CN108846805A - A kind of thermal-induced imagery two o'clock non-uniform correction method based on scene adaptive - Google Patents
A kind of thermal-induced imagery two o'clock non-uniform correction method based on scene adaptive Download PDFInfo
- Publication number
- CN108846805A CN108846805A CN201810411273.3A CN201810411273A CN108846805A CN 108846805 A CN108846805 A CN 108846805A CN 201810411273 A CN201810411273 A CN 201810411273A CN 108846805 A CN108846805 A CN 108846805A
- Authority
- CN
- China
- Prior art keywords
- temperature
- sur
- thermal
- induced imagery
- clock
- 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
- 238000012937 correction Methods 0.000 title claims abstract description 83
- 238000000034 method Methods 0.000 title claims abstract description 45
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 31
- 239000000523 sample Substances 0.000 claims description 23
- 239000011159 matrix material Substances 0.000 claims description 18
- 230000005457 Black-body radiation Effects 0.000 claims description 6
- 230000007547 defect Effects 0.000 abstract description 5
- 238000001931 thermography Methods 0.000 abstract description 5
- 230000001737 promoting effect Effects 0.000 abstract description 2
- 238000013528 artificial neural network Methods 0.000 description 9
- 230000005855 radiation Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000001052 transient effect Effects 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000571 coke Substances 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000001259 photo etching Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000002123 temporal effect Effects 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
-
- 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
- G06T2207/10048—Infrared image
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Radiation Pyrometers (AREA)
- Transforming Light Signals Into Electric Signals (AREA)
- Photometry And Measurement Of Optical Pulse Characteristics (AREA)
Abstract
A kind of thermal-induced imagery two o'clock non-uniform correction method based on scene adaptive is claimed in the present invention, belongs to infrared thermal imaging field.The present invention is directed to the defects of traditional two o'clock non-uniform correction method cannot be with the variation adaptive correction deviation ratio and gain coefficient of environment temperature, so that nonuniformity correction error is larger.The present invention first at different ambient temperatures, finds out the gain coefficient and deviation ratio of thermal-induced imagery two o'clock nonuniformity correction；Then show that gain coefficient and deviation ratio correspond to the expression formula of environment temperature using Polynomial Fitting Technique；The gain coefficient and deviation ratio real time correction thermal-induced imagery found out finally by expression formula, and compared with traditional two o'clock non-uniform correction method, it obtains calibration result of the invention more preferably and the residual non-uniformity of thermal-induced imagery is lower.The present invention has many advantages, such as easy to operate, and algorithm complexity is low and calibration result is good, therefore the present invention has a good application prospect and promotional value.
Description
Technical field
The invention belongs to infrared thermal imaging field, the non-homogeneous of the thermal-induced imagery based on scene adaptive is particularly belonged to
Correction process.
Background technique
Under black matrix homogeneous radiation, the output response of each probe unit of infrared focal plane array answer it is completely the same, and it is real
The output response of each probe unit has certain difference on border, shows as heterogeneity.It is this non-from the point of view of noise point
Uniformity is as caused by spatial noise and transient noise, and spatial noise is mainly by the semiconductor material of infrared focal plane array
Expect uneven (doping concentration, crystal defect and internal structure inhomogeneities etc.) and manufacture craft (exposure mask unevenness and photoetching error
Deng) influence, transient noise is mainly shown as dark current noise caused by circuit design noise and environment temperature.Usual transient state
Noise can multipass average to reduce its influence, but spatial noise cannot use the method, it is necessary to using non-homogeneous
Alignment technique reduces the influence of spatial noise to greatest extent.
On the asymmetric correction method of thermal-induced imagery, correcting algorithm common at present is divided into based on blackbody calibration
Be based on scene adaptive both of which.Algorithm complexity based on calibration is lower, can realize on existing hardware platform real
When nonuniformity correction, therefore in engineering using relatively wide.Such algorithm assumes the sound of each probe unit of infrared focal plane array
It should be linear time invariant, and be demarcated by homogeneous radiation source (black matrix), finally obtain the correction coefficient of each probe unit.
The algorithm Typical Representative：It a little corrects, two point correction and Supplements.In nineteen ninety-five, Schulz M and Caldwell L are directed to
The heterogeneity of thermal-induced imagery uses some correcting algorithms, but this algorithm only corrects and causes coke by doping unevenness
The heterogeneity of planar array does not correct heterogeneity caused by the noises such as dark current.Based on this, there has been proposed infrared
The two o'clock non-uniformity correction algorithm and multiple spot non-uniformity correction algorithm of thermal image, these algorithms reduce red to a certain extent
The heterogeneity of outer thermal image improves the quality of thermal-induced imagery.With the development and a large amount of experiment of infrared thermal imaging technique
Research shows that：Each probe unit response of focal plane arrays (FPA) is nonlinear time-varying, therefore in different environments, non-homogeneous school
Positive coefficient is different, repeatedly to be calibrated.In order to save cumbersome calibration process, Wermer Gross et al. proposes base
In scene adaptive non-uniformity correction algorithm, the main thought of such algorithm is continuously to carry out non-homogeneous school to thermal-induced imagery
Just, until reaching convergence threshold.It studies for a long period of time by expert and scholar, such algorithm is common at present has：Neural network school
It executes, temporal high pass filter method, kalman filter method scheduling algorithm.For these algorithms in the effect of nonuniformity correction, there are subtle
Difference, generally speaking such algorithm all has the advantages that adaptively correcting ability is strong and few with residual non-uniformity, still
There is also convergence rate is slow and the defects of complexity is high for the algorithm.These disadvantages seriously limit non-based on scene adaptive
Application of the even correcting algorithm in engineering practice.
Show that two o'clock non-uniform correction method cannot be with the variation adaptive modified gain system of environment according to the above analysis
The defects of several and deviation ratio, so that nonuniformity correction error is larger.Although also showing that corrected neural network algorithm can be adaptive
Modified gain and deviation ratio, but this algorithm needs a large amount of picture frame that could restrain, therefore complexity is higher, is difficult in work
The defects of being used in journey.Based on this, the invention proposes the thermal-induced imagery two o'clock nonuniformity correction sides based on scene adaptive
Method, the method, come adaptive modified gain coefficient and deviation ratio, improve thermal-induced imagery using Polynomial Fitting Technique
Correction accuracy, and then improve the quality of thermal-induced imagery, while the method has easy to operate, it is excellent that algorithm complexity is low etc.
Point, therefore the present invention has a good application prospect and promotional value.
Summary of the invention
Present invention seek to address that the above problem of the prior art.Propose a kind of quality for improving thermal-induced imagery, behaviour
Make simple, the low thermal-induced imagery two o'clock non-uniform correction method based on scene adaptive of algorithm complexity.Skill of the invention
Art scheme is as follows：
A kind of thermal-induced imagery two o'clock non-uniform correction method based on scene adaptive comprising following steps：
1), firstly, passing through the thermal-induced imagery of acquisition low temperature black matrix and high temperature blackbody, the non-homogeneous school of traditional two o'clock is established
Positive model obtains the gain coefficient G of each probe uniti,jWith deviation ratio Oi,jExpression formula；
2) it, secondly, carrying out two o'clock nonuniformity correction under n different environment temperatures, and records under corresponding environment temperature
Each probe unit gain coefficient Gi,j(Tsur) and deviation ratio Oi,j(Tsur), then respectively at a temperature of drafting varying environment
Gain coefficient Gi,j(Tsur) and deviation ratio Oi,j(Tsur) scatter plot；Then it is counted using quadratic polynomial fitting technique
According to fitting, and obtain gain coefficient Gi,j(Tsur) and deviation ratio Oi,j(Tsur) and environment temperature between expression formula；Last root
According to gain coefficient Gi,j(Tsur) and deviation ratio Oi,j(Tsur) and environment temperature between expression formula, obtain based on scene it is adaptive
The thermal-induced imagery two o'clock nonuniformity correction formula answered, and real-time nonuniformity correction is carried out according to this formula.
Further, the traditional two o'clock nonuniformity correction model of step 1) foundation is：
Yi,j=Gi,j*Xi,j+Oi,j
In formula, Xi,jAnd Yi,jIt is gray value after the sum of the grayscale values of probe unit (i, j) output corrects, G respectivelyi,jAnd Oi,j
The respectively gain coefficient and deviation ratio of probe unit (i, j).
Further, the gain coefficient G of the probe unit (i, j)i,jWith deviation ratio Oi,jThe following institute of calculating process
Show：1) blackbody temperature, is adjusted to low temperature TL, and adjust position and make black body radiation face that infrared focal plane array be completely covered, when
Blackbody temperature is stablized in TLWhen, it acquires black matrix and saves the gray value X of each pixel of thermal-induced imageryi,j(TL), and calculate this temperature
Under average gray valueIt is shown below
In formula, M and N respectively indicate the total line number and columns of thermal-induced imagery, i and j respectively indicate row where pixel and
Column.
2) blackbody temperature, is adjusted to high temperature TH, and adjust position and make black body radiation face that infrared focus plane battle array be completely covered
Column, when blackbody temperature is stablized in THWhen, it acquires black matrix and saves the gray value X of each pixel of thermal-induced imageryi,j(TH), and calculate
Average gray value at a temperature of thisIt is shown below：
3), gain coefficient Gi,jWith deviation ratio Oi,j
In formulaExpression blackbody temperature is high temperature THWhen thermal-induced imagery average gray value,Indicate black matrix
Temperature is low temperature TLWhen thermal-induced imagery average gray value, Xi,j(TH) expression blackbody temperature be high temperature THWhen thermal-induced imagery in
Pixel position is the gray value of (i, j), Xi,j(TL) expression blackbody temperature be low temperature TLWhen thermal-induced imagery in pixel position be
The gray value of (i, j), TLAnd THRespectively indicate the cryogenic temperature and high-temperature temperature of black matrix, while TLAnd THIt must be chosen to be at response
The linear zone of curve.
Further, the step 2) using quadratic polynomial fitting obtain gain coefficient at a temperature of varying environment and
Deviation ratio, and finally acquire the thermal-induced imagery two o'clock nonuniformity correction formula based on scene adaptive, the following institute of process
Show：
1) two o'clock nonuniformity correction, is carried out under n different environment temperatures, and is recorded and respectively visited under corresponding environment temperature
Survey the gain coefficient G of uniti,j(Tsur) and deviation ratio Oi,j(Tsur), T heresurRepresent environment temperature；
2), gain coefficient G at a temperature of drafting varying environmenti,j(Tsur) scatter plot, and using quadratic polynomial be fitted skill
Art carries out the fitting of data, obtains following expression：
Gi,j(Tsur)=Ai,jTsur 2+Bi,jTsur+Ci,j
3), deviation ratio O at a temperature of drafting varying environmenti,j(Tsur) scatter plot, and using quadratic polynomial be fitted skill
Art carries out the fitting of data, obtains following expression：
Oi,j(Tsur)=Di,jTsur 2+Ei,jTsur+Fi,j
2) and 3) 4), according to the gain coefficient and deviation ratio acquired under corresponding temperature, obtain non-based on scene adaptive
Uniformity correction formula, as follows：
Yi,j(T)=(Ai,jTsur 2+Bi,jTsur+Ci,j)Xi,j+(Di,jTsur 2+Ei,jTsur+Fi,j)
A in formulai,j, Bi,j, Ci,j, Di,j, Ei,j, Fi,jFor the correction coefficient of probe unit (i, j).
Further, the environment temperature for carrying out two o'clock nonuniformity correction is 6 DEG C -40 DEG C, wherein between environment temperature
It is divided into 2 DEG C.
It advantages of the present invention and has the beneficial effect that：
The innovation of the invention consists in that step 2), first by carrying out two o'clock nonuniformity correction at different ambient temperatures,
And gain coefficient and deviation ratio under corresponding environment temperature are acquired, gain coefficient is then acquired using Polynomial Fitting Technique
Expression formula between deviation ratio and environment temperature finally carries out the non-homogeneous of thermal-induced imagery using the expression formula found out
Correction.It can be seen that The present invention reduces traditional two o'clock nonuniformity corrections using fixed gain coefficient and deviation ratio
The error of method, and improve the quality of thermal-induced imagery, at the same the present invention have it is easy to operate low with algorithm complexity etc. excellent
Point can be applied in engineering and popularization.
Detailed description of the invention
Fig. 1 is that the present invention provides preferred embodiment two o'clock nonuniformity correction schematic diagram；
Fig. 2 is flow diagram of the invention；
The contrast schematic diagram of each non-uniformity correction algorithm black effects of Fig. 3；
The contrast schematic diagram of each non-uniformity correction algorithm outdoor scene effect of Fig. 4.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, detailed
Carefully describe.Described embodiment is only a part of the embodiments of the present invention.
The present invention solve above-mentioned technical problem technical solution be：
Initially set up two o'clock nonuniformity correction model (as shown in Figure 1).Then at a temperature of finding out varying environment according to model
Each probe unit gain coefficient Gi,j(Tsur) and deviation ratio Oi,j(Tsur), and found out by quadratic polynomial fitting technique
Gain coefficient Gi,j(Tsur) and deviation ratio Oi,j(Tsur) and environment temperature between expression formula.It finally obtains based on scene certainly
The thermal-induced imagery two o'clock nonuniformity correction formula (as shown in Figure 2) of adaptation.It is specifically described the implementation process of each step below
1, establishing traditional two o'clock nonuniformity correction model is：
Yi,j=Gi,j*Xi,j+Oi,j
In formula, Xi,jAnd Yi,jIt is gray value after the sum of the grayscale values of probe unit (i, j) output corrects, G respectivelyi,jAnd Oi,j
The respectively gain coefficient and deviation ratio of probe unit (i, j).Its Gi,jAnd Oi,jCalculating process is as follows：
1) blackbody temperature, is adjusted to low temperature TL, and adjust position and make black body radiation face that infrared focus plane battle array be completely covered
Column, when blackbody temperature is stablized in TLWhen, it acquires black matrix and saves the gray value X of each pixel of thermal-induced imageryi,j(TL), and calculate
Average gray value at a temperature of thisIt is shown below
In formula, M and N respectively indicate the total line number and columns of thermal-induced imagery, i and j respectively indicate row where pixel and
Column.
2) blackbody temperature, is adjusted to high temperature TH, and adjust position and make black body radiation face that infrared focus plane battle array be completely covered
Column, when blackbody temperature is stablized in THWhen, it acquires black matrix and saves the gray value X of each pixel of thermal-induced imageryi,j(TH), and calculate
Average gray value at a temperature of thisIt is shown below：
3), gain coefficient Gi,jWith deviation ratio Oi,j
Pay attention to：TLAnd THIt must be chosen to be at the linear zone of response curve, while THWith TLDifference as far as possible big so that correction
Range is wide.
2, the gain coefficient G at a temperature of varying environment is found out according to two o'clock nonuniformity correction modeli,j(Tsur) and offset system
Number Oi,j(Tsur), and gain coefficient G is found out by quadratic polynomial fitting techniquei,j(Tsur) and deviation ratio Oi,j(Tsur) with
Expression formula between environment temperature finally obtains the thermal-induced imagery two o'clock nonuniformity correction formula based on scene adaptive.Its
Process is as follows：
1) two o'clock nonuniformity correction is carried out at being, 6 DEG C -40 DEG C in environment temperature, and records and is respectively visited under corresponding environment temperature
Survey the gain coefficient G of uniti,j(Tsur) and deviation ratio Oi,j(Tsur), 2 DEG C are wherein divided between environment temperature, T heresur
Represent environment temperature.
2), gain coefficient G at a temperature of drafting varying environmenti,j(Tsur) scatter plot, and using quadratic polynomial be fitted skill
Art carries out the fitting of data, obtains following expression：
Gi,j(Tsur)=Ai,jTsur 2+Bi,jTsur+Ci,j
3), deviation ratio O at a temperature of drafting varying environmenti,j(Tsur) scatter plot, and using quadratic polynomial be fitted skill
Art carries out the fitting of data, obtains following expression：
Oi,j(Tsur)=Di,jTsur 2+Ei,jTsur+Fi,j
2) and 3) 4), according to the gain coefficient and deviation ratio that can be acquired under corresponding temperature, obtain based on scene adaptive
Nonuniformity correction formula, as follows：
Yi,j(T)=(Ai,jTsur 2+Bi,jTsur+Ci,j)Xi,j+(Di,jTsur 2+Ei,jTsur+Fi,j)
A in formulai,j, Bi,j, Ci,j, Di,j, Ei,j, Fi,jFor the correction coefficient of probe unit (i, j).
3, simulating, verifying is carried out by thermal-induced imagery
It chooses the thermal-induced imagery that resolution ratio is 640 × 512 and carries out simulating, verifying, and use the non-homogeneous school of traditional two o'clock
Normal operation method, neural network non-uniformity correction algorithm and algorithm proposed by the present invention carry out real-time nonuniformity correction.And from real-time school
Multiple dimensions such as positive speed, black matrix homogeneous radiation, residual non-uniformity and specific scenery compare the property of these three correcting algorithms
Energy.
1), the comparison of real time correction speed
The speed of real-time nonuniformity correction is to evaluate the important indicator of correcting algorithm, what the present invention used
It is 54MHZ that TMS320DM6437 video, which handles front-end configuration, and under identical experiment environment, for three kinds of non-homogeneous schools of difference
Normal operation method is compared, and comparing result is as shown in table 1
Velocity contrast's table of the different correcting algorithms of table 1
As known from Table 1, algorithm proposed by the present invention is better than neural network non-uniformity correction algorithm in real-time, simultaneously
It is not much different on display frame rate compared with two o'clock non-uniformity correction algorithm.This is because two o'clock nonuniformity correction is needed to every
A data carry out a multiplication and add operation, while algorithm proposed by the present invention needs to carry out multiple multiplication to each data
And add operation, and neural network nonuniformity correction needs the thermal-induced imagery of 1000 frames or more that can just make algorithmic statement, algorithm
Complexity is high.
2) comparison, based on each correcting algorithm of black matrix homogeneous radiation
First respectively in the black matrix thermal-induced imagery that environment temperature is 10 DEG C and 35 DEG C 70 DEG C of acquisitions, such as Fig. 3 (a) and 3
(b) shown in.Then collected two images are used into two o'clock nonuniformity correction, as shown in Fig. 3 (c) and 3 (d).Then it will adopt
The thermal-induced imagery collected uses neural network nonuniformity correction, as shown in Fig. 3 (e) and 3 (f).Finally by the infrared heat of acquisition
Image proposes that algorithm carries out nonuniformity correction using the present invention, as shown in Fig. 3 (m) and 3 (n).
As can be seen from Figure 3, all there is apparent nicking in figure (a) and figure (b), these nickings are due to infrared burnt flat
Caused by each probe unit non_uniform response in face, heterogeneity is shown as in the picture.Notice that the brightness of figure (b) is high simultaneously
In figure (a), and nicking has certain difference, this is because caused by the variation of environment temperature leads to gray scale value drift.Figure
(c) and figure (d) is using the thermal-induced imagery after two o'clock nonuniformity correction, and nicking significantly reduces, but due to gain system
Several and deviation ratio is fixed value, cannot effectively correct heterogeneity caused by temperature drift, so that scheming (c) and scheming the residual of (d)
Remaining heterogeneity is appointed so larger.Scheming (e) and figure (f) is using the thermal-induced imagery after neural network nonuniformity correction, vertical bar
Line there's almost no, only a small amount of non-uniform point, this is because neural network is by continuous iteration come modified gain coefficient
And deviation ratio, energy adaptive environment variation, but the method complexity is larger, in the higher system of requirement of real-time not
It is applicable in.Figure (m) and figure (n) they are the thermal-induced imageries after method correction proposed by the present invention, and nicking there's almost no, two
The residual non-uniformity of width image is smaller and approximate consistent, this is because gain coefficient and deviation ratio according to environment temperature not
With carrying out quadratic polynomial fitting, can adaptive environment variation, very good solution environment temperature brings the non-of thermal-induced imagery
Uniformity, while the complexity of this algorithm is smaller, can apply in the higher infra-red thermal imaging system of real-time.
3), the comparison of residual non-uniformity
The residual non-uniformity of each thermal-induced imagery can be obtained according to Fig. 3, as shown in table 2.
Table 2 passes through the thermal-induced imagery residual non-uniformity of different correcting algorithms
As seen from the above table, the thermal-induced imagery after two o'clock nonuniformity correction, residual non-uniformity have obtained preferably changing
It is kind, but the result corrected at different ambient temperatures is inconsistent.Infrared chart after algorithm proposed by the present invention correction
Picture residual non-uniformity is smaller and relatively stable at different ambient temperatures, this is because gain coefficient and deviation ratio can be certainly
Adapt to the variation of environment.Although neural network non-uniformity correction algorithm has smaller residual non-uniformity and good stability
The advantages that, but the complexity of this algorithm is higher, it can not be using in the high infra-red thermal imaging system of requirement of real-time.
4) comparison, based on each correcting algorithm of outdoor scene
By the advantage and disadvantage of more each correcting algorithm of thermal-induced imagery of the above black matrix homogeneous radiation, and after combination correction
The residual non-uniformity of thermal-induced imagery, which demonstrates algorithm proposed by the present invention, has calibration result good and energy adaptive environment
The advantages that variation.Next the advantage and disadvantage of the more intuitive more each correcting algorithm of specific outdoor scene will be directed to, as shown in figure 4, point
It Biao Shi be not 10 DEG C and 35 DEG C of original thermal-induced imagery in environment temperature, the thermal-induced imagery after traditional two point correction, this hair
Thermal-induced imagery after bright correction.
As can be seen from Figure 4, larger in Fig. 4 (c) and the remaining heterogeneity of Fig. 4 (d) and distribution has a certain difference, such as
Residual non-uniformity is greater than the residual non-uniformity on the left side Fig. 4 (c) on the right of Fig. 4 (d).It is remaining in Fig. 4 (e) and Fig. 4 (f)
Heterogeneity is less and distribution is substantially similar.This is because two o'clock nonuniformity correction is using fixed gain coefficient and offset system
Number, can not correct heterogeneity caused by temperature drift, but the present invention propose the gain coefficient of algorithm and deviation ratio be by
Environment temperature fitting generates, can adaptive environment variation.
The above embodiment is interpreted as being merely to illustrate the present invention rather than limit the scope of the invention.
After the content for having read record of the invention, technical staff can be made various changes or modifications the present invention, these are equivalent
Variation and modification equally fall into the scope of the claims in the present invention.
Claims (5)
1. a kind of thermal-induced imagery two o'clock non-uniform correction method based on scene adaptive, which is characterized in that including following step
Suddenly：
1), firstly, passing through the thermal-induced imagery of acquisition low temperature black matrix and high temperature blackbody, traditional two o'clock nonuniformity correction mould is established
Type obtains the gain coefficient G of each probe uniti,jWith deviation ratio Oi,jExpression formula；
2) it, secondly, carrying out two o'clock nonuniformity correction under n different environment temperatures, and records each under corresponding environment temperature
The gain coefficient G of probe uniti,j(Tsur) and deviation ratio Oi,j(Tsur), gain system at a temperature of varying environment is then drawn respectively
Number Gi,j(Tsur) and deviation ratio Oi,j(Tsur) scatter plot；Then data fitting is carried out using quadratic polynomial fitting technique,
And obtain gain coefficient Gi,j(Tsur) and deviation ratio Oi,j(Tsur) and environment temperature between expression formula；Finally according to gain system
Number Gi,j(Tsur) and deviation ratio Oi,j(Tsur) and environment temperature between expression formula, obtain the infrared heat based on scene adaptive
Image two o'clock nonuniformity correction formula, and real-time nonuniformity correction is carried out according to this formula.
2. the thermal-induced imagery two o'clock non-uniform correction method according to claim 1 based on scene adaptive, feature
It is, the step 1) establishes traditional two o'clock nonuniformity correction model and is：
Yi,j=Gi,j*Xi,j+Oi,j
In formula, Xi,jAnd Yi,jIt is gray value after the sum of the grayscale values of probe unit (i, j) output corrects, G respectivelyi,jAnd Oi,jRespectively
The gain coefficient and deviation ratio of probe unit (i, j).
3. the thermal-induced imagery two o'clock non-uniform correction method according to claim 2 based on scene adaptive, feature
It is, the gain coefficient G of the probe unit (i, j)i,jWith deviation ratio Oi,jCalculating process it is as follows：1), by black matrix
Temperature is adjusted to low temperature TL, and adjust position and make black body radiation face that infrared focal plane array be completely covered, when blackbody temperature is stablized
In TLWhen, it acquires black matrix and saves the gray value X of each pixel of thermal-induced imageryi,j(TL), and calculate the average gray at a temperature of this
ValueIt is shown below
In formula, M and N respectively indicate the total line number and columns of thermal-induced imagery, and i and j respectively indicate the row and column where pixel；
2) blackbody temperature, is adjusted to high temperature TH, and adjust position and make black body radiation face that infrared focal plane array be completely covered, when
Blackbody temperature is stablized in THWhen, it acquires black matrix and saves the gray value X of each pixel of thermal-induced imageryi,j(TH), and calculate this temperature
Under average gray valueIt is shown below：
3), gain coefficient Gi,jWith deviation ratio Oi,j
In formulaExpression blackbody temperature is high temperature THWhen thermal-induced imagery average gray value,Indicate that blackbody temperature is
Low temperature TLWhen thermal-induced imagery average gray value, Xi,j(TH) expression blackbody temperature be high temperature THWhen thermal-induced imagery in pixel position
It is set to the gray value of (i, j), Xi,j(TL) expression blackbody temperature be low temperature TLWhen thermal-induced imagery in pixel position be (i, j) ash
Angle value, TLAnd THRespectively indicate the cryogenic temperature and high-temperature temperature of black matrix, while TLAnd THIt must be chosen to be at the linear of response curve
Area.
4. the thermal-induced imagery two o'clock non-uniform correction method described in one of -3 based on scene adaptive according to claim 1,
It is characterized in that, the step 2) show that the gain coefficient at a temperature of varying environment is with offset using quadratic polynomial fitting
Number, and the thermal-induced imagery two o'clock nonuniformity correction formula based on scene adaptive is finally acquired, process is as follows：
1) two o'clock nonuniformity correction, is carried out under n different environment temperatures, and records each probe unit under corresponding environment temperature
Gain coefficient Gi,j(Tsur) and deviation ratio Oi,j(Tsur), T heresurRepresent environment temperature；
2), gain coefficient G at a temperature of drafting varying environmenti,j(Tsur) scatter plot, and using quadratic polynomial fitting technique into
The fitting of row data, obtains following expression：
Gi,j(Tsur)=Ai,jTsur 2+Bi,jTsur+Ci,j
3), deviation ratio O at a temperature of drafting varying environmenti,j(Tsur) scatter plot, and using quadratic polynomial fitting technique into
The fitting of row data, obtains following expression：
Oi,j(Tsur)=Di,jTsur 2+Ei,jTsur+Fi,j
2) and 3) 4), according to the gain coefficient and deviation ratio acquired under corresponding temperature, obtain non-homogeneous based on scene adaptive
Updating formula, as follows：
Yi,j(T)=(Ai,jTsur 2+Bi,jTsur+Ci,j)Xi,j+(Di,jTsur 2+Ei,jTsur+Fi,j)
A in formulai,j, Bi,j, Ci,j, Di,j, Ei,j, Fi,jFor the correction coefficient of probe unit (i, j).
5. the thermal-induced imagery two o'clock non-uniform correction method according to claim 4 based on scene adaptive, feature
It is, the environment temperature for carrying out two o'clock nonuniformity correction is 6 DEG C -40 DEG C, and 2 DEG C are wherein divided between environment temperature.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810411273.3A CN108846805B (en) | 2018-05-02 | 2018-05-02 | Infrared thermal image two-point non-uniform correction method based on scene self-adaption |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810411273.3A CN108846805B (en) | 2018-05-02 | 2018-05-02 | Infrared thermal image two-point non-uniform correction method based on scene self-adaption |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108846805A true CN108846805A (en) | 2018-11-20 |
| CN108846805B CN108846805B (en) | 2021-12-17 |
Family
ID=64212530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810411273.3A Active CN108846805B (en) | 2018-05-02 | 2018-05-02 | Infrared thermal image two-point non-uniform correction method based on scene self-adaption |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108846805B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109655162A (en) * | 2018-11-30 | 2019-04-19 | 诺仪器（中国）有限公司 | Thermal infrared imager temperature measurement correction system and method |
| CN109798982A (en) * | 2019-03-07 | 2019-05-24 | 杭州新瀚光电科技有限公司 | A kind of no baffle thermal imaging system and its thermometric correction algorithm |
| CN110260991A (en) * | 2019-06-06 | 2019-09-20 | 武汉高德智感科技有限公司 | A kind of method and device of adaptive acquisition temperature drift compensation data amount |
| CN110487412A (en) * | 2019-08-14 | 2019-11-22 | 北京环境特性研究所 | Infrared high spectrum image non-uniform correction method, device and computer equipment |
| CN110850500A (en) * | 2019-11-24 | 2020-02-28 | 北京长峰科威光电技术有限公司 | Infrared image multi-section single-point correction parameter correction method |
| CN111369552A (en) * | 2020-03-13 | 2020-07-03 | 烟台艾睿光电科技有限公司 | Infrared blind pixel detection method and device and computer readable storage medium |
| CN111562012A (en) * | 2020-05-22 | 2020-08-21 | 烟台艾睿光电科技有限公司 | Infrared image non-uniformity correction method and system |
| CN112710397A (en) * | 2020-12-16 | 2021-04-27 | 电子科技大学 | Two-point correction method and system based on temperature substitution |
| CN112752041A (en) * | 2019-10-31 | 2021-05-04 | 合肥美亚光电技术股份有限公司 | CMOS image sensor correction method, system and image processing equipment |
| CN113096041A (en) * | 2021-04-08 | 2021-07-09 | 杭州海康消防科技有限公司 | Image correction method and device |
| CN113252180A (en) * | 2021-05-20 | 2021-08-13 | 浙江宇松科技有限公司 | Temperature calibration method for infrared temperature measurement system and infrared temperature measurement system |
| CN113421220A (en) * | 2021-05-11 | 2021-09-21 | 武汉博宇光电系统有限责任公司 | Method for removing pot cover by infrared image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050247867A1 (en) * | 2004-05-06 | 2005-11-10 | Matthlas Volgt | System and methods for determining nonuniformity correction parameters in detector-array imaging |
| CN101515987A (en) * | 2008-12-30 | 2009-08-26 | 中国资源卫星应用中心 | Method for radiometric correction of remote sensing image taken by rotary scan multiple parallel-scan infrared camera |
| US20130112848A1 (en) * | 2011-11-07 | 2013-05-09 | The Johns Hopkins University | Flexible Readout and Signal Processing in a Computational Sensor Array |
| CN106500846A (en) * | 2016-09-22 | 2017-03-15 | 电子科技大学 | A kind of asymmetric correction method of infrared imaging system |
| CN106768383A (en) * | 2017-01-21 | 2017-05-31 | 浙江红相科技股份有限公司 | A kind of automatic blind element detection of infrared focal plane array and compensation method |
-
2018
- 2018-05-02 CN CN201810411273.3A patent/CN108846805B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050247867A1 (en) * | 2004-05-06 | 2005-11-10 | Matthlas Volgt | System and methods for determining nonuniformity correction parameters in detector-array imaging |
| CN101515987A (en) * | 2008-12-30 | 2009-08-26 | 中国资源卫星应用中心 | Method for radiometric correction of remote sensing image taken by rotary scan multiple parallel-scan infrared camera |
| US20130112848A1 (en) * | 2011-11-07 | 2013-05-09 | The Johns Hopkins University | Flexible Readout and Signal Processing in a Computational Sensor Array |
| CN106500846A (en) * | 2016-09-22 | 2017-03-15 | 电子科技大学 | A kind of asymmetric correction method of infrared imaging system |
| CN106768383A (en) * | 2017-01-21 | 2017-05-31 | 浙江红相科技股份有限公司 | A kind of automatic blind element detection of infrared focal plane array and compensation method |
Non-Patent Citations (3)
| Title |
|---|
| 代少升等: "基于非线性响应的红外焦平面阵列非均匀性自适应校正算法", 《光学精密工程》 * |
| 徐文文: "二次曲线拟合实现红外图像非均匀性校正", 《光电技术应用》 * |
| 樊凡: "基于场景的红外非均匀性校正算法研究", 《中国博士学位论文全文数据库信息科技辑》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109655162A (en) * | 2018-11-30 | 2019-04-19 | 诺仪器（中国）有限公司 | Thermal infrared imager temperature measurement correction system and method |
| CN109798982A (en) * | 2019-03-07 | 2019-05-24 | 杭州新瀚光电科技有限公司 | A kind of no baffle thermal imaging system and its thermometric correction algorithm |
| CN110260991A (en) * | 2019-06-06 | 2019-09-20 | 武汉高德智感科技有限公司 | A kind of method and device of adaptive acquisition temperature drift compensation data amount |
| CN110487412A (en) * | 2019-08-14 | 2019-11-22 | 北京环境特性研究所 | Infrared high spectrum image non-uniform correction method, device and computer equipment |
| CN112752041A (en) * | 2019-10-31 | 2021-05-04 | 合肥美亚光电技术股份有限公司 | CMOS image sensor correction method, system and image processing equipment |
| CN110850500A (en) * | 2019-11-24 | 2020-02-28 | 北京长峰科威光电技术有限公司 | Infrared image multi-section single-point correction parameter correction method |
| CN110850500B (en) * | 2019-11-24 | 2022-02-08 | 北京长峰科威光电技术有限公司 | Infrared image multi-section single-point correction parameter correction method |
| CN111369552A (en) * | 2020-03-13 | 2020-07-03 | 烟台艾睿光电科技有限公司 | Infrared blind pixel detection method and device and computer readable storage medium |
| CN111369552B (en) * | 2020-03-13 | 2023-07-14 | 烟台艾睿光电科技有限公司 | Infrared blind pixel detection method and device and computer readable storage medium |
| CN111562012A (en) * | 2020-05-22 | 2020-08-21 | 烟台艾睿光电科技有限公司 | Infrared image non-uniformity correction method and system |
| CN111562012B (en) * | 2020-05-22 | 2021-09-03 | 烟台艾睿光电科技有限公司 | Infrared image non-uniformity correction method and system |
| CN112710397A (en) * | 2020-12-16 | 2021-04-27 | 电子科技大学 | Two-point correction method and system based on temperature substitution |
| CN113096041A (en) * | 2021-04-08 | 2021-07-09 | 杭州海康消防科技有限公司 | Image correction method and device |
| CN113421220A (en) * | 2021-05-11 | 2021-09-21 | 武汉博宇光电系统有限责任公司 | Method for removing pot cover by infrared image |
| CN113421220B (en) * | 2021-05-11 | 2022-03-15 | 武汉博宇光电系统有限责任公司 | Method for removing pot cover by infrared image |
| CN113252180A (en) * | 2021-05-20 | 2021-08-13 | 浙江宇松科技有限公司 | Temperature calibration method for infrared temperature measurement system and infrared temperature measurement system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108846805B (en) | 2021-12-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108846805A (en) | A kind of thermal-induced imagery two o'clock non-uniform correction method based on scene adaptive | |
| CN107255521B (en) | A kind of Infrared Image Non-uniformity Correction method and system | |
| CN106197673B (en) | A kind of adaptive wide temperature range non-uniform correction method and system | |
| CN108871588A (en) | A kind of infrared imaging system various dimensions joint asymmetric correction method | |
| CN105841821B (en) | The Nonuniformity Correction devices and methods therefor without baffle based on calibration | |
| CN106197690B (en) | Image calibrating method and system under the conditions of a kind of wide temperature range | |
| CN109060140A (en) | Infrared Image Non-uniformity Correction method based on multi-point calibration and fitting | |
| CN106855435B (en) | Heterogeneity real-time correction method on long wave linear array infrared camera star | |
| CN102778296A (en) | Total variation-based self-adaptation non-uniformity correction method for infrared focal plane | |
| CN109813442A (en) | A kind of internal stray radiation asymmetric correction method based on multi-frame processing | |
| CN109903235A (en) | A kind of removing method of infrared image fringes noise | |
| CN109738072A (en) | A kind of cross blind element detection of InGaAs short-wave infrared imager and means for correcting and method | |
| CN105466566A (en) | An infrared nonuniformity correction real time compensation method | |
| CN108665425A (en) | Infrared Image Non-uniformity Correction method based on interframe registration and adaptive step | |
| US20190158823A1 (en) | Measurement method for measuring display panel and apparatus thereof | |
| CN106846275B (en) | A kind of real-time removing method of Infrared video image strip noise | |
| CN204286600U (en) | A kind of heteropical module of two point correction thermal infrared imager | |
| CN105092043B (en) | A kind of asymmetric correction method of the change time of integration based on scene | |
| CN110361094B (en) | Non-uniformity correction method and device for staring type focal plane array | |
| CN108226059A (en) | A kind of satellite EO-1 hyperion CO2The in-orbit Calibration Method of survey meter | |
| CN110363714A (en) | The asymmetric correction method based on scene interframe registration of adjusting learning rate | |
| WO2019183843A1 (en) | Interframe registration and adaptive step size-based non-uniformity correction method for infrared image | |
| CN106815820B (en) | A kind of infrared image strip noise cancellation method | |
| CN103868601B (en) | The bilateral full variational regularization bearing calibration of the non-homogeneous response of IRFPA detector | |
| CN110553739B (en) | Non-barrier-piece non-uniformity correction method for infrared thermal imaging |
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 |