CN110363714B - Non-uniformity correction method based on scene interframe registration of self-adaptive learning rate - Google Patents
Non-uniformity correction method based on scene interframe registration of self-adaptive learning rate Download PDFInfo
- Publication number
- CN110363714B CN110363714B CN201910532213.1A CN201910532213A CN110363714B CN 110363714 B CN110363714 B CN 110363714B CN 201910532213 A CN201910532213 A CN 201910532213A CN 110363714 B CN110363714 B CN 110363714B
- Authority
- CN
- China
- Prior art keywords
- frame
- pixel
- infrared image
- row
- learning rate
- 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
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000006073 displacement reaction Methods 0.000 claims abstract description 20
- 230000003044 adaptive effect Effects 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 17
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 4
- 239000013598 vector Substances 0.000 abstract description 2
- 238000004422 calculation algorithm Methods 0.000 description 5
- 230000004044 response Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000005314 correlation function Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G06T5/94—
-
- G06T5/80—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/38—Registration of image sequences
-
- 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
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20004—Adaptive image processing
- G06T2207/20012—Locally adaptive
Abstract
The invention discloses a scene frame registration-based non-uniformity correction method with adaptive learning rate, which comprises the steps of performing cross-correlation operation on row and column projection vectors of two adjacent frames of images to obtain the displacement of the images in the row and column directions so as to obtain an overlapping area of the two frames of images, and performing iterative updating on correction parameters according to the error of the overlapping area of the two frames of images and the learning rate. The learning rate depends on the spatial features of the image overlapping region, a higher learning rate is used in a flat region, and a lower learning rate is used in a region with more details and edges. Compared with a scene-based non-uniformity correction method using a fixed learning rate, the method has higher convergence rate and can effectively inhibit the generation of a ghost phenomenon.
Description
Technical Field
The invention belongs to the field of infrared image non-uniformity correction, and particularly relates to a non-uniformity correction method based on scene interframe registration and capable of self-adapting to learning rate.
Background
At present, infrared images are widely used in the fields of industry, medicine, military and the like to perform low visibility detection. Under ideal conditions, for infrared light which is uniformly radiated, the gray value of each pixel point on the obtained digital image should be completely the same. However, in practice, the process of manufacturing solid-state electrons is limited, the photosensitive elements (pixels) on the detector often have the problems of uneven impurity concentration, unequal thickness, no absolute average of effective photosensitive area, and the like, the photoelectric conversion efficiency between the pixels is different, and the imaging of uniformly radiated scenery is uneven. In addition, differences between the channels of the image data read-out circuit may cause fixed stripe noise in the image in a column distribution. This requires non-uniformity correction of the image to achieve better visual effect.
The commonly used infrared image non-uniformity correction technology mainly comprises a calibration method and a scene method.
The calibration-based method is a technology that has been adapted at present. However, with the change of the external environment, such as the change of the focal plane temperature and the bias voltage, the response characteristic of the infrared focal plane may drift, so that a new FPN appears on the calibrated and corrected image, and the imaging quality is reduced. In practice, periodic scaling is required to update the correction parameters.
Therefore, in recent years, many scholars have started to research a scene-based nonuniformity correction technique. The correction algorithm based on the scene not only omits a reference radiation source, simplifies the system processing flow, improves the stability of the system, but also can effectively eliminate the influence of parameter characteristic drift and realize the self-adaptive non-uniform correction with high precision and large dynamic range. Algorithms based on scene classes are divided into constant statistical method, neural network algorithm, interframe registration method and the like. However, the neural network non-uniformity correction algorithm is difficult to ensure the correction effect and has higher convergence speed, the engineering application difficulty is higher, and a serious decline phenomenon can occur when the image is static; the constant statistical method has strong dependence on scene random motion and is easily interfered by the scene, and the stability and the convergence speed of the algorithm cannot be considered at the same time; according to the characteristic that the response of each detector element to the same scene in a short time is consistent, the method obtains the response of different pixel elements to the same scene radiation by using an image registration mode. The calculation is simple, engineering application is easy, but convergence speed and stability cannot be considered.
Disclosure of Invention
The invention aims to provide a scene interframe registration-based non-uniformity correction method with a self-adaptive learning rate, which is used for solving the problems that the general scene interframe registration-based non-uniformity correction method is low in convergence speed and easy to generate ghost.
The technical solution for realizing the purpose of the invention is as follows: a non-uniformity correction method based on scene interframe registration of self-adaptive learning rate is realized by the following steps:
step 2, respectively calculating the mean value of all pixel values of each row in the two frames of infrared images, and respectively reducing the mean value of the gray values of the whole image to obtain the row projection value of each row; respectively calculating the mean value of the gray values of all the pixel values of each row, and respectively reducing the mean value of the gray values of the whole image to obtain the row projection value of each row;
step 3, performing cross-correlation operation on the projection values of the rows and the columns in the two frames of infrared images;
and 4, finding out the displacement in the row direction and the column direction when the cross correlation function in the row direction and the column direction is maximum, namely the relative displacement of the two frames of infrared images.
And 5, obtaining the self-adaptive learning rate of each pixel during correction according to the local variance of the pixel.
And 6, obtaining an overlapping area of the two frames of infrared images according to the row and column displacement values of the two frames of infrared images obtained in the step 4, and constructing an error matrix for the overlapping area.
Step 7, updating the correction parameter matrix;
and 8, adding the correction parameters to the original infrared image to obtain an output image.
And 9, executing the operations from the step 2 to the step 8 on all two adjacent frames of the infrared image sequence, namely the 1 st frame and the 2 nd frame, the 2 nd frame and the 3 rd frame, \ 8230, and the k-1 st frame and the k-th frame, \8230, in chronological sequence, and continuously carrying out iterative updating on the correction parameter matrix.
Compared with the prior art, the invention has the remarkable advantages that:
(1) The workload of a calibration method is reduced, and repeated calibration is avoided.
(2) Convergence can be realized only by dozens of frames or even dozens of frames, so that the correction rate is greatly improved;
(3) Because different updating rates are used for different areas, the problems of serious ghost and slow convergence caused by the adoption of a fixed learning rate in a common scene method are solved.
Drawings
FIG. 1 is a flow chart of a method for correcting non-uniformity based on scene frame-to-frame registration of adaptive learning rate according to the present invention.
Fig. 2 (a) is a 100 th frame image in an original infrared image sequence having non-uniformity; FIG. 2 (b) is a 100 th frame of image after processing an original image sequence by using a fixed learning rate non-uniformity correction method based on scene frame-to-frame registration; fig. 2 (c) is the 100 th frame image after being processed by the non-uniformity correction method based on scene frame-to-frame registration using the adaptive learning rate of the present invention.
Detailed Description
The following is further detailed in conjunction with the accompanying drawings.
The invention relates to a non-uniformity correction method based on scene interframe registration and capable of realizing self-adaptive learning rate, which can remove non-uniformity in an infrared image sequence through multi-frame iteration. The principle is as follows: the method comprises the steps of firstly calculating cross-correlation information for line projection vectors of front and back two frames of images in an infrared image sequence, then respectively searching horizontal and vertical displacement values which enable the cross-correlation information to be maximum, namely displacement values of the images, obtaining overlapped areas, and updating correction parameters of the images in the overlapped areas according to learning rate to finish non-uniformity correction. An improved method is provided on the basis of a general non-uniformity correction method based on scene frame registration, a self-adaptive learning rate is calculated according to the local variance of each pixel, and ghost can be avoided while the convergence speed is improved.
With reference to fig. 1, a method for correcting heterogeneity based on scene frame-to-frame registration with adaptive learning rate includes the following steps:
step 2, respectively calculating the mean value of all pixel values of each row in the infrared images of the kth frame and the (k + 1) th frame, and respectively subtracting the mean value of the gray values of the corresponding whole infrared images to obtain the row projection value of each row; respectively calculating the mean value of all pixel values of each column, and respectively subtracting the mean value of the gray values of the corresponding whole infrared image to obtain the column projection values of each column, wherein the calculation formula is as follows:
where i denotes the row coordinates of the pixel, j denotes the column coordinates of the pixel,
represents the line projection value of the ith line of the infrared image, is greater than>
Column projection value, X, representing the jth column of the infrared image n (i, j) represents a gray value of a pixel having the coordinate (i, j) in the nth frame; n represents the number of frames, let n = k, or n = k +1;
step 3, performing cross-correlation operation on the projection values of the rows and columns of the infrared images of the k frame and the k +1 frame:
wherein the displacement h ∈ [0,2 × Δ ] in the horizontal direction col ]The displacement v ∈ [0,2 × Δ ] in the vertical direction row ]，Δ col Is the upper limit of the displacement of the k +1 th frame infrared image relative to the k 1 th frame infrared image in the horizontal direction, delta row Setting the upper limit of the displacement of the (k + 1) th frame infrared image relative to the k frame infrared image in the vertical direction; c row (v) As a function of the line-related information of two adjacent frames of infrared images, C col (h) Column cross-correlation information functions of two adjacent frames of infrared images;
step 4, finding out the general formula C row (v) And C col (h) And taking the v, h value corresponding to the maximum value, and calculating the displacement di in the row direction and the displacement dj in the column direction of the infrared image of the (k + 1) th frame relative to the infrared image of the k-th frame:
step 5, the common scene-based inter-frame registration non-uniformity correction method adopts the same fixed learning rate for all pixels, when the learning rate is set to be high, a ghost phenomenon can be generated, and when the learning rate is set to be low, the convergence speed is very low, so that the convergence speed and the stability cannot be well balanced; the invention calculates the self-adaptive learning rate for each pixel according to the local variance of each pixel, and gives consideration to convergence speed and stability, and the method comprises the following specific steps:
5-1) calculating a local mean value m (i, j) of each pixel, where the local mean value refers to a region with a window size of (2l + 1) × (2l + 1) centered on the coordinate (i, j) of the pixel, where l is a preset integer, and X (a, b) represents a gray value of the pixel, and the calculation formula is as follows:
5-2) calculating the local variance σ of each pixel 2 (i, j) the calculation formula is as follows:
5-3) calculating the learning rate of each pixel according to the local variance, wherein the calculation formula is as follows:
wherein alpha is st As the initial learning rate, α (i, j) is the learning rate of the pixel point whose coordinate is (i, j). The learning rate calculated by the formula is higher in an image flat area, the convergence rate can be improved, and is lower in image details and an edge area, so that ghost can be avoided.
Step 6, obtaining the overlapping area of the infrared images of two adjacent frames according to the line and column displacement values di and dj of the infrared images of the k frame and the k +1 frame obtained in the step 4,constructing an error matrix ERR for the overlapping area k+1 (i,j) 2 ：
Wherein, X k (i, j) represents the gray value of the pixel point with the coordinate (i, j) in the k frame infrared image, O k (i, j) represents the value of coordinate (i, j) in the correction parameter matrix calculated with the k-1 and k-th frames, where k-1 is equal to 1 at minimum.
And 7, updating the correction parameter matrix:
O k+1 (i,j)＝O k (i,j)-2×α(i,j)×ERR k+1 (i,j)
step 8, adding the (k + 1) th frame infrared image to a correction parameter matrix to obtain a (k + 1) th frame corrected output image Y k (i,j)：
Y k+1 (i,j)＝X k+1 (i,j)+O k+1 (i,j)
And 9, executing 8230for all two adjacent frames of the infrared image sequence, namely the 1 st frame and the 2 nd frame, the 2 nd frame and the 3 rd frame, and executing 8230for the k-1 st frame and the k-1 rd frame, and continuously carrying out iterative updating on the correction parameter matrix.
Taking the 100 th frame image in the original infrared image sequence with non-uniformity as shown in fig. 2 (a) as an example, processing the original image sequence by using a non-uniformity correction method based on scene frame-to-frame registration with a fixed learning rate to obtain an image as shown in fig. 2 (b), it is obvious that the non-uniformity is still obviously remained, and a ghost is generated. The image obtained by processing the non-uniformity correction method based on scene frame registration with the adaptive learning rate provided by the invention is shown in fig. 2 (c), and compared with fig. 2 (b), the non-uniformity removal effect is better and no ghost phenomenon exists.
Claims (1)
1. A scene frame-to-frame registration-based nonuniformity correction method based on adaptive learning rate is characterized by comprising the following steps:
step 1, obtaining a real-time infrared image sequence through a thermal infrared imager, wherein the resolution of an infrared image is MxN, M is the number of lines, and N is the number of columns; constructing a correction parameter matrix O with the size of M multiplied by N, storing correction parameters of each pixel point, wherein the initial value of the correction parameter matrix O is 0, and selecting two adjacent frames of infrared images for a real-time infrared image sequence every time, such as the kth frame and the (k + 1) th frame, wherein k is more than or equal to 1;
step 2, respectively calculating the mean value of all pixel values of each row in the infrared images of the kth frame and the (k + 1) th frame, and respectively subtracting the mean value of the gray values of the corresponding whole infrared images to obtain the row projection value of each row; respectively calculating the mean value of all pixel values of each column, and respectively subtracting the mean value of the gray values of the corresponding whole infrared image to obtain the column projection value of each column, wherein the calculation formula is as follows:
where i denotes the row coordinate of the pixel, j denotes the column coordinate of the pixel,
a line projection value representing the ith line of the infrared image,
column projection value, X, representing the jth column of the infrared image n (i, j) represents a gray value of a pixel having the coordinate (i, j) in the nth frame; n represents the number of frames, let n = k, or n = k +1;
step 3, performing cross-correlation operation on the projection values of the rows and columns of the infrared images of the k frame and the k +1 frame:
wherein the displacement h ∈ [0,2 × Δ ] in the horizontal direction col ]The displacement v ∈ [0,2 × Δ ] in the vertical direction row ]，Δ col Is the upper limit of the displacement of the k +1 th frame infrared image relative to the k 1 th frame infrared image in the horizontal direction, delta row The upper limit of the displacement of the (k + 1) th frame infrared image relative to the (k) th frame infrared image in the vertical direction is set; c row (v) As a function of the line-related information of two adjacent frames of infrared images, C col (h) Column cross-correlation information functions of two adjacent frames of infrared images;
step 4, finding out the general formula C row (v) And C col (h) And (3) taking the corresponding v, h value when the value is maximum, and calculating the displacement di in the row direction and the displacement dj in the column direction of the (k + 1) th frame infrared image relative to the k frame infrared image:
step 5, obtaining the adaptive learning rate alpha (i, j) when the pixel updates the correction parameter according to the local variance of each pixel, and the specific steps are as follows:
5-1) calculating a local mean value m (i, j) of each pixel, where the local mean value refers to a region with a window size of (2l + 1) × (2l + 1) centered on the coordinate (i, j) of any pixel, where l is a preset integer, and X (a, b) represents a gray value of the pixel, and the calculation formula is as follows:
5-2) calculating the local variance σ of each pixel 2 (i, j), the calculation formula is as follows:
5-3) calculating the learning rate of each pixel according to the local variance, wherein the calculation formula is as follows:
wherein alpha is st α (i, j) is the learning rate of the pixel point of the coordinate (i, j);
step 6, obtaining the overlapping area of the infrared images of two adjacent frames according to the line and column displacement values di and dj of the infrared images of the k frame and the k +1 frame obtained in the step 4, and constructing an error matrix ERR for the overlapping area k+1 (i,j) 2 ：
Wherein, X k (i, j) represents the gray value of a pixel point with coordinates (i, j) in the k frame infrared image, O k (i, j) represents the value of coordinate (i, j) in the correction parameter matrix calculated with the k-1 and k-th frames, where k-1 is at least equal to 1;
and 7, updating the correction parameter matrix:
O k+1 (i,j)＝O k (i,j)-2×α(i,j)×ERR k+1 (i,j)
step 8, adding the infrared image of the (k + 1) th frame and the correction parameter matrix to obtain a corrected output image Y of the (k + 1) th frame k (i,j)：
Y k+1 (i,j)＝X k+1 (i,j)+O k+1 (i,j)
And (3) according to the time sequence, executing the operations of the steps 2 to 8 on all two adjacent frames of the infrared image sequence, namely the 1 st frame and the 2 nd frame, the 2 nd frame and the 3 rd frame, \8230, the k-1 st frame and the k-1 th frame, and the k +1 th frame, \8230, and continuously carrying out iterative updating on the correction parameter matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910532213.1A CN110363714B (en) | 2019-06-19 | 2019-06-19 | Non-uniformity correction method based on scene interframe registration of self-adaptive learning rate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910532213.1A CN110363714B (en) | 2019-06-19 | 2019-06-19 | Non-uniformity correction method based on scene interframe registration of self-adaptive learning rate |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110363714A CN110363714A (en) | 2019-10-22 |
| CN110363714B true CN110363714B (en) | 2023-03-28 |
Family
ID=68216392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910532213.1A Active CN110363714B (en) | 2019-06-19 | 2019-06-19 | Non-uniformity correction method based on scene interframe registration of self-adaptive learning rate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110363714B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111507915B (en) * | 2020-04-13 | 2024-03-29 | 烟台艾睿光电科技有限公司 | Real-time infrared non-uniformity correction method, device and medium based on fuzzy registration |
| US20220319018A1 (en) * | 2021-03-30 | 2022-10-06 | Distech Controls Inc. | Method and computing device using a neural network to localize an overlap between two thermal images respectively generated by two infrared sensors |
| CN113375808B (en) * | 2021-05-21 | 2023-06-02 | 武汉博宇光电系统有限责任公司 | Infrared image non-uniformity correction method based on scene |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106570829A (en) * | 2015-10-09 | 2017-04-19 | 南京理工大学 | Infrared image non-uniformity correction method |
| CN108665425A (en) * | 2018-03-28 | 2018-10-16 | 西安电子科技大学 | Infrared Image Non-uniformity Correction method based on interframe registration and adaptive step |
| CN109741267A (en) * | 2018-12-05 | 2019-05-10 | 西安电子科技大学 | Infrared Image Non-uniformity Correction method based on three sides filtering and neural network |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102538973B (en) * | 2011-12-31 | 2014-04-02 | 南京理工大学 | Rapidly converged scene-based non-uniformity correction method |
| CN104296876B (en) * | 2013-07-15 | 2017-08-04 | 南京理工大学 | Scene asymmetric correction method and its device based on FPGA |
| EP3099044B1 (en) * | 2014-05-15 | 2018-12-26 | Huawei Technologies Co. Ltd. | Multi-frame noise reduction method and terminal |
| CN105318971B (en) * | 2014-07-07 | 2018-09-04 | 南京理工大学 | The adaptive nonuniformity correction method of image registration is used to infrared video sequence |
| US10692251B2 (en) * | 2017-01-13 | 2020-06-23 | Canon Medical Systems Corporation | Efficient variance-reduced method and apparatus for model-based iterative CT image reconstruction |
-
2019
- 2019-06-19 CN CN201910532213.1A patent/CN110363714B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106570829A (en) * | 2015-10-09 | 2017-04-19 | 南京理工大学 | Infrared image non-uniformity correction method |
| CN108665425A (en) * | 2018-03-28 | 2018-10-16 | 西安电子科技大学 | Infrared Image Non-uniformity Correction method based on interframe registration and adaptive step |
| CN109741267A (en) * | 2018-12-05 | 2019-05-10 | 西安电子科技大学 | Infrared Image Non-uniformity Correction method based on three sides filtering and neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110363714A (en) | 2019-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110363714B (en) | Non-uniformity correction method based on scene interframe registration of self-adaptive learning rate | |
| CN110211056B (en) | Self-adaptive infrared image de-striping algorithm based on local median histogram | |
| CN105318971B (en) | The adaptive nonuniformity correction method of image registration is used to infrared video sequence | |
| US6507018B2 (en) | Ditherless non-uniformity compensation for infrared detector arrays with recursive spatial low pass filtering | |
| CN109741267B (en) | Infrared image non-uniformity correction method based on trilateral filtering and neural network | |
| CN106934771B (en) | Infrared image stripe noise removing method based on local correlation | |
| CN109272520B (en) | Self-adaptive infrared focal plane non-uniform correction method combining motion guidance and edge detection | |
| Zeng et al. | Adaptive image-registration-based nonuniformity correction algorithm with ghost artifacts eliminating for infrared focal plane arrays | |
| CN111161172B (en) | Infrared image column direction stripe eliminating method, system and computer storage medium | |
| CN103475828A (en) | Method for rectifying missing pixels and image sensor | |
| CN102778296B (en) | Total variation-based self-adaptation non-uniformity correction method for infrared focal plane | |
| CN108230249B (en) | Anisotropic-based L1 norm total variation regularization non-uniformity correction method | |
| CN108665425A (en) | Infrared Image Non-uniformity Correction method based on interframe registration and adaptive step | |
| CN109813442B (en) | Multi-frame processing-based internal stray radiation non-uniformity correction method | |
| CN110033414B (en) | Infrared image non-uniformity correction method and system based on averaging processing | |
| WO2017185441A1 (en) | Infrared image stripe filtering method based on statistical relative stripe removal method | |
| JP6830712B1 (en) | Random sampling Consistency-based effective area extraction method for fisheye images | |
| CN110782403A (en) | Infrared image non-uniformity correction method | |
| US11195254B2 (en) | Interframe registration and adaptive step size-based non-uniformity correction method for infrared image | |
| CN111932478A (en) | Self-adaptive non-uniform correction method for uncooled infrared focal plane | |
| CN108225570B (en) | Short wave infrared focal plane self-adaptive non-uniform correction algorithm | |
| CN109636740B (en) | Infrared image multi-scale intelligent non-uniformity correction method | |
| Liu et al. | A time-domain projection-based registration-scene-based nonuniformity correction technology and its detailed hardware realization | |
| Dong et al. | Summarize on the scene-based nonuniformity correction algorithms for IRFPA | |
| CN113375808B (en) | Infrared image non-uniformity correction method based on scene |
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 |