CN109523472B - Retinex color image enhancement method and computer vision processing system - Google Patents
Retinex color image enhancement method and computer vision processing system Download PDFInfo
- Publication number
- CN109523472B CN109523472B CN201811188303.5A CN201811188303A CN109523472B CN 109523472 B CN109523472 B CN 109523472B CN 201811188303 A CN201811188303 A CN 201811188303A CN 109523472 B CN109523472 B CN 109523472B
- Authority
- CN
- China
- Prior art keywords
- image
- mean value
- mid
- value
- enhancement
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000012545 processing Methods 0.000 title claims abstract description 8
- 230000000694 effects Effects 0.000 claims description 31
- 230000002708 enhancing effect Effects 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 abstract description 24
- 238000011156 evaluation Methods 0.000 abstract description 3
- 230000000007 visual effect Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 9
- 230000006872 improvement Effects 0.000 description 5
- 241000282414 Homo sapiens Species 0.000 description 3
- 238000009795 derivation Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000012938 design process Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 238000013441 quality evaluation Methods 0.000 description 2
- 230000003920 cognitive function Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 125000001475 halogen functional group Chemical group 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000002207 retinal effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
-
- 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
- G06T2207/30168—Image quality inspection
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the technical field of image enhancement or restoration, and discloses a mean value adjustable Retinex color image enhancement method and a computer vision processing system; firstly, dividing pixel points of a reflection image r (x, y) into two intervals [ min mid ] and [ mid max ]; secondly, calculating an image gray level d pre-stretched by a median [ mid ] of a reflection image r (x, y) by setting a pre-achieved image enhancement mean value m; and finally, a piecewise linear stretching method is adopted to realize different stretching of the image r (x, y) in two intervals, realize the enhancement of the image and enable the mean value of the enhanced image to reach a preset value. The invention effectively solves the problem of image quality uncertainty of mean value evaluation by setting a pre-achieved enhanced image mean value and adopting a Retinex algorithm and a segmented linear stretching color image enhancement algorithm, and an application sets a proper mean value according to requirements to obtain a required enhanced image.
Description
Technical Field
The invention belongs to the technical field of image enhancement or restoration, and particularly relates to a mean value adjustable Retinex color image enhancement method and a computer vision processing system.
Background
Currently, the current state of the art commonly used in the industry is such that: the image enhancement is an image processing method which changes an original unclear image into clear or emphasizes certain interesting features, inhibits the uninteresting features, improves the image quality, enriches the information quantity and strengthens the image interpretation and identification effects. In recent years, various color image enhancement algorithms are proposed at home and abroad, a large number of documents on color image enhancement are published, and certain effects are achieved, so that the color image enhancement algorithms are widely applied to image enhancement such as rainy images, haze images, night images, bright light halo images and the like. The color image enhancement algorithm is summarized as follows: a histogram modification based color image enhancement method; enhancing a color image based on modes of image filtering, sharpening and the like; color image enhancement based on visual effects. Among them, color image enhancement algorithms based on visual effect are the hot spot of the current color image enhancement research. Color constancy is a typical visual effect-based color image enhancement method. Color constancy is an important cognitive function of human vision, and enables human beings to obtain stable perception of object colors regardless of changes in ambient light. Color constancy based on computer vision can be defined as the computer vision system automatically obtaining a stable description of the color of an image object under unknown ambient lighting conditions. The prior art, namely the retinal cortex Theory (Retinex Theory), is the most influential color constant perception calculation Theory. For the Retinex algorithm, researchers propose many improved algorithms, which can be summarized as: retinex improvement based on scale parameters, retinex improvement based on stretch functions, improvement based on a Retinex model framework, and Retinex improvement based on a center-surround function. In general, researchers use indicators such as image mean, variance, softness, etc. to evaluate enhanced image quality. The image mean value is the most common and important measurement index for enhancing image quality evaluation, and conventionally, the image mean value is counted from the result of enhancing the image R (x, y), and the effectiveness of an image enhancement algorithm is evaluated (the image mean value is preferably in the level of [100150 ]). At present, researchers only use the image mean value as an index for evaluating an enhancement algorithm, but do not purposely design the algorithm according to the image mean value, so that the uncertainty of the image mean value is caused, namely the image mean value cannot effectively evaluate the image enhancement effect.
In summary, the problems of the prior art are as follows: at present, in the image enhancement process, the image mean value is the image mean value counted from the result of image enhancement, and is an important performance index for evaluating the image enhancement algorithm, but because the image mean value parameter is not considered in the enhancement algorithm design process, namely the influence problem of the image pixel value on the enhanced image mean value is not considered, the enhanced image mean value has certain randomness, and therefore, the image mean value cannot effectively measure the image enhancement quality.
The difficulty and significance for solving the technical problems are as follows:
as an important performance index for evaluating the image enhancement quality, the image mean evaluation index should have certain certainty. That is, in the design process of the enhancement algorithm, a researcher should consider the influence of the pixel values of the original image on the mean value of the enhanced image, that is, the researcher should consider the distribution of the pixel values of the original image to design the enhancement algorithm. The mean value adjustable Retinex color image enhancement algorithm fully considers the distribution condition of the pixel values of the original image, and realizes the enhancement of the image by setting the mean value which is pre-reached by the enhanced image according to the certainty of the mean value index of the image. Because the size of the enhanced image mean value can influence human visual induction, in the invention, people can set a proper mean value according to visual effect requirements to achieve the purpose of improving image quality effect, that is, in the image enhancement process, an user can set different mean values according to the image enhanced visual effect requirements to select a proper mean value to achieve the enhancement effect requirements, and the method has good application value in the aspects of military affairs, security protection, visual monitoring and the like.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a mean value adjustable Retinex color image enhancement method and a computer vision processing system.
The implementation process of the invention is shown in fig. 1, and comprises the following steps: firstly, carrying out multi-scale Gaussian filtering on an original color image; secondly, expressing the reflection image r (x, y) by adopting a logarithmic model, and dividing the median of the pixel values of the image r (x, y) into two intervals by taking the median of the pixel values of the image r (x, y) as an interval point; then setting a pre-achieved image enhancement mean value, and calculating the image gray level pre-stretched by the mean value of the reflection image r (x, y); and finally, stretching the image r (x, y) by adopting piecewise linear stretching to realize image enhancement.
Further, the piecewise linear stretching specifically comprises: dividing pixel points of the image r (x, y) into two intervals [ min mid ] and [ mid max ] by taking the median min of the pixel values of the reflected image r (x, y) as a separation point, wherein min is the minimum value of the image r (x, y), mid is the median of the image r (x, y), and max is the maximum value of the image r (x, y). Secondly, an image gray level d pre-stretched to the median of the image r (x, y) is calculated by setting a pre-enhancement image mean m. And finally, performing linear stretching enhancement on the pixel points in two intervals of the image r (x, y) by adopting segmented linear stretching.
Further, the enhancement image mean value is adjustable and specifically comprises: by setting different pre-achieved enhanced image mean values (mean values are adjustable), and respectively calculating the image gray level d pre-stretched by the median [ mid ] of the reflection image r (x, y), the image enhancement effect of different mean values is realized.
Further, the piecewise linear stretching calculation formula is as follows:
further, the mean value m of the enhanced image R (x, y) of the mean value adjustable Retinex color image enhancement method is in a linear relation with the parameter d; the median mid stretched gray value d in image R (x, y) can be calculated from the mean value m of image R (x, y):
another objective of the present invention is to provide a computer vision processing system using the method for enhancing a color image with tunable mean value.
In summary, the advantages and positive effects of the invention are as follows: the image mean value is an important index for measuring the quality of the enhanced image. In the invention, a pre-achieved enhanced image mean value is set, and the Retinex algorithm is adopted to perform piecewise linear stretching, so that the color image enhancement is realized, the problem of uncertainty of image quality evaluation effect of the enhanced image mean value can be effectively solved, and the enhancement requirement of an application on the visual effect requirement of the enhanced image mean value is met.
Drawings
Fig. 1 is a flowchart of a method for enhancing a mean-value-adjustable Retinex color image according to an embodiment of the present invention.
Fig. 2 is a flowchart of an implementation of the method for enhancing a reference-value-adjustable Retinex color image according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a piecewise linear stretching process provided by the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The mean value adjustable Retinex color image enhancement algorithm is an improved algorithm based on the Retinex algorithm, and the image enhancement algorithm is optimized by setting the mean value pre-reached by an enhanced image, so that the color image is enhanced. In the invention, the image mean value is an index for evaluating the image enhancement quality and is a parameter for representing the image enhancement effect, and the required color image enhancement effect can be obtained by setting one image mean value.
The following detailed description of the principles of the invention is provided in connection with the accompanying drawings.
As shown in fig. 2, the method for enhancing a mean-value-adjustable Retinex color image according to an embodiment of the present invention includes the following steps:
s101: an original color image R (red component), G (green component) and B (blue component) are respectively subjected to multi-scale Gaussian filtering (in the experimental process, high, middle and low scales are included), and an incident image is obtained;
s102: a logarithmic model to represent the original image and the incident image;
s103: subtracting the incident image from the original image to obtain a reflection image r (x, y);
s104: division of the reflected image r (x, y) into intervals: the pixel points of the reflection image r (x, y) are divided into two intervals [ min mid ] and [ mid max ], wherein min is the minimum value of the image r (x, y), mid is the median value of the image r (x, y), and max is the maximum value of the image r (x, y);
s105: setting a pre-reached mean value m of an enhanced image R (x, y), wherein the mean value m =127.5 can be pre-set in the experimental process;
s106: calculating the gray value d to which the median value of the image r (x, y) is pre-stretched according to the formula (5), and particularly in the derivation process;
s107: according to a formula (1), performing linear stretching on image pixel values in two intervals of an image r (x, y) by adopting a segmented linear stretching mode to obtain an enhanced color image;
s108: depending on the user's visual needs for enhanced color images, a mean value m may be reset (if the visual effect is perceived to be darker, the newly set mean value is greater than the previously set mean value, and if the visual effect is more brighter, the newly set mean value is less than the previously set mean value), and the enhancement effect is shown in fig. 3.
S109: steps S105-S108 can be designed as a computer vision processing system, wherein the predicted mean value m of the enhanced image R (x, y) is a parameter, and the user can set a reasonable parameter m according to the requirement of the visual effect of the enhanced image to obtain the desired enhanced color image.
The invention adopts the sectional linear stretching to respectively perform linear stretching enhancement on pixel points in two sections of an image r (x, y), as shown in formula (1), and a schematic diagram of the sectional linear stretching is shown in figure 3.
And (3) derivation process:
defining: the r (x, y) pixel value of the reflection image is located at [ min mid []The number of the pixels in the interval is N 1 Average value of m 1 (ii) a At [ mid max]The number of the pixels in the interval is N 2 Average value of m 2 。
Inference 1: from equation (1), the image R (x, y) is in the interval [0d ] after the piecewise linear stretching]Is then m 1 ', in [ d 255]The average value of the interval is m 2 ’:
Inference 2: the mean value of the image R (x, y) is m
From the formula (4), the mean value m of the enhanced image R (x, y) is in a linear relationship with the parameter d. Therefore, the median mid-stretched gray value d in the image R (x, y) can be calculated from the mean value m of the image R (x, y).
The effect of the present invention will be described in detail with reference to experiments.
In the experimental process, the algorithm is adopted to carry out enhancement verification on the low-contrast image, and experiments show that the method is an effective method for enhancing the original image by setting the mean value of the enhanced image and adopting a piecewise linear stretching mode. Wherein the preset range of the mean gray value of the enhanced image is [80150].
(1) The softness of an original image is poor, and the color information is not rich; the enhancement effect with the mean value m =80 is rich in color information, and the softness and comfort of the image are achieved; with the increase of the preset average value m, the enhancement effect of setting the average value m =100, the enhancement effect of setting the average value m =110, the enhancement effect of setting the average value m =127.5 and the enhancement effect of setting the average value m =150 are more and more poor in color saturation, the image softness is gradually reduced, and the image quality is reduced.
(2) The contrast of the original image is low, and the definition is poor; the contrast of the enhanced effect image with the mean value m =80 is increased, the definition is improved, and partial detail information in the image is still fuzzy; the image contrast of the enhancement effect image with the set mean value m =100 is obvious, the definition is good, and the detail information is rich; as the preset mean value m increases, the image content of the enhancement effect image with the set mean value m =110, the enhancement effect image with the set mean value m =127.5 and the enhancement effect image with the set mean value m =150 is enhanced in definition, but the effect of the 'white cloud part' in the image is distorted.
(3) The contrast of the original image is low, and the definition is poor; the contrast of the content of the image part is poor and cannot be identified due to the small setting value of the mean value m of the enhancement effect graph with the mean value m =80, the enhancement effect graph with the mean value m =100 and the enhancement effect graph with the mean value m = 110; as the mean m setting increases; the enhancement effect image content of the mean value m =127.5, near and far, has good contrast and image recognition; in the enhancement effect map with the mean value m =150, the contrast of the near content is relatively good and the image recognition degree is relatively high due to the excessively large mean value m, but the enhancement effect of the far content is reduced.
From (1), (2), (3) we can conclude that: 1) The image mean value is an important index for measuring the image enhancement quality, but the evaluation effect has uncertainty; 2) The mean value adjustable Retinex color image enhancement algorithm can effectively evaluate the image quality by setting a pre-achieved enhanced image mean value m; 3) The mean value adjustable Retinex color image enhancement algorithm can meet the image enhancement quality effect required by different users by purposefully and properly setting the pre-achieved enhanced image mean value m.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (3)
1. A method for enhancing a Retinex color image with a tunable mean value is characterized in that the method for enhancing the Retinex color image with the tunable mean value comprises the following steps: the original color image is subjected to Gaussian filtering, and a reflection image r (x, y) is represented by a logarithmic model; stretching an image r (x, y) by setting a pre-reached image enhancement mean value m and adopting piecewise linear stretching to obtain an enhanced image;
the piecewise linear stretching specifically comprises: dividing pixel points of a reflection image r (x, y) into two intervals [ min mid ] and [ mid max ]; calculating a gray value d to which a median mid is pre-stretched according to a preset enhanced image mean value m; pixel point values in different intervals in the image r (x, y) are respectively subjected to linear stretching; the piecewise linear stretching calculation formula is as follows:
where min is the minimum value of image r (x, y), mid is the median value of image r (x, y), max is the maximum value of image r (x, y), and d is the grayscale value to which mid is pre-stretched in image r (x, y);
the mean value m of the enhanced image R (x, y) is in a straight line relationship with the parameter d, i.e. the gray value d after stretching the median mid of the reflection image R (x, y) can be calculated from the mean value m of the enhanced image R (x, y):
where min is the minimum value of image r (x, y), mid is the median value of image r (x, y), max is the maximum value of image r (x, y), m is 1 Is the r (x, y) interval [ min mid) of the image]Mean value of pixel values, N 1 Is the image r (x, y) interval [ min mid]Total number of pixels, m 2 Is the image r (x, y) interval [ mid max]Mean value of pixel values, N 2 Is the image r (x, y) interval [ mid max]And (4) the total number of pixel points.
2. The method of claim 1, wherein the step of improving the mean-value-adjustable Retinex color image specifically comprises: calculating an image gray value d pre-stretched by a median [ mid ] of an image r (x, y) by setting a pre-achieved enhanced image mean value m, and realizing the enhancement of the image r (x, y) by adopting segmented linear stretching; different enhancement effects are realized by presetting different mean values m, and the required enhancement requirements are obtained.
3. A computer vision processing system using the method for enhancing a color image with a tunable mean value Retinex according to any one of claims 1-2.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811188303.5A CN109523472B (en) | 2018-10-12 | 2018-10-12 | Retinex color image enhancement method and computer vision processing system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811188303.5A CN109523472B (en) | 2018-10-12 | 2018-10-12 | Retinex color image enhancement method and computer vision processing system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109523472A CN109523472A (en) | 2019-03-26 |
CN109523472B true CN109523472B (en) | 2023-04-14 |
Family
ID=65771901
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811188303.5A Active CN109523472B (en) | 2018-10-12 | 2018-10-12 | Retinex color image enhancement method and computer vision processing system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109523472B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109978794B (en) * | 2019-03-29 | 2021-03-23 | 中山爱瑞科技有限公司 | Method and system for processing mammary gland dual-energy image |
CN110570381B (en) * | 2019-09-17 | 2022-04-29 | 合肥工业大学 | Semi-decoupling image decomposition dark light image enhancement method based on Gaussian total variation |
CN112381736A (en) * | 2020-11-17 | 2021-02-19 | 深圳市歌华智能科技有限公司 | Image enhancement method based on scene block |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102682436A (en) * | 2012-05-14 | 2012-09-19 | 陈军 | Image enhancement method on basis of improved multi-scale Retinex theory |
CN105096273A (en) * | 2015-08-20 | 2015-11-25 | 北京邮电大学 | Automatic adjustment method of color image brightness |
CN106846282A (en) * | 2017-03-28 | 2017-06-13 | 华侨大学 | A kind of enhancement method of low-illumination image of use adaptively correcting |
CN107133937A (en) * | 2017-04-27 | 2017-09-05 | 北京环境特性研究所 | A kind of self-adapting enhancement method of infrared image |
CN108062746A (en) * | 2016-11-09 | 2018-05-22 | 深圳市优朋普乐传媒发展有限公司 | A kind of method of video image processing and device, video coding system |
-
2018
- 2018-10-12 CN CN201811188303.5A patent/CN109523472B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102682436A (en) * | 2012-05-14 | 2012-09-19 | 陈军 | Image enhancement method on basis of improved multi-scale Retinex theory |
CN105096273A (en) * | 2015-08-20 | 2015-11-25 | 北京邮电大学 | Automatic adjustment method of color image brightness |
CN108062746A (en) * | 2016-11-09 | 2018-05-22 | 深圳市优朋普乐传媒发展有限公司 | A kind of method of video image processing and device, video coding system |
CN106846282A (en) * | 2017-03-28 | 2017-06-13 | 华侨大学 | A kind of enhancement method of low-illumination image of use adaptively correcting |
CN107133937A (en) * | 2017-04-27 | 2017-09-05 | 北京环境特性研究所 | A kind of self-adapting enhancement method of infrared image |
Non-Patent Citations (3)
Title |
---|
An Improved Multi-scale Retinex Fog and Haze Image Enhancement Method;Tian He Yu et al.;《2016 International Conference on Information System and Artificial Intelligence》;20100116;第557-560页 * |
Image Enhancement Using Piecewise Linear Contrast Stretch Methods based on Unsharp Masking Algorithms for Leather Image Processing;Murinto et al.;《2017 3rd International Conference on Science in Information Technology》;20180115;第669-673页 * |
低照度夜视成像的自然感彩色化及增强方法;朱进等;《光子学报》;20180430;第165-174页 * |
Also Published As
Publication number | Publication date |
---|---|
CN109523472A (en) | 2019-03-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109523472B (en) | Retinex color image enhancement method and computer vision processing system | |
CN109191390A (en) | A kind of algorithm for image enhancement based on the more algorithm fusions in different colours space | |
KR101448164B1 (en) | Method for Image Haze Removal Using Parameter Optimization | |
CN109685742A (en) | A kind of image enchancing method under half-light environment | |
CN108711140B (en) | Image brightness uniformity real-time recovery method based on inter-class variance description | |
CN106846282A (en) | A kind of enhancement method of low-illumination image of use adaptively correcting | |
CN105809643B (en) | A kind of image enchancing method based on adaptive block channel extrusion | |
CN106503644B (en) | Glasses attribute detection method based on edge projection and color characteristic | |
CN105741245B (en) | Adaptive contrast enhancement algorithm based on greyscale transformation | |
CN108154492B (en) | A kind of image based on non-local mean filtering goes haze method | |
CN106157264B (en) | Large area image uneven illumination bearing calibration based on empirical mode decomposition | |
CN101599171A (en) | Auto contrast's Enhancement Method and device | |
CN108288258A (en) | A kind of low-quality images Enhancement Method under severe weather conditions | |
CN103839245B (en) | The Retinex colour-image reinforcing method at night of Corpus--based Method rule | |
CN106971153A (en) | A kind of facial image illumination compensation method | |
CN111080568A (en) | Tetrolet transform-based near-infrared and color visible light image fusion algorithm | |
CN117082690B (en) | Control method and system of intelligent table lamp | |
CN115631350B (en) | Method and device for identifying colors of canned image | |
CN107516083A (en) | A kind of remote facial image Enhancement Method towards identification | |
Xue et al. | Video image dehazing algorithm based on multi-scale retinex with color restoration | |
CN108154490A (en) | Based on the high-voltage transmission line insulator image enchancing method for improving fuzzy set theory | |
CN109859138B (en) | Infrared image enhancement method based on human visual characteristics | |
CN115953824A (en) | Face skin image processing method and system | |
CN110223253B (en) | Defogging method based on image enhancement | |
CN110427868A (en) | A kind of pedestrian identify again in feature extracting method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |