CN105139364A - Image enhancement method and application thereof - Google Patents
Image enhancement method and application thereof Download PDFInfo
- Publication number
- CN105139364A CN105139364A CN201510654050.6A CN201510654050A CN105139364A CN 105139364 A CN105139364 A CN 105139364A CN 201510654050 A CN201510654050 A CN 201510654050A CN 105139364 A CN105139364 A CN 105139364A
- Authority
- CN
- China
- Prior art keywords
- image
- target region
- interesting target
- interest
- enhancing
- 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
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses an image enhancement method which is used for carrying out enhancement on grayscale images. The method specially comprises the following steps: 1) extracting target regions of interest in the grayscale images to be enhanced; 2) carrying out nonlinear transformation on the target regions of interest to obtain target regions of interest after intensity mapping; 3) carrying out linear stretching on the target regions of interest obtained after intensity mapping to obtain target regions of interest after enhancement; and 4) carrying out registering and covering on the target regions of interest after enhancement and the target regions of interest in the grayscale images to be enhanced, and thus enhance images can be obtained. The invention also discloses an application of the method in the image target recognition and detection. The method can effectively enhance images and compress image gray scale dynamic range, so that detail parts of the regions of interest in the images are allowed to be clearer, textural features are more obvious, image quality is improved obviously, and follow-up object identification and detection can be done favorably.
Description
Technical field
The invention belongs to digital image processing techniques field, be specifically related to a kind of image enchancing method and application thereof.
Background technology
In some field; the image of detector image-forming is because be subject to the impact of various disturbing factor; usually there will be skew, as shake, scattering and the phenomenon such as image blurring; detection system signal to noise ratio (S/N ratio) is caused to reduce and picture quality reduction; therefore often need correct image process, so that the process to pictures subsequent.Such as, in infrared detection image field, due to the Pneumatic optical impact in air, the whirlpool stray light wave propagation randomly of atmospheric turbulence, the light radiation value entering sensor distorts, and the quantity of information be reflected on image is changed, focal plane picture point intensity distributions diffusion, peak value is reduced, produces average picture skew, as shake, scattering and image blurring, can cause greatly reducing the detection accuracy of imageable target.
In carrying out detection image correcting, because the image pixel gray level dynamic range after correcting is excessive, when observing or show, the gray-scale pixels of dark space will be covered by the gray-scale pixels in clear zone, thus make a large amount of image detail informations smudgy, cause picture quality very poor.Therefore needing the image information to obtaining to carry out post-processed, improving the resolution characteristic to image, thus reaching the effect of image enhaucament, being convenient to identification and the detection of target in image.
Summary of the invention
The object of the invention is to propose a kind of image enchancing method, it utilizes nonlinear transformation to carry out a process to the gray level image of area-of-interest, the gray level of its each pixel is carried out intensity to map to strengthen image, the method makes the detail section of area-of-interest in image more clear, outstanding textural characteristics, effectively improves picture quality.
Realize a kind of image enchancing method of the object of the invention, it specifically comprises the steps:
(1) obtain gray level image f (x, y) to be reinforced, extract interesting target region f ' (x, y);
(2) in interesting target region f ' (x, y) step (1) obtained, the gray level of each pixel substitutes into non-linear transform function g ' (x, y)=kf ' (x, y)
γ, calculate interesting target region g ' (x, y) after intensity mapping, wherein, k is non-linear transform function g ' (x, y)=kf ' (x, y)
γmiddle scale-up factor k, γ are index;
(3) determine the tonal range [min, max] in interesting target region g ' (x, y) after linear stretch scope [a, b] and intensity mapping, linear stretch is carried out in region g ' (x, y), and wherein stretch function is as follows:
Calculate interesting target region g " (x, y) after enhancing;
(4) by interesting target region the g " (x; y) carry out registration with interesting target region f ' (x, y) in gray level image f (x, y) to be reinforced and cover; the image g (x, y) after can being enhanced after enhancing.
As improvement of the present invention, also enhancing image g (x, y) and image f (x, y) to be reinforced can be contrasted, observe and strengthen effect: if to enhancing good results, then export and strengthen image g (x, y), enhancing process terminates; If be unsatisfied with enhancing effect, then returned step (2), again choose parameter k and the γ of non-linear transform function, proceed enhancing process.
Method of the present invention can strengthen image effectively, compressed image gray scale dynamic range, and make the detail section of area-of-interest in image more clear, textural characteristics is more outstanding, and picture quality significantly improves, and is conducive to target identification and the detection in later stage; In addition, the method can reduce the time required for actual enhancing, especially large in the background area of image, when target area is little.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the nonlinear transformation image enchancing method of the embodiment of the present invention;
Fig. 2 is the input-output curve of the non-linear transform function of the embodiment of the present invention.
Embodiment
For making object of the present invention, technical scheme and technique effect clearly understand, be described in more detail the present invention below in conjunction with the drawings and specific embodiments, following examples are only for explaining the present invention, do not form limitation of the invention.
The nonlinear transformation image enchancing method based on region of interest region constraint of the present embodiment specifically comprises the steps:
(1) first obtain gray level image f (x, y) to be reinforced, and extract interesting target region f ' (x, y) wherein.
In one embodiment, interesting target region can be extracted according to concrete conditions such as target shape, position or other features.
(2) non-linear transform function g ' (x, y)=kf ' (x, y) is determined
γ, wherein, k is scale-up factor, and γ is the value of index.
In a preferred embodiment, first preferably k=2 is chosen, γ=0.8.Owing to having carried out linear stretch in non-linear Enhancement Method, therefore the value of scale-up factor k is with to strengthen effect relation little, preferably chooses k=2, but be not limited to above-mentioned value in the present invention in the present embodiment, can according to also getting other values, as 1-5.
The value of exponent gamma specifically sets according to picture quality and actual enhancing demand, can according to enhancing effect dynamic conditioning in concrete assignment procedure.Such as preferentially can select exponent gamma=0.8, then observe and strengthen effect, when strong DeGrain, then exponent gamma is deducted a fixed threshold (such as 0.2), i.e. γ=γ-0.2, re-starts enhancing.Obviously strengthen effect if existing, then exponent gamma is finely tuned in threshold range [γ-0.2, γ+0.2].Such as can adjust from the near to the remote in the left and right of current γ value, to reach satisfied enhancing effect.If strengthen excessively, then by exponent gamma micro-increasing in threshold range [γ, γ+0.2], to reach satisfied enhancing effect, until reach satisfied enhancing effect, whole adjustment process ensures 0 < γ < 1, such as can preferred γ=0.45.
(3) in interesting target region f ' (x, y) step (1) obtained, the gray level of each pixel substitutes into non-linear transform function:
g′(x,y)=kf′(x,y)
γ
Calculate interesting target region g ' (x, y) after intensity mapping, in formula, f ' (x, y) is input gray grade, and g ' (x, y) is output gray level.
(4) linear stretch is carried out, target area g " (x, y) after being enhanced in interesting target region g ' (x, y) after intensity being mapped.
In one embodiment, the method for preferred linear stretch is as follows:
(4.1) determine linear stretch scope [a, b], generally choose a=0, b=2
n-1, n is image figure place.Namely when f (x, y) is 16 gray level images, when a=0, b=65535, f (x, y) they are 8 gray level images, a=0, b=255.
(4.2) intensity value ranges [min, max] of the area-of-interest g ' (x, y) after calculating strength mapping.Min and max is respectively minimum and maximum gradation value in interesting target region g ' (x, y) after the mapping of described intensity.
Can in the following way during actual computation: appoint and get 1 g ' (x in g ' (x, y)
0, y
0) make min=max=g ' (x
0, y
0), to get successively in g ' (x, y) gray-scale value a little and min and max compare one by one, if min > g ' (x, y), then min=g ' (x, y); If max < g ' (x, y), then max=g ' (x, y).I.e. max=g ' (x, y)
max, min=g ' (x, y)
min;
(4.3) the area-of-interest g ' (x, y) after intensity being mapped substitutes into stretch function:
Calculate interesting target region g " (x, y) after enhancing.
(5) by interesting target region the g " (x; y) carry out registration with interesting target region f ' (x, y) in gray level image f (x, y) to be reinforced and cover; can be enhanced image g (x, y) after enhancing.
The present embodiment method is by application example, and carry out com-parison and analysis to the gray level image before and after strengthening, can find that the interesting target region details strengthening rear image is clear, texture is given prominence to, and picture quality obviously promotes.
Enhancement Method of the present invention is suitable for the image post-processed that aero-optical effect corrects, the enhancing of sky target image, the image processing process that in also can be applicable to images steganalysis and detecting, target area strengthens.
Those skilled in the art will readily understand; the foregoing is only preferred embodiment of the present invention; not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.
Claims (7)
1. an image enchancing method, for strengthening gray level image, it is characterized in that, the method specifically comprises the steps:
(1) the interesting target region in gray level image to be reinforced is extracted;
(2) nonlinear transformation is carried out to described interesting target region, thus obtain the interesting target region after intensity mapping;
(3) linear stretch is carried out to the interesting target region after described intensity mapping, obtain the interesting target region after strengthening;
(4) the interesting target region in the interesting target region after described enhancing and described gray level image to be reinforced is carried out to registration and covered, can be enhanced image;
Wherein, described nonlinear transformation is realized by non-linear transform function, substitutes into this non-linear transform function by the gray level of each pixel in described interesting target region, thus obtains the interesting target region after intensity mapping, wherein, described non-linear transform function is:
g′(x,y)=k·f′(x,y)
γ
In formula, f ' (x, y) is interesting target region, and g ' (x, y) is the interesting target region after intensity mapping, and k is scale-up factor, and γ is index.
2. a kind of image enchancing method according to claim 1, wherein, described linear stretch is realized by stretch function, and this stretch function is specially:
In formula, min and max is respectively minimum and maximum gradation value in interesting target region g ' (x, y) after intensity mapping, a and b is respectively minimum value and the maximal value of linear stretch scope, " (x, y) is the interesting target region after strengthening to g.
3. a kind of image enchancing method according to claim 2, wherein, the minimum value a of described linear stretch scope and maximal value b value are a=0, b=2
n-1, in formula, n is image figure place.
4. a kind of image enchancing method according to any one of claim 1-3, wherein, described scale-up factor k value is 1-5, and be preferably k=2, exponent gamma meets 0 < γ < 1, preferred γ=0.8.
5. a kind of image enchancing method according to any one of claim 1-4, wherein, also can for the enhancing image obtained, by again choosing scale-up factor k and the exponent gamma of non-linear transform function, perform step (2)-(4) to circulate, thus acquisition better strengthens image.
6. a kind of image enchancing method according to claim 5, wherein, again choosing of described exponent gamma can be carried out according to following principle:
When enhancing DeGrain, then exponent gamma is deducted a fixed threshold, re-start enhancing;
Obviously strengthen effect if existing, then exponent gamma is finely tuned in threshold range [γ-0.2, γ+0.2];
If strengthen excessively, then by exponent gamma micro-increasing in threshold range [γ, γ+0.2].
7. the image enchancing method that one of claim 1-6 is described is in images steganalysis and the application in detection.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510654050.6A CN105139364A (en) | 2015-10-10 | 2015-10-10 | Image enhancement method and application thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510654050.6A CN105139364A (en) | 2015-10-10 | 2015-10-10 | Image enhancement method and application thereof |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105139364A true CN105139364A (en) | 2015-12-09 |
Family
ID=54724697
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510654050.6A Pending CN105139364A (en) | 2015-10-10 | 2015-10-10 | Image enhancement method and application thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105139364A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107545552A (en) * | 2017-09-08 | 2018-01-05 | 四川理工学院 | A kind of image rendering method |
CN108053538A (en) * | 2017-12-29 | 2018-05-18 | 潘彦伶 | A kind of intelligent access control system based on cloud storage |
CN108230412A (en) * | 2018-01-19 | 2018-06-29 | 浙江大华技术股份有限公司 | A kind of IR image compression method and device |
CN109377462A (en) * | 2018-10-23 | 2019-02-22 | 上海鹰瞳医疗科技有限公司 | Method for processing fundus images and equipment |
CN109447964A (en) * | 2018-10-23 | 2019-03-08 | 上海鹰瞳医疗科技有限公司 | Method for processing fundus images and equipment |
US10417770B2 (en) | 2015-05-29 | 2019-09-17 | Alibaba Group Holding Limited | Efficient acquisition of a target image from an original image |
CN114353880A (en) * | 2022-01-21 | 2022-04-15 | 国网河南省电力公司电力科学研究院 | Strain insulator string wind-induced vibration online monitoring system and method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102129675A (en) * | 2011-02-24 | 2011-07-20 | 中国兵器工业***总体部 | Nonlinear adaptive infrared image enhancing method |
CN102722871A (en) * | 2012-05-24 | 2012-10-10 | 中山大学 | Quick and effective image enhancing method |
CN103268598A (en) * | 2013-06-13 | 2013-08-28 | 武汉大学 | Retinex-theory-based low-illumination low-altitude remote sensing image enhancing method |
CN103530848A (en) * | 2013-09-27 | 2014-01-22 | 中国人民解放军空军工程大学 | Double exposure implementation method for inhomogeneous illumination image |
CN104112133A (en) * | 2014-07-30 | 2014-10-22 | 福州大学 | Face detection preprocessing method under complex illumination |
-
2015
- 2015-10-10 CN CN201510654050.6A patent/CN105139364A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102129675A (en) * | 2011-02-24 | 2011-07-20 | 中国兵器工业***总体部 | Nonlinear adaptive infrared image enhancing method |
CN102722871A (en) * | 2012-05-24 | 2012-10-10 | 中山大学 | Quick and effective image enhancing method |
CN103268598A (en) * | 2013-06-13 | 2013-08-28 | 武汉大学 | Retinex-theory-based low-illumination low-altitude remote sensing image enhancing method |
CN103530848A (en) * | 2013-09-27 | 2014-01-22 | 中国人民解放军空军工程大学 | Double exposure implementation method for inhomogeneous illumination image |
CN104112133A (en) * | 2014-07-30 | 2014-10-22 | 福州大学 | Face detection preprocessing method under complex illumination |
Non-Patent Citations (1)
Title |
---|
李绘卓等: "一种非线性变换的双直方图红外图像增强方法", 《计算机工程与应用》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10417770B2 (en) | 2015-05-29 | 2019-09-17 | Alibaba Group Holding Limited | Efficient acquisition of a target image from an original image |
CN107545552A (en) * | 2017-09-08 | 2018-01-05 | 四川理工学院 | A kind of image rendering method |
CN108053538A (en) * | 2017-12-29 | 2018-05-18 | 潘彦伶 | A kind of intelligent access control system based on cloud storage |
CN108230412A (en) * | 2018-01-19 | 2018-06-29 | 浙江大华技术股份有限公司 | A kind of IR image compression method and device |
CN109377462A (en) * | 2018-10-23 | 2019-02-22 | 上海鹰瞳医疗科技有限公司 | Method for processing fundus images and equipment |
CN109447964A (en) * | 2018-10-23 | 2019-03-08 | 上海鹰瞳医疗科技有限公司 | Method for processing fundus images and equipment |
CN114353880A (en) * | 2022-01-21 | 2022-04-15 | 国网河南省电力公司电力科学研究院 | Strain insulator string wind-induced vibration online monitoring system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105139364A (en) | Image enhancement method and application thereof | |
Wang et al. | Biologically inspired image enhancement based on Retinex | |
CN101950412B (en) | Method for enhancing details and compressing dynamic range of infrared image | |
EP3206185B1 (en) | Image processing method, image processing device and display device | |
Negru et al. | Exponential contrast restoration in fog conditions for driving assistance | |
CN102693531B (en) | Adaptive double-platform based infrared image enhancement method | |
CN102682436B (en) | A kind of image enchancing method theoretical improved multiple dimensioned Retinex | |
CN107784669A (en) | A kind of method that hot spot extraction and its barycenter determine | |
CN107680054A (en) | Multisource image anastomosing method under haze environment | |
CN104063848A (en) | Enhancement method and device for low-illumination image | |
CN103295191A (en) | Multi-scale vision self-adaptation image enhancing method and evaluating method | |
CN104966285B (en) | A kind of detection method of salient region | |
CN103177429A (en) | FPGA (field programmable gate array)-based infrared image detail enhancing system and method | |
CN107203981A (en) | Image defogging method based on fog concentration characteristics | |
CN104574293A (en) | Multiscale Retinex image sharpening algorithm based on bounded operation | |
CN104574312A (en) | Method and device of calculating center of circle for target image | |
CN103489168A (en) | Enhancing method and system for infrared image being converted to pseudo color image in self-adaptive mode | |
CN113450272B (en) | Image enhancement method based on sinusoidal variation and application thereof | |
CN102999903A (en) | Method for quantitatively evaluating illumination consistency of remote sensing images | |
CN116452447A (en) | Low-illumination high-definition image processing method | |
CN113610863B (en) | Multi-exposure image fusion quality assessment method | |
CN110020999A (en) | A kind of uncooled ir thermal imagery self organizing maps method based on homomorphic filtering | |
CN105303529A (en) | Digital image haze removing method based on histogram equalization | |
CN108537807A (en) | One kind enhancing detection method based on the improved edge of No. four satellite images of high score | |
CN113888420A (en) | Underwater image restoration method and device based on correction model and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151209 |