CN114140354A - A real-time method for removing vertical streaks from infrared images using improved local moment matching - Google Patents
A real-time method for removing vertical streaks from infrared images using improved local moment matching Download PDFInfo
- Publication number
- CN114140354A CN114140354A CN202111426106.4A CN202111426106A CN114140354A CN 114140354 A CN114140354 A CN 114140354A CN 202111426106 A CN202111426106 A CN 202111426106A CN 114140354 A CN114140354 A CN 114140354A
- Authority
- CN
- China
- Prior art keywords
- image
- infrared
- column
- local
- moment matching
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000012937 correction Methods 0.000 claims abstract description 8
- 230000008901 benefit Effects 0.000 claims abstract description 4
- 230000000903 blocking effect Effects 0.000 claims abstract description 3
- 238000001914 filtration Methods 0.000 claims abstract description 3
- 230000000694 effects Effects 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000003672 processing method Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 5
- 230000005855 radiation Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 3
- 238000007796 conventional method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000005316 response function Methods 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- 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)
- Image Processing (AREA)
Abstract
The invention discloses a method for removing vertical stripes of infrared images in real time by improving local moment matching, which is suitable for products of gun-type sights and gun-type sights of uncooled infrared detectors, and in terms of hardware, in infrared core hardware of the uncooled infrared detector, the uncooled infrared detector has the advantages that the image acquisition frame frequency is 50Hz, and according to the video frequency 25Hz which can be identified by human eyes, the time domain algorithm processing advantage is carried out by combining infrared images under the condition of not influencing the finally output observation video frame rate, and the image acquisition frame frequency is determined to be 50Hz; the infrared movement hardware has a blocking piece correction function and can correct and compensate the non-uniformity of the infrared image; the infrared focal plane detector adopts the same column and shares one amplifier; compared with the prior art, the method disclosed by the invention is used for filtering the vertical stripe noise of the infrared detector in real time on the movement assembly of the uncooled infrared detector, so that the infrared image detection capability is improved.
Description
Technical Field
The invention relates to the technical field of infrared imaging, in particular to a method for removing vertical stripes of an infrared image in real time by improving local moment matching.
The algorithm is suitable for products of guns and cannon sights of uncooled infrared detectors, and by means of a method for removing vertical stripes of infrared images through local matching, stripe-shaped non-uniform noise caused by the fact that the infrared detectors are not completely consistent in response characteristics of detector units in an infrared focal plane array due to the limitation of the existing manufacturing process level and materials can be automatically removed in real time, and therefore the infrared detection capability is improved.
Background
Uncooled infrared detectors have been widely used in sights and other viewing instruments. However, due to the limitations of the existing manufacturing process level and materials, the infrared focal plane adopts an integration form, and the infrared focal plane detector contains a plurality of amplifiers, but usually in order to save cost, the same row or the same column output shares one amplifier (the detector adopted by the hardware of the invention shares one amplifier in the same column direction), that is, the integrated current of each column of pixels corresponds to the same bias voltage VFID, because the bias voltage has noise, the gate voltages obtained by the gates of the FETs are not the same when the two columns are integrated, which causes that even if the same infrared radiation is obtained, the integrated current ip flowing through each column of pixels is different, which is the root cause of stripe noise in the infrared image output by the infrared detector. The detection photosensitive elements in different columns generate different output signals for the same infrared radiation, so that stripe-shaped non-uniform noise appears in the infrared image. The vertical stripes of the image can seriously affect the use of products and are not beneficial to the identification of an observer to a target, so that the vertical stripes of the infrared image need to be removed in real time. The conventional method for removing vertical stripes mainly takes a method of moment matching as a main method, however, the conventional method for moment matching introduces a 'banding effect' as shown in fig. 1 below, and in order to solve the 'banding effect', the invention proposes an improved method for local moment matching to remove the vertical stripes, and simultaneously greatly eliminate the 'banding effect'.
Disclosure of Invention
Aiming at the defect that the conventional traditional moment matching method is used for processing the vertical stripes of the infrared image, the invention aims to provide a method for removing the vertical stripes of the infrared image in real time by improving local moment matching.
In order to achieve the purpose, the technical scheme of the invention is as follows: a method for removing vertical stripes of infrared images in real time by improving local moment matching is suitable for products of gun type sighting devices and gun type sighting devices of uncooled infrared detectors, and comprises the following hardware aspects:
in infrared core hardware of the uncooled infrared detector, the image acquisition frame frequency is 50Hz, and according to the video frequency 25Hz recognizable to human eyes, the uncooled infrared detector combines the advantage of time domain algorithm processing of infrared images under the condition of not influencing the final output observation video frame rate and determines the uncooled infrared detector to be 50Hz;
the infrared movement hardware has a blocking piece correction function and can correct and compensate the non-uniformity of the infrared image; the infrared focal plane detector adopts the same column and shares one amplifier;
the original data acquired by the 14-bit AD is used as the original data for image processing, so that the image details can be effectively reserved, and the analysis of the original data of the infrared image is facilitated;
in the aspect of software, the traditional moment matching is improved, the moment matching method of a local window is used for filtering vertical stripe noise, the effect of removing the vertical stripe noise of an image by the moment matching method can be ensured, the complex environment image is responded, and the 'banding effect' is eliminated;
in combination with the practical application scene, the column mean distribution between the columns adjacent to the local window is approximately equal, especially the column mean is more approximately uniform when the local window is smaller, so the method adopts the adjacent two columns as the reference column, namely the first two columns of the current column, effectively avoids the phenomenon of gray level distortion of the image caused by improper selection of the reference column, particularly filters the image for the following analysis,
let the image window size be R, then there are: n belongs to [1, row/R ], image data X (i, j) are traversed twice before and after the image is subjected to basic non-uniform correction, and a final output image gray value Y (i, j) is calculated according to the following steps, wherein i, j is a row-column index, delta (n, j), u (n, j) is a local column mean square error and expectation of a jth column, row is a total row number of the infrared image, n is an nth window for dividing the image into row/R local windows, and the specific steps are as follows:
1) determining an n value by traversing the i index of the data X (i, j), wherein the value of the n value is obtained according to the following formula:
2) calculating the local mean and variance of the nth window according to the following formula:
3) and traversing the data X (i, j) again, and calculating the final output according to the following processing:
when j is greater than 2, the calculation formula is:
eliminating the 'banding effect' caused by selecting a fixed reference column by a moment matching method of a local window, and effectively eliminating the vertical stripe noise;
when j is less than 2, the calculation formula is:
according to the processing method of the two formulas, the first two columns at the edge of the image can be effectively processed.
Drawings
The following describes in further detail embodiments of the present invention with reference to the accompanying drawings.
FIG. 1 shows the banding effect caused before and after the conventional moment matching method processes the image (the left side is the image before processing, and the right side is the image after processing).
FIG. 2 shows the image contrast before and after processing by the improved local moment matching method of the present invention (left side is the image before processing, right side is the image after processing).
Detailed Description
FIG. 1 is a schematic diagram of a matrix matching method for processing vertical stripes according to the following process in a conventional manner:
the vertical stripe noise is mainly caused by the fact that response functions of all detection units in a spectral response area are inconsistent, the vertical stripe noise is generally distributed on an original image and has no periodicity, when a local table signal is weak, the noise is more obvious, and the noise is generally superposed on the signal, so that the noise is separated according to statistical results of the image on different scales;
assuming that the scene detected by each detector is ideally in the same equalized radiation distribution, the variation of the data recorded on the detector is linear with the gain coefficient and the offset coefficient of the radiation correction and has shift invariance. If let Ci be the ith detection unit, the spectral response function of Ci can be expressed as:
where Yi represents an output value of Ci, i.e., a gradation value of an image; x represents the radiation value received by the system, i.e. the real scene value; gi represents a gain coefficient; oi represents a bias coefficient; ε i represents random noise. In the case of high snr, the influence of ei can be neglected, (1) formula can be written as:
as can be seen from the equation (2), when the radiation intensity X is the same, if the gain coefficient Gi and the offset coefficient Oi take different values, the obtained gray value Yi is also different, which causes the appearance of a vertical stripe image, and therefore, if the gray value Yi is normalized to the same value, the vertical stripe noise can be effectively eliminated;
under the premise of assuming uniform scene, the mean and variance of the incident radiation intensity of each row are approximately equal, the matrix matching method is to select a certain row as a reference, and then the mean and variance of other detector rows are respectively adjusted to the radiance of the reference row according to the formula (2):
in the formula, Xi and Yi respectively represent the gray values of the pixels in the ith column in the image before and after correction; μ r, σ r represent mean and standard deviation of the reference row, respectively; μ i, σ i represent mean and standard deviation, respectively, of the ith row.
According to the algorithm, the moment matching method has a good effect of removing the image stripe noise, but is based on the condition that the image scene distribution is uniform. Under the condition of uneven gray distribution caused by complex scenes or rich image contents, the moment matching method usually generates a 'banding effect', and the basic reason for the generation of the phenomenon is that all column mean values of an image are approximately equal after the moment matching method, the distribution condition of the image column mean values is changed, so that the approximate distribution of the image column mean values is a straight line, and the gray distribution condition of the straight line is difficult to achieve due to the fact that the image is generally impossible to be uniformly distributed, so if the reference column is selected unreasonably, the gray information of an original image is changed, the phenomenon of gray distortion occurs, and the 'banding effect' occurs, as shown in figure 1.
FIG. 2 is an embodiment of the present invention, disclosing a method for removing vertical stripes in an infrared image in real time by improving local moment matching,
let the image window size be R =16, then: n belongs to [1, row/16], image data X (i, j) are traversed twice before and after the image is subjected to basic non-uniform correction, and a final output image gray value Y (i, j) is calculated according to the following steps, wherein i, j is a row-column index, delta (n, j), u (n, j) is a local column mean square error and expectation of a jth column, row is a total row number of the infrared image, n is an nth window for dividing the image into row/R local windows, and the specific steps are as follows:
1) determining an n value by traversing the i index of the data X (i, j), wherein the value of the n value is obtained according to the following formula:
2) calculating the local mean and variance of the nth window according to the following formula:
3) and traversing the data X (i, j) again, and calculating the final output according to the following processing:
when j is greater than 2, the calculation formula is:
eliminating the 'banding effect' caused by selecting a fixed reference column by a moment matching method of a local window, and effectively eliminating the vertical stripe noise;
when j is less than 2, the calculation formula is:
the processing method according to the above two formulas can effectively process the first two columns at the edge of the image, and the "banding effect" in the image shown in fig. 2 is basically eliminated.
The above description is only a preferred embodiment of the present invention, and the above specific embodiments are not intended to limit the present invention, and modifications, modifications or equivalents thereof, which may occur to those skilled in the art, are included within the scope of the present invention.
Claims (1)
1. A method for removing the vertical stripes of infrared images in real time by improving local moment matching is suitable for the products of gun-type sights and gun-type sights of uncooled infrared detectors, and is characterized in that:
in terms of hardware, in infrared core hardware of the uncooled infrared detector, the image acquisition frame frequency is 50Hz, and according to the video frequency 25Hz recognizable to human eyes, the uncooled infrared detector combines the advantage of time domain algorithm processing of infrared images under the condition of not influencing the finally output observation video frame rate and determines the uncooled infrared detector to be 50Hz;
the infrared movement hardware has a blocking piece correction function and can correct and compensate the non-uniformity of the infrared image; the infrared focal plane detector adopts the same column and shares one amplifier;
the original data acquired by the 14-bit AD is used as the original data for image processing, so that the image details can be effectively reserved, and the analysis of the original data of the infrared image is facilitated;
in the aspect of software, the traditional moment matching is improved, the moment matching method of a local window is used for filtering vertical stripe noise, the effect of removing the vertical stripe noise of an image by the moment matching method can be ensured, the complex environment image is responded, and the 'banding effect' is eliminated;
in combination with the practical application scene, the column mean distribution between the columns adjacent to the local window is approximately equal, especially the column mean is more approximately uniform when the local window is smaller, so the method adopts the adjacent two columns as the reference column, namely the first two columns of the current column, effectively avoids the phenomenon of gray level distortion of the image caused by improper selection of the reference column, particularly filters the image for the following analysis,
let the image window size be R, then there are: n belongs to [1, row/R ], image data X (i, j) are traversed twice before and after the image is subjected to basic non-uniform correction, and a final output image gray value Y (i, j) is calculated according to the following steps, wherein i, j is a row-column index, delta (n, j), u (n, j) is a local column mean square error and expectation of a jth column, row is a total row number of the infrared image, n is an nth window for dividing the image into row/R local windows, and the specific steps are as follows:
1) determining an n value by traversing the i index of the data X (i, j), wherein the value of the n value is obtained according to the following formula:
2) calculating the local mean and variance of the nth window according to the following formula:
3) and traversing the data X (i, j) again, and calculating the final output according to the following processing:
when j is greater than 2, the calculation formula is:
eliminating the 'banding effect' caused by selecting a fixed reference column by a moment matching method of a local window, and effectively eliminating the vertical stripe noise;
when j is less than 2, the calculation formula is:
according to the processing method of the two formulas, the first two columns at the edge of the image can be effectively processed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111426106.4A CN114140354A (en) | 2021-11-27 | 2021-11-27 | A real-time method for removing vertical streaks from infrared images using improved local moment matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111426106.4A CN114140354A (en) | 2021-11-27 | 2021-11-27 | A real-time method for removing vertical streaks from infrared images using improved local moment matching |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114140354A true CN114140354A (en) | 2022-03-04 |
Family
ID=80388151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111426106.4A Pending CN114140354A (en) | 2021-11-27 | 2021-11-27 | A real-time method for removing vertical streaks from infrared images using improved local moment matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114140354A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN104200430A (en) * | 2014-08-20 | 2014-12-10 | 北京空间机电研究所 | Image region stray light eliminating device based on moment matching and method thereof |
| CN104580937A (en) * | 2015-01-21 | 2015-04-29 | 中国科学院上海技术物理研究所 | Stripe noise removal method for infrared imaging system |
| US20150319387A1 (en) * | 2014-04-30 | 2015-11-05 | Ulis | Method of infrared image processing for non-uniformity correction |
| CN106780403A (en) * | 2017-01-19 | 2017-05-31 | 中国科学院上海技术物理研究所 | A kind of push-broom type thermal infrared high-spectrum remote sensing asymmetric correction method |
| CN108830808A (en) * | 2018-06-01 | 2018-11-16 | 北京航空航天大学 | Infrared image fringes noise minimizing technology on star based on the compensation of similar linear window mean value |
| WO2020224518A1 (en) * | 2019-05-06 | 2020-11-12 | 南京理工大学 | Self-adaptive infrared image defringe algorithm based on local median histogram |
-
2021
- 2021-11-27 CN CN202111426106.4A patent/CN114140354A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| US20150319387A1 (en) * | 2014-04-30 | 2015-11-05 | Ulis | Method of infrared image processing for non-uniformity correction |
| CN104200430A (en) * | 2014-08-20 | 2014-12-10 | 北京空间机电研究所 | Image region stray light eliminating device based on moment matching and method thereof |
| CN104580937A (en) * | 2015-01-21 | 2015-04-29 | 中国科学院上海技术物理研究所 | Stripe noise removal method for infrared imaging system |
| CN106780403A (en) * | 2017-01-19 | 2017-05-31 | 中国科学院上海技术物理研究所 | A kind of push-broom type thermal infrared high-spectrum remote sensing asymmetric correction method |
| CN108830808A (en) * | 2018-06-01 | 2018-11-16 | 北京航空航天大学 | Infrared image fringes noise minimizing technology on star based on the compensation of similar linear window mean value |
| WO2020224518A1 (en) * | 2019-05-06 | 2020-11-12 | 南京理工大学 | Self-adaptive infrared image defringe algorithm based on local median histogram |
Non-Patent Citations (1)
| Title |
|---|
| 康一飞: "资源一号０２Ｃ影像条带噪声去除 的改进矩匹配方法", 《武汉大学学报· 信息科学版》, vol. 40, no. 12, 30 November 2015 (2015-11-30), pages 1582 - 1587 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110211056B (en) | Adaptive infrared image stripping algorithm based on local median histogram | |
| KR101955498B1 (en) | Infrared image correction apparatus using neural network structure and method thereof | |
| CN106197690B (en) | Image calibrating method and system under the conditions of a kind of wide temperature range | |
| Wang | Stripe noise removal for infrared image by minimizing difference between columns | |
| CN109813442B (en) | Multi-frame processing-based internal stray radiation non-uniformity correction method | |
| CN109934790A (en) | Nonuniformity Correction Method for Infrared Imaging System with Adaptive Threshold | |
| WO2019183843A1 (en) | Interframe registration and adaptive step size-based non-uniformity correction method for infrared image | |
| CN111080561A (en) | Time domain high-pass filtering method | |
| Lee et al. | Dual-branch structured de-striping convolution network using parametric noise model | |
| Li et al. | Scene-based nonuniformity correction based on bilateral filter with reduced ghosting | |
| Liu et al. | Interframe phase-correlated registration scene-based nonuniformity correction technology | |
| Li et al. | Infrared scene-based non-uniformity correction based on deep learning model | |
| Li et al. | Interior radiation noise reduction method based on multiframe processing in infrared focal plane arrays imaging system | |
| CN114140354A (en) | A real-time method for removing vertical streaks from infrared images using improved local moment matching | |
| Yu et al. | An improved retina-like nonuniformity correction for infrared focal-plane array | |
| CN110567585B (en) | Real-time infrared image pot cover effect inhibition method | |
| CN115002360B (en) | A non-uniformity correction method for infrared video based on robust estimation | |
| CN111932478A (en) | Self-adaptive non-uniform correction method for uncooled infrared focal plane | |
| CN108470325B (en) | Space-time three-dimensional noise identification and compensation method for area array staring infrared remote sensing image | |
| Guadagnoli et al. | Thermal imager non-uniformity sources modeling | |
| CN117853381A (en) | On-orbit image color-homogenizing correction processing method for satellite-borne optical camera | |
| Bekhtin et al. | Algorithmic supply of IR sensors with FPN using texture homogeneity levels | |
| Zhou et al. | Local spatial correlation-based stripe non-uniformity correction algorithm for single infrared images | |
| CN111076815B (en) | A method for correcting non-uniformity of hyperspectral images | |
| Chen et al. | Two-dimensional calibration for fixed-pattern noise reduction of thermal images |
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 |