CN106952243A - UUV Layer Near The Sea Surface infrared image self adaptation merger histogram stretches Enhancement Method - Google Patents
UUV Layer Near The Sea Surface infrared image self adaptation merger histogram stretches Enhancement Method Download PDFInfo
- Publication number
- CN106952243A CN106952243A CN201710151287.1A CN201710151287A CN106952243A CN 106952243 A CN106952243 A CN 106952243A CN 201710151287 A CN201710151287 A CN 201710151287A CN 106952243 A CN106952243 A CN 106952243A
- Authority
- CN
- China
- Prior art keywords
- merger
- histogram
- gray
- infrared image
- value
- 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
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000006978 adaptation Effects 0.000 title claims abstract description 11
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 230000002708 enhancing effect Effects 0.000 description 7
- 206010057249 Phagocytosis Diseases 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 230000008782 phagocytosis Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 230000033001 locomotion Effects 0.000 description 1
- 238000005070 sampling 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
- G06T5/40—Image enhancement or restoration using histogram techniques
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
Enhancement Method is stretched the present invention is to provide a kind of UUV Layer Near The Sea Surfaces infrared image self adaptation merger histogram.(1) original infrared image, is obtained;(2) infrared image conventional histogram, is obtained;(3), merger threshold value G is obtained using Otsu methods;(4), the corresponding pixel of gray-scale value less than threshold value T is integrated into its right adjacent gray level;(5) gray scale stretching, is carried out to obtained new merger histogram;(6) enhanced infrared image, is obtained.The present invention can preferably strengthen target, highlight wave of the sea texture, and can show the profile of scenery in blurred picture very well.In addition, Otsu methods are applied in the selection of merger threshold value by the present invention, the self adaptation for realizing threshold value is chosen.
Description
Technical field
The present invention relates to a kind of infrared imagery technique, more particularly, to a kind of enhancing of UUV Layer Near The Sea Surfaces infrared image
Method.
Background technology
Because itself imaging reason can cause infrared image to have, signal to noise ratio is low, scheme in imaging process for infrared imaging system
As the features such as fuzzy, low contrast.When UUV shoots Layer Near The Sea Surface infrared image, influence, the photographed temperature difference due to UUV motions
The factors such as difference, sea wave disturbance so that its infrared image quality shot is worse compared to general infrared image.Therefore, in order to
Improve UUV shoot Layer Near The Sea Surface infrared image quality so that the information in image preferably can be captured and utilize, it is necessary to
Enhancing processing is carried out to infrared image.
Histogram equalization be it is a kind of effectively, it is simple, using extensive image enchancing method, its realization is with gray scale
" phagocytosis " is cost.Scholar Hu Dianhai etc. exists《A kind of improved histogram-equalized image Enhancement Method》In for image because
Loss in detail and propose the problem of fog to increase the method for image gray levels after a kind of equalization processing.This method can be improved
Image overall contrast is with retaining the more details of original image.But this method is in the processing extremely fuzzy and unclear figure of profile
During picture, effect is undesirable.Scholar Wang Ping Jian etc. exists《Infrared image self-adaptive enhancement algorithm based on Plateau histogram》It is middle to propose
Adaptive algorithm overcome the problem of selection of platform threshold value is difficult, and can preferably strengthen target, reduce operand, but
It is that his selection mainly to platform threshold value carries out Improvement and do not carry out correlation to plateau equalization algorithm and grind
Study carefully and improve.It can be seen from histogram equalization principle, when number of pixels proportion [0,1/N) (N is grey level range,
Generally 255) between when, Pixel-level may be merged, but when number of pixels proportion be more than or equal to 1/N when,
Pixel-level will not be necessarily integrated into adjacent gray level.Therefore histogram equalization can not realize contrast it is great in or
The merging of Pixel-level equal to 1/N.But in some extremely fuzzy images, although the proportion of some Pixel-levels be more than or
Equal to 1/N, but its contribution to picture quality is simultaneously little, you can so that its gray level is changed to improve picture quality.
Now histogram equalization principle just can not be played a role to the part gray level.
The content of the invention
It is an object of the invention to provide a kind of definition that can improve infrared image, the profile for highlighting target and texture
UUV Layer Near The Sea Surface infrared image self adaptation merger histogram stretches Enhancement Method.
The object of the present invention is achieved like this:
(1) original infrared image, is obtained;
(2) infrared image conventional histogram, is obtained;
(3), merger threshold value G is obtained using Otsu methods;
(4), the corresponding pixel of gray-scale value less than threshold value T is integrated into its right adjacent gray level;
(5) gray scale stretching, is carried out to obtained new merger histogram;
(6) enhanced infrared image, is obtained.
The present invention can also include
1st, the step of utilization Otsu methods obtain merger threshold value G includes:
(1), the gray-scale value of conventional histogram is stored into an array;
(2) the threshold value T of numerical value in the array, is asked for using Otsu methods;
(3), merger threshold value takes threshold value T half, therefore G=0.5*T.
2nd, the corresponding pixel of gray-scale value by less than merger threshold value T is integrated into the step in its adjacent gray level
Including:
If infrared image gray-scale value is less than merger threshold value G and gray level is less than 255, by the corresponding pixel of the gray level
Value is set to the size of the right adjacent gray levels of the gray level;If infrared image gray-scale value is less than merger threshold value G and gray level etc.
In 255, the corresponding pixel point value of the gray level is set to the size of the left adjacent gray levels of the gray level;If infrared image gray scale
Level value is more than merger threshold value G, then all holdings of corresponding pixel points are constant.
3rd, the step of described pair of obtained new merger histogram carries out gray scale stretching includes:
For obtained merger histogram, line translation is entered using following formula, realizes that merger histogram is stretched:
Above formula is carried out to simplify:
Wherein, I (i, j) is image pixel value;X is the pixel value of I (i, j) corresponding pixel points after stretching conversion;255 be figure
As maximum gray scale;GminFor gray level minimum in merger histogram;GmaxFor gray level maximum in merger histogram.
To solve the present invention of problem present in prior art the adaptive of merger threshold value is realized using maximum variance between clusters
Should choose, break conventional histogram equalization can only by number of pixels proportion [0,1/N) it is interval in pixel merger limitation,
From main regulation number of pixels proportion merger scope, give full play to details " phagocytosis " enhancing is fuzzy, profile unclear image when
Effect, experiment proves that this method image fuzzy to extreme has very good result.The invention is to sacrifice details as generation
Valency exchanges the definition of infrared image for, has highlighted the profile and texture of target.
The beneficial effects of the invention are as follows:(1) pixel is realized using Otsu methods (maximum variance between clusters or Da-Jin algorithm)
The selection of number threshold value.And the half of threshold value is taken as merger threshold value according to actual experiment experience, realize enhancing algorithm from
Adapt to.(2) by the inspiration of histogram equalization " phagocytosis " details, proposition determines " to swallow " condition and range by setting threshold value,
The interference of sky exposure can be suppressed to a certain extent and target is protruded, make wave that there is obvious shape, no longer join together
Lighted region, so that the texture of wave becomes more fully apparent, makes the image-region belonged in same tonal range become phase
To smooth.
Brief description of the drawings
Fig. 1 is flow chart of the invention;
Fig. 2 is the original image of one embodiment of the invention.
Fig. 3 a- Fig. 3 b (scheme for the conventional histogram (Fig. 3 a) and merger histogram schematic diagram of one embodiment of the invention
3b);
Fig. 4 strengthens result figure for the histogram stretching of one embodiment.
Fig. 5 is the histogram equalization result figure of one embodiment.
Fig. 6 is the result figure of one embodiment of the invention.
Embodiment
Illustrate below in conjunction with the accompanying drawings and more detailed description done to the present invention, but protection scope of the present invention be not limited to it is following
It is described.
As shown in figure 1, a kind of UUV Layer Near The Sea Surfaces infrared image self adaptation merger histogram stretching Enhancement Method, including it is following
Step:
1st, traditional grey level histogram of UUV Layer Near The Sea Surface infrared images is shown using Matlab Programming with Pascal Language, and obtains figure
The corresponding gray-scale value of each gray level and stored it in as in an array;
2nd, data segmentation is carried out to the array in step 1 using traditional Otsu methods, obtains adaptivenon-uniform sampling threshold value T, make G
=0.5*T realizes merger threshold value T selection.Known by actual experiment experience, merger threshold value is better when taking T half and to figure
Picture quality influences smaller, therefore G=0.5*T.
3rd, according to merger threshold value T size, according to rule:If infrared image gray-scale value is less than merger threshold value T and gray scale
Level is less than 255, then the corresponding pixel point value of the gray level is set to the size of the right adjacent gray levels of the gray level;If infrared
Image gray level value is less than merger threshold value T and gray level is equal to 255, then the corresponding pixel point value of the gray level is set into this
The size of the left adjacent gray levels of gray level;If infrared image gray-scale value is more than merger threshold value T, all guarantors of corresponding pixel points
Hold constant.Merger histogram results are as shown in Figure 3 b.It is as follows that profit is formulated rule:
Wherein, h represents the gray level of image, and h scopes are [0,255];hmaxFor gray level maximum in merger histogram;
PG(h) be image merger histogram value;P (h) is conventional histogram value;G is merger threshold value.
4th, realize that enhancing is handled to infrared image on the basis of obtained merger histogram using image stretch enhancing algorithm,
Line translation is entered to each pixel using following formula:
Above formula is carried out to simplify:
Wherein, I (i, j) is image pixel value;X is the pixel value of I (i, j) corresponding pixel points after stretching conversion;255 be figure
As maximum gray scale;GminFor gray level minimum in merger histogram;GmaxFor gray level maximum in merger histogram.
5th, the infrared image after enhancing algorithm process is finally given, as a result as shown in Figure 6.
The step 2 comprises the following steps:
2.1 by merger histogram obtain gray-scale value non-zero array H [count], wherein count scopes for [0,
255]。
2.2 statistics array H [count] histogram, obtains the non-zero array S [differ] of array histogram value,
Differ be array H in different elements number.Finally, according to maximum variance between clusters, optimal segmenting threshold T is asked for, from
And obtain merger threshold value G=0.5*T.
If 2.3 non-zero array S [differ] store the number of same grey level value, it is general that gray-scale value differ occurs
Rate is:
Wherein differ=0,1 ..., ∞, and
Gray-scale value is divided into two classes by 2.4 according to threshold value T, i.e., rudimentary value class and senior value class, Otsu points of correspondence tradition
Background classes and target class in segmentation method.The span of low class is [0, T], and the span of high class is [T, ∞].
The probability that rudimentary value class occurs is:
The probability that senior value class occurs is:
Wherein, ω0+ω1=1.
The average gray-level value of 2.5 rudimentary value classes is:
The average gray-level value of senior value class is:
The average gray-level value of 2.6 statistic histograms:
δ2(T)=ω0*(u0-u)2+ω1*(u1-u)2 (10)
Wherein, u is whole array S average value, w0For the probability shared by the data amount check less than merger threshold value, w1To be big
In with the probability shared by the data amount check equal to merger threshold value, u0For the average value of the data less than merger threshold value, u1For more than with
Equal to the average value of the data of merger threshold value, T value changes from 0~L, works as δ2When taking maximum, the T values taken are to obtain
The optimal threshold arrived.
Claims (5)
1. a kind of UUV Layer Near The Sea Surfaces infrared image self adaptation merger histogram stretches Enhancement Method, it is characterized in that:
(1) original infrared image, is obtained;
(2) infrared image conventional histogram, is obtained;
(3), merger threshold value G is obtained using Otsu methods;
(4), the corresponding pixel of gray-scale value less than threshold value T is integrated into its right adjacent gray level;
(5) gray scale stretching, is carried out to obtained new merger histogram;
(6) enhanced infrared image, is obtained.
2. UUV Layer Near The Sea Surfaces infrared image self adaptation merger histogram according to claim 1 stretches Enhancement Method, its feature
It is that the step of utilization Otsu methods obtain merger threshold value G includes:
(1), the gray-scale value of conventional histogram is stored into an array;
(2) the threshold value T of numerical value in the array, is asked for using Otsu methods;
(3), merger threshold value takes threshold value T half, therefore G=0.5*T.
3. UUV Layer Near The Sea Surfaces infrared image self adaptation merger histogram according to claim 1 or 2 stretches Enhancement Method, its
It is characterized in the step bag that the corresponding pixel of gray-scale value by less than merger threshold value T is integrated into its adjacent gray level
Include:
If infrared image gray-scale value is less than merger threshold value G and gray level is less than 255, the corresponding pixel point value of the gray level is set
It is set to the size of the right adjacent gray levels of the gray level;If infrared image gray-scale value is less than merger threshold value G and gray level is equal to
255, the corresponding pixel point value of the gray level is set to the size of the left adjacent gray levels of the gray level;If infrared image gray level
Value is more than merger threshold value G, then all holdings of corresponding pixel points are constant.
4. UUV Layer Near The Sea Surfaces infrared image self adaptation merger histogram according to claim 1 or 2 stretches Enhancement Method, its
It is characterized in that the step of described pair of obtained new merger histogram carries out gray scale stretching includes:
For obtained merger histogram, line translation is entered using following formula, realizes that merger histogram is stretched:
Above formula is carried out to simplify:
Wherein, I (i, j) is image pixel value;X is the pixel value of I (i, j) corresponding pixel points after stretching conversion;255 be image most
High-gray level level;GminFor gray level minimum in merger histogram;GmaxFor gray level maximum in merger histogram.
5. UUV Layer Near The Sea Surfaces infrared image self adaptation merger histogram according to claim 3 stretches Enhancement Method, its feature
It is that the step of described pair of obtained new merger histogram carries out gray scale stretching includes:
For obtained merger histogram, line translation is entered using following formula, realizes that merger histogram is stretched:
Above formula is carried out to simplify:
Wherein, I (i, j) is image pixel value;X is the pixel value of I (i, j) corresponding pixel points after stretching conversion;255 be image most
High-gray level level;GminFor gray level minimum in merger histogram;GmaxFor gray level maximum in merger histogram.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710151287.1A CN106952243A (en) | 2017-03-14 | 2017-03-14 | UUV Layer Near The Sea Surface infrared image self adaptation merger histogram stretches Enhancement Method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710151287.1A CN106952243A (en) | 2017-03-14 | 2017-03-14 | UUV Layer Near The Sea Surface infrared image self adaptation merger histogram stretches Enhancement Method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106952243A true CN106952243A (en) | 2017-07-14 |
Family
ID=59468313
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710151287.1A Pending CN106952243A (en) | 2017-03-14 | 2017-03-14 | UUV Layer Near The Sea Surface infrared image self adaptation merger histogram stretches Enhancement Method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106952243A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108986124A (en) * | 2018-06-20 | 2018-12-11 | 天津大学 | In conjunction with Analysis On Multi-scale Features convolutional neural networks retinal vascular images dividing method |
CN109801246A (en) * | 2019-01-10 | 2019-05-24 | 华侨大学 | A kind of color histogram equalization methods of adaptive threshold |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102129675A (en) * | 2011-02-24 | 2011-07-20 | 中国兵器工业系统总体部 | Nonlinear adaptive infrared image enhancing method |
-
2017
- 2017-03-14 CN CN201710151287.1A patent/CN106952243A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102129675A (en) * | 2011-02-24 | 2011-07-20 | 中国兵器工业系统总体部 | Nonlinear adaptive infrared image enhancing method |
Non-Patent Citations (4)
Title |
---|
N.OTSU: "Threshold Selection Method from Gray-Level Histograms", 《IEEE》 * |
渠文平: "基于数字图像处理技术的岩石细观量化试验研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
许蓓蕾 等: "基于对象的多级图像增强法", 《计算机应用》 * |
陈天华: "《数字图像处理》", 30 June 2007 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108986124A (en) * | 2018-06-20 | 2018-12-11 | 天津大学 | In conjunction with Analysis On Multi-scale Features convolutional neural networks retinal vascular images dividing method |
CN109801246A (en) * | 2019-01-10 | 2019-05-24 | 华侨大学 | A kind of color histogram equalization methods of adaptive threshold |
CN109801246B (en) * | 2019-01-10 | 2022-11-01 | 华侨大学 | Global histogram equalization method for adaptive threshold |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109754377B (en) | Multi-exposure image fusion method | |
CN107527332B (en) | Low-illumination image color retention enhancement method based on improved Retinex | |
Lv et al. | Real-time dehazing for image and video | |
Gao et al. | Sand-dust image restoration based on reversing the blue channel prior | |
Xu et al. | Structure-texture aware network for low-light image enhancement | |
CN112164011B (en) | Motion image deblurring method based on self-adaptive residual error and recursive cross attention | |
US20100098331A1 (en) | System and method for segmenting foreground and background in a video | |
CN110570360B (en) | Retinex-based robust and comprehensive low-quality illumination image enhancement method | |
CN112507617B (en) | Training method of SRFlow super-resolution model and face recognition method | |
CN110807742B (en) | Low-light-level image enhancement method based on integrated network | |
CN111047543A (en) | Image enhancement method, device and storage medium | |
JP7353803B2 (en) | Image processing device, image processing method, and program | |
Liu et al. | Learning hadamard-product-propagation for image dehazing and beyond | |
CN104318529A (en) | Method for processing low-illumination images shot in severe environment | |
Yu et al. | Image and video dehazing using view-based cluster segmentation | |
Swami et al. | Candy: Conditional adversarial networks based fully end-to-end system for single image haze removal | |
CN103489168A (en) | Enhancing method and system for infrared image being converted to pseudo color image in self-adaptive mode | |
CN109859138B (en) | Infrared image enhancement method based on human visual characteristics | |
CN113379861B (en) | Color low-light-level image reconstruction method based on color recovery block | |
CN106952243A (en) | UUV Layer Near The Sea Surface infrared image self adaptation merger histogram stretches Enhancement Method | |
KR102277005B1 (en) | Low-Light Image Processing Method and Device Using Unsupervised Learning | |
CN117611467A (en) | Low-light image enhancement method capable of balancing details and brightness of different areas simultaneously | |
CN117422653A (en) | Low-light image enhancement method based on weight sharing and iterative data optimization | |
Zhang et al. | Dehazing with improved heterogeneous atmosphere light estimation and a nonlinear color attenuation prior model | |
CN104966271B (en) | Image de-noising method based on biological vision receptive field mechanism |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170714 |
|
RJ01 | Rejection of invention patent application after publication |