CN109389579A - A kind of combination is entropy constrained anti-light according to interference method with the textile acetes chinensis of standard deviation weighting - Google Patents
A kind of combination is entropy constrained anti-light according to interference method with the textile acetes chinensis of standard deviation weighting Download PDFInfo
- Publication number
- CN109389579A CN109389579A CN201810994862.9A CN201810994862A CN109389579A CN 109389579 A CN109389579 A CN 109389579A CN 201810994862 A CN201810994862 A CN 201810994862A CN 109389579 A CN109389579 A CN 109389579A
- Authority
- CN
- China
- Prior art keywords
- image
- standard deviation
- sub
- textile
- entropy
- 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
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30124—Fabrics; Textile; Paper
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
It is entropy constrained anti-light according to interference method with the textile acetes chinensis of standard deviation weighting that the present invention provides a kind of combination, it is anti-light according to perturbation technique field to be related to textile acetes chinensis, first with entropy constrained method, lighting process is carried out to textile image, then by treated, image is divided into several sub-blocks, calculates tri- channel average and standard deviations of each sub-block R, G, B；If standard deviation is big, illustrate that color is relatively abundant in the sub-block, because the difference of color is relatively large, and standard deviation is also big when there are different objects in a certain sub-block, similarly, when color is relatively simple in sub-block, standard deviation is smaller；By carrying out standard deviation weighting (giving up the small sub-block of correlation, the big sub-block of retention relationship) to each sub-block, the influence of bulk solid color can reduce, being equivalent to the image " conversion " of bulk solid color is the image with rich colors.
Description
Technical field
It is anti-light according to perturbation technique field that the invention belongs to textile acetes chinensis, and in particular to a kind of combination is entropy constrained and marks
The textile acetes chinensis of quasi- difference weighting is anti-light to shine interference method.
Background technique
In textile printing and dyeing industry, the color difference between measurement dyeing cloth has considerable meaning in the industrial production
Color difference in justice, especially printing and dyeing mill between production sample and standard sample is an important indicator of quality management.Cloth exists
The color presented under different light sources is different.Different light environment, it will between the color for leading to the cloth image of acquisition
There are a degree of deviation, this deviation will will affect the accuracy of subsequent image analysis.Therefore seek color correction algorithm
The influence to cloth Color development such as elimination or decrease light environment.
Gray world algorithm is based on being assumed by gray world, it is assumed that thinks there are a large amount of color changes for a width
The average value of tri- components of image R, G, B tends to the same gray value.But when number of colors is less in image or bulk occurs
When solid color, which can usually fail.
(1) the technical issues of solving
For the deficiency of gray world algorithm penalty in the case where scene color is less, the invention proposes one kind
The textile acetes chinensis weighted in conjunction with entropy constrained and standard deviation is anti-light according to interference method, constrains gain coefficient using image entropy
To prevent correction from deteriorating, i.e., gray world method entropy constrained based on image.
(2) technical solution
In order to achieve the above object, the present invention is achieved by the following technical programs:
A kind of combination is entropy constrained anti-light according to interference method, including following step with the textile acetes chinensis of standard deviation weighting
It is rapid:
S1, the average value for calculating tri- channels R, G, BEnable the average gray value of image
S2, the gain coefficient for asking tri- channels R, G, B；
S3, the one-dimensional discrete entropy for calculating separately three channels,Wherein: k=R, G, B；
Pk,iIndicate pixel that the gray value of k component is i shared ratio in the picture；
S4, by treated, image is divided into the sub-block B that size is 16X16i,j(1≤i≤M, 1≤i≤N), is divided into M
× N number of sub-block；
S5, for each sub-block Bi,j, calculate R, tri- channel average value (R of G, Bi,j、Gi,j、Bi,j) and standard deviation (SRi,j、
SGi,j、SBi,j)；
S6, for each sub-block Bi,j, calculate its related coefficient Di,j；
DI, j=(| SRI, j-SRI, j+1|+|SGI, j-SGI, j+1|+|SBI, j-SBI, j+1|)/(|SRI, j+SRI, j+1|+|SGI, j+
SGI, j+1|+|SBI, j+SBI, j+1|)
SR in formulai,j、SRi,j+1For the standard deviation of adjacent sub-blocks；
S7, for each sub-block Bi,j, setting mark Fi,j
In formula: θ is preset threshold value；
S8, the R for calculating standard weighting, G, B average value
S9, R, G, the gain coefficient of B triple channel are corrected are as follows:
Gain coefficient correction is carried out by above formula, and by each pixel R of imagenew、Gnew、BnewBe adjusted to displayable range it
It is interior, i.e., [0,255], the image after output calibration；
Image after S10, output calibration calculates value of chromatism Δ E compared with standard picture.
Further, in the step S1Calculation formula it is as follows: Wherein N is the sum of all pixels of image, Ri、Gi、BiIth pixel before respectively correcting
Three components of RGB.
Further, the following formula of the step S2 gain coefficient:
Further, the one-dimensional discrete entropy in tri- channels the step S3, specific implementation process can be divided into four steps:
S3.1, one-dimensional discrete relative entropy is calculated separately to three Color Channels of image first
Wherein: k=R, G, B；Pk,iThe pixel for indicating that the gray value of k component is i is being schemed
The shared ratio as in；
S3.2, R, G, " constraint " gain coefficient of B triple channel are calculated；
S3.3, the correction of progress " constraint " gain coefficient；
S3.4, tri- components of R', G', B' of each pixel of image are adjusted within the range [0,255] that can be shown,
The maximum value MAXval for first having to find out all R', G', B' in image, enables factor=MAXval/255, if factor > 1,
Then pixel each in image is readjusted, the image after last output adjustment.
Further, in the step S3.2 triple channel " constraint " gain coefficient,
Wherein, kr、kg、kbFor three channel gain coefficients of gray world algorithm.
Further, the step S3.3 gain coefficient correction, for each of image pixel C, after adjustment
Three channels be R', G', B'；
Further, pixel each in image is readjusted in the step S3.4, R', G', B' obtain new
R ", G ", B " are calculated with following formula:
Further, threshold θ=0.1 in the step S7.
Further, color difference Δ E < 0.5 indicates small color difference in the step S10；0.5≤Δ E < 1.5 indicates small color
Difference；1.5≤Δ E < 3 indicates smaller color difference；3≤Δ E < 6 indicates larger color difference；6≤Δ E indicates big color difference.
(3) beneficial effect
Beneficial effects of the present invention: a kind of combination is entropy constrained anti-light according to interference with the textile acetes chinensis of standard deviation weighting
Method is first calculated the one-dimensional discrete relative entropy of three Color Channels of image, is carried out about using gain coefficient of the entropy to triple channel
Beam carries out constraint gain coefficient correction, image after last output calibration to image later；The present invention constrains increasing using image entropy
Beneficial coefficient avoids between the color of image acquired under different light environments that there are deviations, effectively to prevent correction from deteriorating
Improve the accuracy of subsequent image analysis；The present invention is used to be calculated based on entropy constrained gray world, and color of image is abundanter, thenCloser to 1,I.e. to the gain coefficient in channel almost without doing any adjustment；
Otherwise color of image is fewer,Closer to 0,Color of image does not do any correction, to prevent " mistake
The phenomenon that correction ".
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the method for the present invention flow chart；
Fig. 2 is the color difference contrast curve chart of original image, processing result image and standard picture.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
Color is a kind of important feature of image, for fields such as image segmentation, object detection and recognition, image retrievals
Research is of great significance.The color that object is presented under different light sources is different.For gray world algorithm in scene face
The deficiency of penalty in the case that color is less, devise it is a kind of constrained using image entropy gain coefficient to prevent correction dislike
The algorithm of change, i.e., gray world algorithm entropy constrained based on image.Gray world algorithm is the color of image according to actual photographed
Information does color correction, therefore image color information has conclusive influence.Scene number of color is more, then color school
Positive error is smaller；And color is fewer, error is bigger.Current gray world algorithm, the situation less to color, correction ratio is not
The error of correction is also big.Therefore need to find the mechanism for detecting this special circumstances to reduce its influence, entropy is exactly one
Applicable tool.
In conjunction with Fig. 1, a kind of combination is entropy constrained anti-light according to interference method with the textile acetes chinensis of standard deviation weighting, including
Following steps:
S1, the average value for calculating tri- channels R, G, B(N is the sum of all pixels of image, Ri、Gi、BiPoint
Three components of RGB of ith pixel before Wei not correcting)；Enable the average gray value of imageIts
In,Calculation formula it is as follows:
S2, the gain coefficient for asking tri- channels R, G, B, with following formula:
S3, the one-dimensional discrete entropy for calculating separately three channels export image；Specific implementation process can be divided into four steps:
S3.1, one-dimensional discrete relative entropy is calculated separately to three Color Channels of image first
Wherein: k=R, G, B；Pk,iThe pixel for indicating that the gray value of k component is i is being schemed
The shared ratio as in.
S3.2, calculate R, G, " constraint " gain coefficient of B triple channel:
Wherein, kr、kg、kbFor three channel gain coefficients of gray world algorithm.
S3.3, the correction of progress " constraint " gain coefficient.For each of image pixel C, three after correction
Channel is R', G', B'；
S3.4, tri- components of R', G', B' of each pixel of image are adjusted within the range [0,255] that can be shown,
The maximum value MAXval for first having to find out all R', G', B' in image, enables factor=MAXval/255, if factor > 1,
Then pixel each in image is readjusted.Its R', G', B' obtain new R ", G ", B ", are calculated with following formula:
Image after output adjustment.
S4, by treated, image is divided into the sub-block B that size is 16X16i,j(1≤i≤M, 1≤i≤N), is divided into M
× N number of sub-block.Since the size of sub-block has a great impact to result, sub-block is too big or too small cannot all obtain good effect
Fruit.Since the width and height of real image are not necessarily 16 integral multiple, so in practical piecemeal, it is possible to give up figure
As some pixels at edge, this processing influences result little.
S5, for each sub-block Bi,j, calculate R, tri- channel average value (R of G, Bi,j、Gi,j、Bi,j) and standard deviation (SRi,j、
SGi,j、SBi,j)。
S6, for each sub-block Bi,j, calculate its related coefficient Di,j；
SR in formulai,j、SRi,j+1For the standard deviation of adjacent sub-blocks.According to Di,jThe numerical value sub-block small to correlation give up
It abandons, to reduce the influence of bulk solid color.
S7, for each sub-block Bi,j, setting mark Fi,j
In formula: θ is preset threshold value, θ=0.1.
S8, the R for calculating standard weighting, G, B average value
S9, R, G, the gain coefficient of B triple channel are corrected are as follows:
Gain coefficient correction is carried out by above formula, and by each pixel R of imagenew、Gnew、BnewBe adjusted to displayable range it
It is interior, i.e., [0,255]；Image after output calibration.
Image after S10, output calibration calculates value of chromatism Δ E compared with standard picture.
In implementation process of the present invention, step S2 proposes entropy constrained thought, and using based on entropy constrained gray scale generation
Boundary's algorithm.In the algorithm, color of image is abundanter, thenCloser to 1,I.e. to logical
The gain coefficient in road is almost without doing any adjustment；Otherwise color of image is fewer,Closer to 0,Figure
As color does not do any correction, thus the phenomenon that preventing " overcorrect ".
Embodiment:
A kind of combination is entropy constrained anti-light according to interference method, including following step with the textile acetes chinensis of standard deviation weighting
It is rapid:
S1, image is read, calculates the average value in tri- channels R, G, BWherein,'s
Calculation formula is as follows:
S2, the gain coefficient for asking tri- channels R, G, B, with following formula:
S3, the one-dimensional discrete entropy for calculating separately three channels export image；Specific implementation process can be divided into four steps:
Entropy in S3.1, information theory be describe variable information amount number, is defined as: assuming that collection of random event
{Xi, i=1,2 ..., N they occur probability be respectively Pi, and meet condition:
Define comentropy are as follows:
It is assumed that image data has a nonnegative value, i.e. f (x, y) >=0, for the image of a width M × N,
Image entropy is a kind of statistical form of the included information of image, it reflects the number of average information in image.
The one-dimensional entropy of image indicates the information content that the aggregation characteristic of intensity profile in image is included, and enables PiIndicate that gray value is in image
Ratio shared by the pixel of i then defines the one-dimensional discrete entropy of gray level image are as follows:
Image entropy is asked for color image, the frequency distribution of three its gray values of channel is counted first, then calculates one
Discrete entropy is tieed up, finally calculates the average value of three channel one-dimensional discrete entropys as image entropy.For real image, the size of entropy is not
It is only dependent upon picture material, additionally depending on is to express grayscale with how many number of bits.Relative entropy H is defined hereinR, it reflects
The abundant degree of color of image.Calculation formula is as follows:
In formula, NbitIt is the number of bits of grayscale.If it is bianry image, then Nbit=1；Such as
Fruit is 256 gray scale images, then Nbit=8.Color of image quantity is more, abundanter, then its relative entropy is bigger；Conversely, then smaller.
Wherein HRFor the average value of three channel relative entropies.
S3.2, calculate R, G, " constraint " gain coefficient of B triple channel:
Wherein, kr、kg、kbFor three channel gain coefficients of gray world algorithm.
S3.3, the correction of progress " constraint " gain coefficient.For each of image pixel C, three after correction
Channel is R', G', B'；
S3.4, tri- components of R', G', B' of each pixel of image are adjusted within the range [0,255] that can be shown,
The maximum value MAXval for first having to find out all R', G', B' in image, enables factor=MAXval/255, if factor > 1,
Then pixel each in image is readjusted.Its R', G', B' obtain new R ", G ", B ", are calculated with following formula:
Image after output adjustment.
S4, by treated, image is divided into the sub-block B that size is 16X16i,j(1≤i≤M, 1≤i≤N), is divided into M
× N number of sub-block.Since the size of sub-block has a great impact to result, sub-block is too big or too small cannot all obtain good effect
Fruit.Since the width and height of real image are not necessarily 16 integral multiple, so in practical piecemeal, it is possible to give up figure
As some pixels at edge, this processing influences result little.
S5, for each sub-block Bi,j, calculate R, tri- channel average value (R of G, Bi,j、Gi,j、Bi,j) and standard deviation (SRi,j、
SGi,j、SBi,j)。
S6, for each sub-block Bi,j, calculate its related coefficient Di,j；
DI, j=(| SRI, j-SRI, j+1|+|SGI, j-SGI, j+1|+|SBI, j-SBI, j+1|)/(|SRI, j+SRI, j+1|+|SGI, j+
SGI, j+1|+|SBI, j+SBI, j+1|)
SR in formulai,j、SRi,j+1For the standard deviation of adjacent sub-blocks.According to Di,jThe numerical value sub-block small to correlation give up
It abandons, to reduce the influence of bulk solid color.
S7, for each sub-block Bi,j, setting mark Fi,j
In formula: θ is preset threshold value, θ=0.1.
S8, the R for calculating standard weighting, G, B average value
S9, R, G, the gain coefficient of B triple channel are corrected are as follows:
Gain coefficient correction is carried out by above formula, and by each pixel R of imagenew、Gnew、BnewBe adjusted to displayable range it
Interior, i.e., [0,255], specific method is the same as step S3.4.Image after last output calibration.
Image after S10, output calibration calculates value of chromatism Δ E, wherein Δ E < 0.5 indicates small compared with standard picture
Color difference；0.5≤Δ E < 1.5 indicates small color difference；1.5≤Δ E < 3 indicates smaller color difference；3≤Δ E < 6 indicates larger color difference；6
≤ Δ E indicates big color difference.
Using c1 as standard sample, c2-c10 is test sample, obtains original image, processing result image and standard picture
Color difference correlation data is as shown in table 1, and Fig. 2 gives the color difference correlation curve of original image, processing result image and standard picture,
From table 1 and Fig. 2 as can be seen that under the influence of varying environment light, there are biggish color difference between test sample and standard sample,
After the algorithm process, the influence of ambient lighting can be preferably eliminated, vision subjective sensation and target image are almost without difference
It is different.Color difference △ E is calculated by Lab colour difference formula, by data as can be seen that the present invention has relatively by force different illumination conditions
Robustness.
Table 1: the color difference correlation data of original image, processing result image and standard picture
In conclusion the embodiment of the present invention, the textile acetes chinensis weighted in conjunction with entropy constrained and standard deviation is anti-light according to dry
Method is disturbed, the one-dimensional discrete relative entropy of three Color Channels of image is first calculated, is carried out using gain coefficient of the entropy to triple channel
Constraint carries out constraint gain coefficient correction, image after last output calibration to image later；The present invention is constrained using image entropy
Gain coefficient avoids between the color of image acquired under different light environments that there are deviations, effectively to prevent correction from deteriorating
Ground improves the accuracy of subsequent image analysis；The present invention is used to be calculated based on entropy constrained gray world, and color of image is abundanter,
ThenCloser to 1, I.e. to the gain coefficient in channel almost without doing any adjustment；
Otherwise color of image is fewer,Closer to 0,Color of image does not do any correction, to prevent " mistake
The phenomenon that correction ".
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification or
Replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (9)
1. a kind of combination is entropy constrained anti-light according to interference method with the textile acetes chinensis of standard deviation weighting, which is characterized in that packet
Include following steps:
S1, the average value for calculating tri- channels R, G, BEnable the average gray value of image
S2, the gain coefficient for asking tri- channels R, G, B；
S3, the one-dimensional discrete entropy for calculating separately three channels,Wherein: k=R, G, B；Pk,iIt indicates
The pixel that the gray value of k component is i shared ratio in the picture；
S4, by treated, image is divided into the sub-block B that size is 16X16i,j(1≤i≤M, 1≤i≤N) is divided into M × N number of
Sub-block；
S5, for each sub-block Bi,j, calculate R, tri- channel average value (R of G, Bi,j、Gi,j、Bi,j) and standard deviation (SRi,j、
SGi,j、SBi,j)；
S6, for each sub-block Bi,j, calculate its related coefficient Di,j；
Di,j=(| SRi,j-SRi,j+1|+|SGi,j-SGi,j+1|+|SBi,j-SBi,j+1|)/
(|SRi,j+SRi,j+1|+|SGi,j+SGi,j+1|+|SBi,j+SBi,j+1|)
SR in formulai,j、SRi,j+1For the standard deviation of adjacent sub-blocks；
S7, for each sub-block Bi,j, setting mark Fi,j
In formula: θ is preset threshold value；
S8, the R for calculating standard weighting, G, B average value
S9, R, G, the gain coefficient of B triple channel are corrected are as follows:
Gain coefficient correction is carried out by above formula, and by each pixel R of imagenew、Gnew、BnewIt is adjusted within displayable range, i.e.,
[0,255], the image after output calibration；
Image after S10, output calibration calculates value of chromatism Δ E compared with standard picture.
2. a kind of combination as described in claim 1 is entropy constrained anti-light according to disturber with the textile acetes chinensis of standard deviation weighting
Method, it is characterised in that: in the step S1Calculation formula it is as follows:Wherein N is the sum of all pixels of image, Ri、Gi、BiRespectively
Three components of RGB of ith pixel before correcting.
3. a kind of combination as described in claim 1 is entropy constrained anti-light according to disturber with the textile acetes chinensis of standard deviation weighting
Method, which is characterized in that the following formula of the step S2 gain coefficient:
4. a kind of combination as described in claim 1 is entropy constrained anti-light according to disturber with the textile acetes chinensis of standard deviation weighting
Method, which is characterized in that the one-dimensional discrete entropy in tri- channels the step S3, specific implementation process can be divided into four steps:
S3.1, one-dimensional discrete relative entropy is calculated separately to three Color Channels of image first
Wherein: k=R, G, B；Pk,iIndicate the gray value of k component for the pixel institute in the picture of i
The ratio accounted for；
S3.2, R, G, " constraint " gain coefficient of B triple channel are calculated；
S3.3, the correction of progress " constraint " gain coefficient；
S3.4, tri- components of R', G', B' of each pixel of image are adjusted within the range [0,255] that can be shown, first
The maximum value MAXval for finding out all R', G', B' in image, enables factor=MAXval/255, right if factor > 1
Each pixel is readjusted in image, the image after last output adjustment.
5. a kind of combination as claimed in claim 4 is entropy constrained anti-light according to disturber with the textile acetes chinensis of standard deviation weighting
Method, which is characterized in that " constraint " gain coefficient of triple channel in the step S3.2,
Wherein, kr、kg、kbFor three channel gain coefficients of gray world algorithm.
6. a kind of combination as claimed in claim 4 is entropy constrained anti-light according to disturber with the textile acetes chinensis of standard deviation weighting
Method, which is characterized in that step S3.3 gain coefficient correction, for each of image pixel C, adjusted three
A channel is R', G', B '；
7. a kind of combination as claimed in claim 4 is entropy constrained anti-light according to disturber with the textile acetes chinensis of standard deviation weighting
Method, which is characterized in that in the step S3.4 for pixel each in image readjust, R', G', B' obtain new R ",
G ", B " are calculated with following formula:
8. a kind of combination as described in claim 1 is entropy constrained anti-light according to disturber with the textile acetes chinensis of standard deviation weighting
Method, which is characterized in that threshold θ=0.1 in the step S7.
9. a kind of combination as described in claim 1 is entropy constrained anti-light according to disturber with the textile acetes chinensis of standard deviation weighting
Method, which is characterized in that color difference Δ E < 0.5 indicates small color difference in the step S10；0.5≤Δ E < 1.5 indicates small color difference；
1.5≤Δ E < 3 indicates smaller color difference；3≤Δ E < 6 indicates larger color difference；6≤Δ E indicates big color difference.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810994862.9A CN109389579A (en) | 2018-08-29 | 2018-08-29 | A kind of combination is entropy constrained anti-light according to interference method with the textile acetes chinensis of standard deviation weighting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810994862.9A CN109389579A (en) | 2018-08-29 | 2018-08-29 | A kind of combination is entropy constrained anti-light according to interference method with the textile acetes chinensis of standard deviation weighting |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109389579A true CN109389579A (en) | 2019-02-26 |
Family
ID=65418603
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810994862.9A Pending CN109389579A (en) | 2018-08-29 | 2018-08-29 | A kind of combination is entropy constrained anti-light according to interference method with the textile acetes chinensis of standard deviation weighting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109389579A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110148095A (en) * | 2019-05-14 | 2019-08-20 | 北京科技大学 | A kind of underwater picture Enhancement Method and enhancement device |
| CN110646352A (en) * | 2019-10-18 | 2020-01-03 | 江南大学 | Method for identifying different-color spun yarn bobbin |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107144353A (en) * | 2017-04-12 | 2017-09-08 | 江南大学 | A kind of textile chromatism measurement method based on digital camera |
| CN107248154A (en) * | 2017-05-27 | 2017-10-13 | 江苏理工学院 | A kind of cloth aberration real-time on-line detecting method |
-
2018
- 2018-08-29 CN CN201810994862.9A patent/CN109389579A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107144353A (en) * | 2017-04-12 | 2017-09-08 | 江南大学 | A kind of textile chromatism measurement method based on digital camera |
| CN107248154A (en) * | 2017-05-27 | 2017-10-13 | 江苏理工学院 | A kind of cloth aberration real-time on-line detecting method |
Non-Patent Citations (2)
| Title |
|---|
| 侯小平等著: "《凹版印刷工》", 30 June 2015, 文化发展出版社 * |
| 徐晓昭 等: "改进灰度世界颜色校正算法", 《光子学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110148095A (en) * | 2019-05-14 | 2019-08-20 | 北京科技大学 | A kind of underwater picture Enhancement Method and enhancement device |
| CN110148095B (en) * | 2019-05-14 | 2021-05-07 | 北京科技大学 | Underwater image enhancement method and enhancement device |
| CN110646352A (en) * | 2019-10-18 | 2020-01-03 | 江南大学 | Method for identifying different-color spun yarn bobbin |
| CN110646352B (en) * | 2019-10-18 | 2021-07-23 | 江南大学 | Method for identifying different-color spun yarn bobbin |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109063694B (en) | Video target detection and identification method | |
| Lissner et al. | Image-difference prediction: From grayscale to color | |
| US8295606B2 (en) | Device and method for detecting shadow in image | |
| CN100551081C (en) | A kind of method and device of realizing white balance correction | |
| CN110490914A (en) | It is a kind of based on brightness adaptively and conspicuousness detect image interfusion method | |
| CN104834912A (en) | Weather identification method and apparatus based on image information detection | |
| CN115018844B (en) | Plastic film quality evaluation method based on artificial intelligence | |
| CN113554629A (en) | Strip steel red rust defect detection method based on artificial intelligence | |
| CN104504722B (en) | Method for correcting image colors through gray points | |
| CN110275975A (en) | A kind of method for quickly retrieving of similar pictures | |
| CN109389579A (en) | A kind of combination is entropy constrained anti-light according to interference method with the textile acetes chinensis of standard deviation weighting | |
| CN114878595B (en) | Book printing quality detection method | |
| CN115512224A (en) | Road crack detection method with enhanced curve fitting contrast | |
| CN109447915A (en) | The line scan image increased quality method with gamma gray correction is established based on characteristic model | |
| CN112637593A (en) | Video coding optimization method based on artificial intelligence and video analysis | |
| CN108965749B (en) | Defect pixel detection and correction device and method based on texture recognition | |
| CN110223253B (en) | Defogging method based on image enhancement | |
| Sazzad et al. | Establishment of an efficient color model from existing models for better gamma encoding in image processing | |
| CN111862244A (en) | Plastic sheet intelligent color sorting method based on image processing | |
| CN114998346B (en) | Waterproof cloth quality data processing and identifying method | |
| CN115147410B (en) | Multicolor block packaging film covering defect detection method | |
| CN113610940B (en) | Ocean vector file and image channel threshold based coastal area color homogenizing method | |
| Ning et al. | The optimization and design of the auto-exposure algorithm based on image entropy | |
| CN107782447A (en) | Space dimension automatic identifying method and system in imaging spectrometer spectral calibration | |
| Saberi et al. | Adaptive contrast enhancement of satellite images based on histogram and non-linear transfer function methods |
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 |
Application publication date: 20190226 |
|
| RJ01 | Rejection of invention patent application after publication |