzzy Processing, obtains the target image after Fuzzy Processing.
Optionally, computer executable instructions also include for carrying out following computer executable instructions when executed：
For each described image sub-block, according to the blur radius of the image subblock, respectively to the image subblock
Carry out image Fuzzy Processing further includes after obtaining the target image after Fuzzy Processing：
According to the location information of each described image sub-block, the location information of sub-block intersection in the target image is determined；
According to the location information of the sub-block intersection, the target pixel points and the object pixel for waiting for smoothing processing are determined
The fuzzy rear rgb value of point；
According to the fuzzy rear rgb value of each target pixel points, the smooth rear RGB of the target pixel points is determined respectively
Value；
Rgb value after the rgb value of each target pixel points is adjusted to smooth by fuzzy rear rgb value, is obscured and is handed over
Target image after boundary is smooth.
Optionally, computer executable instructions when executed, it is described according to each target pixel points it is fuzzy after RGB
Value determines the smooth rear rgb value of the target pixel points respectively, including：
According to the location information of each target pixel points, target pixel points are divided into multiple targets for waiting for smoothing processing
Pixel collection；
For each object pixel point set, using linear interpolation method according to each target in the object pixel point set
The fuzzy rear rgb value of pixel, determines the smoothing factor of each target pixel points respectively；
It is true respectively according to the smoothing factor of the fuzzy rear rgb value and the target pixel points of each target pixel points
The smooth rear rgb value of fixed each target pixel points.
Optionally, computer executable instructions are when executed, described to carry out region division processing to the target image,
Multiple image subblocks for waiting for image Fuzzy Processing are obtained, including：
Image recognition processing is carried out to the target image, extracts the image information in the target image；
According to for the target image partition information and the described image information that extracts, determine to be divided
The location information in multiple images region, wherein the partition information is to ask to determine according to region division input by user
, it is being determined according to division rule corresponding with the target image or the division option that automatically determines is directed to according to user
Selection operation determine；
According to the location information in multiple images region, region division processing is carried out to the target image, is obtained more
A image subblock for waiting for image Fuzzy Processing.
Optionally, computer executable instructions are when executed, described to determine each described image for dividing and obtaining respectively
The corresponding blur radius of block, including：
Blur radius setting request of the user for each image subblock is received, and the image is determined according to setting request
The blur radius of sub-block；
Alternatively,
According to the relevant information of each described image sub-block, the blur radius of described image sub-block is determined, wherein the correlation
Information includes：At least one of image content information, location information, fuzzy preceding rgb value distribution；
Alternatively,
The selection operation that user is directed to the multiple blur radius options automatically determined is received, and true according to the selection operation
The blur radius of fixed each image subblock.
Optionally, computer executable instructions are when executed, described to be directed to each described image sub-block, according to the image
The blur radius of sub-block carries out image Fuzzy Processing to the image subblock respectively, obtains the target image after Fuzzy Processing,
Including：
Present Fuzzy processing is chosen for each described image sub-block, in the multiple pixels for including in the image subblock
Target pixel points；
According to the corresponding blur radius of image subblock where the target pixel points, the multiple of the target pixel points are determined
Neighbor pixel；
According to the fuzzy preceding rgb value of the fuzzy preceding rgb value of the target pixel points and each neighbor pixel, institute is determined
The fuzzy rear rgb value for stating target pixel points, the fuzzy rear rgb value until determining each pixel in described image sub-block；
RGB after the rgb value of each pixel in the target image is adjusted to fuzzy by the fuzzy preceding rgb value of the pixel
Value, obtains the target image after Fuzzy Processing.
Optionally, computer executable instructions are when executed, described to obscure preceding rgb value according to the target pixel points
With the fuzzy preceding rgb value of each neighbor pixel, the fuzzy rear rgb value of the target pixel points is determined, including：
According to the fuzzy preceding rgb value of the fuzzy preceding rgb value of the target pixel points and each neighbor pixel, institute is determined
State the fuzzy preceding rgb value matrix of target pixel points；
Using specified fuzzy algorithmic approach according to the blur radius of image subblock where the target pixel points, the target is determined
The weight coefficient of the weight coefficient of pixel and each neighbor pixel obtains the fuzzy weighted values coefficient of the target pixel points
Matrix；
According to the fuzzy preceding rgb value matrix and the fuzzy weighted values coefficient matrix, the mould of the target pixel points is determined
Rgb value after paste.
Image processing equipment in the embodiment of the present application, after receiving the request of image Fuzzy Processing, acquisition waits for image mould
Paste the target image of processing；Region division processing is carried out to the target image, obtains multiple image for waiting for image Fuzzy Processing
Block；The corresponding blur radius of each image subblock for dividing and obtaining is determined respectively；For each image subblock, according to the image
The blur radius of sub-block carries out image Fuzzy Processing to the image subblock respectively, obtains the target image after Fuzzy Processing.It can
See, by the image processing equipment in the embodiment of the present application, by the way of carrying out fuzzy partition processing to target image, together
When, Fuzzy Processing is carried out using corresponding blur radius for different image subblocks, so that image is fuzzy to have more needle
To property, the stereovision of the target image after obscuring is improved, meets the variety of images Fuzzy Processing demand of user, improves user
Usage experience.
Preferably, the embodiment of the present application also provides a kind of image processing equipment, including processor 901, and memory 902 is deposited
The computer program that can be run on memory 902 and on processor 901 is stored up, which is executed by processor 902
Each process of the above-mentioned image processing method embodiments of Shi Shixian, and identical technique effect can be reached, to avoid repeating, here
It repeats no more.
Further, corresponding above-mentioned Fig. 1 is to method shown in Fig. 7, and based on the same technical idea, the embodiment of the present application is also
A kind of computer readable storage medium is provided, computer program is stored on computer readable storage medium, the computer journey
Each process of above-mentioned image processing method embodiment is realized when sequence is executed by processor, and can reach identical technique effect,
To avoid repeating, which is not described herein again.Wherein, the computer readable storage medium, such as read-only memory (Read-Only
Memory, abbreviation ROM), random access memory (Random Access Memory, abbreviation RAM), magnetic disc or CD etc..
Each embodiment in this specification is described in a progressive manner, identical similar portion between each embodiment
Point just to refer each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying example, since it is substantially similar to the method embodiment, so description is fairly simple, related place is referring to embodiment of the method
Part explanation.
Above is only an example of the present application, it is not intended to limit this application.For those skilled in the art
For, the application can have various modifications and variations.It is all within spirit herein and principle made by any modification, equivalent
Replace, improve etc., it should be included within the scope of claims hereof.
Claims (13)
1. a kind of image processing method, which is characterized in that including：
After receiving the request of image Fuzzy Processing, the target image for waiting for image Fuzzy Processing is obtained；
Region division processing is carried out to the target image, obtains multiple image subblocks for waiting for image Fuzzy Processing；
Determine respectively and divide obtained each corresponding blur radius of described image sub-block, wherein the blur radius according to
The relevant information of information-setting by user and/or image subblock determines, for indicating the corresponding fog-level of image subblock；
Figure is carried out to the image subblock according to the blur radius of the image subblock respectively for each described image sub-block
As Fuzzy Processing, the target image after Fuzzy Processing is obtained.
2. according to the method described in claim 1, it is characterized in that, for each described image sub-block, according to image
The blur radius of block, respectively to the image subblock carry out image Fuzzy Processing, obtain the target image after Fuzzy Processing it
Afterwards, further include：
According to the location information of each described image sub-block, the location information of sub-block intersection in the target image is determined；
According to the location information of the sub-block intersection, the target pixel points and the target pixel points for waiting for smoothing processing are determined
Rgb value after fuzzy；
According to the fuzzy rear rgb value of each target pixel points, the smooth rear rgb value of the target pixel points is determined respectively；
Rgb value after the rgb value of each target pixel points is adjusted to smooth by fuzzy rear rgb value, is obscured and is had a common boundary flat
Target image after cunning.
3. according to the method described in claim 2, it is characterized in that, the fuzzy rear RGB according to each target pixel points
Value determines the smooth rear rgb value of the target pixel points respectively, including：
According to the location information of each target pixel points, target pixel points are divided into multiple object pixels for waiting for smoothing processing
Point set；
For each object pixel point set, using linear interpolation method according to each object pixel in the object pixel point set
The fuzzy rear rgb value of point, determines the smoothing factor of each target pixel points respectively；
According to the smoothing factor of the fuzzy rear rgb value and the target pixel points of each target pixel points, determine respectively each
The smooth rear rgb value of the target pixel points.
4. according to the method described in claim 1, it is characterized in that, it is described to the target image carry out region division processing,
Multiple image subblocks for waiting for image Fuzzy Processing are obtained, including：
Image recognition processing is carried out to the target image, extracts the image information in the target image；
According to for the target image partition information and the described image information that extracts, determine to be divided multiple
The location information of image-region, wherein the partition information is according to region division input by user request determination, root
Selection determined according to division rule corresponding with the target image or that the division option automatically determined is directed to according to user
Operation determination；
According to the location information in multiple images region, region division processing is carried out to the target image, obtains multiple wait for
The image subblock of image Fuzzy Processing.
5. according to the method described in claim 1, it is characterized in that, described determine each described image sub-block for dividing and obtaining respectively
Corresponding blur radius, including：
Blur radius setting request of the user for each image subblock is received, and the image subblock is determined according to setting request
Blur radius；
Alternatively,
According to the relevant information of each described image sub-block, the blur radius of described image sub-block is determined, wherein the relevant information
Including：At least one of image content information, location information, fuzzy preceding rgb value distribution；
Alternatively,
The selection operation that user is directed to the multiple blur radius options automatically determined is received, and is determined respectively according to the selection operation
The blur radius of image subblock.
6. according to the method described in claim 1, it is characterized in that, described be directed to each described image sub-block, according to the image
The blur radius of sub-block carries out image Fuzzy Processing to the image subblock respectively, obtains the target image after Fuzzy Processing,
Including：
For each described image sub-block, the mesh of present Fuzzy processing is chosen in the multiple pixels for including in the image subblock
Mark pixel；
According to the corresponding blur radius of image subblock where the target pixel points, the multiple adjacent of the target pixel points is determined
Pixel；
According to the fuzzy preceding rgb value of the fuzzy preceding rgb value of the target pixel points and each neighbor pixel, the mesh is determined
The fuzzy rear rgb value for marking pixel, the fuzzy rear rgb value until determining each pixel in described image sub-block；
Rgb value after the rgb value of each pixel in the target image is adjusted to fuzzy by the fuzzy preceding rgb value of the pixel,
Obtain the target image after Fuzzy Processing.
7. according to the method described in claim 6, it is characterized in that, the fuzzy preceding rgb value according to the target pixel points
With the fuzzy preceding rgb value of each neighbor pixel, the fuzzy rear rgb value of the target pixel points is determined, including：
According to the fuzzy preceding rgb value of the fuzzy preceding rgb value of the target pixel points and each neighbor pixel, the mesh is determined
Mark the fuzzy preceding rgb value matrix of pixel；
Using specified fuzzy algorithmic approach according to the blur radius of image subblock where the target pixel points, the object pixel is determined
The weight coefficient of the weight coefficient and each neighbor pixel of point, obtains the fuzzy weighted values coefficient square of the target pixel points
Battle array；
According to it is described it is fuzzy before rgb value matrix and the fuzzy weighted values coefficient matrix, determine the target pixel points it is fuzzy after
Rgb value.
8. a kind of image processing apparatus, which is characterized in that including：
Image collection module, for after receiving the request of image Fuzzy Processing, obtaining the target image for waiting for image Fuzzy Processing；
Image division module obtains multiple image Fuzzy Processings of waiting for for carrying out region division processing to the target image
Image subblock；
Blur radius determining module, for determining each corresponding blur radius of described image sub-block for dividing and obtaining respectively,
Wherein, the blur radius is determined according to the relevant information of information-setting by user and/or image subblock, for indicating image subblock
Corresponding fog-level；
Image obfuscation module is right respectively according to the blur radius of the image subblock for being directed to each described image sub-block
The image subblock carries out image Fuzzy Processing, obtains the target image after Fuzzy Processing.
9. according to the method described in claim 8, it is characterized in that, further including：Sub-block boundary Leveling Block, wherein the son
Block boundary Leveling Block, is used for：
According to the location information of each described image sub-block, the location information of sub-block intersection in the target image is determined；
According to the location information of the sub-block intersection, the target pixel points and the target pixel points for waiting for smoothing processing are determined
Rgb value after fuzzy；
According to the fuzzy rear rgb value of each target pixel points, the smooth rear rgb value of the target pixel points is determined respectively；
Rgb value after the rgb value of each target pixel points is adjusted to smooth by fuzzy rear rgb value, is obscured and is had a common boundary flat
Target image after cunning.
10. according to the method described in claim 9, it is characterized in that, sub-block boundary Leveling Block, is specifically used for：
According to the location information of each target pixel points, target pixel points are divided into multiple object pixels for waiting for smoothing processing
Point set；
For each object pixel point set, using linear interpolation method according to each object pixel in the object pixel point set
The fuzzy rear rgb value of point, determines the smoothing factor of each target pixel points respectively；
According to the smoothing factor of the fuzzy rear rgb value and the target pixel points of each target pixel points, determine respectively each
The smooth rear rgb value of the target pixel points.
11. according to the method described in claim 8, it is characterized in that, described image division module, is specifically used for：
Image recognition processing is carried out to the target image, extracts the image information in the target image；
According to for the target image partition information and the described image information that extracts, determine to be divided multiple
The location information of image-region, wherein the partition information is according to region division input by user request determination, root
Selection determined according to division rule corresponding with the target image or that the division option automatically determined is directed to according to user
Operation determination；
According to the location information in multiple images region, region division processing is carried out to the target image, obtains multiple wait for
The image subblock of image Fuzzy Processing.
12. according to the method described in claim 8, it is characterized in that, the blur radius determining module, is specifically used for：
Blur radius setting request of the user for each image subblock is received, and the image subblock is determined according to setting request
Blur radius；
Alternatively,
According to the relevant information of each described image sub-block, the blur radius of described image sub-block is determined, wherein the relevant information
Including：At least one of image content information, location information, fuzzy preceding rgb value distribution；
Alternatively,
The selection operation that user is directed to the multiple blur radius options automatically determined is received, and is determined respectively according to the selection operation
The blur radius of image subblock.
13. according to the method described in claim 8, it is characterized in that, described image obfuscation module, is specifically used for：
For each described image sub-block, the mesh of present Fuzzy processing is chosen in the multiple pixels for including in the image subblock
Mark pixel；
According to the corresponding blur radius of image subblock where the target pixel points, the multiple adjacent of the target pixel points is determined
Pixel；
According to the fuzzy preceding rgb value of the fuzzy preceding rgb value of the target pixel points and each neighbor pixel, the mesh is determined
The fuzzy rear rgb value for marking pixel, the fuzzy rear rgb value until determining each pixel in described image sub-block；
Rgb value after the rgb value of each pixel in the target image is adjusted to fuzzy by the fuzzy preceding rgb value of the pixel,
Obtain the target image after Fuzzy Processing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810266817.1A CN108564541B (en) | 2018-03-28 | 2018-03-28 | Image processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810266817.1A CN108564541B (en) | 2018-03-28 | 2018-03-28 | Image processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108564541A true CN108564541A (en) | 2018-09-21 |
| CN108564541B CN108564541B (en) | 2022-04-15 |
Family
ID=63533202
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810266817.1A Active CN108564541B (en) | 2018-03-28 | 2018-03-28 | Image processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108564541B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110728632A (en) * | 2019-09-04 | 2020-01-24 | 北京奇艺世纪科技有限公司 | Image blurring processing method, image blurring processing device, computer device and storage medium |
| CN111199514A (en) * | 2019-12-31 | 2020-05-26 | 无锡宇宁智能科技有限公司 | Image background blurring method, device and equipment and readable storage medium |
| CN112488964A (en) * | 2020-12-18 | 2021-03-12 | 深圳市镜玩科技有限公司 | Image processing method for sliding list, related device, equipment and medium |
| CN113225606A (en) * | 2021-04-30 | 2021-08-06 | 上海哔哩哔哩科技有限公司 | Video barrage processing method and device |
| WO2023174137A1 (en) * | 2022-03-17 | 2023-09-21 | 北京字跳网络技术有限公司 | Image processing method and apparatus, and computer device and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101375588A (en) * | 2006-01-27 | 2009-02-25 | 微软公司 | Region-based image denoising |
| CN103945118A (en) * | 2014-03-14 | 2014-07-23 | 华为技术有限公司 | Picture blurring method and device and electronic equipment |
| CN104375797A (en) * | 2014-11-17 | 2015-02-25 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| CN104704820A (en) * | 2012-09-26 | 2015-06-10 | 富士胶片株式会社 | Image processing device, method, program, printer, and display device |
| US20160105601A1 (en) * | 2014-09-26 | 2016-04-14 | SZ DJI Technology Co., Ltd. | System and method for automatic focusing based on statistical data |
| CN105787888A (en) * | 2014-12-23 | 2016-07-20 | 联芯科技有限公司 | Human face image beautifying method |
| CN107169920A (en) * | 2017-04-24 | 2017-09-15 | 深圳市金立通信设备有限公司 | A kind of intelligence repaiies drawing method and terminal |
| CN107454332A (en) * | 2017-08-28 | 2017-12-08 | 厦门美图之家科技有限公司 | Image processing method, device and electronic equipment |
| CN107578459A (en) * | 2017-08-31 | 2018-01-12 | 北京麒麟合盛网络技术有限公司 | Expression is embedded in the method and device of candidates of input method |
-
2018
- 2018-03-28 CN CN201810266817.1A patent/CN108564541B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101375588A (en) * | 2006-01-27 | 2009-02-25 | 微软公司 | Region-based image denoising |
| CN104704820A (en) * | 2012-09-26 | 2015-06-10 | 富士胶片株式会社 | Image processing device, method, program, printer, and display device |
| CN103945118A (en) * | 2014-03-14 | 2014-07-23 | 华为技术有限公司 | Picture blurring method and device and electronic equipment |
| US20160105601A1 (en) * | 2014-09-26 | 2016-04-14 | SZ DJI Technology Co., Ltd. | System and method for automatic focusing based on statistical data |
| CN104375797A (en) * | 2014-11-17 | 2015-02-25 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| CN105787888A (en) * | 2014-12-23 | 2016-07-20 | 联芯科技有限公司 | Human face image beautifying method |
| CN107169920A (en) * | 2017-04-24 | 2017-09-15 | 深圳市金立通信设备有限公司 | A kind of intelligence repaiies drawing method and terminal |
| CN107454332A (en) * | 2017-08-28 | 2017-12-08 | 厦门美图之家科技有限公司 | Image processing method, device and electronic equipment |
| CN107578459A (en) * | 2017-08-31 | 2018-01-12 | 北京麒麟合盛网络技术有限公司 | Expression is embedded in the method and device of candidates of input method |
Non-Patent Citations (1)
| Title |
|---|
| 张向泽: "基于图像的景深渲染算法的研究与实现", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110728632A (en) * | 2019-09-04 | 2020-01-24 | 北京奇艺世纪科技有限公司 | Image blurring processing method, image blurring processing device, computer device and storage medium |
| CN110728632B (en) * | 2019-09-04 | 2022-07-12 | 北京奇艺世纪科技有限公司 | Image blurring processing method, image blurring processing device, computer device and storage medium |
| CN111199514A (en) * | 2019-12-31 | 2020-05-26 | 无锡宇宁智能科技有限公司 | Image background blurring method, device and equipment and readable storage medium |
| CN111199514B (en) * | 2019-12-31 | 2022-11-18 | 无锡宇宁智能科技有限公司 | Image background blurring method, device and equipment and readable storage medium |
| CN112488964A (en) * | 2020-12-18 | 2021-03-12 | 深圳市镜玩科技有限公司 | Image processing method for sliding list, related device, equipment and medium |
| CN112488964B (en) * | 2020-12-18 | 2024-04-16 | 深圳市镜玩科技有限公司 | Image processing method, related device, equipment and medium for sliding list |
| CN113225606A (en) * | 2021-04-30 | 2021-08-06 | 上海哔哩哔哩科技有限公司 | Video barrage processing method and device |
| WO2023174137A1 (en) * | 2022-03-17 | 2023-09-21 | 北京字跳网络技术有限公司 | Image processing method and apparatus, and computer device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108564541B (en) | 2022-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108564541A (en) | A kind of image processing method and device | |
| Fang et al. | Bottom-up saliency detection model based on human visual sensitivity and amplitude spectrum | |
| Celik et al. | Automatic image equalization and contrast enhancement using Gaussian mixture modeling | |
| US20170249743A1 (en) | Method and apparatus for dividing image area | |
| CN103473799B (en) | The method for dynamically processing of a kind of picture and device, terminal unit | |
| CN108197618B (en) | Method and device for generating human face detection model | |
| US20150326845A1 (en) | Depth value restoration method and system | |
| CN109087377B (en) | Method and apparatus for handling image | |
| CN109101919A (en) | Method and apparatus for generating information | |
| CN109472264B (en) | Method and apparatus for generating an object detection model | |
| Lee et al. | Real-time tracking of visually attended objects in virtual environments and its application to LOD | |
| CN104580883B (en) | Image processing apparatus, image processing method and recording medium | |
| CN110502974A (en) | A kind of methods of exhibiting of video image, device, equipment and readable storage medium storing program for executing | |
| CN107168672A (en) | Display methods and device | |
| CN108934016A (en) | Division methods, device, computer equipment and the storage medium of cell scenario classification | |
| CN108549836A (en) | Reproduction detection method, device, equipment and the readable storage medium storing program for executing of photo | |
| CN110321892A (en) | A kind of picture screening technique, device and electronic equipment | |
| CN109145783A (en) | Method and apparatus for generating information | |
| CN114365190A (en) | Spleen tumor identification method based on VRDS 4D medical image and related device | |
| CN109409979A (en) | Virtual cosmetic method, device and equipment | |
| CN107704128A (en) | Data processing method and device, storage medium and processor | |
| CN106340024B (en) | A kind of image partition method, application and calculate equipment | |
| Li et al. | Directive local color transfer based on dynamic look-up table | |
| CN112445704A (en) | Test method, test device and storage medium | |
| CN116468857A (en) | Electromagnetic simulation and optimization platform construction method and system based on brain modeling |
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 |