CN107146225A - A kind of method and apparatus for obtaining intrinsic image - Google Patents
A kind of method and apparatus for obtaining intrinsic image Download PDFInfo
- Publication number
- CN107146225A CN107146225A CN201710311828.2A CN201710311828A CN107146225A CN 107146225 A CN107146225 A CN 107146225A CN 201710311828 A CN201710311828 A CN 201710311828A CN 107146225 A CN107146225 A CN 107146225A
- Authority
- CN
- China
- Prior art keywords
- image
- parameter value
- spd
- tristimulus values
- shadow region
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- 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
- G06T2207/10004—Still image; Photographic image
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention discloses a kind of method and apparatus for obtaining intrinsic image, by the SPD for calculating daylight and skylight, i.e. the linear RGB tristimulus values of image non-hatched area and the linear RGB tristimulus values of image shadow region, can calculate the corresponding parameter values of the SPD；Using entropy optimization algorithm, targeted parameter value corresponding with image to be tested can be chosen from the parameter value；The targeted parameter value is substituted into Orthogonal Decomposition formula, the intrinsic image of the image to be tested just can be obtained.By way of calculating SPD, corresponding parameter value under any zenith angle can be drawn, the targeted parameter value accuracy selected from the parameter value is also higher, so that the intrinsic image being worth to according to the target component can be very good to eliminate the influence of illumination, namely intrinsic image accuracy is higher.In addition, obtaining parameter value by way of calculating, artificial operation is reduced, human resources are saved, and avoid the error produced during manual measurement.
Description
Technical field
The present invention relates to technical field of computer vision, more particularly to a kind of method and apparatus for obtaining intrinsic image.
Background technology
Image can be the scene information recorded by video camera or camera under certain illumination condition, and it is meter
Calculation machine vision, Image Processing and Pattern Recognition are analyzed and complete the source to environment sensing, cognitive data.Illumination is imaging
Indispensable fundamental, for Same Scene, under different illumination, its image effect produced is not quite similar.One width
Image can be broken down into luminance picture and albedo image, wherein, albedo image is to maintain in the case of illumination variation
Constant, it can be understood as it is the image for removing external factor influence, external factor is such as the position of light source and color, reflection
Object included in image indeformable in itself, can referred to as intrinsic image.
, can be by when we have obtained the intrinsic image that is not influenceed by illumination condition, or perhaps during illumination invariant image
It is applied to many computer vision fields, such as image recognition, Road Detection, self-navigation, target following.For example,
During target following, obtained image under the conditions of a width outdoor natural light shines, because illumination generates shade, shade for target with
Track can produce very big interference.If the intrinsic image not by illumination effect can be obtained, it is possible to avoid shade to target
Influence, improves the accuracy of target following.
The shadow region of outdoor light image and non-hatched area are linear, can be by parameter value KH(KR,KG,KB)
Represent, parameter value KHAcquisition be from original image obtain albedo image key., can be with using spectrometer in traditional approach
Measurement obtains under clear sky the spectral power distribution (Spectral of the daylight and skylight zenith angle is from 20 degree to 80 degree
Power Distribution, SPD), or referred to as spectral irradiance.It can be calculated according to SPD and obtain corresponding parameter value KH,
, can be from parameter value K for the image to be analyzedHIn select the suitable K of the width imageH, according to the KHCorrespondence can be obtained
Albedo image.
But, traditional approach measures under clear sky the partial sun light zenith angle is from 20 degree to 80 degree by spectrometer
With the SPD of skylight, then calculate and obtain KHValue.KHThe calculating of value is actually what is together decided on by zenith angle and weather turbidity,
Above conventional method hardly results in daylight and day under accurate different zenith angles and different weather turbidity by optical spectrum instrumentation
The SPD of empty light, can only obtain limited a series of measurements.Thus the K calculatedHBe worth optional scope smaller, accuracy compared with
It is low.
, can only be from existing K for any piece imageHIn approximate one group of parameter K of selectionH, due to KHThe optional model of value
Limit is with, this group of K selectedHIt may can not accurately reflect the actual parameter value of the width image, this is easy for causing
Obtained intrinsic image accuracy is relatively low.And using needing manually to be operated during spectrometer measurement, because measurement data is numerous
It is many, great human resources will be consumed, in addition, manual measurement can also produce certain error, so as to influence the essence of intrinsic image
Exactness.
It can be seen that, the accuracy for obtaining intrinsic image how is improved, is those skilled in the art's urgent problem to be solved.
The content of the invention
The purpose of the embodiment of the present invention is to provide a kind of method and apparatus for obtaining intrinsic image, can effectively improve acquisition
The accuracy of intrinsic image.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of method for obtaining intrinsic image, including：
Calculate the SPD of daylight and skylight；Wherein, the SPD includes the linear RGB tristimulus of image non-hatched area
Value, and image shadow region linear RGB tristimulus values；
According to the linear RGB tristimulus values and the linear RGB tri- of described image shadow region of described image non-hatched area
Values, calculates the corresponding parameter values of the SPD；The parameter value is used to represent image non-hatched area and image shadow region
The proportionate relationship in domain；
Using entropy optimization algorithm, targeted parameter value corresponding with image to be tested is chosen from the parameter value；
The targeted parameter value is substituted into Orthogonal Decomposition formula, the intrinsic image of the image to be tested is obtained.
Optionally, it is described according to the linear RGB tristimulus values of described image non-hatched area and described image shadow region
Linear RGB tristimulus values, calculating the corresponding parameter values of the SPD includes：
Using image non-hatched area and the linear formula of image shadow region, the corresponding parameter values of the SPD are calculated；
Wherein, the linear formula is as follows：
KH=((FH+14)/(fH+14))2.4
Wherein, KHFor the corresponding parameter values of the SPD, FHFor the linear RGB tristimulus values of image non-hatched area,
fHFor the linear RGB tristimulus values of image shadow region.
Optionally, the utilization entropy optimization algorithm, chooses target ginseng corresponding with image to be tested from the parameter value
Numerical value includes：
The parameter value is substituted into the corresponding linear formula of the image to be tested, the image after the shade that is eliminated,
And selecting the target image for meeting preparatory condition from described image, the corresponding parameter value of the target image is target component
Value.
The embodiment of the present invention additionally provide it is a kind of obtain intrinsic image device, including computing unit, choose unit and
To unit,
The computing unit, the SPD for calculating daylight and skylight；Wherein, the SPD includes image non-hatched area
Linear RGB tristimulus values, and image shadow region linear RGB tristimulus values；
The computing unit is additionally operable to linear RGB tristimulus values and described image the moon according to described image non-hatched area
The linear RGB tristimulus values in shadow zone domain, calculates the corresponding parameter values of the SPD；The parameter value is used to represent the non-the moon of image
The proportionate relationship in shadow zone domain and image shadow region；
The selection unit, for choosing targeted parameter value corresponding with image to be tested from the parameter value；
It is described to obtain unit, for the targeted parameter value to be substituted into Orthogonal Decomposition formula, obtain the image to be tested
Intrinsic image.
Optionally, the computing unit is specifically for the linear public affairs using image non-hatched area and image shadow region
Formula, calculates the corresponding parameter values of the SPD；Wherein, the linear formula is as follows：
KH=((FH+14)/(fH+14))2.4
Wherein, KHFor the corresponding parameter values of the SPD, FHFor the linear RGB tristimulus values of image non-hatched area,
fHFor the linear RGB tristimulus values of image shadow region.
Optionally, the selection unit by the parameter value specifically for substituting into the corresponding linear public affairs of the image to be tested
In formula, the image after the shade that is eliminated, and the target image for meeting preparatory condition, the target are selected from described image
The corresponding parameter value of image is targeted parameter value.
By calculating the SPD of daylight and skylight, i.e. image non-hatched area line it can be seen from above-mentioned technical proposal
Property RGB tristimulus values and the linear RGB tristimulus values of image shadow region, can calculate the corresponding parameter values of the SPD；Profit
Entropy optimization algorithm is used, targeted parameter value corresponding with image to be tested can be chosen from the parameter value；The target is joined
Numerical value substitutes into Orthogonal Decomposition formula, just can obtain the intrinsic image of the image to be tested., can by way of calculating SPD
To draw corresponding parameter value under any zenith angle, the targeted parameter value accuracy selected from the parameter value is also higher, from
And the intrinsic image being worth to according to the target component can be very good to eliminate the influence of illumination, namely intrinsic image accuracy is got over
It is high.In addition, obtaining parameter value by way of calculating, artificial operation is reduced, human resources are saved, and avoid artificial
The error produced during measurement.
Brief description of the drawings
In order to illustrate the embodiments of the present invention more clearly, the required accompanying drawing used in embodiment will be done simply below
Introduce, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for ordinary skill people
For member, on the premise of not paying creative work, other accompanying drawings can also be obtained according to these accompanying drawings.
Fig. 1 is a kind of flow chart for the method for obtaining intrinsic image provided in an embodiment of the present invention；
Fig. 2 is a kind of structural representation for the device for obtaining intrinsic image provided in an embodiment of the present invention；
Fig. 3 a are a kind of entropy optimization method schematic diagram for choosing optimal objective parameter value provided in an embodiment of the present invention；
Fig. 3 b are a kind of schematic diagram of non-optimal targeted parameter value provided in an embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Based on this
Embodiment in invention, those of ordinary skill in the art are not under the premise of creative work is made, and what is obtained is every other
Embodiment, belongs to the scope of the present invention.
In order that those skilled in the art more fully understand the present invention program, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.
Next, a kind of method for acquisition intrinsic image that the embodiment of the present invention is provided is discussed in detail.Fig. 1 is the present invention
A kind of flow chart of the method for acquisition intrinsic image that embodiment is provided, this method includes：
S101：Calculate the SPD of daylight and skylight.
Outdoor light image includes non-hatched area and shadow region, and daylight can regard the nonshaded area of image as
Domain, its corresponding SPD is the linear RGB tristimulus values of image non-hatched area；Skylight can regard the shade of image as
Region, its corresponding SPD is the linear RGB tristimulus values of image shadow region.
S102：According to the linear RGB tristimulus values of image non-hatched area and the linear RGB tristimulus of image shadow region
Value, calculates the corresponding parameter values of the SPD.
Parameter value can be used to indicate that the proportionate relationship of image non-hatched area and image shadow region.
K can be used in embodiments of the present inventionHRepresent the corresponding parameter values of SPD, FHRepresent the linear of image non-hatched area
RGB tristimulus values, fHRepresent the linear RGB tristimulus values of image shadow region.Image non-hatched area and image shadow region
With linear relationship, the linear relationship of image non-hatched area and image shadow region can be represented using linear formula.
Wherein, the linear formula is as follows：
KH=((FH+14)/(fH+14))2.4
By FHAnd fHThe linear formula is substituted into, corresponding K just can be calculatedH。
S103：Using entropy optimization algorithm, targeted parameter value corresponding with image to be tested is chosen from the parameter value.
For any one image to be tested, often because zenith angle and sky turbidity are unknown, it is impossible to directly count
The parameter value of this image to be tested is calculated, accordingly, it would be desirable to from the series of parameters value under any zenith angle calculated
In select targeted parameter value.Targeted parameter value can be used to indicate that the parameter value for meeting this image to be tested.
When choosing, series of parameters value can be substituted into the corresponding linear formula of the image to be tested, be disappeared
Except the image after shade (illumination effect), and the target image for meeting preparatory condition, the target are selected from described image
The corresponding parameter value of image is targeted parameter value.
Target image can eliminate shadow removing most clean piece image in the image after shade.Preparatory condition can be with
It is the condition that target image is chosen from the image after shade is eliminated, the preparatory condition can be understood as after elimination shade
A width shadow removing most clean piece image is selected in image.
In the specific implementation, being f by the SPD values of the testing image shadow regionHAnd calculated in above-mentioned S102
A series of K arrivedHValue, can be F in the hope of a series of SPD values after this region shadow-freeH.The F now obtainedHIt should be same material
Matter, goes to eliminate the image-region SPD values of illumination effect.Each pixel in this region can be fitted in log B/R, log G/R
Straight line, and be vertically projected on gray level image.We select the minimum corresponding K of that class value of projected areaHValue, be
Targeted parameter value corresponding with image to be tested.
Two width schematic diagrames as shown in Figure 3 a and Figure 3 b shows, will eliminate the image after illumination effect for a width, and its is each
Pixel is transformed into 2D chrominance spaces (B/R, G/R) from 3d space (RGB), that is, solves two ratio B/R of its RGB triple channel,
G/R.Then, then log rgb spaces (i.e. log B/R, log G/R) are converted it to.Now, identical material is in same illumination bar
Each point under part can be fitted to straight line, and perpendicular to grey scale image planes.When we choose be optimum parameter value KH's
When, the projected area area on grey scale image planes tends to be minimum, as shown in Figure 3 a, is defined as targeted parameter value；And its
He is parameter value KHThe point tried to achieve then can be at random project on grey scale image planes, as shown in Figure 3 b.
S104：The targeted parameter value is substituted into Orthogonal Decomposition formula, the intrinsic image of the image to be tested is obtained.
By taking an image to be tested as an example, it can determine that each pixel is each in the image to be tested by above-mentioned steps
Self-corresponding targeted parameter value.
Wherein, Orthogonal Decomposition formula is as follows：
U=up+αu0
u0Freely solved to be normalized, upFor particular solution, and meet upPerpendicular to u0.For given piece image, its freedom
Solve u0(K is uniquely determined by photoenvironment parameterHDetermine).upThe influence that different illumination are caused is eliminated, is the colored intrinsic figure of a width
Picture.
By calculating the SPD of daylight and skylight, i.e. image non-hatched area line it can be seen from above-mentioned technical proposal
Property RGB tristimulus values and the linear RGB tristimulus values of image shadow region, can calculate the corresponding parameter values of the SPD；Profit
Entropy optimization algorithm is used, targeted parameter value corresponding with image to be tested can be chosen from the parameter value；The target is joined
Numerical value substitutes into Orthogonal Decomposition formula, just can obtain the intrinsic image of the image to be tested., can by way of calculating SPD
To draw corresponding parameter value under any zenith angle, the targeted parameter value accuracy selected from the parameter value is also higher, from
And the intrinsic image being worth to according to the target component can be very good to eliminate the influence of illumination, namely intrinsic image accuracy is got over
It is high.In addition, obtaining parameter value by way of calculating, artificial operation is reduced, human resources are saved, and avoid artificial
The error produced during measurement.
Fig. 2 is a kind of structural representation for the device for obtaining intrinsic image provided in an embodiment of the present invention, including calculates single
First 21, choose unit 22 and obtain unit 23,
The computing unit 21, the SPD for calculating daylight and skylight；Wherein, the SPD includes image nonshaded area
The linear RGB tristimulus values in domain, and image shadow region linear RGB tristimulus values.
The computing unit 21 is additionally operable to linear RGB tristimulus values and described image according to described image non-hatched area
The linear RGB tristimulus values of shadow region, calculates the corresponding parameter values of the SPD；The parameter value is used to represent that image is non-
The proportionate relationship of shadow region and image shadow region.
The selection unit 22, for choosing targeted parameter value corresponding with image to be tested from the parameter value；
It is described to obtain unit 23, for the targeted parameter value to be substituted into Orthogonal Decomposition formula, obtain described to be measured attempting
The intrinsic image of picture.
Optionally, the computing unit is specifically for the linear public affairs using image non-hatched area and image shadow region
Formula, calculates the corresponding parameter values of the SPD；Wherein, the linear formula is as follows：
KH=((FH+14)/(fH+14))2.4
Wherein, KHFor the corresponding parameter values of the SPD, FHFor the linear RGB tristimulus values of image non-hatched area,
fHFor the linear RGB tristimulus values of image shadow region.
Optionally, the selection unit by the parameter value specifically for substituting into the corresponding linear public affairs of the image to be tested
In formula, the image after the shade that is eliminated, and the target image for meeting preparatory condition, the target are selected from described image
The corresponding parameter value of image is targeted parameter value.
The explanation of feature may refer to the related description of embodiment corresponding to Fig. 1 in embodiment corresponding to Fig. 2, here no longer
Repeat one by one.
By calculating the SPD of daylight and skylight, i.e. image non-hatched area line it can be seen from above-mentioned technical proposal
Property RGB tristimulus values and the linear RGB tristimulus values of image shadow region, can calculate the corresponding parameter values of the SPD；Profit
Entropy optimization algorithm is used, targeted parameter value corresponding with image to be tested can be chosen from the parameter value；The target is joined
Numerical value substitutes into Orthogonal Decomposition formula, just can obtain the intrinsic image of the image to be tested., can by way of calculating SPD
To draw corresponding parameter value under any zenith angle, the targeted parameter value accuracy selected from the parameter value is also higher, from
And the intrinsic image being worth to according to the target component can be very good to eliminate the influence of illumination, namely intrinsic image accuracy is got over
It is high.In addition, obtaining parameter value by way of calculating, artificial operation is reduced, human resources are saved, and avoid artificial
The error produced during measurement.A kind of method and apparatus of the acquisition intrinsic image provided above the embodiment of the present invention is carried out
It is discussed in detail.The embodiment of each in specification is described by the way of progressive, and what each embodiment was stressed is and other
Between the difference of embodiment, each embodiment identical similar portion mutually referring to.For device disclosed in embodiment
For, because it is corresponded to the method disclosed in Example, so description is fairly simple, related part is said referring to method part
It is bright.It should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention,
Some improvement and modification can also be carried out to the present invention, these are improved and modification also falls into the protection domain of the claims in the present invention
It is interior.
Professional further appreciates that, with reference to the unit of each example of the embodiments described herein description
And algorithm steps, can be realized with electronic hardware, computer software or the combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software, generally describes the composition and step of each example according to function in the above description.These
Function is performed with hardware or software mode actually, depending on the application-specific and design constraint of technical scheme.Specialty
Technical staff can realize described function to each specific application using distinct methods, but this realization should not
Think beyond the scope of this invention.
Directly it can be held with reference to the step of the method or algorithm that the embodiments described herein is described with hardware, processor
Capable software module, or the two combination are implemented.Software module can be placed in random access memory (RAM), internal memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Claims (6)
1. a kind of method for obtaining intrinsic image, it is characterised in that including：
Calculate the SPD of daylight and skylight；Wherein, the SPD includes the linear RGB tristimulus values of image non-hatched area, and
The linear RGB tristimulus values of image shadow region；
According to the linear RGB tristimulus values of described image non-hatched area and the linear RGB tristimulus of described image shadow region
Value, calculates the corresponding parameter values of the SPD；The parameter value is used to represent image non-hatched area and image shadow region
Proportionate relationship；
Using entropy optimization algorithm, targeted parameter value corresponding with image to be tested is chosen from the parameter value；
The targeted parameter value is substituted into Orthogonal Decomposition formula, the intrinsic image of the image to be tested is obtained.
2. according to the method described in claim 1, it is characterised in that the linear RGB according to described image non-hatched area
Tristimulus values and the linear RGB tristimulus values of described image shadow region, calculating the corresponding parameter values of the SPD includes：
Using image non-hatched area and the linear formula of image shadow region, the corresponding parameter values of the SPD are calculated；Its
In, the linear formula is as follows：
KH=((FH+14)/(fH+14))2.4
Wherein, KHFor the corresponding parameter values of the SPD, FHFor the linear RGB tristimulus values of image non-hatched area, fHFor
The linear RGB tristimulus values of image shadow region.
3. method according to claim 2, it is characterised in that the utilization entropy optimization algorithm, is selected from the parameter value
Targeted parameter value corresponding with image to be tested is taken to include：
The parameter value is substituted into the corresponding linear formula of the image to be tested, the image after the shade that is eliminated, and from
Select the target image for meeting preparatory condition in described image, the corresponding parameter value of the target image is targeted parameter value.
4. a kind of device for obtaining intrinsic image, it is characterised in that including computing unit, choose unit and obtain unit,
The computing unit, the SPD for calculating daylight and skylight；Wherein, the SPD includes the line of image non-hatched area
Property RGB tristimulus values, and image shadow region linear RGB tristimulus values；
The computing unit is additionally operable to according to the linear RGB tristimulus values of described image non-hatched area and described image shadow region
The linear RGB tristimulus values in domain, calculates the corresponding parameter values of the SPD；The parameter value is used to represent image nonshaded area
The proportionate relationship in domain and image shadow region；
The selection unit, for choosing targeted parameter value corresponding with image to be tested from the parameter value；
It is described to obtain unit, for the targeted parameter value to be substituted into Orthogonal Decomposition formula, obtain the sheet of the image to be tested
Levy image.
5. device according to claim 4, it is characterised in that the computing unit is specifically for utilizing image nonshaded area
Domain and the linear formula of image shadow region, calculate the corresponding parameter values of the SPD；Wherein, the linear formula is as follows：
KH=((FH+14)/(fH+14))2.4
Wherein, KHFor the corresponding parameter values of the SPD, FHFor the linear RGB tristimulus values of image non-hatched area, fHFor
The linear RGB tristimulus values of image shadow region.
6. device according to claim 5, it is characterised in that the selection unit is specifically for the parameter value is substituted into
In the corresponding linear formula of the image to be tested, the image after the shade that is eliminated, and select from described image satisfaction
The target image of preparatory condition, the corresponding parameter value of the target image is targeted parameter value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710311828.2A CN107146225A (en) | 2017-05-05 | 2017-05-05 | A kind of method and apparatus for obtaining intrinsic image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710311828.2A CN107146225A (en) | 2017-05-05 | 2017-05-05 | A kind of method and apparatus for obtaining intrinsic image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107146225A true CN107146225A (en) | 2017-09-08 |
Family
ID=59777003
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710311828.2A Pending CN107146225A (en) | 2017-05-05 | 2017-05-05 | A kind of method and apparatus for obtaining intrinsic image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107146225A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184403A (en) * | 2011-05-20 | 2011-09-14 | 北京理工大学 | Optimization-based intrinsic image extraction method |
| US20130089263A1 (en) * | 2011-10-11 | 2013-04-11 | Tandent Vision Science, Inc. | System and method for digital image signal compression using intrinsic images |
| CN104484884A (en) * | 2014-12-30 | 2015-04-01 | 天津大学 | Intrinsic image decomposition method based on multi-scale L0 sparse constraint |
| CN104700109A (en) * | 2015-03-24 | 2015-06-10 | 清华大学 | Method and device for decomposing hyper-spectral intrinsic images |
| CN106355601A (en) * | 2016-08-31 | 2017-01-25 | 上海交通大学 | Intrinsic image decomposition method |
-
2017
- 2017-05-05 CN CN201710311828.2A patent/CN107146225A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184403A (en) * | 2011-05-20 | 2011-09-14 | 北京理工大学 | Optimization-based intrinsic image extraction method |
| US20130089263A1 (en) * | 2011-10-11 | 2013-04-11 | Tandent Vision Science, Inc. | System and method for digital image signal compression using intrinsic images |
| CN104484884A (en) * | 2014-12-30 | 2015-04-01 | 天津大学 | Intrinsic image decomposition method based on multi-scale L0 sparse constraint |
| CN104700109A (en) * | 2015-03-24 | 2015-06-10 | 清华大学 | Method and device for decomposing hyper-spectral intrinsic images |
| CN106355601A (en) * | 2016-08-31 | 2017-01-25 | 上海交通大学 | Intrinsic image decomposition method |
Non-Patent Citations (3)
| Title |
|---|
| JIANDONG TIAN等: "Linearity of Each Channel Pixel Values from a Surface in and out of Shadows and its application", 《CVPR》 * |
| 段志刚: "室外自然光照成像计算与阴影处理技术研究", 《中国科学院机构知识库网格》 * |
| 段志刚等: "基于正交分解的室外光照阴影检测", 《光学学报》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103824544B (en) | The bearing calibration of LED display, Apparatus and system | |
| DE102013114631B4 (en) | Calibration of lens shading for cameras | |
| CN102456142B (en) | A kind of analysis method for smoke blackness based on computer vision | |
| Stumpfel et al. | Direct HDR capture of the sun and sky | |
| US8619153B2 (en) | Radiometric calibration using temporal irradiance mixtures | |
| DE102009029391B4 (en) | Image processing apparatus and image processing method | |
| CN103578429A (en) | LED display screen system and rectifying method for bright and dark lines thereof | |
| CN106705855A (en) | High-dynamic performance three-dimensional measurement method based on adaptive grating projection | |
| CN107894215A (en) | HDR optical grating projection method for three-dimensional measurement based on fully automatic exposure | |
| CN109520706A (en) | Automobile fuse box assembly detection system, image-recognizing method and screw hole positioning mode | |
| CN104994376B (en) | Method and system for automatically simulating projection colors of projector | |
| CN104121850A (en) | Canopy density measurement method and device | |
| JP5076055B2 (en) | Image illumination detection | |
| CN108088658A (en) | A kind of dazzle measuring method and its measuring system | |
| CN111105365A (en) | Color correction method, medium, terminal and device for texture image | |
| CN107566689A (en) | System and method for changing input image data | |
| CN101185103A (en) | Image processing methods | |
| CN110231018A (en) | A kind of structure light measurement method, apparatus and computer readable storage medium | |
| CN103218820B (en) | A kind of camera calibration error compensating method based on multidimensional characteristic | |
| CN107833223A (en) | A kind of fruit high-spectrum image segmentation method based on spectral information | |
| CN101815190B (en) | Projection display system and characteristic analysis method thereof | |
| KR20100032618A (en) | Method and apparatus for correcting white balance | |
| US20160070025A1 (en) | Weather forecasting apparatus and weather forecasting method | |
| Tu et al. | 2D in situ method for measuring plant leaf area with camera correction and background color calibration | |
| CN107146225A (en) | A kind of method and apparatus for obtaining intrinsic image |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170908 |