CN107369144A - Based on the multiple dimensioned Retinex images defogging method for improving homomorphic filtering - Google Patents
Based on the multiple dimensioned Retinex images defogging method for improving homomorphic filtering Download PDFInfo
- Publication number
- CN107369144A CN107369144A CN201710565491.8A CN201710565491A CN107369144A CN 107369144 A CN107369144 A CN 107369144A CN 201710565491 A CN201710565491 A CN 201710565491A CN 107369144 A CN107369144 A CN 107369144A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- msup
- image
- frequency
- 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
- 238000001914 filtration Methods 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000012546 transfer Methods 0.000 claims abstract description 5
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 7
- 239000003595 mist Substances 0.000 description 7
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000001678 irradiating effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005728 strengthening Methods 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
- G06T5/73—Deblurring; Sharpening
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 based on the multiple dimensioned Retinex images defogging method for improving homomorphic filtering, belong to technical field of image processing, comprise the following steps：Step 01：Input foggy image I（x）；Step 02：To traditional homomorphic filtering transfer function H（U, v）It is improved；Step 03：By foggy image I（x）Tri- color channel images of R, G, B are decomposed into, the homomorphic filtering processing after being improved respectively to each Color Channel；Step 04：Multi-Scale Retinex Algorithm processing is carried out respectively to the filtered channel image obtained in step 03；Step 05：The channel image that step 04 is obtained merges, and obtains the image after defogging.It is an advantage of the invention that：Improve defog effect so that the picture contrast height after defogging, details are clear, and computation complexity is low, speed is fast.
Description
Technical field
It is more particularly to a kind of based on the multiple dimensioned of improvement homomorphic filtering the present invention relates to technical field of image processing
Retinex image defogging methods.
Background technology
Haze occurs in China frequently and distributed areas are wide, is a kind of common weather phenomenon.Influenceed by haze, imaging is set
It is standby to be difficult to and extract characteristics of image, our life therefore can be by a certain degree of influence, such as under haze weather
Traffic accident incidence increases, and the picture contrast that monitoring system is shot is low, is difficult to extract feature of personage etc..
In defogging method based on image enhaucament, homomorphic filtering and Retinex algorithm are two kinds of representative defoggings
Algorithm, both algorithms are all based on irradiating reflection model, it is intended to remove irradiation light and retain reflected light so as to obtain image original
This color, but Homomorphic Filtering Algorithm defog effect is bad, contrast is poor after Retinex algorithm defogging, details is unintelligible, and
The problems such as cross-color occurs.
The content of the invention
The technical problems to be solved by the invention are the defects of being directed to background technology, are proposed a kind of based on improvement homomorphic filtering
Multiple dimensioned Retinex images defogging method, how realization be efficiently modified that mist elimination image mist residual is excessive, details is unintelligible
The problem of.
The present invention above-mentioned technical problem technical scheme is that：
A kind of multiple dimensioned Retinex images defogging method based on improvement homomorphic filtering, comprises the following steps：
Step 01：Input foggy image I (x)；
Step 02：Traditional homomorphic filtering transfer function H (u, v) is improved, traditional H (u, v) expression formula is：It is as follows to improve step：
a：Cut-off frequency D is seto；
b：To cut-off frequency in DoWithin part suppressed；
c：To cut-off frequency in DoPart above is strengthened；
d：Setting removes cut-off frequency DoOther specification numerical value in addition, obtain improved homomorphic filtering transmission function expression
Formula：
Wherein rHRepresent high-frequency gain, rLLow-frequency gain is represented, and meets rH>1,rL<1, D (u, v) represents frequency (u, v)
To filter center (uo,vo) distance, DoCut-off frequency is represented, coefficient c is used to control sharpness, and size is between rHAnd rLIt
Between；
Step 03：Foggy image I (x) is decomposed into tri- color channel images of R, G, B, each Color Channel is entered respectively
Homomorphic filtering after row improves is handled；
Step 04：Multi-Scale Retinex Algorithm processing is carried out respectively to the filtered channel image obtained in step 03；
Step 05：The channel image that step 04 is obtained merges, and obtains the image after defogging.
Further, wherein, rH=1.5, rL=0.5, c=1, Do=4.
Further, multi-Scale Retinex Algorithm expression formula is in step 04：
Wherein ri(x, y) represents the output in i-th of passage, Si(x, y) represents the foggy image of i-th of passage input, i
Value be 1,2,3；N represents the number of scale parameter；ωnThe weight of each scale parameter is represented, and is metFn
(x, y) is to surround function, and * represents convolution.
Further, around function FnThe expression formula of (x, y) is：
Wherein λnIt is normalization factor, b represents scale parameter.
Further, ∫ ∫ Fn(x, y) dxdy=1.
Further, wherein N value takes 3, and the scale parameter is set to 15,100,240.The present invention uses above skill
Art scheme compared with prior art, has following technique effect：
It is proposed by the present invention a kind of based on the multiple dimensioned Retinex images defogging method for improving homomorphic filtering, using filter
Ripple has used improved new forestry when handling foggy image, while carries out high LPF, to cut-off frequency in DoWithin
Part suppressed, that is, inhibit the stronger part of irradiation component；To cut-off frequency in DoPart above is increased
By force, that is, the details of image is enhanced；The foggy image I (x) of input tri- Color Channels of RGB are carried out at homomorphic filtering respectively
Reason, reach the effect for suppressing irradiation component and compression of dynamic range；Multi-Scale Retinex Algorithm processing is carried out again to each passage；
Each channel image after processing is merged, obtains final defogging result.Defogging speed of the present invention is fast, improves defogging effect
Fruit, the problem of multiple dimensioned Retinex defoggings mist residual is excessive is improved, details picture rich in detail contrast is strong after defogging, has real
The property used.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the present invention；
Fig. 2 positions present invention and homomorphic filtering and the comparison diagram of multi-Scale Retinex Algorithm.
Embodiment
The present invention is described in further detail below in conjunction with accompanying drawing.
First, as shown in figure 1, a kind of multiple dimensioned Retinex images based on improvement homomorphic filtering provided by the invention are gone
Mist method, it specifically includes following step：
Step 01：Input foggy image I (x)；
Step 02：Traditional homomorphic filtering transfer function H (u, v) is improved, traditional H (u, v) expression formula is：It is as follows to improve step：
a：Cut-off frequency D is seto；
b：To cut-off frequency in DoWithin part suppressed；
c：To cut-off frequency in DoPart above is strengthened；
d：Setting removes cut-off frequency DoOther specification numerical value in addition, obtain improved homomorphic filtering transmission function expression
Formula：
Wherein rHRepresent high-frequency gain, rLLow-frequency gain is represented, and meets rH>1,rL<1, D (u, v) represents frequency (u, v)
To filter center (uo,vo) distance, DoCut-off frequency is represented, coefficient c is used to control sharpness, and size is between rHAnd rLIt
Between；
Step 03：Foggy image I (x) is decomposed into tri- color channel images of R, G, B, each Color Channel is entered respectively
Homomorphic filtering after row improves is handled；
Step 04：Multi-Scale Retinex Algorithm processing is carried out respectively to the filtered channel image obtained in step 03；
Step 05：The channel image that step 04 is obtained merges, and obtains the image after defogging.
Wherein rHRepresent high-frequency gain, rLLow-frequency gain is represented, D (u, v) represents that frequency (u, v) arrives filter center (uo,
vo) distance, DoCut-off frequency is represented, coefficient c is used to control sharpness, and size is between rHAnd rLBetween.It is improved new same
State filter transfer function is respectively to cut-off frequency DoThe part of both sides carries out low pass and high-pass filtering processing, to cut-off frequency in Do
Within part suppressed, that is, inhibit the stronger part of irradiation component；To cut-off frequency in DoPart above is carried out
Enhancing, that is, enhance the details of image.
Step 03：Foggy image I (x) is decomposed into tri- color channel images of R, G, B, each Color Channel is entered respectively
Homomorphic filtering after row improves is handled；
Specifically, tri- color channel images of R, G, B for being decomposed into of foggy image I (x) for I (:,:,1)、I(:,:,2)、I
(:,:, 3), be filtered respectively processing after result for I_filt (:,:,1)、I_filt(:,:,2)、I_filt(:,:,3).
Step 04：Multi-Scale Retinex Algorithm processing is carried out respectively to the filtered channel image obtained in step 03；
Step 05：The channel image that step 04 is obtained merges, and obtains the image after defogging.
Specifically, the expression formula of multi-Scale Retinex Algorithm is：
Wherein ri(x, y) represents the output in i-th of passage, Si(x, y) represents the foggy image of i-th of passage input, i
Value be 1,2,3.N represents the number of scale parameter, usually 3, it is divided into 3 small yardstick, mesoscale, large scale yardsticks ginsengs
Number, wherein scale parameter are set to 15,100,240 and allow multi-Scale Retinex Algorithm to have single scale Retinex calculations concurrently
The advantages of when method takes small, middle and high yardstick.ωnThe weight of each scale parameter is represented, and is metFn(x, y) is ωn
Corresponding n-th around function, * expression convolution.
Wherein, around function FnThe expression formula of (x, y) is：
Wherein λnIt is normalization factor, b represents scale parameter, ∫ ∫ Fn(x, y) dxdy=1.
After filtering process channel image I_filt (:,:,1)、I_filt(:,:,2)、I_filt(:,:, 3) and carry out more chis
Spend Retinex algorithm processing after result be I_msr (:,:,1)、I_msr(:,:,2)、I_msr(:,:, 3), finally by merging
After obtain defogging result I_msr=unit (I_msr).
Fig. 2 shows the mist elimination image after different several defogging algorithm process.(a), (e) have mist figure for original in Fig. 2
Picture, we carry out defogging using the method for traditional homomorphic filtering, multi-Scale Retinex Algorithm and the present invention to image respectively, its
In (b), (f) be using the result after traditional homomorphic filtering defogging, it can be seen that the residual of mist is excessive after defogging, picture contrast
Difference；Figure (c), (g) they are to use the result after multi-Scale Retinex Algorithm defogging, it can be seen that there is preferable defog effect, but it is special
Determine region such as white construction thing not obtaining strengthening defog effect well, the profile and details of scenery are still relatively fuzzyyer；Figure
(d), (h) is to use the result after the inventive method defogging, it can be seen that picture contrast obtains good improvement, the wheel of scenery
Wide and texture becomes apparent from, stereovision is clearly demarcated.
This specific embodiment is only explanation of the invention, and it is not limitation of the present invention, people in the art
Member can make the modification of no creative contribution to the present embodiment as needed after this specification is read, but as long as at this
All protected in the right of invention by Patent Law.
Claims (6)
1. it is a kind of based on the multiple dimensioned Retinex images defogging method for improving homomorphic filtering, it is characterized in that：Comprise the following steps：
Step 01：Input foggy image I (x)；
Step 02：Traditional homomorphic filtering transfer function H (u, v) is improved, traditional H (u, v) expression formula is：It is as follows to improve step：
a：Cut-off frequency D is seto；
b：To cut-off frequency in DoWithin part suppressed；
c：To cut-off frequency in DoPart above is strengthened；
d：Setting removes cut-off frequency DoOther specification numerical value in addition, obtain improved homomorphic filtering transmission function expression formula：
<mrow>
<mi>H</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>r</mi>
<mi>H</mi>
</msub>
<mo>-</mo>
<msub>
<mi>r</mi>
<mi>L</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>{</mo>
<mfrac>
<mn>1</mn>
<msup>
<mrow>
<mo>&lsqb;</mo>
<mn>1</mn>
<mo>+</mo>
<msub>
<mi>D</mi>
<mi>o</mi>
</msub>
<mo>/</mo>
<mi>c</mi>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
</mrow>
<mn>2</mn>
</msup>
</mfrac>
<mo>+</mo>
<mfrac>
<mn>1</mn>
<msup>
<mrow>
<mo>&lsqb;</mo>
<mn>1</mn>
<mo>+</mo>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>/</mo>
<msub>
<mi>D</mi>
<mi>o</mi>
</msub>
<mo>&rsqb;</mo>
</mrow>
<mn>2</mn>
</msup>
</mfrac>
<mo>}</mo>
<mo>+</mo>
<msub>
<mi>r</mi>
<mi>L</mi>
</msub>
</mrow>
Wherein rHRepresent high-frequency gain, rLLow-frequency gain is represented, and meets rH>1,rL<1, D (u, v) represents frequency (u, v) to filtering
Device center (uo,vo) distance, DoCut-off frequency is represented, coefficient c is used to control sharpness, and size is between rHAnd rLBetween；
Step 03：Foggy image I (x) is decomposed into tri- color channel images of R, G, B, each Color Channel is changed respectively
Homomorphic filtering processing after entering；
Step 04：Multi-Scale Retinex Algorithm processing is carried out respectively to the filtered channel image obtained in step 03；
Step 05：The channel image that step 04 is obtained merges, and obtains the image after defogging.
2. the method according to claim 11, it is characterized in that：Wherein, rH=1.5, rL=0.5, c=1, Do=4.
3. the method according to claim 11, it is characterized in that：Multi-Scale Retinex Algorithm expression formula is in step 04：
<mrow>
<msub>
<mi>r</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>n</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>N</mi>
</munderover>
<msub>
<mi>&omega;</mi>
<mi>n</mi>
</msub>
<mo>{</mo>
<mi>log</mi>
<mi> </mi>
<msub>
<mi>S</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mi>l</mi>
<mi>o</mi>
<mi>g</mi>
<mo>&lsqb;</mo>
<msub>
<mi>F</mi>
<mi>n</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>*</mo>
<msub>
<mi>S</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
<mo>}</mo>
</mrow>
Wherein ri(x, y) represents the output in i-th of passage, Si(x, y) represents the foggy image of i-th of passage input, and i's takes
It is worth for 1,2,3；N represents the number of scale parameter；ωnThe weight of each scale parameter is represented, and is metFn(x,y)
It is to surround function, * represents convolution.
4. the method according to claim 11, it is characterized in that：Around function FnThe expression formula of (x, y) is：
<mrow>
<msub>
<mi>F</mi>
<mi>n</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>&lambda;</mi>
<mi>n</mi>
</msub>
<msup>
<mi>e</mi>
<mrow>
<mo>-</mo>
<mfrac>
<mrow>
<mo>(</mo>
<msup>
<mi>x</mi>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<mi>y</mi>
<mn>2</mn>
</msup>
<mo>)</mo>
</mrow>
<mrow>
<msup>
<msub>
<mi>b</mi>
<mi>n</mi>
</msub>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
</mrow>
</msup>
</mrow>
Wherein λnIt is normalization factor, b represents scale parameter.
5. the method according to claim 11, it is characterized in that：∫∫Fn(x, y) dxdy=1.
6. the method according to claim 11, it is characterized in that：Wherein N value takes 3, the scale parameter is set to 15,
100、240。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710565491.8A CN107369144A (en) | 2017-07-12 | 2017-07-12 | Based on the multiple dimensioned Retinex images defogging method for improving homomorphic filtering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710565491.8A CN107369144A (en) | 2017-07-12 | 2017-07-12 | Based on the multiple dimensioned Retinex images defogging method for improving homomorphic filtering |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107369144A true CN107369144A (en) | 2017-11-21 |
Family
ID=60306911
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710565491.8A Pending CN107369144A (en) | 2017-07-12 | 2017-07-12 | Based on the multiple dimensioned Retinex images defogging method for improving homomorphic filtering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107369144A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108288258A (en) * | 2018-04-23 | 2018-07-17 | 电子科技大学 | A kind of low-quality images Enhancement Method under severe weather conditions |
| CN109753878A (en) * | 2018-12-06 | 2019-05-14 | 北京科技大学 | Imaging recognition methods and system under a kind of bad weather |
| CN109829862A (en) * | 2019-01-22 | 2019-05-31 | 三峡大学 | A kind of solar flare image based on multiple dimensioned Retinex removes cloud method |
| CN110276729A (en) * | 2019-06-10 | 2019-09-24 | 浙江工业大学 | A kind of Enhancement Method of low-luminance color image |
| CN111161360A (en) * | 2019-12-17 | 2020-05-15 | 天津大学 | Retinex theory-based image defogging method for end-to-end network |
| CN112613458A (en) * | 2020-12-29 | 2021-04-06 | 安徽创世科技股份有限公司 | Image preprocessing method and device for face recognition |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101783012A (en) * | 2010-04-06 | 2010-07-21 | 中南大学 | Automatic image defogging method based on dark primary colour |
| CN104200437A (en) * | 2014-09-04 | 2014-12-10 | 北京工业大学 | Image defogging method |
| CN105844601A (en) * | 2016-05-20 | 2016-08-10 | 中国矿业大学(北京) | Mine image enhancement method based on bilateral filtering and multi-scale Retinex algorithm |
| CN106355563A (en) * | 2016-08-31 | 2017-01-25 | 河南工业大学 | Image defogging method and device |
-
2017
- 2017-07-12 CN CN201710565491.8A patent/CN107369144A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101783012A (en) * | 2010-04-06 | 2010-07-21 | 中南大学 | Automatic image defogging method based on dark primary colour |
| CN104200437A (en) * | 2014-09-04 | 2014-12-10 | 北京工业大学 | Image defogging method |
| CN105844601A (en) * | 2016-05-20 | 2016-08-10 | 中国矿业大学(北京) | Mine image enhancement method based on bilateral filtering and multi-scale Retinex algorithm |
| CN106355563A (en) * | 2016-08-31 | 2017-01-25 | 河南工业大学 | Image defogging method and device |
Non-Patent Citations (2)
| Title |
|---|
| 周小军等: "基于改进同态滤波的遥感图像去云算法", 《无线电工程》 * |
| 汪秦峰等: "基于同态滤波和Retinex的图像去雾算法", 《火控雷达技术》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108288258A (en) * | 2018-04-23 | 2018-07-17 | 电子科技大学 | A kind of low-quality images Enhancement Method under severe weather conditions |
| CN108288258B (en) * | 2018-04-23 | 2021-08-10 | 电子科技大学 | Low-quality image enhancement method under severe weather condition |
| CN109753878A (en) * | 2018-12-06 | 2019-05-14 | 北京科技大学 | Imaging recognition methods and system under a kind of bad weather |
| CN109829862A (en) * | 2019-01-22 | 2019-05-31 | 三峡大学 | A kind of solar flare image based on multiple dimensioned Retinex removes cloud method |
| CN110276729A (en) * | 2019-06-10 | 2019-09-24 | 浙江工业大学 | A kind of Enhancement Method of low-luminance color image |
| CN110276729B (en) * | 2019-06-10 | 2021-12-17 | 浙江工业大学 | Low-illumination color image enhancement method |
| CN111161360A (en) * | 2019-12-17 | 2020-05-15 | 天津大学 | Retinex theory-based image defogging method for end-to-end network |
| CN111161360B (en) * | 2019-12-17 | 2023-04-28 | 天津大学 | Image defogging method of end-to-end network based on Retinex theory |
| CN112613458A (en) * | 2020-12-29 | 2021-04-06 | 安徽创世科技股份有限公司 | Image preprocessing method and device for face recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107369144A (en) | Based on the multiple dimensioned Retinex images defogging method for improving homomorphic filtering | |
| CN105225210B (en) | A kind of self-adapting histogram enhancing defogging method based on dark | |
| CN102802005B (en) | Method for 3d video content generation | |
| CN103942758B (en) | Dark channel prior image dehazing method based on multiscale fusion | |
| CN104200437B (en) | A kind of image defogging processing method | |
| CN108564549A (en) | A kind of image defogging method based on multiple dimensioned dense connection network | |
| CN103268598B (en) | Based on the low-light (level) low altitude remote sensing image Enhancement Method of Retinex theory | |
| CN104252700B (en) | A kind of histogram equalization method of infrared image | |
| CN106815821A (en) | The denoising method and device of near-infrared image | |
| CN105160647B (en) | A kind of panchromatic multispectral image fusion method | |
| CN103886565A (en) | Nighttime color image enhancement method based on purpose optimization and histogram equalization | |
| CN105046658A (en) | Low-illumination image processing method and device | |
| CN103955905A (en) | Rapid wavelet transformation and weighted image fusion single-image defogging method | |
| CN107358585A (en) | Misty Image Enhancement Method based on fractional order differential and dark primary priori | |
| CN111179189A (en) | Image processing method and device based on generation countermeasure network GAN, electronic equipment and storage medium | |
| CN104796582A (en) | Video image denoising and enhancing method and device based on random ejection retinex | |
| CN109523474A (en) | A kind of enhancement method of low-illumination image based on greasy weather degradation model | |
| CN115272303A (en) | Textile fabric defect degree evaluation method, device and system based on Gaussian blur | |
| CN111127340A (en) | Image defogging method | |
| CN112233019A (en) | ISP color interpolation method and device based on self-adaptive Gaussian kernel | |
| CN103955900A (en) | Image defogging method based on biological vision mechanism | |
| CN104463810B (en) | The adaptive diffusing filter image de-noising method flowed based on TV | |
| CN111179293B (en) | Bionic contour detection method based on color and gray level feature fusion | |
| CN102497576A (en) | Full-reference image quality assessment method based on mutual information of Gabor features (MIGF) | |
| CN106803236A (en) | Asymmetric correction method based on fuzzy field singular value decomposition |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20171121 |
|
| WD01 | Invention patent application deemed withdrawn after publication |