CN102750682A - Image preprocessing method for processing nonuniform illumination of miner face image and coal surface - Google Patents
Image preprocessing method for processing nonuniform illumination of miner face image and coal surface Download PDFInfo
- Publication number
- CN102750682A CN102750682A CN2012102461145A CN201210246114A CN102750682A CN 102750682 A CN102750682 A CN 102750682A CN 2012102461145 A CN2012102461145 A CN 2012102461145A CN 201210246114 A CN201210246114 A CN 201210246114A CN 102750682 A CN102750682 A CN 102750682A
- Authority
- CN
- China
- Prior art keywords
- image
- radius
- component
- frequency
- low
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000005286 illumination Methods 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000007781 pre-processing Methods 0.000 title claims abstract description 22
- 239000003245 coal Substances 0.000 title claims abstract description 20
- 238000012545 processing Methods 0.000 title claims abstract description 17
- 238000001914 filtration Methods 0.000 claims abstract description 59
- 230000005764 inhibitory process Effects 0.000 claims abstract description 10
- 230000001815 facial effect Effects 0.000 claims description 10
- 230000002708 enhancing effect Effects 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 5
- 230000000694 effects Effects 0.000 abstract description 16
- 230000002401 inhibitory effect Effects 0.000 abstract 1
- 238000013461 design Methods 0.000 description 7
- 230000003313 weakening effect Effects 0.000 description 7
- 238000012544 monitoring process Methods 0.000 description 5
- 238000012546 transfer Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000000354 decomposition reaction Methods 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 229920000742 Cotton Polymers 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003818 cinder Substances 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000005242 forging Methods 0.000 description 1
- 230000003760 hair shine Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000002703 mutagenesis Methods 0.000 description 1
- 231100000350 mutagenesis Toxicity 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Abstract
The invention discloses an image preprocessing method for processing nonuniform illumination of a miner face image and a coal surface. According to the image preprocessing method, by using a homomorphic filtering principle algorithm, illumination components are weakened and reflection components are enhanced, and meanwhile excessive inhibition to low frequency components is avoided. A homomorphic filtering function of the preprocessing method is used for separately processing the low frequency components and the high frequency components, and processing the low frequency components at different levels by segments, the inhibiting radius of the low frequency components is determined to be five pixel numbers according to the image filtering effect when the limiting frequency radius of the traditional low pass filter is five. The image preprocessing method has strong pertinence and remarkable effects to the processing of strong illumination influence in the miner face image and the coal surface.
Description
Technical field
The present invention relates to the image preprocessing technical field of coal mine underground monitoring system, is a kind of processing low irradiance borehole image uneven illumination, strengthens the image pre-processing method of picture contrast.
Background technology
In the coal mine underground monitoring system, because circumstance complication in the coal mine roadway, background is dark; Its illumination is mainly derived from illuminating lamp, and the light zone is very big with the backlight area luminance contrast, adds the factors such as intense flash lamp of ccd video camera; Cause image irradiation uneven, have a strong impact on the acquisition quality of image.To the conditions down-hole of complicacy, the images acquired in the coal mine underground monitoring system must just can be used for the analysis identification of image through sufficient pre-service.The image irradiation uneven phenomenon mainly comprises: the 1. whole low visibility of image capture environment, the light zone is very big with the backlight area luminance contrast, add that light source sends the illumination ray stronger, the demonstration of possible interfering picture characteristic; 2. illumination causes reflective, the high optical phenomenon of the local appearance of body surface in the image, causes integral image or local uneven illumination, has a strong impact on the acquisition quality of image.Therefore, to the image of uneven illumination, need carry out necessary pre-service and just can be used for Computer Analysis identification.
Handle at present the uneven image enhancement technique of image irradiation and comprise that mainly spatial domain Enhancement Method such as histogram equalization, gradient strengthen method; The frequency domain Enhancement Method is as based on the homomorphic filtering method of illumination-reflection model and based on Retinex method of illumination compensation etc., and the image enchancing method that homomorphic filtering method in addition and Retinex method combine with wavelet transformation also is suggested.Wherein histogram equalization has effect preferably for handling the big image of gray scale contrast, particularly can improve the overall gray value of the dark night illumination image of background, but not good for illumination and reflective high-brightness region treatment effect; To handling strong Gao Guang, the retroreflective regions of illumination effect effect is preferably arranged based on the homomorphic filtering method of illumination-reflection model and based on the Retinex method of illumination compensation, therefore receiving publicity more aspect the uneven illumination handling.
One Chinese patent application numbers 200910021318.7, an open day 2009.07.29 discloses a kind of method of forging image based on the detection of wavelet field homomorphic filtering.This invention is carried out one deck wavelet decomposition to the coloured image luminance component; And design wavelet field homomorphic filtering function carries out filtering; Coefficient of wavelet decomposition under the different resolution is carried out the processing of similar high-pass filtering, thereby utilize boundary operators to carry out the enhancing that Boundary Extraction realizes image boundary again.This method stresses image boundary and strengthens, and does not handle the uneven influence of image irradiation but be suitable for.
One Chinese patent application 201110274647.X, open day 2012.02.15 discloses the two-dimensional barcode image recognition methods on a kind of high reflective cylinder metal.This invention utilizes the gradient ramp of six order polynomial interpolation formula simulate for lateral pixels, and utilizes the interpolation formula of the Newton interpolating method calculating pixels across of equidistance node.Thereby the uneven illumination of cylinder correction is realized the location to the two-dimensional bar code zone.This invention is mainly used in the image recognition of the slight or moderate uneven illumination on the high reflective cylinder metal.
One Chinese patent application numbers 201110316982.1, an open day 2012.01.25 discloses cotton puppet foreign fiber colour-image reinforcing method and system a kind of uneven illumination under.This invention is carried out photo-irradiation treatment to the coloured image luminance component, adopts the gradiometer algorithm to obtain new luminance picture, realizes the gradient homogenising of luminance component, replaces original luminance component with new luminance component, thereby handles the uneven illumination of image.This invention belongs to spatial domain and strengthens method, compares with frequency domain enhancing method, and its computing method are complicated, and cycle index is more, and calculated amount is big.
Given this, the present invention in conjunction with homomorphic filtering principle algorithm, designs a kind of special in the miner's facial image in the processing coal mine underground monitoring system and the image pre-processing method of coal face uneven illumination from the frequency domain image enchancing method.Compare with traditional homographic filtering method, the present invention avoids the excessive inhibition to low frequency component in further refinement aspect the processing of irradiation component when suppressing illumination effect, stronger for the uneven illumination phenomenon specific aim of handling the low irradiance borehole image.
Summary of the invention
The invention provides the image pre-processing method of a kind of miner's of processing facial image and coal face uneven illumination.In image pre-processing method; Homomorphic filtering principle algorithm based on illumination-reflection model; Design special to weakening low irradiance bad border hypograph intense light irradiation and strong reflective influence; The realization picture contrast strengthens, and avoids handling the dark excessively homomorphic filtering function of back image simultaneously, thereby image is carried out pre-service.Compare with traditional homographic filtering method; The homomorphic filtering function that the present invention designed is directed against the photo-irradiation treatment of low irradiance borehole image especially; The low-frequency range frequency component is carried out segmentation processing in various degree, when weakening the irradiation component, avoid excessive inhibition as far as possible low frequency component in a big way.The present invention is for the low irradiance image of handling uneven illumination phenomenons such as the Gao Guang that caused by the strong illumination light source, reflective and light and shade contrast, realizes weakening illumination effect, strengthens picture contrast remarkable result is arranged.
Thinking of the present invention is the algorithm for image enhancement of utilization homomorphic filtering principle, handles the influence of the illumination ray of mine lamp lighting source and the direct projection of ccd video camera intense flash lamp, reflection to low irradiance down-hole collection image quality.In image pre-processing method,, coloured image is converted into gray level image and takes the logarithm (if original image is a gray level image according to the homomorphic filtering principle; Then directly take the logarithm), carry out the center Fourier transform again, obtain after the frequency-domain function; Design special to weakening low irradiance bad border hypograph intense light irradiation and strong reflective influence; The realization picture contrast strengthens, and avoids excessively suppressing the homomorphic filtering function of illumination component simultaneously, thereby image is carried out pre-service.For the image after when reducing illumination effect, avoiding handling is dark excessively, the homomorphic filtering function is designed to:
Wherein, (u v) is that the arbitrfary point is to the pixel distance of initial point in the frequency domain scope to D, and unit is a pixel count; D
O1For low frequency component suppresses radius (0＜D
O1≤15), unit is a pixel count; D
O2For high fdrequency component strengthens radius (D
O2>=15), unit is a pixel count; Alpha
LControl Radius D
O1Inhibition degree (0＜α with interior low frequency component
L＜1); Alpha
HControl Radius D
O2Enhancing degree (the α of above high fdrequency component
H＞0); Constant c scalable function shape (c＞0); N is filter order (best value is 2, to reduce ringing effect); At radius D
O1To D
O2Between the lower low-frequency range of frequency in, keep basically 1 times of initial value constant.For avoiding excessively suppressing the illumination component, the homomorphic filtering function of design is realized to low-frequency range component segmentation processing in various degree, to radius D
O1Low frequency component with interior suppresses, and radius D
O1To D
O2Between the low scope of frequency in low frequency component keep original value constant basically.For avoiding the excessive inhibition to wider low frequency component, radius D
O1Value is unsuitable excessive; Because when cutoff frequency point radius is 5; The image that conventional lowpass filter (like Butterworth low-pass filter, gauss low frequency filter) filtering obtains is very fuzzy, can't see any details basically, only stays the information of bright dark variation; Can be similar to the illumination contribution of regarding main as, so the low frequency of clear and definite homomorphic filtering function suppresses radius D
O1Best value be 5.Thereby the homomorphic filtering function that utilization is designed carries out photo-irradiation treatment.
The invention has the beneficial effects as follows:
1. with strong points.For low irradiance miner facial image and the coal face image of handling uneven illumination phenomenons such as the Gao Guang that causes by the strong illumination light source, reflective and light and shade contrast, realize weakening illumination effect, the enhancing picture contrast has remarkable result;
2. it is dark excessively to avoid handling the back image when handling uneven illumination.Low frequency component segmentation processing had in various degree both been realized the inhibition to the uneven illumination influence, and the image after having avoided again simultaneously handling is dark excessively;
3. the appropriateness to the illumination composition suppresses.Confirm D
O1Best value is 5, the low frequency component scope of handling is quantitatively estimated, to realize the appropriateness inhibition to the illumination composition;
4. controllability is strong.Low frequency, high fdrequency component are handled radius D respectively
O1Control low frequency process range, radius D
O2Control high frequency process range, thus controllability increased to irradiation component and reflecting component control.
Description of drawings
Below in conjunction with accompanying drawing and embodiment the present invention is further specified.
Fig. 1 is a homomorphic filtering implementation procedure in the image pre-processing method.
Fig. 2 is a homomorphic filtering transfer curve of the present invention.
Fig. 3 is the coal face original image that has local reflective phenomenon.
Fig. 4 is the mine workman's of head wearing mine light a original image.
Fig. 5 is that second order Butterworth LPF Cutoff radius is got the result images that Fig. 3 is carried out LPF at 5 o'clock.
Fig. 6 is that gauss low frequency filter ends radius and gets the result images that Fig. 3 is carried out LPF at 5 o'clock.
Fig. 7 utilizes the image pre-processing method that is designed Fig. 3 to be carried out the result images of photo-irradiation treatment.
Fig. 8 utilizes the image pre-processing method that is designed Fig. 4 to be carried out the result images of photo-irradiation treatment.
Wherein, Can find out from Fig. 5, Fig. 6; When cutoff frequency point radius is 5; Conventional lowpass filter Butterworth low-pass filter, gauss low frequency filter are respectively to the LPF that carries out embodiment illustrated in fig. 3, and the filtered that obtains the only light and shade of surplus image changes, and can be similar to and regard the main contribution of shining component as.In the employed homomorphic filtering function, parameter is provided with as follows among Fig. 7 and Fig. 8: n=2, c=1, D
O1=5, α
L=0.3, D
O2=15, α
H=4.
Embodiment
According to the homomorphic filtering implementation procedure among Fig. 1, by image processing software (like matlab software) embodiment Fig. 3 and Fig. 4 are carried out photo-irradiation treatment, embodiment is following:
(1) coloured image Fig. 3 and Fig. 4 are converted into gray level image, and the gray level image function is taken the logarithm, formula is following:
z(x，y)＝lnf(x，y)
Wherein (x y) is the gray level image function to f, and (x y) is the gained function of taking the logarithm to z;
(2) to function z (x y) carries out the center Fourier transform, obtain frequency-domain function Z (u, v);
(3) design homomorphic filtering function.The crucial part of Homomorphic Filtering Algorithm is that the homomorphic filtering transfer function H (control irradiation component and reflecting component respectively through transport function by u, selection v).The illumination composition of image irradiation component representative image, the primitive character when on behalf of it, reflecting component do not receive illumination effect.The irradiation component of considering image is changed to characteristic with spatial domain usually slowly; And reflecting component mutagenesis often; The low-frequency component of the Fourier transform after these characteristics cause image to be taken the logarithm interrelates with the irradiation component; And HFS and reflecting component interrelate, so can be similar to the control that is converted into image low frequency and high fdrequency component to the control of irradiation component and reflecting component.In order to suppress the influence of irradiation component as far as possible, highlight reflecting component simultaneously, to reach the alleviation uneven illumination, realize the purpose of figure image intensifying, consider to suppress the contribution of low frequency (illumination), and strengthen the contribution of high frequency (reflection).
1. shine the inhibition of component: suppress the irradiation component by traditional low-pass filter function.Because low-frequency component interrelates with the irradiation component,, need suppress the low-frequency range component in order to weaken the influence of irradiation component as far as possible.If low frequency inhibition scope is positioned at radius D
O1In, can fall radius D by low pass filter filters out
O1Frequency component in addition only keeps radius D
O1With interior low frequency part, it is reduced to handle.Because when cutoff frequency point radius is 5, conventional lowpass filter Butterworth low-pass filter, gauss low frequency filter are respectively to the LPF that carries out embodiment illustrated in fig. 3 again, and result images Fig. 5 and Fig. 6 of obtaining are very fuzzy; Basically can't see any details; Only stay the information of bright dark variation, can be similar to the illumination contribution of regarding main as, so consider mainly radius 5 to be suppressed with interior low-frequency component; Promptly on the basis of original frequency size, deduct the α of primary frequency component
LDoubly.
2. the enhancing of reflecting component: strengthen reflecting component by traditional high-pass filtering function.Primitive character when on behalf of it, the image reflection component do not receive illumination effect, the approximate contribution of HFS being regarded as reflecting component.In order to strengthen high fdrequency component, establish high frequency enhancing scope and be positioned at radius D
O2Frequency range in addition can filter out radius D by Hi-pass filter
O2With interior frequency component, only keep radius D
O2HFS in addition carries out enhancement process to it.High fdrequency component on the basis of original frequency size, is added original α
HDoubly, can reach the purpose that strengthens characteristics of image.In order to improve the contrast of image, reservation minutia as much as possible, filtering is by radius D
O2Unsuitable excessive, so that the high-frequency range that is kept is big as far as possible.
3. make up homomorphic filtering transfer function H (u; V): especially to the miner's facial image identification of head wearing mine light and the mine car monitoring and the coal cinder identification of coal face; In the image pre-processing method that is designed; The homomorphic filtering function is realized the low-frequency range frequency component has been carried out segmentation processing in various degree, and when strengthening image, it is dark excessively to avoid handling the back image.According to the homomorphic filtering principle, the contribution of irradiation component is positioned at low-frequency range, and the contribution of reflecting component is positioned at high-frequency range, so the homomorphic filtering function that is designed is to radius D
O1Low frequency component with interior suppresses, to radius D
O2Above high fdrequency component strengthens, and radius D
O1To D
O2Between the low scope of frequency in low frequency component keep original value constant basically, thereby realize reducing the intense light irradiation influence, improve picture contrast, avoid excessive inhibition simultaneously to low frequency component in a big way.
In image pre-processing method, the mentality of designing of homomorphic filtering function is low with Butterworth, the high-pass filtering function combines, and designs following homomorphic filtering function:
Wherein, (u v) is that the arbitrfary point is to the distance of initial point in the frequency domain scope to D, and unit is a pixel count; D
O1For low frequency component suppresses radius (0＜D
O1≤15), unit is a pixel count; D
O2For high fdrequency component strengthens radius (D
O2>=15), unit is a pixel count; Alpha
LControl Radius D
O1Inhibition degree (0＜α with interior low frequency component
L＜1); Alpha
HControl Radius D
O2Enhancing degree (the α of above high fdrequency component
H＞0); Constant c scalable function shape (c＞0); N is filter order (best value is 2, to reduce ringing effect); At radius D
O1To D
O2Between the lower low-frequency range of frequency in, keep basically 1 times of initial value constant.
The transfer function H that is designed (u, v) in, the low frequency component of image reduces α on original basis
LDoubly, the high fdrequency component of image increases α on original basis
HDoubly.
(4) the homomorphic filtering function parameter in the step (3) is carried out assignment, and embodiment Fig. 3, Fig. 4 are carried out homomorphic filtering at the frequency-domain function that step (2) obtains.
Analysis according in the step (3) the irradiation component being suppressed obtains: homomorphic filtering function low frequency suppresses radius D
O1Get 5 the bests.Require to choose other parameters in the homomorphic filtering function according to the illumination characteristic of original image with to treatment of picture.Wherein, 0＜α
L＜1, α
LValue is big more, and image irradiation weakening degree is big more, and the image after the processing is dark more, therefore shines darker image, α for integral light
LValue should be less than 0.5, to reduce the weakening to overall brightness; α
HValue is big more, and it is big more that picture contrast strengthens degree, α
HMore little, the enhancing degree is not obvious more, usually α
HValue is more preferably greater than 1.High frequency strengthens radius D
O2Decide according to the processing needs, can select with reference to the filter effect of Butterworth Hi-pass filter.If need to strengthen the integral image contrast, show the more details characteristic, radius D
O2Should choose smaller value, enlarging the scope of the high fdrequency component strengthened, otherwise, change place greatly, radius D if only want to strengthen shade of gray such as image boundary
O2Should choose higher value.Filter order n should get 2, to reduce ringing effect.Coefficient c＞0, it selects that the shape of homomorphic filter function is had certain influence, should get 0 to 5 with interior number.
According to above analysis, n=2 is set, c=1, D
O1=5, α
L=0.3, D
O2=15, α
H=4, embodiment Fig. 3, Fig. 4 are carried out homomorphic filtering at the frequency-domain function that step (2) obtains: with frequency-domain function Z (u, v) with the homomorphic filtering transfer function H that is designed (u v) multiplies each other, and formula is following:
S(u，v)＝Z(u，v)*H(u，v)；
Wherein, (u, v) for through the frequency-domain function of Fourier transform, (u v) is the homomorphic filtering transport function to H to Z, S (u, the Fourier transform of the image of v) exporting for filtering.
(5) the homomorphic filtering result is carried out the center inverse fourier transform, obtain function s (x, y);
(6) step (5) gained result is got index, obtain result images Fig. 7 and Fig. 8 after the homomorphic filtering, wherein Fig. 7 is the homomorphic filtering result of Fig. 3, and Fig. 8 is the homomorphic filtering result of Fig. 4.Formula is following:
g(x，y)＝e
s(x，y)，
Wherein (x y) is step (5) gained result to s, and (x y) is the homomorphic filtering result images to g.
Claims (5)
1. an image pre-processing method of handling miner's facial image and coal face uneven illumination is characterized in that comprising the steps:
(1) if original image is a coloured image, then be translated into gray level image earlier and take the logarithm again, if original image is a gray level image, then directly take the logarithm;
(2) carry out the center Fourier transform, obtain frequency-domain function;
(3) make up the homomorphic filtering function, carry out homomorphic filtering, weaken intense light irradiation and strong reflective influence under the bad border of low irradiance, strengthen picture contrast, avoid excessively suppressing the irradiation component simultaneously image;
(4) center inverse fourier transform and get index obtains result images.
2. a kind of image pre-processing method of handling miner's facial image and coal face uneven illumination according to claim 1, it is characterized in that: the homomorphic filtering function does in this preprocess method
Wherein, (u v) is that the arbitrfary point is to the distance of initial point in the frequency domain scope to D, and unit is a pixel count; D
O1For low frequency component suppresses radius, 0＜D
O1≤15, unit is a pixel count; D
O2For high fdrequency component strengthens radius, D
O2>=15, unit is a pixel count; Alpha
LControl Radius D
O1With the inhibition degree of interior low frequency component, 0＜α
L＜1; Alpha
HControl Radius D
O2The enhancing degree of above high fdrequency component, α
H＞0; Constant c scalable function shape, c＞0; N is a filter order, n=2; At radius D
O1To D
O2Between the lower low-frequency range of frequency in, keep 1 times of initial value constant.
3. a kind of image pre-processing method of handling miner's facial image and coal face uneven illumination according to claim 2 is characterized in that: the homomorphic filtering function is realized to low-frequency range component segmentation processing in various degree, to radius D
O1Low frequency component with interior suppresses, and radius D
O1To D
O2Between the low scope of frequency in low frequency component keep original value constant.
4. a kind of image pre-processing method of handling miner's facial image and coal face uneven illumination according to claim 3 is characterized in that: the low frequency of homomorphic filtering function suppresses radius D
O1Value be 5.
5. a kind of image pre-processing method of handling miner's facial image and coal face uneven illumination according to claim 1; It is characterized in that: the homomorphic filtering function is handled respectively low frequency, high fdrequency component in this preprocess method; Increased controllability, radius D to irradiation component and reflecting component control
O1Control low frequency process range, radius D
O2Control high frequency process range, and at radius D
O1To D
O2Between the low-frequency range component keep former letter constant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210246114.5A CN102750682B (en) | 2012-07-17 | 2012-07-17 | A kind of image pre-processing method processing miner face image and coal face uneven illumination |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210246114.5A CN102750682B (en) | 2012-07-17 | 2012-07-17 | A kind of image pre-processing method processing miner face image and coal face uneven illumination |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102750682A true CN102750682A (en) | 2012-10-24 |
| CN102750682B CN102750682B (en) | 2016-01-27 |
Family
ID=47030832
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210246114.5A Expired - Fee Related CN102750682B (en) | 2012-07-17 | 2012-07-17 | A kind of image pre-processing method processing miner face image and coal face uneven illumination |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102750682B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106485674A (en) * | 2016-09-20 | 2017-03-08 | 天津大学 | A kind of low light image Enhancement Method based on integration technology |
| CN108564923A (en) * | 2018-01-30 | 2018-09-21 | 武汉华星光电技术有限公司 | High dynamic contrast image display method and apparatus based on subregion backlight |
| CN109816609A (en) * | 2019-01-31 | 2019-05-28 | 领航基因科技（杭州）有限公司 | A kind of digital pcr image restoring method and its application based on Fourier transformation |
| CN110197487A (en) * | 2019-03-06 | 2019-09-03 | 上海商路网络科技有限公司 | A method of color spot is detected based on skin image |
| CN110276729A (en) * | 2019-06-10 | 2019-09-24 | 浙江工业大学 | A kind of Enhancement Method of low-luminance color image |
| CN110874823A (en) * | 2018-09-03 | 2020-03-10 | 中国矿业大学(北京) | Mine fog image enhancement method based on dark primary color prior and homomorphic filtering |
| 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 |
| CN112651350A (en) * | 2020-12-29 | 2021-04-13 | 安徽创世科技股份有限公司 | Video processing method and device |
| CN115240240A (en) * | 2022-04-29 | 2022-10-25 | 清远蓄能发电有限公司 | Infrared face recognition method and system based on YOLO network |
| CN115359044A (en) * | 2022-10-18 | 2022-11-18 | 南通群富金属科技有限公司 | Metal part surface scratch detection method based on image enhancement |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030161549A1 (en) * | 2002-02-26 | 2003-08-28 | Zhichun Lei | Contrast enhancement of digital images |
| CN101488218A (en) * | 2008-12-19 | 2009-07-22 | 四川虹微技术有限公司 | Image self-adapting airspace homographic filtering method |
-
2012
- 2012-07-17 CN CN201210246114.5A patent/CN102750682B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030161549A1 (en) * | 2002-02-26 | 2003-08-28 | Zhichun Lei | Contrast enhancement of digital images |
| CN101488218A (en) * | 2008-12-19 | 2009-07-22 | 四川虹微技术有限公司 | Image self-adapting airspace homographic filtering method |
Non-Patent Citations (1)
| Title |
|---|
| 罗海霞等: "同态滤波在光照补偿中的应用", 《湖南工业大学学报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106485674A (en) * | 2016-09-20 | 2017-03-08 | 天津大学 | A kind of low light image Enhancement Method based on integration technology |
| CN108564923A (en) * | 2018-01-30 | 2018-09-21 | 武汉华星光电技术有限公司 | High dynamic contrast image display method and apparatus based on subregion backlight |
| CN110874823A (en) * | 2018-09-03 | 2020-03-10 | 中国矿业大学(北京) | Mine fog image enhancement method based on dark primary color prior and homomorphic filtering |
| CN109816609A (en) * | 2019-01-31 | 2019-05-28 | 领航基因科技（杭州）有限公司 | A kind of digital pcr image restoring method and its application based on Fourier transformation |
| CN109816609B (en) * | 2019-01-31 | 2021-12-21 | 领航基因科技（杭州）有限公司 | Fourier transform-based digital PCR image reduction method and application thereof |
| CN110197487A (en) * | 2019-03-06 | 2019-09-03 | 上海商路网络科技有限公司 | A method of color spot is detected based on skin image |
| CN110276729A (en) * | 2019-06-10 | 2019-09-24 | 浙江工业大学 | A kind of Enhancement Method of low-luminance color image |
| 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 |
| CN112651350A (en) * | 2020-12-29 | 2021-04-13 | 安徽创世科技股份有限公司 | Video processing method and device |
| CN115240240A (en) * | 2022-04-29 | 2022-10-25 | 清远蓄能发电有限公司 | Infrared face recognition method and system based on YOLO network |
| CN115359044A (en) * | 2022-10-18 | 2022-11-18 | 南通群富金属科技有限公司 | Metal part surface scratch detection method based on image enhancement |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102750682B (en) | 2016-01-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102750682A (en) | Image preprocessing method for processing nonuniform illumination of miner face image and coal surface | |
| CN107153816B (en) | Data enhancement method for robust face recognition | |
| CN108734670B (en) | Method for restoring single night weak-illumination haze image | |
| CN103020920B (en) | Method for enhancing low-illumination images | |
| CN101783963B (en) | Nighttime image enhancing method with highlight inhibition | |
| CN102789635B (en) | Image enhancement method and image enhancement device | |
| CN105488793B (en) | Image display method and image processing method | |
| CN104268843A (en) | Image self-adaptation enhancing method based on histogram modification | |
| CN102289792A (en) | Method and system for enhancing low-illumination video image | |
| CN103034986A (en) | Night vision image enhancement method based on exposure fusion | |
| CN105427257A (en) | Image enhancement method and apparatus | |
| CN108288258A (en) | A kind of low-quality images Enhancement Method under severe weather conditions | |
| CN109272461A (en) | Infrared image enhancing method based on median filtering and color histogram | |
| CN105205794A (en) | Synchronous enhancement de-noising method of low-illumination image | |
| CN104463804A (en) | Image enhancement method based on intuitional fuzzy set | |
| CN112184646B (en) | Image fusion method based on gradient domain oriented filtering and improved PCNN | |
| CN102999890B (en) | Based on the image light dynamic changes of strength bearing calibration of environmental factor | |
| CN115660964A (en) | Nighttime road image enhancement method based on threshold partition weighted brightness component | |
| CN109584182A (en) | A kind of image processing method and system | |
| CN103489168A (en) | Enhancing method and system for infrared image being converted to pseudo color image in self-adaptive mode | |
| Mandal et al. | Real-time automotive night-vision system for drivers to inhibit headlight glare of the oncoming vehicles and enhance road visibility | |
| CN105426847A (en) | Nonlinear enhancing method for low-quality natural light iris images | |
| CN111311503A (en) | Night low-brightness image enhancement system | |
| Zhang et al. | Infrared and visible image fusion based on contrast enhancement guided filter and infrared feature decomposition | |
| CN116823686B (en) | Night infrared and visible light image fusion method based on image enhancement |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C53 | Correction of patent of invention or patent application | ||
| CB03 | Change of inventor or designer information |
Inventor after: Liu Xiaoyang Inventor after: Cheng Menglin Inventor after: Tian Zijian Inventor after: Jia Xufen Inventor before: Tian Zijian Inventor before: Jia Xufen |
|
| COR | Change of bibliographic data |
Free format text: CORRECT: INVENTOR; FROM: TIAN ZIJIAN JIA XUFEN TO: LIU XIAOYANG CHENG MENGLIN TIAN ZIJIAN JIA XUFEN |
|
| CB03 | Change of inventor or designer information |
Inventor after: Liu Xiaoyang Inventor after: Cheng Menglin Inventor after: Wang Wanlin Inventor after: Tian Zijian Inventor after: Jia Xufen Inventor before: Liu Xiaoyang Inventor before: Cheng Menglin Inventor before: Tian Zijian Inventor before: Jia Xufen |
|
| COR | Change of bibliographic data | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160127 Termination date: 20200717 |