CN104748865B - A kind of multi-point combination bearing calibration for infrared image - Google Patents
A kind of multi-point combination bearing calibration for infrared image Download PDFInfo
- Publication number
- CN104748865B CN104748865B CN201510145168.6A CN201510145168A CN104748865B CN 104748865 B CN104748865 B CN 104748865B CN 201510145168 A CN201510145168 A CN 201510145168A CN 104748865 B CN104748865 B CN 104748865B
- Authority
- CN
- China
- Prior art keywords
- temperature
- blackbody
- original data
- correction
- parameter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Radiation Pyrometers (AREA)
Abstract
The present invention relates to a kind of multi-point combination bearing calibration for infrared image, this method comprises the following steps：The temperature of blackbody radiation source is respectively set to T1, T2, T3, and T1<T2<T3, three groups of blackbody temperature two dimension original data sets, respectively blackbody temperature T1 two-dimentional original data set X are gathered from blackbody radiation sourceijT1, blackbody temperature T2 two-dimentional original data set XijT2, blackbody temperature T3 two-dimentional original data set XijT3, according to three two-dimentional original data set XijT1、XijT2、XijT3With three temperature value Y set in advanceT1、YT2、YT3, calculate Supplements parameter Aij、Bij、Cij, and three groups of correction parameters are stored in flash memory Flash, by the temperature value Y after correctionijN times progressive average and predetermined temperature response y are taken, calculates single-point parameter bij, finally use correction parameter Aij、Bij、Cij、bijEstablish Nonuniformity Correction equation.
Description
Technical field
The invention belongs to infrared image to correct field, is related to a kind of multi-point combination bearing calibration for infrared image, should
Method needs to build a set of hardware handles system, using PLD and digital signal processor, completes correction parameter
Calculating and infrared calibration image real-time display.
Background technology
Infrared focal plane array (IRFPA) due to the influence of material, manufacturing process, internal reading circuit and working environment,
So that each pixel of focal plane arrays (FPA) can also export the response voltage differed under the irradiation of identical radiation flux, this
The inconsistency of kind response is referred to as the heterogeneity of infrared image, also known as fixed pattern noise.In addition, with focal plane array
The change of row bias voltage and surrounding environment, the output of the same pixel of focal plane arrays (FPA) can also change, and here it is temperature drift to show
As.The heterogeneity of infrared image is an important factor for restricting infrared focal plane array development and application.It is therefore necessary to infrared
Focal plane arrays (FPA) carries out Nonuniformity Correction.
At present, infrared focal plane array heterogeneity bearing calibration can be divided into two major classes:Scaling method and base based on black matrix
In the statistic law of scene.In general, the scaling method principle based on black matrix is simple, occupancy hardware resource is few, is easy to hardware realization,
And although the statistic law based on scene has very strong adaptive ability, but its algorithm is complicated, is not suitable for hardware and handles in good time, because
This, it is use the scaling method based on black matrix in engineer applied more.And apply the precondition of blackbody calibration method to have two, a side
It is linear that face, which assumes that focal plane array is listed in certain temperature range,；On the other hand assume that focal plane arrays (FPA) has the stabilization of time
Property.
Blackbody calibration method includes single point correction, two point correction and Supplements again, and every kind of bearing calibration has been required for standard
Blackbody radiation source, and be to infrared focal plane array output initial data calculate, obtain correction parameter, it is every kind of
The correction parameter of method is separate, mutually not general, if necessary to change correction parameter, is required to re-scale parameter, original
Correction parameter be not fully utilized.
The content of the invention
The technical problem of the solution of the present invention：Overcome prior art insufficient, propose a kind of multiple spot group for infrared image
Bearing calibration is closed, on the premise of existing correction parameter is made full use of, by being combined correction to the data after correction to reach
To the effect of real time correction.Because the assumed condition of focal plane arrays (FPA) is extremely difficult to, pass through the unlimited condition of Supplements
Go to approach it, then make full use of the parameter of Supplements to be combined correction.
The present invention includes following technical scheme：A kind of multi-point combination bearing calibration for infrared image, including step is such as
Under：
(1) temperature of blackbody radiation source is respectively set to T1, T2, T3, and T1<T2<T3, from blackbody radiation source collection three
Group blackbody temperature two dimension original data set, respectively blackbody temperature T1 two-dimentional original data set XijT1, blackbody temperature T2 two
Tie up original data set XijT2, blackbody temperature T3 two-dimentional original data set XijT3, according to blackbody temperature T1 two-dimentional initial data
Group XijT1, blackbody temperature T2 two-dimentional original data set XijT2, blackbody temperature T3 two-dimentional original data set XijT3With preset
Three temperature value YT1、YT2、YT3, calculate Supplements parameter Aij、Bij、Cij, group of equations is as follows：
In formula, YT3>YT2>YT1；
(2) by the correction parameter A of step (1)ij、Bij、Cij, deposit flash memory Flash；
(3) the two-dimentional original data set X of uniform target is gatheredij, with the correction parameter A of step (2) storageij、Bij、Cij,
Calculate the temperature value Y after correctionij, formula is as follows：
(4) the temperature value Y after being corrected to step (3)ijAfter taking n times cumulative mean, average temperature value is obtained
(5) average temperature value obtained according to step (2)With the temperature-responsive value y of setting, single-point parameter b is obtainedij,
Formula is as follows：
(6) the correction parameter A of step (1) is usedij、Bij、CijWith the single-point parameter b of step (5)ij, establish nonuniformity correction side
Journey is as follows：
In formula, NijFor target scene data, i.e., non-homogeneous data, MijFor heterogeneity corresponding to the target scene data
The result of correction.
The N values that n times in the step (3) add up are 0 to 128.
The step (1) Y set in advanceT1、YT2、YT3Respectively 0 to 256.
Step (5) the predetermined temperature response y is 0 to 256.
Compared with the prior art, the invention has the advantages that：
(1) the conventional scaling method based on black matrix have be it is independent, it is mutually not general, be all the original image of focal plane
Calculating is acquired, if necessary to change correction parameter, is required to re-scale parameter, and the present invention is the base in calibrating parameters
On plinth, calculating is no longer acquired to initial data, calculating is acquired to the data after correction by step (3), so may be used
To make full use of existing calibrating parameters, method is more agile and all-purpose.
(2) present invention is carried out on the basis of Supplements, can be more by effective demarcation of black-body reference
Close to the response curve of focal plane, along with combination correction, the parameter of Supplements can be made full use of, real-time is more preferable.
(3) for the present invention compared with the bearing calibration based on scene, algorithm is relatively easy, and practicality is stronger, is more suitable for hard
Handled in good time on part platform, and scene bearing calibration is typically difficult processing in good time on a hardware platform, even if can handle, also can
Substantial amounts of hardware resource is taken, improves the power consumption of system indirectly.
(4) present invention does not need scene objects constantly to change, just energy normal work, and scene bearing calibration is to be based on sequence
Iterative estimate, it needs the continuous changes of scene objects to realize adaptive algorithm, if scene objects do not change, nothing
Method realizes the correction based on scene.
(5) the cumulative N values of the n times in step (3) of the present invention are 0 to 128, step (1) Y set in advanceT1、YT2、YT3Point
Not Wei 0 to 256, step (5) predetermined temperature response y be 0 to 256, be to be obtained by lot of experiments, wherein, N values
Can neither be too high, can not be too low, too high arithmetic speed can reduce, too low and can not eliminate error well；System operations
Digit is 8, so YT1、YT2、YT3, y takes is numerical value between 0 to 256, be advantageous to the calculating of parameter.
Brief description of the drawings
Fig. 1 is a kind of multi-point combination bearing calibration workflow diagram for infrared image of the invention；
Fig. 2 is demonstration and verification platform of the present invention；
Fig. 3 is the composition frame chart of focal plane component circuit；
Fig. 4 is the implementation process figure that Nonuniformity Correction establishes equation.
Embodiment
Just it is further described below with reference to specific implementation of the accompanying drawing to the present invention：
The present invention proposes a kind of multi-point combination bearing calibration for infrared image, and this method has corrected in abundant contrast
The advantage and disadvantage of method, by constantly summarizing and putting into practice what is drawn.In order to preferably complete the checking of combination correction method, need
A set of demonstration and verification platform is built, the platform specifically includes standard black body radiation source, optical system, focal plane component, image
Collection and processing system, its composition frame chart are as shown in Figure 2.
The parameter of multi-point combination correction is by focal plane component circuit by constantly gathering, storing, being calculated, Fig. 3
It is the schematic diagram of focal plane component circuit, it includes AD Acquisition Circuit, FPGA, DSP, FLASH are stored, CameraLink images are defeated
Go out circuit, 6 sram cache circuits, external control circuits etc., wherein FPGA is the main control chip of focal plane component circuit, is responsible for
AD collection, FLASH storages, the output of CameraLink images, SRAM read-write and with DSP communication etc., DSP is responsible for parameter
Calculating, receive external control instruction etc., SRAM1, SRAM2, SRAM3 are used for caching facing in high/low temperature data and calculating process
Variations per hour, SRAM4, SRAM5, SRAM6 are used for caching the parameter of combination correction, every time after upper electricity, system from SRAM4, SRAM5,
Correction parameter A is directly invoked in SRAM6ij、Bij、CijIt is combined correction.
The specific workflow of the present invention is as shown in Figure 1：
The temperature in standard black body radiation source is respectively set to T1 first, collection low temperature is then sent by external control module
Instruct, the two-dimentional original data set of the FPGA collection blackbody radiation sources in focal plane component circuit, and the data group that will be collected
It is stored in SRAM1；Then, the temperature in standard black body radiation source is respectively set to T2, is then sent and adopted by external control module
Collect medium temperature instruction, the two-dimentional original data sets of the FPGA collection blackbody radiation sources in focal plane component circuit, and will collect
In data group deposit SRAM2；Finally, the temperature in standard black body radiation source is respectively set to T3, then by external control module
Send collection high temperature to instruct, the two-dimentional original data set of the FPGA collection blackbody radiation sources in focal plane component circuit, and will adopt
In the data group deposit SRAM3 collected.
After the completion of three temperature acquisitions, multiple spot parameter computations, at this moment, focal plane group are sent by external control module
The DSP of part circuit controls SRAM1, SRAM2, SRAM3 respectively by FPGA, and is calculated according to equation below：
In formula, YT1、YT2、YT3It is the definite value pre-set, it is desirable to YT3>YT2>YT1, XijT1、XijT2、XijT3It is not equality of temperature
The two-dimentional original data set collected under degree, by FPGA respectively by XijT1、XijT2、XijT3Read from SRAM1, SRAM2, SRAM3
Take out, for calculating Aij、Bij、Cij.Due to the calculating for not having enough memory spaces to be used for carrying out big data quantity in DSP, lead to
The mode for crossing " one, which reads a calculation one, deposits " completes the calculating of parameter, so-called " one, which reads a calculation one, deposits ", exactly reads one group, calculates one
Group, one group is stored, this calculation is completed by DSP, and will calculate Aij、Bij、CijBe stored in respectively by FPGA SRAM4,
In SRAM5, SRAM6, after the completion of storage, the parameter in SRAM4, SRAM5, SRAM6 is stored in FLASH by FPGA respectively again, this
Sample is to not lose after ensureing parameter power down, when system is upper electric again, directly can call A from FLASHij、Bij、Cij。
When having calculated Aij、Bij、CijAfterwards, SRAM1, SRAM2, SRAM3 can free out calculated for subsequent other parameters
, b is described belowijCalculating process.
The A calculated is read from SRAM4, SRAM5, SRAM6 firstij、Bij、Cij, pass through formulaThe output of image after correcting is completed, can so pass through the data Y after FPGA acquisition correctionsij,
To the view data Y after SupplementsijN cumulative means are taken to obtain：
It is secondary cumulative to gather N (N=64), and accumulated result is put into SRAM1, while cumulative end letter is sent to DSP
Number.After DSP receives cumulative end signal, just take N (N=64) secondary average accumulation result in SRAM1 by FPGA, equally
Using the method for " one, which reads a calculation one, deposits ", by average result YijIt is put into SRAM2.
Assuming that the normal response value of single point correction is y, had according to the model of single point correction：
After average result is stored in SRAM2, correction instruction again is sent by outside control, such DSP is by FPGA from SRAM2
Middle reading average valueAnd calculated using equation below：
In formula, y is a definite value set, is still calculated in calculating process by the way of " one, which reads a calculation one, deposits "
Go out the parameter b corrected againij, and by bijIt is stored in SRAM3.
As shown in figure 4, last correction parameter is A respectivelyij、Bij、Cij、bij, by AijIt is stored in SRAM4, by BijDeposit
Storage is in SRAM5, by CijIt is stored in SRAM6, by bijIt is stored in SRAM3, by the reading to 4 groups of parameters, establishes final
Nonuniformity correction equation：
Bearing calibration is changed into Mij=Nij(AijNij+Bij)+Cij+ b, the implementation process of correction equation are as shown in Figure 4：
The access to SRAM3, SRAM4, SRAM5, SRAM6 is completed by master control module controls address decoding module first, from
Middle reading parameter Aij、Bij、Cij、bij。
The first parameter A readijWith initial data NijThe product of the two is completed by multiplier, then passes through adder
With parameter BijIt is added, then passes through multiplier and N againijBe multiplied, by the result of multiplication respectively with Cij、bijAdded up by adder,
Accumulated result is exactly heterogeneity combination correction.
It is described above, it is only the optimal embodiment of the present invention, but this bright protection domain is not limited thereto, and is appointed
How those familiar with the art is in the technical scope of this bright exposure, the variations and alternatives that can be readily occurred in, all should
It is included within the scope of the present invention.
Unspecified part of the present invention belongs to general knowledge as well known to those skilled in the art.
Claims (1)
1. a kind of multi-point combination bearing calibration for infrared image, it is characterised in that comprise the following steps：
(1) temperature of blackbody radiation source is respectively set to T1, T2, T3, and T1<T2<T3 is black from three groups of blackbody radiation source collection
The two-dimentional original data set X of temperature two dimension original data set, respectively blackbody temperature T1ijT1, blackbody temperature T2 two dimension it is former
Beginning data group XijT2, blackbody temperature T3 two-dimentional original data set XijT3, according to blackbody temperature T1 two-dimentional original data set
XijT1, blackbody temperature T2 two-dimentional original data set XijT2, blackbody temperature T3 two-dimentional original data set XijT3With it is set in advance
Three temperature value YT1、YT2、YT3, calculate Supplements parameter Aij、Bij、Cij, group of equations is as follows：
In formula, YT3>YT2>YT1, and Y set in advanceT1、YT2、YT3Respectively 0 to 256；
(2) by the correction parameter A of step (1)ij、Bij、Cij, deposit flash memory Flash；
(3) the two-dimentional original data set X of uniform target is gatheredij, with the correction parameter A of step (2) storageij、Bij、Cij, calculate
The temperature value Y gone out after correctionij, formula is as follows：
(4) the temperature value Y after being corrected to step (3)ijAfter taking n times cumulative mean, average temperature value is obtainedN values are 0 to 128；
(5) average temperature value obtained according to step (2)With the temperature-responsive value y of setting, single-point parameter b is obtainedij, formula is such as
Under：
The temperature-responsive value y set is 0 to 256；
(6) the correction parameter A of step (1) is usedij、Bij、CijWith the single-point parameter b of step (5)ij, nonuniformity correction equation is established,
It is as follows：
In formula, NijFor target scene data, i.e., non-homogeneous data, MijFor Nonuniformity Correction corresponding to the target scene data
As a result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510145168.6A CN104748865B (en) | 2015-03-30 | 2015-03-30 | A kind of multi-point combination bearing calibration for infrared image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510145168.6A CN104748865B (en) | 2015-03-30 | 2015-03-30 | A kind of multi-point combination bearing calibration for infrared image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104748865A CN104748865A (en) | 2015-07-01 |
| CN104748865B true CN104748865B (en) | 2018-01-05 |
Family
ID=53588886
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510145168.6A Active CN104748865B (en) | 2015-03-30 | 2015-03-30 | A kind of multi-point combination bearing calibration for infrared image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104748865B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106296627B (en) * | 2016-09-22 | 2019-04-16 | 电子科技大学 | A kind of heteropical method correcting infrared imaging system |
| CN109060140A (en) * | 2018-07-19 | 2018-12-21 | 中国科学院西安光学精密机械研究所 | Infrared Image Non-uniformity Correction method based on multi-point calibration and fitting |
| CN111811694B (en) * | 2020-07-13 | 2021-11-30 | 广东博智林机器人有限公司 | Temperature calibration method, device, equipment and storage medium |
| CN113514155B (en) * | 2021-04-13 | 2023-04-28 | 武汉华中数控股份有限公司 | Non-uniform correction method without shutter |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IL107828A (en) * | 1992-12-07 | 1996-09-04 | Hughes Aircraft Co | Wide dynamic range non-uniformity compensation for infrared focal plane arrays |
| CN101776487A (en) * | 2009-12-31 | 2010-07-14 | 华中科技大学 | Infrared focal plane non-uniformity fingerprint extraction and image correction method |
| CN102564598A (en) * | 2012-01-06 | 2012-07-11 | 电子科技大学 | Method for calibrating and correcting temperature measurement of infrared probe as well as corresponding temperature measuring method |
-
2015
- 2015-03-30 CN CN201510145168.6A patent/CN104748865B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IL107828A (en) * | 1992-12-07 | 1996-09-04 | Hughes Aircraft Co | Wide dynamic range non-uniformity compensation for infrared focal plane arrays |
| CN101776487A (en) * | 2009-12-31 | 2010-07-14 | 华中科技大学 | Infrared focal plane non-uniformity fingerprint extraction and image correction method |
| CN102564598A (en) * | 2012-01-06 | 2012-07-11 | 电子科技大学 | Method for calibrating and correcting temperature measurement of infrared probe as well as corresponding temperature measuring method |
Non-Patent Citations (2)
| Title |
|---|
| 基于改进多项式拟合的红外焦平面非均匀性校正方法;黄英东;《红外》;20110330;第32卷(第3期);第29-33页 * |
| 红外成像系统非均匀性校正方法;胡贵红;《中国优秀博硕士学位论文全文数据库（硕士）信息科技辑 I135-20》;20040315(第01期);正文第28-34页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104748865A (en) | 2015-07-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104748865B (en) | A kind of multi-point combination bearing calibration for infrared image | |
| CN103914860B (en) | Crystal bar location lookup table generating method and device | |
| CN104657958B (en) | A kind of infrared image fringes noise removing method | |
| CN103268594B (en) | A kind of blind element replacement method of infrared thermal imagery instrument system | |
| CN103369192A (en) | Method and device for Full-hardware splicing of multichannel video images | |
| CN103308178A (en) | Non-uniformity correction method for non-refrigeration infrared focal plane array | |
| CN110595630B (en) | Multi-point non-uniform correction method based on detector temperature | |
| CN107976255A (en) | A kind of modification method and device of infrared detector Nonuniformity Correction coefficient | |
| CN108846805A (en) | A kind of thermal-induced imagery two o'clock non-uniform correction method based on scene adaptive | |
| CN106197673A (en) | A kind of self adaptation wide temperature range non-uniform correction method and system | |
| CN105841821A (en) | Calibration-based barrier sheet-free non-uniformity correction device and method thereof | |
| CN103942807A (en) | Real-time processing fast image block fusion system and fusion method | |
| CN112393807B (en) | Infrared image processing method, device, system and computer readable storage medium | |
| CN106408518A (en) | Image fusion method and apparatus, and terminal device | |
| Sheng et al. | Calibration-based NUC method in real-time based on IRFPA | |
| CN106855435B (en) | Heterogeneity real-time correction method on long wave linear array infrared camera star | |
| CN108122280A (en) | The method for reconstructing and device of a kind of three-dimensional point cloud | |
| CN107860312A (en) | A kind of the embedded of structural light three-dimensional Measurement Algorithm realizes system and method | |
| CN106780633B (en) | A kind of method for correcting image, device and binocular vision system | |
| CN103124334A (en) | Lens distortion correction method | |
| CN103778637B (en) | A kind of in-orbit 90 ° of radiation calibration processing methods based on histogram balance | |
| CN103310446B (en) | A kind of solid matching method that instructs filtering based on iteration | |
| CN106851141B (en) | A kind of asymmetric correction method of remote sensing images | |
| CN109974857A (en) | A kind of extensive infrared focus plane Real-time Nonuniformity Correction system and method | |
| CN104537614A (en) | Orthographic correction method of CCD image of HJ-1 satellite |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |