CN106530262A - Image synthesizing method and device - Google Patents
Image synthesizing method and device Download PDFInfo
- Publication number
- CN106530262A CN106530262A CN201510586747.4A CN201510586747A CN106530262A CN 106530262 A CN106530262 A CN 106530262A CN 201510586747 A CN201510586747 A CN 201510586747A CN 106530262 A CN106530262 A CN 106530262A
- Authority
- CN
- China
- Prior art keywords
- image
- light
- light image
- histogram
- synthesized
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
- Studio Devices (AREA)
Abstract
The invention provides an image synthesizing method and a device. The method comprises steps of acquiring a first image and a second image; decomposing the first image into a first illumination image and a first reflection image; according to the first illumination image, converting the second image into an image under an illumination condition of the first image; and synthesizing the converted second image into the first image. According to the invention, two image obtained under different illumination conditions can be synthesized together, and the synthesized image produces a natural effect.
Description
Technical field
The disclosure relates in general to computer vision field, and in particular to image combining method and equipment.
Background technology
Image synthesis refers to for multiple image to synthesize the new image of a width, and makes the image after synthesis
Seem natural reality as far as possible.For example, (the hereinafter letter of foreground object image is extracted from piece image
Claim foreground image), and to be synthesized to another width background image to form the new image of a width be exactly an allusion quotation
The image synthesis of type.Image synthesis has great importance in image with video editing application, and which causes
User can seamlessly be synthesized to the object extracted in piece image in other piece image, thus
Such as augmented reality, video conference etc. need many for synthesizing people or other objects and virtual background
Have a wide range of applications in occasion.
Synthesize with regard to image, the common method in this area includes Poisson picture editting, color transfer etc..
Image composition problem is expressed as interpolation problem using Poisson's equation by the former, but using the method synthesis
Color displacement problem is there may be in image.The latter is applied in the MatchColor functions of PhotoShop,
Which passes through the average of the color histogram of such as foreground image and background image to be synthesized and variance are alignd
To be synthesized.But, as the illumination of image and reflectivity properties are combined process by the method,
Therefore foreground image can be subject to the interference of background image and cause the result after synthesis that factitious effect is presented
Really.On the other hand, each width image generally to be synthesized is obtained under different time, place, scene
, thus with different illumination conditions.It is straight so if lighting issues are not considered when image synthesizes
Connect and a sub-picture is added in another piece image, then it is inconsistent due to illumination condition, will cause to generate
Composograph have obvious synthetic vestige, it appears it is unnatural untrue.Above-mentioned two in this area
Plant conventional synthetic method to have not been able to solve this problem well.
The content of the invention
In view of problem above and propose the disclosure.
According to the embodiment of an aspect of this disclosure, there is provided a kind of image combining method, including：Obtain
Take the first image and the second image；It is the first light image and the first reflected image by the first picture breakdown；
Second image is transformed to by the image under the illumination condition of the first image according to the first light image；Will conversion
The second image afterwards is synthesized in the first image.
According to the embodiment of another aspect of the disclosure, there is provided a kind of image composition equipment, including：
Obtaining widget, is configured to obtain the first image and the second image；Knock-Down Component, is configured to the first image
It is decomposed into the first light image and the first reflected image；Transform component, is configured to according to the first light image
Second image is transformed to into the image under the illumination condition of the first image；Compound component, is configured to convert
The second image afterwards is synthesized in the first image.
According to the embodiment of another aspect of the disclosure, there is provided a kind of image composition equipment, including：
Processor；Memorizer；With storage computer program instructions in which memory, in the computer
Programmed instruction performs following steps when being run by the processor：Obtain the first image and the second image；Will
First picture breakdown is the first light image and the first reflected image；According to the first light image by the second figure
As the image being transformed under the illumination condition of the first image；The second image after by conversion is synthesized to the first figure
As in.
By the image combining method and equipment of the above-mentioned aspect according to the disclosure, can be by different illumination bars
Two width images under part are synthesized together, and make the image after synthesis that natural effect is presented.
Description of the drawings
The embodiment of the present disclosure is described in more detail by combining accompanying drawing, the disclosure above-mentioned and its
Its purpose, feature and advantage will be apparent from.Accompanying drawing is used for providing entering one to the embodiment of the present disclosure
Step understands, and constitutes a part for description, is used for explaining the disclosure together with the embodiment of the present disclosure,
Restriction of this disclosure is not constituted.In the accompanying drawings, identical reference number typically represent same parts or
Step.
Fig. 1 schematically shows the possible applied field of image composing technique according to embodiments of the present invention
Scape.
The flow chart that Fig. 2 shows the image combining method according to the embodiment of the present disclosure.
Fig. 3 decomposes the background illumination figure for obtaining exemplified with exemplary background image and by the background image
Picture and background reflected image.
Fig. 4 show according to the embodiment of the present disclosure, foreground image is transformed to according to background illumination image
The flow chart of the process of the image under the illumination condition of background image.
Fig. 5 decomposes the prospect illumination pattern for obtaining exemplified with exemplary foreground image and by the foreground image
Picture and prospect reflected image.
Fig. 6 show according to the embodiment of the present disclosure, according to background illumination image by prospect light image become
The flow chart of the process of the image being changed under the illumination condition of background image.
Fig. 7 shows that the prospect light image after the conversion by example and prospect reflected image synthesize and is become
The illustrative case of the foreground image after changing.
Fig. 8 shows that the foreground image after the conversion by example is synthesized to the illustrative case in background image.
Fig. 9 shows the functional configuration block diagram of the image composition equipment according to the embodiment of the present disclosure.
Figure 10 shows the calculating for realization according to the example images synthesis device of the embodiment of the present disclosure
Equipment block diagram.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present disclosure, the technical scheme in the embodiment of the present disclosure is carried out
Clearly and completely describe, it is clear that described embodiment is only a part of embodiment of the disclosure, and
It is not all, of embodiment.Based on the embodiment in the disclosure, those of ordinary skill in the art are not doing
The every other embodiment obtained under the premise of going out creative work, belongs to the scope of disclosure protection.
Fig. 1 schematically shows the possible applied field of image composing technique according to embodiments of the present invention
Scape.As shown in figure 1, subscriber station is in the image pickup scope of photographic head 101, the photographic head 101 is to user
Shot, the photographic head 101 can be 2D or 3D photographic head.The processing equipment 102 of such as computer
Image to being shot by photographic head 101 is analyzed, and extracts foreground object, by foreground object and a certain back of the body
Scape image is synthesized, and generates new composograph.Projector 103 and projection screen 104 are used to show
The other display equipment of purpose, such as display, TV etc. also can used as said purpose.It should be noted that
Fig. 1 illustrate only a kind of possible application scenarios of the present invention, according to practical situation, in application scenarios
Device correspondingly can increase or decrease, and with different configurations.
As it was previously stated, image synthesizes for multiple image is synthesized the new image of a width.Hereinafter,
For the ease of description, by the image of the foreground object extracted from a sub-picture (hereinafter referred to as foreground picture
Picture) with a secondary background image synthesis as a example by, image composing technique of the invention is described, its
In comprising the foreground image image and the background image be under different time, place and scene
Obtain, thus with different illumination conditions.
The image combining method according to the embodiment of the present disclosure is described below with reference to Fig. 2.Fig. 2 is illustrated
According to the flow chart of the image combining method of the embodiment of the present disclosure.
As shown in Fig. 2 in step S210, obtaining foreground image and background image.
In this step, foreground image to be synthesized and background image can be obtained by any mode.Example
Such as, can be from foreground image described in outside input and background image, it is also possible to which shooting obtains the foreground picture
Picture and background image.
In step S220, background image is decomposed into into background illumination image and background reflected image.
Theoretical, the piece image according to the Retinex color constancy that Edward Land were proposed in 1986
Illumination (illumination) image and reflection (reflectance) figure can be broken down into as shown in expression formula (1)
Picture.
I (x, y)=S (x, y) R (x, y) ... (1)
Wherein, S (x, y) represents light image, and R (x, y) represents reflected image, and (x, y) is pixel in image
Coordinate.
At present, in computer vision field, for the technology of picture breakdown has been carried out many researchs,
And propose many specific picture breakdown methods.For example, Arjan Gijsenij were delivered in 2011
“Color constancy using Natural Image Statistics and Scene Semantics”、Nikola
" the Using the random sprays retinex algorithm for global that Banic was delivered in 2013
" the Intrinsic images in that illumination estimation ", Sean Bell was delivered in 2014
Specific picture breakdown method is given in the articles such as wild ".In the present invention, as an example, adopt
The picture breakdown method proposed by Sean Bell.What is be merely to illustrate that below is complete, to the image point
Solution method is simply introduced.
According to the method that Sean Bell are proposed, picture breakdown is found and best suits the anti-of following probability distribution p
Penetrate image R and light image S：
Meet
WhereinRepresent Color Channel c ∈ { r, g, b }
Specifically, in the method that Sean Bell are proposed, first by carrying out k-means to pixel color
Cluster, initializes the set of reflected image R；Then, to each pixel from the set of reflected image R
Choose one to be marked so that and p (R, S | I) maximize；Finally, by minimizing in light image S
Discontinuity, adjusts reflected image R.
Fig. 3 decomposes the background illumination figure for obtaining exemplified with exemplary background image and by the background image
Picture and background reflected image.It can be seen that background image and thus decomposing obtain background illumination image,
Background reflected image is of the same size.
Above-mentioned picture breakdown method is only an example, in step S220, those skilled in the art
Background image can be decomposed into by background illumination image S using any of picture breakdown methodBG(x,y)
With background reflected image RBG(x,y)。
In step S230, foreground image is transformed to according to background illumination image for the illumination condition of background image
Under image.The process in the step is described in detail below in conjunction with Fig. 4.Fig. 4 shows
According to the embodiment of the present disclosure, illumination that foreground image is transformed to background image according to background illumination image
Under the conditions of image process flow chart.
As shown in figure 4, in step S2301, foreground image being decomposed into prospect light image and prospect being anti-
Penetrate image.
It is identical with the decomposition of background image, in this step, can be using any of in this area
Foreground image is decomposed into prospect light image S by picture breakdown methodFG(x, y) and prospect reflected image
RFG(x, y), here is omitted.Fig. 5 is exemplified with exemplary foreground image and by the foreground image point
Prospect light image and prospect reflected image that solution is obtained.
In step S2302, prospect light image is transformed to according to background illumination image for the light of background image
Image according under the conditions of.
As an example, in this step using the rectangular histogram of background illumination image and prospect light image to front
Scape light image enters line translation.On the other hand, as it was previously stated, foreground image is extracted from a sub-picture
Foreground object image, its be normally only whole image a part, and synthesis when, also only
The subregion of background image is covered only.That is, the foreground image being synthesized can only be subject to Background
The impact of local illumination condition as in.Accordingly, as example, in this step, using background illumination figure
As in, the rectangular histogram of regional area enters line translation to the rectangular histogram of prospect light image.
The process in the step is described in detail below in conjunction with Fig. 6.Fig. 6 is shown according to this
Open embodiment, prospect light image is transformed to the illumination bar of background image according to background illumination image
The flow chart of the process of the image under part.
In step S23021, in determining background image, foreground image will be synthesized to region therein.
The region that the step will be covered for the positioning foreground image in background image.For example, with Fig. 4
As a example by the foreground image and background image shown in Fig. 5, it is assumed that the width of foreground image is w, is highly
H, then in background image position a foreground image in this step and will be synthesized to width therein and height
Degree is respectively the region of w and h.
Local Nogata in step S23022, calculating background illumination image with the region corresponding region
Figure.
As it was previously stated, the background illumination image that background image and decomposition are obtained is of the same size.
In the step, the local of the region corresponding region for positioning in calculating background illumination image and in background image
Rectangular histogram.
The rectangular histogram of image is a kind of conventional figure for describing gradation of image distribution, and which generally uses transverse axis
The gray level of image is represented, the number of pixels that each gray level is included is represented with the longitudinal axis.In this step, make
For example, the region for positioning in such as calculating background illumination image shown in expression formula (3) and in background image
The local histogram of corresponding region：
Wherein, rkRepresent kth level gray value, niIn representing image, gray scale is riPixel number, n tables
Show the number of pixel in whole image, P (rk) it is gray value rkThe probability for occurring in the picture.
In step S23023, the rectangular histogram of prospect light image is calculated.
In this step, the rectangular histogram of prospect light image is similarly calculated with step S23022, herein not
Describe in detail again.
In step S23024, using the local histogram of background illumination image as goal histogram, to prospect
The rectangular histogram of light image carries out regulation, with the prospect light image after being converted.
Histogram specification is common image processing method as known in the art.So-called histogram specification,
It is exactly, by a grayscale mapping function, former grey level histogram to be transformed into into the histogram shape of regulation.Directly
Side's figure regulationization is actually the process to the image representated by rectangular histogram, i.e., by carrying out to input picture
Histogram specification process so that output image has the histogram shape of regulation.In specific to the step,
It is then, as shown in expression formula (4), the rectangular histogram of the prospect light image as input picture to be specified
Change is processed, and to make it have local histogram's shape of background illumination image, is thus obtained as output figure
Prospect light image after the conversion of picture：
zk=G-1(sk), wherein
Wherein G is grayscale mapping function, zkThe regulation probability density function that output image has is wished in expression
Discretized values, skRepresent the discretized values of the probability density function of input picture, Pr(rj) represent and wish defeated
Go out the specified probability density function that image has, rjJ-th stage gray value is represented, L is discrete gray levels
Quantity.
Processed by above-mentioned histogram specification, in the prospect light image as input picture
Each pixel, by its original gray value is mapped to new gray value, realizes prospect light image
Conversion.
More than, with reference to Fig. 6 to the rectangular histogram using background illumination image and prospect light image to prospect light
The example process being transformed to according to image under the illumination condition of background image is described.Optionally, at this
In process, can be fritter by prospect light image and background illumination image division, and by little one by one
Carry out histogram specification to realize the conversion of prospect light image block.Specifically, when shown in Fig. 6
The step of S23022 in when calculating local histogram with the region corresponding region in background illumination image,
First by region division corresponding with the region in background illumination image be m × n fritter, then for every
The individual fritter calculates its rectangular histogram；It is similar, when calculating prospect in step S23023 shown in Fig. 6
During the rectangular histogram of light image, prospect light image is equally divided into into the m × n fritter, Ran Houzhen
Rectangular histogram is calculated to fritter each described；Subsequently, when Nogata is carried out in step S23024 shown in Fig. 6
During the process of figure specification processing, for the rectangular histogram of each fritter of prospect light image, with background illumination
The rectangular histogram of the corresponding fritter in image carries out histogram specification as target, is achieved in prospect light
According to the conversion of image.Histogrammic regulation is carried out by as above block-by-block, illumination can be kept well to become
The flatness of change so that the prospect light image after conversion is related to the illumination patterns position of background image,
Also can more preferable effect of the simulation prospect light image under the illumination condition of background image.
Fig. 4 is returned to, the prospect light image S ' after step S2303, synthesis conversionFG(x, y) is anti-with prospect
Penetrate image RFG(x, y) is with the foreground image after being converted.
As noted above, piece image can be broken down into as shown in expression formula (1) light image and
Reflected image.In this step, the expression formula (1), the prospect light after synthesis conversion be again may be by
According to image S 'FG(x, y) and prospect reflected image RFG(x, y) is with the foreground image after being converted.Fig. 7 is illustrated
Foreground picture after being converted synthesized by the prospect light image and prospect reflected image after the conversion of example
The illustrative case of picture.
Optionally, if by prospect light image and background illumination image division in above-mentioned steps S2302
For fritter, and carry out histogram specification to realize the conversion of prospect light image by patch one by one,
Then in step S2303, before synthesis process is carried out, can first to the prospect illumination pattern after conversion
As carrying out the smothing filterings such as gaussian filtering, to eliminate possible " blocking effect ", subsequently by smothing filtering
Prospect light image afterwards is synthesized with prospect reflected image with the foreground image after being converted.
Fig. 2 is returned to, in step S240, the foreground image after conversion is synthesized in background image.
In this step, the foreground image after by conversion directly overlies and is positioned in background image
Region to be covered, obtain synthesize image.Fig. 8 shows the foreground image after the conversion by example
The illustrative case being synthesized in background image.
Image combining method by reference to Description of Drawings according to the embodiment of the present disclosure above.In the method
In, foreground image is transformed to by background image according to the background illumination image obtained by background image decomposition first
Illumination condition under image, the foreground image after conversion is synthesized to into background image then such that it is able to
The two width images obtained under different illumination conditions are synthesized together, and present the image after synthesis
Natural effect.
It should be noted that in the above description as a example by foreground image to be synthesized with background image, it is right
It is described according to the image composing technique of the disclosure.It is understood that the disclosure is not limited to foreground picture
The synthesis of picture and background image, and can be used for the synthesis of any two width image.More particularly, on
Background image and foreground image described in text can be arbitrary first image and the second image respectively, only
Want the size of the second image relatively small so as to whole first image of covering will not be blocked in synthesis.
The image composition equipment according to the embodiment of the present disclosure is described below with reference to Fig. 9.Fig. 9 shows basis
The functional configuration block diagram of the image composition equipment of the embodiment of the present disclosure.As shown in figure 9, image composition equipment
900 can include：Obtaining widget 910, Knock-Down Component 920, transform component 930, and compound component
940.The concrete function of each part and operation are essentially identical with what is described above for Fig. 2-8, because
This hereinafter only carries out brief description, and omits to identical thin in order to avoid repeating to the equipment
The detailed description of section.
Obtaining widget 910 is configured to obtain the first image and the second image.The obtaining widget 910 can be with
Foreground image to be synthesized and background image are obtained by various appropriate modes.For example, the acquisition unit
Part 910 can be image unit, obtain foreground image to be synthesized and background image to shoot, or should
Obtaining widget 910 can be input block, with from foreground image described in outside input and background image.
It is the first light image and the first reflected image that Knock-Down Component 920 is configured to the first picture breakdown.
Specifically, Knock-Down Component 920 can be using any of picture breakdown method in computer vision field
It is the first light image S by the first picture breakdownBG(x, y) and the first reflected image RBG(x, y), it is no longer detailed herein
It is thin to describe.
Transform component 930 is configured to the light that the second image is transformed to the first image according to the first light image
Image according under the conditions of.The transform component 930 can include sub- Knock-Down Component 9301, sub- change as an example
Change part 9302 and sub- compound component 9303.
It is the second light image S that sub- Knock-Down Component 9301 is configured to the second picture breakdownFG(x, y) and second
Reflected image RFG(x,y)。
Second light image is transformed to the first image according to the first light image by sub- transform component 9302
Image under illumination condition.As an example, the sub- transform component 9302 utilizes the first light image and second
The rectangular histogram of light image enters line translation to the second light image.Specifically, sub- transform component 9302 determines
In first image, the second image will be synthesized to region therein, calculate in the first light image with it is described
The local histogram of region corresponding region, and calculate the rectangular histogram of the second light image.Sub- transform component 9302
It is further configured to using the local histogram of the first light image as goal histogram, to the second illumination pattern
The rectangular histogram of picture carries out regulation, with the second light image after being converted.Histogram specification is this
Known common image processing method in field.So-called histogram specification, is exactly reflected by a gray scale
Function is penetrated, former grey level histogram is transformed into into the histogram shape of regulation.Herein, sub- transform component 9302
Professional etiquette is entered to the rectangular histogram of the second light image as input picture by expression formula (4) mentioned above
Determining process, to make it have local histogram's shape of the first light image, thus obtains as output
The second light image after the conversion of image.Optionally, sub- transform component 9302 can be by the second illumination pattern
Picture and the first light image are divided into fritter, and carry out histogram specification come real by patch one by one
The conversion of existing second light image.Specifically, sub- transform component 9302 calculate the first light image in
During the local histogram of the region corresponding region, first will be corresponding with the region in the first light image
Region division is m × n fritter, then calculates its rectangular histogram for fritter each described；Similar,
When calculating the rectangular histogram of the second light image, the second light image is equally divided into by sub- transform component 9302
M × n the fritter, then calculates rectangular histogram for fritter each described；Subsequently, entering column hisgram
During specification processing process, sub- transform component 9302 is for the Nogata of each fritter of the second light image
Figure, the rectangular histogram using the corresponding fritter in the first light image carry out histogram specification as target,
It is achieved in the conversion of the second light image.Histogrammic regulation is carried out by as above block-by-block, can be very
Keep well the flatness of illumination variation so that the illumination of the second light image and the first image after conversion
Distributing position is related, also can more preferable effect of the second light image of simulation under the illumination condition of the first image
Really.
Sub- compound component 9303 be configured to synthesize conversion after the second light image and the second reflected image with
The second image after being converted.Specifically, sub- compound component 9303 can pass through the above table
Synthesize the second light image and the second reflected image after conversion up to formula (1) with second after being converted
Image.Can be selected in, such as fruit transform component 9302 is configured to the second light image and the first illumination pattern
As being divided into fritter, and carry out histogram specification to realize the second light image by patch one by one
Conversion, then sub- compound component 9303 before synthesis process is carried out, can first to conversion after the second illumination
Image carries out the smothing filterings such as gaussian filtering, to eliminate possible " blocking effect ", subsequently by smooth filter
The second light image and the second reflected image after ripple is synthesized with the second image after being converted.
The second image after compound component 940 is configured to conversion is synthesized in the first image.Specifically,
Second image of the compound component 940 after by conversion directly overlie positioned in the first image will
Coated region, obtains the image for synthesizing.
The image composition equipment 900 according to the embodiment of the present disclosure is described by reference to Fig. 9 above.The figure
As the second image is transformed to by synthesis device 900 according to the first light image obtained by the first picture breakdown
Image under the illumination condition of the first image, subsequently by conversion after the second image be synthesized to the first image,
The two width images for being thus able to obtain under different illumination conditions are synthesized together, and after making synthesis
Image is presented natural effect.
Below, can be used for the example images synthesis device of realizing the embodiment of the present disclosure to describe with reference to Figure 10
Computing device block diagram.
As shown in Figure 10, computing device 1000 includes one or more processors 1002, storage device
1004th, photographic head 1006 and output device 1008, these components pass through bus system 1010 and/or other
Bindiny mechanism's (not shown) interconnection of form.It should be noted that computing device 1000 shown in Figure 10
Component and structure are illustrative, and not restrictive, and as needed, computing device 1000 can also
With other assemblies and structure.
Processor 1002 can be CPU (CPU) or have data-handling capacity and/or
The processing unit of the other forms of instruction execution capability, and other in computing device 1000 can be controlled
Component is performing desired function.
Storage device 1004 can include one or more computer programs, and the computer program is produced
Product can include various forms of computer-readable recording mediums, such as volatile memory and/or non-volatile
Property memorizer.The volatile memory can for example include random access memory (RAM) and/or height
Fast buffer storage (cache) etc..The nonvolatile memory can for example include read only memory
(ROM), hard disk, flash memory etc..One or many can be stored on the computer-readable recording medium
Individual computer program instructions, processor 1002 can run described program instruction, mentioned above to realize
The function of embodiment of the disclosure and/or other desired functions.In the computer-readable storage medium
Various application programs and various data can also be stored in matter, such as it is above-mentioned first light image, first anti-
Penetrate image, the second light image, the second reflected image, each rectangular histogram, the second image etc. after conversion
Deng.
Photographic head 1006 is used to shoot described first image and the second image, and by the first captured figure
Picture and the second image are stored in storage device 1004 and use for other components.
Output device 1008 to the various information of outside output, such as image for synthesizing, and be able to can be wrapped
Include the various display devices such as display, projector, TV.
The ultimate principle of the disclosure is described above in association with specific embodiment, however, it is desirable to, it is noted that
The advantage that refers in the disclosure, advantage, effect etc. are only exemplary rather than limiting, it is impossible to think that these are excellent
Point, advantage, effect etc. are that each embodiment of the disclosure is prerequisite.In addition, tool disclosed above
Body details is merely to the effect of example and the effect for readily appreciating, and unrestricted, and above-mentioned details is not limited
The disclosure processed is must to be realized using above-mentioned concrete details.
The device that is related in the disclosure, device, equipment, the block diagram only illustratively example of property of system
And it is not intended to require or imply and must be attached, arrange, configure according to the mode that square frame is illustrated.
As the skilled person will recognize, can be connected, be arranged by any-mode, configure these devices,
Device, equipment, system.Such as " include ", "comprising", the word of " having " etc. are open vocabulary,
Refer to " including but not limited to ", and can be with its used interchangeably.Vocabulary "or" used herein above and " and " refer to vocabulary
"and/or", and can be with its used interchangeably, unless it be not such that context is explicitly indicated.It is used herein above
Vocabulary " such as " refers to phrase " such as, but not limited to ", and can be with its used interchangeably.
The step flow chart and above method description only illustratively example of property and unawareness in the disclosure
The step of figure is required or hint must carry out each embodiment according to the order for providing, some steps can be simultaneously
Row, perform independently of one another or according to other appropriate orders.In addition, such as " thereafter ", " and then ", " connect
Get off " etc. word be not intended to limit the order of step；These words are only used for guiding reader to read over these
The description of method.
It may also be noted that in the apparatus and method of the disclosure, each part or each step can be point
Solve and/or reconfigure.These decompose and/or reconfigure the equivalents that should be regarded as the disclosure.
The above description of disclosed aspect is provided so that any person skilled in the art can make or
Using the disclosure.It is very aobvious and easy to those skilled in the art to the various modifications in terms of these
See, and General Principle defined herein is can apply in terms of other without deviating from the scope of the present disclosure.
Therefore, the disclosure is not intended to be limited to the aspect being shown in which, but according to principle disclosed herein
The widest range consistent with novel feature.Although already discussed above multiple exemplary aspects and embodiment,
But it would be recognized by those skilled in the art that its some modification, modification, change, addition and sub-portfolio.
Claims (9)
1. a kind of image combining method, including：
Obtain the first image and the second image；
It is the first light image and the first reflected image by the first picture breakdown；
Second image is transformed to by the image under the illumination condition of the first image according to the first light image；
The second image after by conversion is synthesized in described first image.
2. image combining method as claimed in claim 1, wherein it is described according to the first light image by
The image that two images are transformed under the illumination condition of the first image is further included：
It is the second light image and the second reflected image by the second picture breakdown；
Second light image is transformed to by the figure under the illumination condition of the first image according to the first light image
Picture；
The second light image and the second reflected image after synthesis conversion is with the second image after being converted.
3. image combining method as claimed in claim 2, wherein it is described according to the first light image by
The image that two light images are transformed under the illumination condition of the first image is further included：
In determining the first image, the second image will be synthesized to region therein；
Calculate the local histogram with the region corresponding region in the first light image；
Calculate the rectangular histogram of the second light image；
Using the local histogram of the first light image as goal histogram, the Nogata to the second light image
Figure carries out regulation, with the second light image after being converted.
4. image combining method as claimed in claim 3, wherein in calculating the first light image with it is described
The local histogram in the corresponding region in region includes：
It is m × n fritter by region division corresponding with the region in the first light image；
Rectangular histogram is calculated for fritter each described.
5. image combining method as claimed in claim 4, wherein calculating the rectangular histogram of the second light image
Including：
Second light image is divided into into the m × n fritter；
Rectangular histogram is calculated for fritter each described.
6. image combining method as claimed in claim 5, wherein with the local Nogata of the first light image
Used as goal histogram, carry out regulation to the rectangular histogram of the second light image includes figure：
For the rectangular histogram of each fritter of the second light image, with the correspondence in the first light image
The rectangular histogram of fritter carries out regulation as goal histogram.
7. image combining method as claimed in claim 6, wherein the second light image after synthesis conversion
Further included with the second image after being converted with the second reflected image：
The second light image after to conversion carries out smothing filtering；
After the second light image and the second reflected image after by smothing filtering is synthesized to be converted
Second image.
8. a kind of image composition equipment, including：
Obtaining widget, is configured to obtain the first image and the second image；
Knock-Down Component, it is the first light image and the first reflected image to be configured to the first picture breakdown；
Transform component, is configured to the second image is transformed to the illumination of the first image according to the first light image
Under the conditions of image；
Compound component, the second image after being configured to conversion are synthesized in the first image.
9. a kind of image composition equipment, including：
Processor；
Memorizer；With
Storage computer program instructions in which memory, it is described in the computer program instructions
Processor performs following steps when running：
Obtain the first image and the second image；
It is the first light image and the first reflected image by the first picture breakdown；
Second image is transformed to by the image under the illumination condition of the first image according to the first light image；
The second image after by conversion is synthesized in the first image.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510586747.4A CN106530262A (en) | 2015-09-15 | 2015-09-15 | Image synthesizing method and device |
| JP2016179761A JP6226050B2 (en) | 2015-09-15 | 2016-09-14 | Image composition method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510586747.4A CN106530262A (en) | 2015-09-15 | 2015-09-15 | Image synthesizing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106530262A true CN106530262A (en) | 2017-03-22 |
Family
ID=58349216
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510586747.4A Pending CN106530262A (en) | 2015-09-15 | 2015-09-15 | Image synthesizing method and device |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP6226050B2 (en) |
| CN (1) | CN106530262A (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110322440B (en) * | 2019-07-08 | 2023-01-13 | 东北大学 | Method for expanding cell microscopic image data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020080148A1 (en) * | 2000-11-09 | 2002-06-27 | Fumiko Uchino | Image processing device and image processing method |
| CN101102399A (en) * | 2007-07-26 | 2008-01-09 | 上海交通大学 | Real time digital image processing and enhancing method with noise removal function |
| CN101102398A (en) * | 2007-07-26 | 2008-01-09 | 上海交通大学 | Fully automatic real-time digital image processing enhancement system |
| CN101247468A (en) * | 2007-02-15 | 2008-08-20 | 北京思比科微电子技术有限公司 | Method and apparatus for regulating image brightness |
| CN101739672A (en) * | 2009-12-02 | 2010-06-16 | 北京中星微电子有限公司 | Method and device for equalizing histogram based on sub-regional interpolation |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3971783B2 (en) * | 2004-07-28 | 2007-09-05 | 松下電器産業株式会社 | Panorama image composition method and object detection method, panorama image composition device, imaging device, object detection device, and panorama image composition program |
| US8606050B2 (en) * | 2011-06-16 | 2013-12-10 | Tandent Vision Science, Inc. | Method for processing multiple images of a same scene |
-
2015
- 2015-09-15 CN CN201510586747.4A patent/CN106530262A/en active Pending
-
2016
- 2016-09-14 JP JP2016179761A patent/JP6226050B2/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020080148A1 (en) * | 2000-11-09 | 2002-06-27 | Fumiko Uchino | Image processing device and image processing method |
| CN101247468A (en) * | 2007-02-15 | 2008-08-20 | 北京思比科微电子技术有限公司 | Method and apparatus for regulating image brightness |
| CN101102399A (en) * | 2007-07-26 | 2008-01-09 | 上海交通大学 | Real time digital image processing and enhancing method with noise removal function |
| CN101102398A (en) * | 2007-07-26 | 2008-01-09 | 上海交通大学 | Fully automatic real-time digital image processing enhancement system |
| CN101739672A (en) * | 2009-12-02 | 2010-06-16 | 北京中星微电子有限公司 | Method and device for equalizing histogram based on sub-regional interpolation |
Non-Patent Citations (2)
| Title |
|---|
| TATSUKI MURAKAMI ET AL.: "Background replacement using chromatic adaptation transform for visual communication", 《J.VIS.COMMUN.IMAGE R.》 * |
| 林忠涛: "光照变化条件下人脸识别技术研究", 《中国优秀硕士学位论文全文数据库（电子期刊）-信息科技辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2017059233A (en) | 2017-03-23 |
| JP6226050B2 (en) | 2017-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11055828B2 (en) | Video inpainting with deep internal learning | |
| KR102134405B1 (en) | System and Method for Improving Low Light Level Image Using Generative Adversarial Network | |
| DE102019216136A1 (en) | METHOD AND SYSTEM FOR 3D RECONSTRUCTION WITH VOLUME-BASED FILTERING FOR IMAGE PROCESSING | |
| Allene et al. | Seamless image-based texture atlases using multi-band blending | |
| US8698796B2 (en) | Image processing apparatus, image processing method, and program | |
| US11875451B2 (en) | Arbitrary view generation | |
| US7593020B2 (en) | Image editing using image-wide matting | |
| US20160196665A1 (en) | Locating and Augmenting Object Features in Images | |
| DE112018007721T5 (en) | Acquire and modify 3D faces using neural imaging and time tracking networks | |
| US7630541B2 (en) | Image-wide matting | |
| KR102530116B1 (en) | Arbitrary view generation | |
| US9324170B2 (en) | Creating a blended image | |
| US9483812B2 (en) | Enhancing motion pictures with accurate motion information | |
| US20190206117A1 (en) | Image processing method, intelligent terminal, and storage device | |
| US20210217225A1 (en) | Arbitrary view generation | |
| US20180225882A1 (en) | Method and device for editing a facial image | |
| CN113039576A (en) | Image enhancement system and method | |
| KR20240089729A (en) | Image processing methods, devices, storage media and electronic devices | |
| Hervieu et al. | Stereoscopic image inpainting: distinct depth maps and images inpainting | |
| US20170287136A1 (en) | Method and apparatus for real-time matting using local color estimation and propagation | |
| CN106530262A (en) | Image synthesizing method and device | |
| KR20130075636A (en) | Apparatus and method for automatic object segmentation for background composition | |
| CN109949400B (en) | Shadow estimation and reconstruction method suitable for AR virtual soft-loading synthesis | |
| TWM625817U (en) | Image simulation system with time sequence smoothness | |
| CN114463213A (en) | Video processing method, video processing device, terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170322 |
|
| RJ01 | Rejection of invention patent application after publication |