CN110415188A - A kind of HDR image tone mapping method based on Multiscale Morphological - Google Patents
A kind of HDR image tone mapping method based on Multiscale Morphological Download PDFInfo
- Publication number
- CN110415188A CN110415188A CN201910622002.7A CN201910622002A CN110415188A CN 110415188 A CN110415188 A CN 110415188A CN 201910622002 A CN201910622002 A CN 201910622002A CN 110415188 A CN110415188 A CN 110415188A
- Authority
- CN
- China
- Prior art keywords
- image
- hdr image
- pyramid
- carries out
- detail
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000013507 mapping Methods 0.000 title claims abstract description 34
- 230000000877 morphologic effect Effects 0.000 title claims abstract description 18
- 238000012423 maintenance Methods 0.000 claims abstract description 43
- 238000007906 compression Methods 0.000 claims abstract description 29
- 230000006835 compression Effects 0.000 claims abstract description 28
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 13
- 230000002708 enhancing effect Effects 0.000 claims abstract description 12
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 238000012937 correction Methods 0.000 claims abstract description 4
- 230000009466 transformation Effects 0.000 claims description 52
- 238000005498 polishing Methods 0.000 claims description 37
- 238000001914 filtration Methods 0.000 claims description 9
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 claims description 6
- 239000010931 gold Substances 0.000 claims description 6
- 229910052737 gold Inorganic materials 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 235000013399 edible fruits Nutrition 0.000 description 8
- 230000000694 effects Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000009499 grossing Methods 0.000 description 5
- 238000013459 approach Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 101000648554 Mus musculus Sushi domain-containing protein 6 Proteins 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- KAATUXNTWXVJKI-UHFFFAOYSA-N cypermethrin Chemical compound CC1(C)C(C=C(Cl)Cl)C1C(=O)OC(C#N)C1=CC=CC(OC=2C=CC=CC=2)=C1 KAATUXNTWXVJKI-UHFFFAOYSA-N 0.000 description 1
- 230000023077 detection of light stimulus Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 125000001475 halogen functional group Chemical group 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- NTHWMYGWWRZVTN-UHFFFAOYSA-N sodium silicate Chemical compound [Na+].[Na+].[O-][Si]([O-])=O NTHWMYGWWRZVTN-UHFFFAOYSA-N 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20208—High dynamic range [HDR] image processing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of HDR image tone mapping method based on Multiscale Morphological, this method comprises: S1, inputs HDR image to be compressed；S2 constructs the logarithmic image of HDR image to be compressed；S3 carries out dynamic range compression multi-resolution decomposition to logarithmic image, obtains the first gaussian pyramid and the first laplacian pyramid；S4 polishes the enhanced levels of detail of the first gaussian pyramid of operator extraction using a maintenance boundary, the levels of detail of the first gaussian pyramid is successively added to the corresponding layer of the first laplacian pyramid, obtains the second laplacian pyramid；S5 carries out dynamic range compression reconstruct to the second laplacian pyramid, and the details after obtaining dynamic range compression enhances image；S6 carries out exponential transform to the image of details enhancing, obtains the first LDR image；S7 carries out color correction to the first LDR image, obtains LDR image to be output.The present invention can be realized being effectively compressed for HDR image dynamic range, and can meet certain real-time application demands.
Description
Technical field
The present invention relates to technical field of image processing, in particular to a kind of HDR image tone based on Multiscale Morphological
Mapping method.
Background technique
HDR (full name in English is " High Dynamic Range ", and Chinese name is " high dynamic range ") image compares LDR
(full name in English is " Low Dynamic Range ", and Chinese name is " low-dynamic range ") image can express more scenes
Details and brightness contrast information, are widely used in Digital image technology.Dedicated HDR shows that equipment manufacturing cost is high
Expensive, cost performance is low.HDR image LDR show equipment on show when, can not true reappearance original scene complete effect of shadow, and
Being effectively compressed for HDR image dynamic range may be implemented in good tone mapping method, solves dynamic range mismatch problem, with full
The more practical application requests of foot.
In general, HDR image field is broadly divided into global tone mapping method and part in relation to the method for tone mapping
Tone mapping method two major classes.Wherein, global tone mapping method all uses together each of HDR image pixel
The mapping function of sample is converted.Such method is one-to-one reflect due to being each pixel using the same mapping function
Relationship is penetrated, therefore the advantages of such method is that method is simple, the speed of service is very fast, but the method does not account for the space of pixel
Position only considered the gray value of pixel, will lead to result images and causes damages in terms of brightness, color and details.Earliest
In 1984, Miller was based on Stevens psychophysical testing data and proposes a kind of global approach.1993, Tumblin and
Rushmeier equally proposes a kind of global approach, this method for brightness domain on the basis of Stevens psychophysical testing
It is nonlinear.1994, Ward view-based access control model sensibility proposed the Linear Mapping that image comparison brightness can be enable to be promoted
Method.2003, Drago proposed a kind of tone mapping method on the basis of logarithmic transformation, because of logarithmic transformation and human eye
It is even more like to the perception of light, but this method loss in detail is serious, it can not efficiently compression of dynamic range.
Local tone mapping method, i.e., for the different pixels point in HDR image be respectively adopted different mapping functions into
Row transformation.Different zones where each pixel are carried out with different transformation, the spatial position of pixel is also accounted for
In range, finally it is possible to become the same value after there is pixel value mapping different before mapping, and the preceding pixel of mapping
It is worth identical point due to the case where becoming different values after the different mappings of spatial position.It is compared with global tone mapping method
Compared with for, local tone mapping method not only allows for the gray value of pixel, further considers the space bit of pixel
It sets, therefore passes through such method treated that result images can obtain better effect.And local tone mapping method is the disadvantage is that institute
The calculation amount needed is larger, and is easy to produce " halation (halo) " effect.1993, propose a kind of part earliest by Chiu et al.
Method, this method obtains local brightness variation coefficient by sensibility of the human visual system to brightness change, but this method exists
The very dark region of incandescent is unable to reach ideal effect.2002, Reinhard et al. was based on photography model and proposes a kind of adaptive tune
The partial approach of brightness is saved, this method speed of service is very fast, and realizes the controllability behaviour of color, contrast and global brightness
Make, but easily causes halation or loss in detail.2002, Fattal et al. was based on the compression of brightness step domain and proposes a kind of part side
Method carries out dynamic range compression operation to image greater brightness gradient region by regulating gradient attenuation function.2011,
Local Laplace filter is applied to HDR image dynamic range compression by SylvainParis et al., and it is general to propose part drawing
The edge of Lars filtering perceives tone mapping method.2016, Li et al. people was based on Image Multiscale decomposition and guiding filtering proposes
A kind of partial approach, this method can preferably keep global contrast and local contrast, while can retain more details
Information.Different levels can be divided the image into this method operating process according to actual needs, and is completed in delaminating process
To the relevant treatment of image, but since the process can not finely be held, there is halation phenomenon when will lead to image synthesis.
Summary of the invention
The purpose of the present invention is to provide it is a kind of come overcome or at least mitigate in the drawbacks described above of the prior art at least one
A HDR image tone mapping method based on Multiscale Morphological, the method achieve effective pressures to HDR image dynamic range
Contracting, avoids halation or the generation of other artifacts, and this method calculation amount is small, and the application that can satisfy certain real-times needs
It asks.
To achieve the above object, the present invention provides a kind of HDR image tone mapping method based on Multiscale Morphological, should
Method includes the following steps:
S1 inputs HDR image to be compressed；
S2 constructs the logarithmic image of the HDR image to be compressed；
S3 carries out dynamic range compression multi-resolution decomposition to the logarithmic image with the first presupposition multiple, it is high to obtain first
This pyramid and the first laplacian pyramid；
S4, using the enhanced levels of detail of the first gaussian pyramid described in maintenance boundary polishing operator extraction, and by institute
It states the levels of detail of the first gaussian pyramid and the corresponding layer of first laplacian pyramid is added to successively with the second presupposition multiple,
Obtain the second laplacian pyramid；
S5 carries out dynamic range compression reconstruct to second laplacian pyramid with third presupposition multiple, is moved
Details after state Ratage Coutpressioit enhances image；
S6 carries out exponential transform to the image of details enhancing, obtains the first LDR image；And
S7 carries out color correction to first LDR image, obtains LDR image to be output；
Wherein, " the enhanced details of the first gaussian pyramid described in operator extraction is polished using a maintenance boundary in S4
The method of layer " specifically includes:
S41, by protect boundary two dimension polishing operator according to the HDR image to be compressed horizontal and vertical both direction into
Row decomposes, and obtains horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator；
S42, according to the type of the HDR image to be compressed in S1, the horizontal maintenance boundary obtained using S41
Polish operator and operator polished on a vertical maintenance boundary, respectively to first gaussian pyramid successively carry out top cap transformation and
Bottom cap transformation, to obtain the bright levels of detail and dark levels of detail of first gaussian pyramid；With
The gamma transformation result of the obtained bright levels of detail of S42 is successively subtracted the gamma for the dark levels of detail that S42 is obtained by S43
Transformation results obtain the enhanced levels of detail of the first gaussian pyramid.
Further, in S4, in the case of the HDR image to be compressed is the HDR image of natural scene, S42 is specific
Include:
S421, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described
First gaussian pyramid successively carries out top cap transformation, by every layer of horizontal direction top cap transformation results and vertical direction top cap transformation knot
Fruit takes small point by point, obtains bright levels of detail；
S422, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described
First gaussian pyramid successively carries out bottom cap transformation, by every layer of horizontal direction bottom cap transformation results and vertical direction bottom cap transformation knot
Fruit takes small point by point, obtains dark levels of detail.
Further, in S4, the HDR image to be compressed is in the case of CT-HDR image, S42 is specifically included:
S421, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described
First gaussian pyramid successively carries out top cap transformation, by every layer of horizontal direction top cap transformation results and vertical direction top cap transformation knot
Fruit takes greatly point by point, obtains bright levels of detail；
S422, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described
First gaussian pyramid successively carries out bottom cap transformation, by every layer of horizontal direction bottom cap transformation results and vertical direction bottom cap transformation knot
Fruit takes greatly point by point, obtains dark levels of detail.
Further, S3 includes: to utilize such as following formula (1), since the bottom of first gaussian pyramid, to described
First gaussian pyramid carries out dynamic range compression multi-resolution decomposition with the first presupposition multiple；
In formula (1), I is the logarithmic image, { G0, G1..., GN-1It is the first gaussian pyramid, G0It is described
The bottom of one gaussian pyramid, GlIt is l+1 layers of first gaussian pyramid, N is first gaussian pyramid
The number of plies, downsample indicate filtering down-sampling operator, β1For first presupposition multiple.
Further, S5 includes: using following formula (2), since the top of second laplacian pyramid, to institute
It states the second laplacian pyramid and dynamic range compression reconstruct is carried out with third presupposition multiple, obtained details enhancing image
G′0；
In formula (2), L 'N-1It is the top of second laplacian pyramid, GN-1It is first gaussian pyramid
Top layer images, G 'lIt is pyramidal l+1 layers of middle Gaussian generated, LlIt is the l+ of second laplacian pyramid
1 layer,It is to G 'l+1Layer filtering up-sampling as a result, β2For the third presupposition multiple, l is the second Laplce gold word
L+1 layers of tower, N are the numbers of plies for decomposing obtained second laplacian pyramid.
Further, the multi-resolution decomposition number of plies of the first gaussian pyramid described in S3 and the first laplacian pyramid with
The size of the HDR image to be compressed is related.
Further, the dynamic range of the second presupposition multiple and third presupposition multiple and input picture described in S4 and S5
It is related, it is set as [0.35,1].
Further, the first presupposition multiple described in S3 is set as [0.1,1].
The present invention has the advantage that the 1, present invention realizes HDR image dynamic model due to using upper technical solution
That encloses is effectively compressed, in addition, this method calculating speed is fast, can meet certain real-time application demands.2, the present invention is in dynamic model
In confining pressure compression process, the generation of halation He other artifacts can be effectively prevented from；3, the method for the present invention is suitable for different types of
Image, HDR image, INDUSTRIAL CT IMAGE such as real scene.
Detailed description of the invention
Fig. 1 is the HDR image of the natural scene of input；
Fig. 2 is using the method for the present invention to the image after Fig. 1 Multiscale Morphological dynamic range compression；
Fig. 3 is industry CT-HDR image of input；
Fig. 4 is using the method for the present invention to the image after Fig. 2 Multiscale Morphological dynamic range compression.
Specific embodiment
In the accompanying drawings, same or similar element is indicated using same or similar label or there is same or like function
Element.The embodiment of the present invention is described in detail with reference to the accompanying drawing.
HDR image tone mapping method provided in this embodiment based on Multiscale Morphological includes the following steps:
S1 inputs HDR image to be compressed.Wherein, HDR image is either the HDR of natural scene shown in Fig. 1 schemes
Picture is also possible to industry CT-HDR image shown in Fig. 3, can also be other types of HDR image.
S2 constructs the logarithmic image of the HDR image I to be compressed.Wherein, " logarithmic image of construction HDR image " tool
Body is to include: firstly, then the luminance component image for calculating the HDR image to be compressed of S1 input is sought luminance picture pixel-by-pixel
Natural logrithm obtains the logarithmic image of the HDR image I to be compressed.
S3 carries out dynamic range compression multi-resolution decomposition to the logarithmic image with the first presupposition multiple, it is high to obtain first
This pyramid { G0, G1..., GN-1, then pass through the first gaussian pyramid { G0, G1..., GN-1Obtain the first Laplce gold
Word tower { L0, L1..., LN-1}.Wherein, " multiple dimensioned point of dynamic range compression is carried out with the first presupposition multiple to the logarithmic image
Solution, obtain the first gaussian pyramid " concrete methods of realizing will become clear from the description below.And " pass through the first Gauss gold word
Tower { G0, G1..., GN-1Obtain the first laplacian pyramid { L0, L1..., LN-1" concrete methods of realizing be existing skill
Art, herein not reinflated description.First presupposition multiple is related with the dynamic range of HDR image to be compressed that S1 is inputted, if
It is set to [0.35,1].
S4 polishes the first gaussian pyramid of operator extraction { G using a maintenance boundary0, G1..., GN-1Enhanced details
Layer, and the first gaussian pyramid { G that will be obtained0, G1..., GN-1Enhanced levels of detail is successively added to the second presupposition multiple
First laplacian pyramid { the L0, L1..., LN-1Corresponding layer, obtain the second laplacian pyramid { L '0, L
′1..., L 'N-1}.Wherein, the value range of the second presupposition multiple is set as [0.1,1], and details enhancing is more severe.S4's is specific
Implementation method will be according to the difference of the type of the HDR image to be compressed in S1, it will there is different implementations, these
Method will be unfolded to illustrate below.
S5 carries out dynamic range compression reconstruct to second laplacian pyramid with third presupposition multiple, is moved
Details after state Ratage Coutpressioit enhances image.Wherein, the third presupposition multiple and the HDR image to be compressed in S1
Dynamic range is related, and test data is shown: the value dynamic range compression effect being set as in [0.35,1] range is preferable." to institute
State the second laplacian pyramid and dynamic range compression reconstruct carried out with third presupposition multiple " it will be unfolded to illustrate below.
S6 carries out exponential transform to the image of details enhancing, obtains the first LDR image I '.Wherein, " exponential transform "
For the prior art, not reinflated explanation herein.
S7 carries out color correction to first LDR image, obtains LDR image I " to be output.Wherein, " color school
It just " is the prior art, herein not reinflated explanation.
The present embodiment is operated using multiple dimensioned dynamic range compression and details enhancing as above, it is possible to prevente effectively from halation
It generates, and method provided by the present embodiment can be applied to the effective dynamic range compression of different image types.
It in one embodiment, is two dimensional image for the HDR image to be compressed of S1 input, if directlyed adopt two-dimensional
Protect edge smoothing operator, it will introduce halation, therefore the present embodiment is one-dimensional at two by two-dimensional guarantor's edge smoothing Decompose operaton
Smoothing operator recycles one-dimensional operator to operate the two dimensional image of input.One-dimensional smoothing operator in the present embodiment is adopted
It is dimension morphological operation, naturally it is also possible to which smooth using one-dimensional gradient zero norm, one-dimensional median filtering etc. protects boundary
One-dimensional smoothing operator.In consideration of it, S4 is specifically included:
S41, by protect boundary two dimension polishing operator according to the HDR image to be compressed horizontal and vertical both direction into
Row decomposes, and obtains horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator.Wherein, a usual maintenance
The size that operator is polished on boundary is 3 elements.
S42, according to the type of the HDR image to be compressed in S1, a horizontal maintenance boundary is polished operator and is erected
Straight maintenance boundary polishing operator is respectively to the first gaussian pyramid { G0, G1..., GN-1Successively carry out top cap transformation and bottom
Cap transformation, to obtain the first gaussian pyramid { G0, G1..., GN-1Bright levels of detail and dark levels of detail.
The gamma transformation result of the obtained bright levels of detail of S42 is successively subtracted the gamma for the dark levels of detail that S42 is obtained by S43
Transformation results obtain the first gaussian pyramid { G0, G1..., GN-1Enhanced levels of detail.Such as: the bright details that S42 is obtained
It is denoted asThe dark details that S42 is obtained is denoted asFirst gaussian pyramid { G0, G1..., GN-1Enhanced
Levels of detail is denoted as detail, this is represented by following formula:
S44, using following formula (3), the first gaussian pyramid { G that will be obtained0, G1..., GN-1Enhanced levels of detail
Details is successively added to the first laplacian pyramid { L with the second presupposition multiple λ0, L1..., LN-1Corresponding layer, it obtains
To the second laplacian pyramid { L '0, L '1..., L 'N-1}。
It should be noted that if input HDR image is two-dimensional structure, then above-mentioned "horizontal" direction refers to image array
Line direction, "vertical" direction refers to the column direction of image array.
In one embodiment, in the case of the HDR image to be compressed in S4 is the HDR image of natural scene, S42 tool
Body includes:
S421, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described
First gaussian pyramid successively carries out top cap transformation, by every layer of horizontal direction top cap transformation results and vertical direction top cap transformation knot
Fruit takes small point by point, obtains bright levels of detail；
S422, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described
First gaussian pyramid successively carries out bottom cap transformation, by every layer of horizontal direction bottom cap transformation results and vertical direction bottom cap transformation knot
Fruit takes small point by point, obtains dark levels of detail.
The HDR image of targeted natural scene in the present embodiment, dynamic range compression result should be with the senses of human eye system
Know and be consistent, the present embodiment passes through the generation for strongly avoiding halation, while enhancing the mode of details, i.e., using to horizontal direction
Take with the details in the two directions of vertical direction small, the ring of light of edge blurry in the HDR image of natural scene is enable to disappear
It removes, i.e. halation in the HDR image of elimination natural scene, to be more in line with human eye system for the vision of true nature scene
Perception.
In one embodiment, in the case of the HDR image to be compressed in S4 is CT-HDR image, S42 is specifically included:
S421, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described
First gaussian pyramid successively carries out top cap transformation, by every layer of horizontal direction top cap transformation results and vertical direction top cap transformation knot
Fruit takes greatly point by point, obtains bright levels of detail；
S422, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described
First gaussian pyramid successively carries out bottom cap transformation, by every layer of horizontal direction bottom cap transformation results and vertical direction bottom cap transformation knot
Fruit takes greatly point by point, obtains dark levels of detail.
Targeted CT-HDR image in the present embodiment, it is bright to the reserving degree and entirety of image detail dark right to be more concerned about
It than the quality of degree, therefore, is taken greatly using the details of horizontal direction and the two directions of vertical direction, is more advantageous to details enhancing,
The LDR image details of generation is more prominent.
In one embodiment, S3 includes:
Using such as following formula (1), since the bottom of the first gaussian pyramid, to first gaussian pyramid with first
Presupposition multiple carries out dynamic range compression multi-resolution decomposition:
In formula (1), I is the logarithmic image, { G0, G1..., GN-1It is the first gaussian pyramid, G0It is described
The bottom of one gaussian pyramid, GlIt is l+1 layers of first gaussian pyramid, N is first gaussian pyramid
The number of plies, downsample indicate filtering down-sampling operator, β1For first presupposition multiple.
In one embodiment, S5 includes:
Using following formula (2), since the top of second laplacian pyramid, using top-down mode,
Recursion is successively carried out, until l=0, carries out dynamic range compression to second laplacian pyramid with third presupposition multiple
Reconstruct, obtained details enhancing image G '0:
In formula (2), LN-1It is the top of second laplacian pyramid, GN-1It is first gaussian pyramid
Top layer images, G 'lIt is pyramidal l+1 layers of middle Gaussian generated, LlIt is the l+1 of second laplacian pyramid
Layer,It is to G 'l+1Layer filtering up-sampling as a result, β2For the third presupposition multiple, l is the second Laplce gold word
L+1 layers of tower, N are the numbers of plies for decomposing obtained second laplacian pyramid.
In one embodiment, the multi-resolution decomposition of the first gaussian pyramid and the first laplacian pyramid described in S3
The number of plies is related to the size of the HDR image to be compressed.Under normal circumstances, HDR image size is bigger, and Decomposition order is more,
Such as: the HDR image that size is 1848*1848 can be 8 layers with Decomposition order.
Comparison diagram 1 and Fig. 2 discovery, Fig. 2 meet perception of the human eye system for real scene, comparison diagram 3 and Fig. 4 discovery,
Details in Fig. 4 is more prominent, is handled convenient for further.The HDR image in natural scene is shown in Fig. 2, but originally
Invention protection scope is not limited to the dynamic range compression to HDR image in natural scene.
Finally it is noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations.This
The those of ordinary skill in field is it is understood that be possible to modify the technical solutions described in the foregoing embodiments or right
Part of technical characteristic is equivalently replaced；These are modified or replaceed, and it does not separate the essence of the corresponding technical solution originally
Invent the spirit and scope of each embodiment technical solution.
Claims (8)
1. a kind of HDR image tone mapping method based on Multiscale Morphological, which comprises the steps of:
S1 inputs HDR image to be compressed；
S2 constructs the logarithmic image of the HDR image to be compressed；
S3 carries out dynamic range compression multi-resolution decomposition to the logarithmic image with the first presupposition multiple, obtains the first Gauss gold
Word tower and the first laplacian pyramid；
S4, using the enhanced levels of detail of the first gaussian pyramid described in maintenance boundary polishing operator extraction, and by described the
The levels of detail of one gaussian pyramid is successively added to the corresponding layer of first laplacian pyramid with the second presupposition multiple, obtains
Second laplacian pyramid；
S5 carries out dynamic range compression reconstruct to second laplacian pyramid with third presupposition multiple, obtains dynamic model
Enclose compressed details enhancing image；
S6 carries out exponential transform to the image of details enhancing, obtains the first LDR image；And
S7 carries out color correction to first LDR image, obtains LDR image to be output；
Wherein, in S4 " using a maintenance boundary polishing operator extraction described in the enhanced levels of detail of the first gaussian pyramid "
Method specifically includes:
S41 will protect boundary two dimension polishing operator and be divided according to the horizontal and vertical both direction of the HDR image to be compressed
Solution obtains horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator；
S42 is polished according to the type of the HDR image to be compressed in S1 using the horizontal maintenance boundary that S41 is obtained
Operator is polished on operator and a vertical maintenance boundary, successively carries out top cap transformation and bottom cap to first gaussian pyramid respectively
Transformation, to obtain the bright levels of detail and dark levels of detail of first gaussian pyramid；With
The gamma transformation result of the obtained bright levels of detail of S42 is successively subtracted the gamma transformation for the dark levels of detail that S42 is obtained by S43
As a result, obtaining the enhanced levels of detail of the first gaussian pyramid.
2. the HDR image tone mapping method based on Multiscale Morphological as described in claim 1, which is characterized in that in S4,
In the case of the HDR image to be compressed is the HDR image of natural scene, S42 is specifically included:
S421, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described first
Gaussian pyramid successively carries out top cap transformation, by every layer of horizontal direction top cap transformation results and vertical direction top cap transformation results by
Point takes small, obtains bright levels of detail；
S422, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described first
Gaussian pyramid successively carries out bottom cap transformation, by every layer of horizontal direction bottom cap transformation results and vertical direction bottom cap transformation results by
Point takes small, obtains dark levels of detail.
3. the HDR image tone mapping method based on Multiscale Morphological as described in claim 1, which is characterized in that in S4,
The HDR image to be compressed is in the case of CT-HDR image, S42 is specifically included:
S421, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described first
Gaussian pyramid successively carries out top cap transformation, by every layer of horizontal direction top cap transformation results and vertical direction top cap transformation results by
Point takes greatly, obtains bright levels of detail；
S422, using horizontal maintenance boundary polishing operator and vertical maintenance boundary polishing operator respectively to described first
Gaussian pyramid successively carries out bottom cap transformation, by every layer of horizontal direction bottom cap transformation results and vertical direction bottom cap transformation results by
Point takes greatly, obtains dark levels of detail.
4. the HDR image tone mapping method based on Multiscale Morphological as claimed any one in claims 1 to 3, special
Sign is that S3 includes: to utilize such as following formula (1), since the bottom of first gaussian pyramid, to the first Gauss gold
Word tower carries out dynamic range compression multi-resolution decomposition with the first presupposition multiple；
In formula (1), I is the logarithmic image, { G0, G1... ..., GN-1It is the first gaussian pyramid, G0It is first Gauss
The pyramidal bottom, GlIt is l+1 layers of first gaussian pyramid, N is the number of plies of first gaussian pyramid,
Downsample indicates filtering down-sampling operator, β1For first presupposition multiple.
5. the HDR image tone mapping method based on Multiscale Morphological as claimed in claim 4, which is characterized in that S5 packet
It includes: using following formula (2), since the top of second laplacian pyramid, to second laplacian pyramid
Dynamic range compression reconstruct is carried out with third presupposition multiple, obtained details enhancing image G '0；
In formula (2), L 'N-1It is the top of second laplacian pyramid, GN-1It is the top of first gaussian pyramid
Tomographic image, G 'lIt is pyramidal l+1 layers of middle Gaussian generated, LlIt is l+1 layers of second laplacian pyramid,It is to G 'l+1Layer filtering up-sampling as a result, β2For the third presupposition multiple, l is second laplacian pyramid
L+1 layers, N is the number of plies for decomposing obtained second laplacian pyramid.
6. the HDR image tone mapping method based on Multiscale Morphological as claimed in claim 5, which is characterized in that in S3
The multi-resolution decomposition number of plies of first gaussian pyramid and the first laplacian pyramid and the HDR image to be compressed
Size is related.
7. the HDR image tone mapping method based on Multiscale Morphological as described in claim 1, which is characterized in that S4 and
Second presupposition multiple described in S5 and third presupposition multiple are related with the dynamic range of input picture, are set as [0.35,1].
8. the HDR image tone mapping method based on Multiscale Morphological as described in claim 1, which is characterized in that in S3
First presupposition multiple is set as [0.1,1].
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910622002.7A CN110415188B (en) | 2019-07-10 | 2019-07-10 | HDR image tone mapping method based on multi-scale morphology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910622002.7A CN110415188B (en) | 2019-07-10 | 2019-07-10 | HDR image tone mapping method based on multi-scale morphology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110415188A true CN110415188A (en) | 2019-11-05 |
| CN110415188B CN110415188B (en) | 2021-08-20 |
Family
ID=68360977
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910622002.7A Expired - Fee Related CN110415188B (en) | 2019-07-10 | 2019-07-10 | HDR image tone mapping method based on multi-scale morphology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110415188B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113674173A (en) * | 2021-08-19 | 2021-11-19 | Oppo广东移动通信有限公司 | Image processing method and device, terminal and readable storage medium |
| CN114463207A (en) * | 2022-01-24 | 2022-05-10 | 哈尔滨理工大学 | Tone mapping method based on global dynamic range compression and local brightness estimation |
Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7146059B1 (en) * | 2003-03-05 | 2006-12-05 | Massachusetts Institute Of Technology | Method of performing fast bilateral filtering and using the same for the display of high-dynamic-range images |
| WO2011008239A1 (en) * | 2009-06-29 | 2011-01-20 | Thomson Licensing | Contrast enhancement |
| CN102637292A (en) * | 2011-02-10 | 2012-08-15 | 西门子公司 | Image processing method and device |
| CN102646269A (en) * | 2012-02-29 | 2012-08-22 | 中山大学 | Image processing method and device based on Laplace pyramid |
| WO2012118961A1 (en) * | 2011-03-02 | 2012-09-07 | Dolby Laboratories Licensing Corporation | Local multiscale tone-mapping operator |
| CN104463820A (en) * | 2014-10-29 | 2015-03-25 | 广东工业大学 | Reverse tone mapping algorithm based on frequency domain |
| CN104504666A (en) * | 2015-01-16 | 2015-04-08 | 成都品果科技有限公司 | Tone mapping method based on Laplacian pyramid |
| US20150170389A1 (en) * | 2013-12-13 | 2015-06-18 | Konica Minolta Laboratory U.S.A., Inc. | Automatic selection of optimum algorithms for high dynamic range image processing based on scene classification |
| US20150181186A1 (en) * | 2013-09-10 | 2015-06-25 | Apple Inc. | Image Tone Adjustment using Local Tone Curve Computation |
| CN105825472A (en) * | 2016-05-26 | 2016-08-03 | 重庆邮电大学 | Rapid tone mapping system and method based on multi-scale Gauss filters |
| CN105869112A (en) * | 2016-04-20 | 2016-08-17 | 西安理工大学 | Method for tone mapping of high dynamic range picture with edge kept minimized |
| CN106506983A (en) * | 2016-12-12 | 2017-03-15 | 天津大学 | A kind of HDR video generation methods suitable for LDR videos |
| US9600741B1 (en) * | 2015-03-18 | 2017-03-21 | Amazon Technologies, Inc. | Enhanced image generation based on multiple images |
| CN106530263A (en) * | 2016-10-19 | 2017-03-22 | 天津大学 | Single-exposure high-dynamic range image generation method adapted to medical image |
| CN107220956A (en) * | 2017-04-18 | 2017-09-29 | 天津大学 | A kind of HDR image fusion method of the LDR image based on several with different exposures |
| CN107274372A (en) * | 2017-06-26 | 2017-10-20 | 重庆名图医疗设备有限公司 | Dynamic image Enhancement Method and device based on pyramid local contrast |
| CN107845128A (en) * | 2017-11-03 | 2018-03-27 | 安康学院 | A kind of more exposure high-dynamics image method for reconstructing of multiple dimensioned details fusion |
| US20190043177A1 (en) * | 2018-06-15 | 2019-02-07 | Intel Corporation | Hybrid tone mapping for consistent tone reproduction of scenes in camera systems |
| CN109493283A (en) * | 2018-08-23 | 2019-03-19 | 金陵科技学院 | A kind of method that high dynamic range images ghost is eliminated |
| CN109685742A (en) * | 2018-12-29 | 2019-04-26 | 哈尔滨理工大学 | A kind of image enchancing method under half-light environment |
-
2019
- 2019-07-10 CN CN201910622002.7A patent/CN110415188B/en not_active Expired - Fee Related
Patent Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7146059B1 (en) * | 2003-03-05 | 2006-12-05 | Massachusetts Institute Of Technology | Method of performing fast bilateral filtering and using the same for the display of high-dynamic-range images |
| WO2011008239A1 (en) * | 2009-06-29 | 2011-01-20 | Thomson Licensing | Contrast enhancement |
| CN102637292A (en) * | 2011-02-10 | 2012-08-15 | 西门子公司 | Image processing method and device |
| WO2012118961A1 (en) * | 2011-03-02 | 2012-09-07 | Dolby Laboratories Licensing Corporation | Local multiscale tone-mapping operator |
| CN102646269A (en) * | 2012-02-29 | 2012-08-22 | 中山大学 | Image processing method and device based on Laplace pyramid |
| US20150181186A1 (en) * | 2013-09-10 | 2015-06-25 | Apple Inc. | Image Tone Adjustment using Local Tone Curve Computation |
| US20150170389A1 (en) * | 2013-12-13 | 2015-06-18 | Konica Minolta Laboratory U.S.A., Inc. | Automatic selection of optimum algorithms for high dynamic range image processing based on scene classification |
| CN104463820A (en) * | 2014-10-29 | 2015-03-25 | 广东工业大学 | Reverse tone mapping algorithm based on frequency domain |
| CN104504666A (en) * | 2015-01-16 | 2015-04-08 | 成都品果科技有限公司 | Tone mapping method based on Laplacian pyramid |
| US9600741B1 (en) * | 2015-03-18 | 2017-03-21 | Amazon Technologies, Inc. | Enhanced image generation based on multiple images |
| CN105869112A (en) * | 2016-04-20 | 2016-08-17 | 西安理工大学 | Method for tone mapping of high dynamic range picture with edge kept minimized |
| CN105825472A (en) * | 2016-05-26 | 2016-08-03 | 重庆邮电大学 | Rapid tone mapping system and method based on multi-scale Gauss filters |
| CN106530263A (en) * | 2016-10-19 | 2017-03-22 | 天津大学 | Single-exposure high-dynamic range image generation method adapted to medical image |
| CN106506983A (en) * | 2016-12-12 | 2017-03-15 | 天津大学 | A kind of HDR video generation methods suitable for LDR videos |
| CN107220956A (en) * | 2017-04-18 | 2017-09-29 | 天津大学 | A kind of HDR image fusion method of the LDR image based on several with different exposures |
| CN107274372A (en) * | 2017-06-26 | 2017-10-20 | 重庆名图医疗设备有限公司 | Dynamic image Enhancement Method and device based on pyramid local contrast |
| CN107845128A (en) * | 2017-11-03 | 2018-03-27 | 安康学院 | A kind of more exposure high-dynamics image method for reconstructing of multiple dimensioned details fusion |
| US20190043177A1 (en) * | 2018-06-15 | 2019-02-07 | Intel Corporation | Hybrid tone mapping for consistent tone reproduction of scenes in camera systems |
| CN109493283A (en) * | 2018-08-23 | 2019-03-19 | 金陵科技学院 | A kind of method that high dynamic range images ghost is eliminated |
| CN109685742A (en) * | 2018-12-29 | 2019-04-26 | 哈尔滨理工大学 | A kind of image enchancing method under half-light environment |
Non-Patent Citations (8)
| Title |
|---|
| B. GU ET AL.: "Local Edge-Preserving Multiscale Decomposition for High Dynamic Range Image Tone Mapping", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 * |
| ERIK REINHARD ET AL.: "Photographic tone reproduction for digital images", 《IN PROCEEDINGS OF THE 29TH ANNUAL CONFERENCE ON COMPUTER GRAPHICS AND INTERACTIVE TECHNIQUES 》 * |
| RAIHAN FIROZ ET AL.: "Medical Image Enhancement Using Morphological Transformation", 《JOURNAL OF DATA ANALYSIS AND INFORMATION PROCESSING》 * |
| 伍世宾 等: "基于多尺度带限的自适应直方图均衡和数学形态学的医学X射线图像对比度增强算法", 《集成技术》 * |
| 张文婷: "基于导向滤波的图像增强算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 梁云 等: "改进拉普拉斯金字塔模型的高动态图像色调映射方法", 《计算机辅助设计与图形学学报》 * |
| 芦碧波 等: "基于双边滤波的多尺度分层色调映射算法", 《液晶与显示》 * |
| 蒲雅蕾: "高动态范围图像色调映射方法的研究与实现", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113674173A (en) * | 2021-08-19 | 2021-11-19 | Oppo广东移动通信有限公司 | Image processing method and device, terminal and readable storage medium |
| CN114463207A (en) * | 2022-01-24 | 2022-05-10 | 哈尔滨理工大学 | Tone mapping method based on global dynamic range compression and local brightness estimation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110415188B (en) | 2021-08-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Efficient single image dehazing and denoising: An efficient multi-scale correlated wavelet approach | |
| US8059911B2 (en) | Depth-based image enhancement | |
| Wu et al. | Contrast-accumulated histogram equalization for image enhancement | |
| US7305144B2 (en) | System and method for compressing the dynamic range of an image | |
| TWI406571B (en) | Image Processing Method for Enhancing Image Edge Resolution | |
| CN105850114A (en) | Method for inverse tone mapping of an image | |
| KR20040044556A (en) | Image processing method, apparatus, and program | |
| JP4862897B2 (en) | Image processing method | |
| WO2016031189A1 (en) | Image processing apparatus, image processing method, recording medium, and program | |
| CN106780417A (en) | A kind of Enhancement Method and system of uneven illumination image | |
| CN101276464A (en) | Picture processing apparatus and method | |
| CN116051428B (en) | Deep learning-based combined denoising and superdivision low-illumination image enhancement method | |
| CN112116536A (en) | Low-illumination image enhancement method and system | |
| Zheng et al. | Ultra-high-definition image hdr reconstruction via collaborative bilateral learning | |
| CN111353955A (en) | Image processing method, device, equipment and storage medium | |
| CN110415188A (en) | A kind of HDR image tone mapping method based on Multiscale Morphological | |
| Ma et al. | Underwater image restoration through a combination of improved dark channel prior and gray world algorithms | |
| CN107358578B (en) | Yin-yang face treatment method and device | |
| Peng et al. | Efficient image resolution enhancement using edge-directed unsharp masking sharpening for real-time ASIC applications | |
| CN112200719B (en) | Image processing method, electronic device, and readable storage medium | |
| CN110175959B (en) | Typhoon cloud picture enhancement method | |
| Hanumantharaju et al. | Adaptive color image enhancement based geometric mean filter | |
| CN116630198A (en) | Multi-scale fusion underwater image enhancement method combining self-adaptive gamma correction | |
| CN114648467B (en) | Image defogging method and device, terminal equipment and computer readable storage medium | |
| CN111754412A (en) | Method and device for constructing data pairs and terminal equipment |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210820 |