CN109671032A - According to the compensation factor equilibrium model of image brilliance feature - Google Patents
According to the compensation factor equilibrium model of image brilliance feature Download PDFInfo
- Publication number
- CN109671032A CN109671032A CN201811546359.3A CN201811546359A CN109671032A CN 109671032 A CN109671032 A CN 109671032A CN 201811546359 A CN201811546359 A CN 201811546359A CN 109671032 A CN109671032 A CN 109671032A
- Authority
- CN
- China
- Prior art keywords
- image
- model
- compensation
- compensation factor
- brilliance
- 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
- 238000000034 method Methods 0.000 claims abstract description 20
- 230000000694 effects Effects 0.000 claims abstract description 7
- 239000000284 extract Substances 0.000 claims abstract 7
- 238000012935 Averaging Methods 0.000 abstract 1
- 238000012545 processing Methods 0.000 description 14
- 238000012937 correction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- XKRFYHLGVUSROY-UHFFFAOYSA-N Argon Chemical compound [Ar] XKRFYHLGVUSROY-UHFFFAOYSA-N 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 241000023320 Luma <angiosperm> Species 0.000 description 1
- 229910052786 argon Inorganic materials 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- OSWPMRLSEDHDFF-UHFFFAOYSA-N methyl salicylate Chemical compound COC(=O)C1=CC=CC=C1O OSWPMRLSEDHDFF-UHFFFAOYSA-N 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization 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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of compensation factor equilibrium model according to image brilliance feature, including image brilliance trend prediction, establishes luminance compensation model, extracts compensation factor.In image brilliance trend prediction, extracts image greyscale and change lesser row gradation data, calculate each column average gray value in effective row gradation data;It establishes in luminance compensation model, corresponding brightness desired value is set, be quotient with averaging of income gradation data, obtain corresponding offset, compensation rate data are fitted to obtain compensation model using approximating method;It extracts in compensation factor, according to the map sheet width of image, the compensation factor of respective width is extracted in luminance compensation model, realizes luminance proportion.The present invention can effectively realize luminance proportion according to image feature, reduce image root-mean-square error, shorten image available gray-scale activity interval, reduce image and radiate poor peak value.
Description
Technical field
The present invention relates to image brilliance equilibrium field, in particular to coefficient of utilization method is realized image brilliance equilibrium field, is answered
For the processing of remote sensing image luminance proportionization.
Technical background
With the development of image processing technique, luminance proportion method has tended to be mature and perfect, wherein passing through coefficient pair
The method that image brilliance difference is handled is quickly grown, and becomes the main direction of studying of present intensity equalization processing.Work as shadow
As changing the original gray value of image under the action of coefficient there are when luminance difference, realizes that image brilliance is balanced, have operation
Simplicity the advantages that limitation by map sheet size and image quantity, is used widely in fields such as remote sensing image processing.
The method that coefficient of utilization realizes luminance proportion at present, including addition correction factor method and traditional increase brightness system
Counting method.Addition correction factor method is by calculating the gray scale difference value between target area and reference zone and being pocessed, to add
The form of method handles image original gray value, such as: the disclosure of the invention of Patent No. CN201510704720.0 one kind
The corresponding brightness correction coefficient in image two sides is calculated, to the method that the brightness value of each pixel of image is modified, defect exists
In not can guarantee computational accuracy, gray-scale level is easily caused to be mutated, when map sheet is larger, calculation amount is increased with it.Traditional increase is bright
Coefficient method is spent, specific model is established according to certain parameter, image original gray value is handled in the form of multiplication,
Such as: the disclosure of the invention of Patent No. CN201710431459.0 is a kind of to resolve specific formulation, luminance factor is formed, to filtering
The method that image afterwards carries out luminance proportion processing, though have a disadvantage in that is not influenced by error is calculated, fixed mould
Type be unable to satisfy image diversity with from particularity.In order to meet the brightness of image, processing accuracy is improved, it is necessary to being
Number methods improve and perfect.
Summary of the invention
The object of the present invention is to provide a kind of compensation factor equilibrium models according to image brilliance feature, can effectively adapt to
Image from particularity, reduce calculation amount, guarantee image precision, avoid generating gray-scale level mutation, realize colors of image equilibrium most
Optimization.
The contents of the present invention are as follows: a kind of compensation factor equilibrium model according to image brilliance feature, including image brilliance become
Gesture prediction model, luminance compensation model, compensation factor equilibrium model.In image brilliance trend prediction model, according to image brilliance
Feature threshold value is arranged, image greyscale is extracted and changes lesser row gradation data by row (or column) difference processing to image,
Resulting row gradation data is subjected to difference again, rejects the biggish data of gray scale difference, remaining gradation data is known as having
Row gradation data is imitated, the average gray value that effective row gradation data respectively arranges is calculated, realizes to image brilliance trend, obtains
To image brilliance trend prediction model;In luminance compensation model, the phase appropriate is selected according to the gray scale activity interval of prediction model
Prestige value, it would be desirable to which value and each column average gray value are quotient, obtain compensation rate and are intended using approximating method compensation rate data
It closes, obtains image brightness compensation model;In compensation factor equilibrium model, according to the map sheet width of image, in luminance compensation model
The middle compensation factor for extracting respective width, obtains compensation factor equilibrium model, product is done with image original gray value, after obtaining processing
Image, realize luminance proportion.
The beneficial effects of the present invention are: using a kind of compensation factor equilibrium model according to image brilliance feature, including shadow
Image brightness trend prediction model, luminance compensation model, compensation factor equilibrium model.Image brilliance trend prediction model includes to shadow
The trend of image brilliance variation is effectively predicted out in the mathematical description of image brightness feature;Luminance compensation model includes according to image
Brightness establishes function model, gets rid of influence and fixed model of the computational accuracy to image to the beam for meeting image brilliance feature
It ties up;Compensation factor equilibrium model includes stabilization of the addition correction factor method to the adaptability of image and traditional brightness coefficient method
Property the advantages of, the present invention establishes compensation factor equilibrium model according to the image of different brightness, and the square of image is effectively reduced
Root error shortens image available gray-scale activity interval.
Detailed description of the invention
Fig. 1 is image brilliance difference schematic diagram of the embodiment of the present invention.
Fig. 2 is flow chart of the present invention.
Fig. 3 is brightness trend prediction flow chart of the present invention.
Fig. 4 is luma prediction model schematic of the embodiment of the present invention.
Fig. 5 is compensation model schematic diagram of the embodiment of the present invention.
Fig. 6 is processing result schematic diagram of the embodiment of the present invention.
Specific embodiment
The following further describes the specific embodiments of the present invention with reference to the drawings.Obviously implementation described in the invention
Example is only a part of the embodiments of the present invention, rather than whole embodiments.Based on the embodiment of the present invention, area research personnel exist
Under the premise of not making creative work, other all embodiments belong to the scope of the present invention.
Embodiment:
Greenland region partial image is chosen in the present embodiment and carries out equalization processing, which is clapped by ARGON satellite
Take the photograph gained.For image due to shooting environmental and artificial mechanism, brightness is extremely unbalanced.As shown in Figure 1, existing at the image lower right corner
Apparent brightness band.
The specific steps of luminance proportion provided by the present invention can be with reference flow sheet (Fig. 2):
Step 1: reading the gradation data of image in MATLAB, and convert double format for gradation data.
Step 2: image is pressed into formula Δ gray=gray (mi+1,n)-gray(mi, n) and carry out difference processing line by line, in formula
M represents the line number of image, and n represents the columns of image, counts all differentiated gradation datas.
Step 3: threshold value T is set according to actual conditions, the line number of Δ gray < T is extracted, as effective row gradation data.
Step 4: establishing image brilliance prediction model in conjunction with Fig. 3, counted in effective row brightness data first in same column
Gray value carries out difference processing again, rejects the biggish gradation data of deviation, to guarantee that modeling accuracy can carry out limited times repetition
Experiment;Next according to weighted mean formulaRemaining gradation data is calculated, x, which is represented, in formula is somebody's turn to do
The intensity-weighted average value of column, p represent the weight that the column pixel respectively represents, and g (m) represents pixel grey scale size, and m represents effective
Line number is carried out each column average gray value data to rearrange a line ordered series of numbers by present position, finally obtains image brilliance prediction
Model, as shown in Figure 4.
Step 5: according to the gray scale activity interval of image brilliance prediction model, desired value P being set, it would be desirable to which value is bright with image
Gradation data in degree prediction model is quotient, finds out the offset of image brilliance prediction model.
Step 6: each offset being fitted using fitting process, obtains matched curve f (x), f (x) is the bright of image
Compensation model is spent, acquired results are as shown in Figure 5.
Step 7: according to striograph breadth degree n, extracting respective width n's in the expression formula f (x) of luminance compensation model
Compensation factor is done product with image original gradation data in the form of multiplication by compensation factor number.If using compensation factor to image
After processing, effect is excessively bright or excessively dark, and compensation factor number is appropriately extended, and according to the actual conditions of image brilliance feature, cuts choosing
Compensation factor identical with striograph breadth degree handles image original gray value.
Step 8: output processing result converts required type for data, result such as Fig. 6 after gained image is balanced
Shown, by treated, image can clearly find that brightness band has been eliminated.
Claims (4)
1. a kind of compensation factor equilibrium model according to image brilliance feature, including image brilliance trend prediction, establish brightness benefit
It repays model, extract compensation factor, it is characterised in that specific steps embody are as follows: in image brilliance trend prediction, according to image brilliance
Feature establishes image brilliance prediction model, predicts image greyscale value size variation trend;It establishes in luminance compensation model, by setting
Determine desired value and effective row gradation data is quotient and obtains offset, establishes luminance compensation model using approximating method;Extract brightness
In the factor, in conjunction with striograph breadth degree, compensation factor is extracted in luminance compensation model, by compensation factor and image original gradation
Value does product.
2. a kind of compensation factor equilibrium model according to image brilliance feature according to claim 1, it is characterised in that institute
Stating image brilliance trend prediction includes: to extract trip gradation data using differential technique, forms effective row gradation data;According to adding
Mean value formula is weighed, the average gray of each column is calculated;It carries out each column average gray value data to rearrange one by present position
Line number column, analyze image feature and variation tendency.
3. a kind of compensation factor equilibrium model according to image brilliance feature according to claim 2, it is characterised in that institute
Stating and establishing luminance compensation model includes, according to the gray scale activity interval of image brilliance prediction model, setting desired value appropriate;It will
Gradation data in desired value and image brilliance prediction model is quotient, finds out the offset of image brilliance prediction model;Using quasi-
It is legal to be fitted each offset, fit the function expression of luminance compensation model.
4. a kind of compensation factor equilibrium model according to image brilliance feature according to claim 3, it is characterised in that institute
State and extract compensation factor and include, according to striograph breadth degree, luminance compensation model expression extract the compensation of respective width because
Son;In the case that treatment effect is excessively bright or excessively dark, compensation factor quantity is appropriately extended, according to the practical feelings of image brilliance feature
Condition is cut the compensation factor that choosing is consistent with striograph breadth degree and is handled image original gray value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811546359.3A CN109671032B (en) | 2018-12-18 | 2018-12-18 | Method for constructing compensation factor equalization model according to image brightness characteristics |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811546359.3A CN109671032B (en) | 2018-12-18 | 2018-12-18 | Method for constructing compensation factor equalization model according to image brightness characteristics |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109671032A true CN109671032A (en) | 2019-04-23 |
CN109671032B CN109671032B (en) | 2023-04-07 |
Family
ID=66143879
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811546359.3A Active CN109671032B (en) | 2018-12-18 | 2018-12-18 | Method for constructing compensation factor equalization model according to image brightness characteristics |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109671032B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
AU2001247611A1 (en) * | 2000-03-20 | 2001-12-13 | Gentex Corporation | System for Controlling Exterior Vehicle Lights |
CN101604509A (en) * | 2008-06-13 | 2009-12-16 | 胜华科技股份有限公司 | Image-displaying method |
CN105761288A (en) * | 2016-02-02 | 2016-07-13 | 华中科技大学 | Real-time star point centroid location method and device based on FPGA |
CN105913401A (en) * | 2016-05-06 | 2016-08-31 | 北京信息科技大学 | Industrial camera photogrammetry image brightness compensation method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6587573B1 (en) * | 2000-03-20 | 2003-07-01 | Gentex Corporation | System for controlling exterior vehicle lights |
-
2018
- 2018-12-18 CN CN201811546359.3A patent/CN109671032B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
AU2001247611A1 (en) * | 2000-03-20 | 2001-12-13 | Gentex Corporation | System for Controlling Exterior Vehicle Lights |
CN101604509A (en) * | 2008-06-13 | 2009-12-16 | 胜华科技股份有限公司 | Image-displaying method |
CN105761288A (en) * | 2016-02-02 | 2016-07-13 | 华中科技大学 | Real-time star point centroid location method and device based on FPGA |
CN105913401A (en) * | 2016-05-06 | 2016-08-31 | 北京信息科技大学 | Industrial camera photogrammetry image brightness compensation method |
Non-Patent Citations (3)
Title |
---|
EVELIINAJUNTUNEN等: "Smart and dynamic route lighting control based on movement tracking", 《ARXIV.ORG》 * |
TONGZHOU ZHAO等: "The Face Recognition Algorithm Based on Wavelet Stretching Transformation Pre-processing", 《2013 INTERNATIONAL CONFERENCE ON COMPUTER SCIENCES AND APPLICATIONS》 * |
董胜光等: "一种高分辨率遥感影像阴影补偿方法", 《测绘科学》 * |
Also Published As
Publication number | Publication date |
---|---|
CN109671032B (en) | 2023-04-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109872285B (en) | Retinex low-illumination color image enhancement method based on variational constraint | |
CN106204447A (en) | The super resolution ratio reconstruction method with convolutional neural networks is divided based on total variance | |
CN110119687A (en) | Detection method based on the road surface slight crack defect that image procossing and convolutional neural networks combine | |
CN107358626A (en) | A kind of method that confrontation network calculations parallax is generated using condition | |
CN104700376A (en) | Gamma correction and smoothing filtering based image histogram equalization enhancing method | |
CN103914699A (en) | Automatic lip gloss image enhancement method based on color space | |
CN109685743A (en) | Image mixed noise removing method based on noise learning neural network model | |
CN103377468A (en) | Image processing device and image processing method | |
CN110349153B (en) | Image method tobacco peduncle content detection method based on multi-feature fusion | |
CN104574328A (en) | Color image enhancement method based on histogram segmentation | |
CN111127360B (en) | Gray image transfer learning method based on automatic encoder | |
CN106127688A (en) | A kind of super-resolution image reconstruction method and system thereof | |
CN105303561A (en) | Image preprocessing grayscale space division method | |
CN110866872B (en) | Pavement crack image preprocessing intelligent selection method and device and electronic equipment | |
CN101211459A (en) | Boundary point processed histogram balancing method | |
CN109740451A (en) | Road scene image semantic segmentation method based on importance weighting | |
CN110009574A (en) | A kind of method that brightness, color adaptively inversely generate high dynamic range images with details low dynamic range echograms abundant | |
CN111882555B (en) | Deep learning-based netting detection method, device, equipment and storage medium | |
CN107610675A (en) | A kind of image processing method and device based on dynamic level | |
CN107958450A (en) | Panchromatic multispectral image fusion method and system based on adaptive Gaussian mixture model | |
CN102306378A (en) | Image enhancement method | |
Majumdar et al. | Modified CLAHE: An adaptive algorithm for contrast enhancement of aerial, medical and underwater images | |
CN104036493A (en) | No-reference image quality evaluation method based on multifractal spectrum | |
CN109003287A (en) | Image partition method based on improved adaptive GA-IAGA | |
CN108225502B (en) | A kind of truck loads ore quality estimation method and system |
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 |