Infrared Image Optical Nonuniformity Correction Method Based on Multi-Scale Frequency Domain Filtering

Abstract

The present application provides an infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering. In this method, the second image information is respectively subjected to low-pass filtering processing at different scales to obtain a third image information set and a fifth image information set, and then the third image information set and the fifth image information set are jointly weighted and transformed to obtain optical non-uniformity image information. The second image information is corrected with the optical non-uniformity image information to obtain de-optical non-uniformity image information. Then, based on the de-optical non-uniformity image information, as well as the maximum and minimum values of the image pixel values in the de-optical non-uniformity image information, de-optical non-uniformity shaping image information is calculated. The infrared image optical non-uniformity correction method provided by the present application can effectively improve the display effect of infrared images, suppress the influence of optical non-uniformity, while enhancing the contrast of details and improving the expressiveness of infrared images under multi-scale frequency domain filtering processing.

Description

Technical Field

[0001] The present application specifically discloses an infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering. Background Art

[0002] An infrared imaging system is used to convert the infrared radiation of a scene into a visual image. In a specific application scenario, when the infrared imaging system is in a complex working environment, for example, when the internal and external environmental temperatures change, stray thermal radiation of the optical system will be introduced, resulting in image degradation.

[0003] To improve the image quality, it is often necessary to correct the infrared image using traditional non-uniformity correction methods. Specifically, the technical principle of the traditional non-uniformity correction method usually uses a calibrated calculation method, that is, by changing the environmental temperature of the test sample through a high and low temperature test chamber. This method requires calibration for each set of products, and usually, it takes more than one day to calibrate a set of products. In addition, the effect on the irregular non-uniformity shown in the image introduced during use is limited and urgently needs improvement. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, the present application aims to provide an infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering.

[0005] The first aspect of the present application provides an infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering.

[0006] An infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering includes the following steps:

[0007] Obtain first infrared image information, which is obtained by an infrared thermal imager;

[0008] Convert the first infrared image information into second image information of a first format type;

[0009] Obtain a spectrogram corresponding to the second image information to obtain first spectrogram image information;

[0010] Perform offset processing on the first spectrogram image information so that the low-frequency part of the first spectrogram image information is moved to the center to obtain second spectrogram image information;

[0011] Adopt at least two Gaussian low-pass filter radii to perform filtering processing on the second spectrogram image information respectively to obtain a first set of low-pass filtered image information;

[0012] Convert the first low-pass filtered image information in the first set of low-pass filtered image information into third image information one by one to obtain a third set of image information;

[0013] Convert the second image information to the fourth image information of the second format type;

[0014] Obtain the spectrogram corresponding to the fourth image information to get the third spectrogram image information;

[0015] Use different homomorphic filtering radii to perform filtering processing on the third spectrogram image information respectively to obtain the second set of low-pass filtered image information;

[0016] Convert the second low-pass filtered image information in the second set of low-pass filtered image information to the fifth image information one by one to obtain the fifth set of image information;

[0017] Perform weighted transformation processing on the third set of image information and the fifth set of image information together to obtain the optical non-uniformity image information;

[0018] Correct the second image information with the optical non-uniformity image information to obtain the de-optically non-uniformity image information;

[0019] Calculate the de-optically non-uniformity shaped image information with the de-optically non-uniformity image information, as well as the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information.

[0020] According to the technical solution provided by the embodiment of the present application, the first format type is double-precision floating-point type;

[0021] The specific steps for converting the first infrared image information to the second image information of double-precision floating-point type are:

[0022] F' = double(F)

[0023] Where: F is the first infrared image information; F' is the second image information.

[0024] According to the technical solution provided by the embodiment of the present application, the specific steps for obtaining the spectrogram corresponding to the second image information to get the first spectrogram image information are:

[0025] Perform two-dimensional Fourier transform on the second image information to obtain the first spectrogram image information

[0026] G = FFT2(F')

[0027] Where: G is the first spectrogram image information; F' is the second image information.

[0028] According to the technical solution provided by the embodiment of the present application, the specific steps for offset processing the first spectrogram image information so that the low-frequency part of the first spectrogram image information moves to the center to obtain the second spectrogram image information are:

[0029] G' = FFTSHIFT(G)

[0030] Where: G' is the second spectral image information, and G is the first spectral image information.

[0031] According to the technical solution provided by the embodiments of the present application, the specific steps of using at least two Gaussian low-pass filter radii to perform filtering processing on the second spectral image information to obtain the first low-pass filtered image information set are as follows:

[0032] Calculate the first distance between each pixel in the second spectral image information and the center of the second spectral image information, that is:

[0033]

[0034] Where:

[0035] i represents the horizontal position of the current pixel on the second spectral image information;

[0036] j represents the vertical position of the current pixel on the second spectral image information;

[0037] m represents the total number of pixels in the horizontal direction of the second spectral image information;

[0038] n represents the total number of pixels in the vertical direction of the second spectral image information;

[0039] Calculate the Gaussian low-pass filter kernel at each Gaussian low-pass filter radius based on the first distance, that is:

[0040]

[0041] Where: r k represents the k-th Gaussian low-pass filter radius, 1 ≤ k ≤ q;

[0042] Calculate the first low-pass filtered image information based on the Gaussian low-pass filter kernel at each Gaussian low-pass filter radius and the second spectral image information, that is:

[0043] Gauss k = GaussKernel k * G'

[0044] Summarize all the first low-pass filtered image information to obtain the first low-pass filtered image information set.

[0045] According to the technical solution provided by the embodiments of the present application, the second format type is a logarithmic type;

[0046] The specific steps of converting the second image information to the fourth image information of the second format type are as follows:

[0047] L = lnF'

[0048] Wherein: L is the fourth image information.

[0049] According to the technical solution provided by the embodiment of the present application, the specific steps for obtaining the spectrogram corresponding to the fourth image information and obtaining the third spectrogram image information are as follows:

[0050] LF = FFT(L)

[0051] Wherein: LF is the third spectrogram image information.

[0052] According to the technical solution provided by the embodiment of the present application, the specific steps for performing filtering processing on the third spectrogram image information respectively by using different homomorphic filtering radii to obtain the set of second low-pass filtered image information are as follows:

[0053] Calculate the second distance between each pixel in the third spectrogram image information and the center of the third spectrogram image information, that is:

[0054]

[0055] Wherein:

[0056] i represents the horizontal position of the current pixel on the third spectrogram image information;

[0057] j represents the vertical position of the current pixel on the third spectrogram image information;

[0058] m represents the total number of pixels in the horizontal direction of the third spectrogram image information;

[0059] n represents the total number of pixels in the vertical direction of the third spectrogram image information;

[0060] Calculate the homomorphic low-pass filter kernel at each homomorphic filtering radius with the second distance, that is:

[0061]

[0062] Wherein:

[0063] D s represents the s-th homomorphic filtering radius, 1 ≤ s ≤ p;

[0064] r H is a homomorphic filtering parameter, r H ≥ 1;

[0065] r L is a homomorphic filtering parameter, r L < 1;

[0066] Calculate the second low-pass filtered image information by using the homomorphic low-pass filter kernel at each homomorphic filtering radius and the third spectrogram image information, that is:

[0067] Homo s = H s *LF

[0068] Summarize all the second low-pass filtered image information to obtain a set of second low-pass filtered image information.

[0069] According to the technical solution provided by the embodiments of the present application, the specific steps for converting the first low-pass filtered image information in the first low-pass filtered image information set into third image information are as follows:

[0070] GaussLow k = IFFT(IFFTSHIFT(Gauss k ))

[0071] Where: GaussLow k is the third image information;

[0072] The specific steps for converting the second low-pass filtered image information in the second low-pass filtered image information set into fifth image information are as follows:

[0073] HomoLow s = exp[IFFT(IFFTSHIFT(Homo s ))]

[0074] Where: HomoLow s is the fifth image information;

[0075] The specific steps for jointly performing weighted transformation processing on the third image information set and the fifth image information set to obtain optical non-uniformity image information are as follows:

[0076]

[0077] According to the technical solution provided by the embodiments of the present application, the specific steps for correcting the second image information with the optical non-uniformity image information to obtain de-optically non-uniformity image information are as follows:

[0078] ImgTemp = F' - NULow

[0079] The specific steps for calculating the de-optically non-uniformity shaped image information with the de-optically non-uniformity image information, as well as the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information are as follows:

[0080] Obtain the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information;

[0081] Calculate the de-optically non-uniformity shaped image information with the de-optically non-uniformity image information, the maximum and minimum values of the image pixel values, that is:

[0082]

[0083] Wherein: min(ImgTemp) is the minimum value of the image pixel values in the image information after removing optical non-uniformity; max(ImgTemp) is the maximum value of the image pixel values in the image information after removing optical non-uniformity.

[0084] The second aspect of the present application provides an infrared image optical non-uniformity correction device based on multi-scale frequency domain filtering, including:

[0085] A first acquisition module configured to acquire first infrared image information captured by an infrared thermal imager;

[0086] A first conversion module configured to convert the first infrared image information into second image information of a first format type;

[0087] A first operation module configured to obtain a spectrogram corresponding to the second image information to obtain first spectrogram image information;

[0088] A second operation module configured to perform offset processing on the first spectrogram image information so that the low-frequency part of the first spectrogram image information is moved to the center to obtain second spectrogram image information;

[0089] A first filtering module configured to perform filtering processing on the second spectrogram image information respectively by using at least two Gaussian low-pass filtering radii to obtain a set of first low-pass filtered image information;

[0090] A second conversion module configured to convert the first low-pass filtered image information in the set of first low-pass filtered image information into third image information one by one to obtain a set of third image information;

[0091] A third conversion module configured to convert the second image information into fourth image information of a second format type;

[0092] A third operation module configured to obtain a spectrogram corresponding to the fourth image information to obtain third spectrogram image information;

[0093] A second filtering module configured to perform filtering processing on the third spectrogram image information respectively by using different homomorphic filtering radii to obtain a set of second low-pass filtered image information;

[0094] A fourth conversion module, configured to convert the second low-pass filtered image information in the second low-pass filtered image information set one by one into fifth image information, thereby obtaining a fifth image information set;

[0095] A fourth operation module, configured to perform a weighted transformation process on the third image information set and the fifth image information set jointly, thereby obtaining optical non-uniformity image information;

[0096] A first correction module, configured to correct the second image information with the optical non-uniformity image information, thereby obtaining de-optically non-uniformity image information;

[0097] A fifth operation module, configured to calculate de-optically non-uniformity shaped image information by using the de-optically non-uniformity image information, as well as the maximum value and the minimum value of the image pixel values in the de-optically non-uniformity image information.

[0098] According to the technical solution provided by the embodiment of the present application, the first format type is a double-precision floating-point type; a first conversion module, configured to convert the first infrared image information into the second image information of the double-precision floating-point type; specifically:

[0099] F' = double(F)

[0100] where: F is the first infrared image information; F' is the second image information.

[0101] According to the technical solution provided by the embodiment of the present application, a first operation module, configured to perform a two-dimensional Fourier transform on the second image information, thereby obtaining first spectral image information

[0102] G = FFT2(F')

[0103] where: G is the first spectral image information; F' is the second image information.

[0104] According to the technical solution provided by the embodiment of the present application, a second operation module, configured to perform an offset process on the first spectral image information, so that the low-frequency part of the first spectral image information is moved to the center, thereby obtaining the second spectral image information. The specific steps are:

[0105] G' = FFTSHIFT(G)

[0106] where: G' is the second spectral image information, and G is the first spectral image information.

[0107] According to the technical solution provided by the embodiment of the present application, the first filtering module includes:

[0108] The first arithmetic unit, which is configured to calculate the first distance between each pixel in the second spectral image information and the center of the second spectral image information, that is:

[0109]

[0110] Where:

[0111] i represents the horizontal position of the current pixel in the second spectral image information;

[0112] j represents the vertical position of the current pixel in the second spectral image information;

[0113] m represents the total number of pixels in the horizontal direction of the second spectral image information;

[0114] n represents the total number of pixels in the vertical direction of the second spectral image information;

[0115] The second arithmetic unit, which is configured to calculate the Gaussian low-pass filter kernel at each Gaussian low-pass filter radius based on the first distance, that is:

[0116]

[0117] Where: r k represents the k-th Gaussian low-pass filter radius, 1 ≤ k ≤ q;

[0118] The third arithmetic unit, which is configured to calculate the first low-pass filtered image information based on the Gaussian low-pass filter kernel at each Gaussian low-pass filter radius and the second spectral image information, that is:

[0119] Gauss k = GaussKernel k * G'

[0120] The fourth arithmetic unit, which is configured to aggregate all the first low-pass filtered image information to obtain a set of the first low-pass filtered image information.

[0121] According to the technical solution provided by the embodiment of the present application, the second format type is a logarithmic type; the third conversion module is configured to convert the second image information into the fourth image information of the second format type; specifically:

[0122] L = lnF'

[0123] Where: L is the fourth image information.

[0124] According to the technical solution provided by the embodiment of the present application, the third arithmetic module, the specific steps for the third arithmetic module to obtain the spectral diagram corresponding to the fourth image information to obtain the third spectral image information are:

[0125] LF = FFT(L)

[0126] Where: LF is the third spectral image information.

[0127] According to the technical solution provided by the embodiments of the present application, the second filtering module includes:

[0128] A fifth arithmetic unit configured to calculate the second distance between each pixel in the third spectral image information and the center of the third spectral image information, that is:

[0129]

[0130] Where:

[0131] i represents the horizontal position of the current pixel on the third spectral image information;

[0132] j represents the vertical position of the current pixel on the third spectral image information;

[0133] m represents the total number of pixels in the horizontal direction on the third spectral image information;

[0134] n represents the total number of pixels in the vertical direction on the third spectral image information;

[0135] A sixth arithmetic unit configured to calculate the homomorphic low-pass filter kernel at each homomorphic filtering radius with the second distance, that is:

[0136]

[0137] Where:

[0138] D s represents the s-th homomorphic filtering radius, 1 ≤ s ≤ p;

[0139] r H is the homomorphic filtering parameter, r H ≥ 1;

[0140] r L is the homomorphic filtering parameter, r L < 1;

[0141] A seventh arithmetic unit configured to calculate the second low-pass filtered image information with the homomorphic low-pass filter kernel at each homomorphic filtering radius and the third spectral image information, that is:

[0142] Homo s = H s * LF

[0143] An eighth operation unit, configured to summarize all the second low-pass filtered image information to obtain a set of second low-pass filtered image information.

[0144] According to the technical solution provided by the embodiment of the present application, the specific steps for the second conversion module to convert the first low-pass filtered image information in the first set of low-pass filtered image information into third image information are as follows:

[0145] GaussLow k = IFFT(IFFTSHIFT(Gauss k ))

[0146] Where: GaussLow k is the third image information;

[0147] A fourth conversion module, the specific steps for the fourth conversion module to convert the second low-pass filtered image information in the set of second low-pass filtered image information into fifth image information are as follows:

[0148] HomoLow s = exp[IFFT(IFFTSHIFT(Homo s ))]

[0149] Where: HomoLow s is the fifth image information;

[0150] A fourth operation module, the specific steps for the fourth operation module to perform a weighted transformation process on the set of third image information and the set of fifth image information to obtain optical non-uniformity image information are as follows:

[0151]

[0152] According to the technical solution provided by the embodiment of the present application, a first correction module, the specific steps for the first correction module to correct the second image information with the optical non-uniformity image information to obtain de-optical non-uniformity image information are as follows:

[0153] ImgTemp = F' - NULow

[0154] The fifth operation module includes:

[0155] A ninth operation unit, configured to obtain the maximum and minimum values of the image pixel values in the de-optical non-uniformity image information;

[0156] A tenth operation unit, configured to calculate de-optical non-uniformity shaped image information with the de-optical non-uniformity image information, the maximum and minimum values of the image pixel values, that is:

[0157]

[0158] Wherein:

[0159] min(ImgTemp) is the minimum value of the image pixel values in the image information after removing optical non-uniformity;

[0160] max(ImgTemp) is the maximum value of the image pixel values in the image information after removing optical non-uniformity.

[0161] The third aspect of the present application provides a processing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering as described above are implemented.

[0162] The fourth aspect of the present application provides a computer-readable storage medium, which has a computer program. When the computer program is executed by a processor, the steps of the infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering as described above are implemented.

[0163] The beneficial effects of the present application are as follows:

[0164] First, convert the first infrared image information into second image information of the first format type; then, obtain a spectrogram corresponding to the second image information to obtain first spectrogram image information; further, perform offset processing on the first spectrogram image information so that the low-frequency part of the first spectrogram image information is moved to the center to obtain second spectrogram image information; finally, use at least two Gaussian low-pass filter radii to perform filtering processing on the second spectrogram image information respectively to obtain a set of first low-pass filtered image information; convert each first low-pass filtered image information in the set of first low-pass filtered image information into third image information one by one to obtain a set of third image information.

[0165] In addition, convert the second image information into fourth image information of the second format type; then, obtain a spectrogram corresponding to the fourth image information to obtain third spectrogram image information; further, use different homomorphic filter radii to perform filtering processing on the third spectrogram image information respectively to obtain a set of second low-pass filtered image information; convert each second low-pass filtered image information in the set of second low-pass filtered image information into fifth image information one by one to obtain a set of fifth image information.

[0166] Finally, a weighted transformation process is jointly performed on the third image information set and the fifth image information set to obtain optical non-uniformity image information; the second image information is corrected with the optical non-uniformity image information to obtain de-optically non-uniformity image information; and de-optically non-uniformity shaped image information is calculated using the de-optically non-uniformity image information, as well as the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information.

[0167] In summary, the optical non-uniformity of an infrared imaging system usually appears as low-frequency noise in an image, with the characteristic of slow change in neighborhood gray values. The technical solution of this application respectively performs low-pass filtering processing on the second image information at different scales to obtain a third image information set and a fifth image information set, and jointly performs a weighted transformation process on the third image information set and the fifth image information set to obtain optical non-uniformity image information. The second image information is corrected with the optical non-uniformity image information to obtain de-optically non-uniformity image information. Finally, de-optically non-uniformity shaped image information is calculated using the de-optically non-uniformity image information, as well as the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information. Compared with the prior art, the infrared image optical non-uniformity correction method provided by this application can effectively improve the display effect of infrared images, suppress the influence of optical non-uniformity, while enhancing the contrast of details and improving the expressiveness of infrared images under multi-scale frequency domain filtering processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0168] Other features, objects, and advantages of the present application will become more apparent by reading the detailed description of the non-limiting embodiments with reference to the following drawings:

[0169] Figure 1 is a schematic flowchart of an infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering;

[0170] Figure 2 is a schematic diagram of the first infrared image information;

[0171] Figure 3 is a schematic diagram of the second image information;

[0172] Figure 4 is a schematic diagram of the second spectral image information;

[0173] Figure 5 is a schematic diagram of the first low-pass filtered image information Gauss1;

[0174] Figure 6 is a schematic diagram of the first low-pass filtered image information Gauss2;

[0175] Figure 7 is a schematic diagram of the second low-pass filtered image information Homo2;

[0176] Figure 8 It is a schematic diagram of the second low-pass filtered image information Homo3;

[0177] Figure 9 It is a schematic diagram of the image information of removing optical non-uniformity;

[0178] Figure 10 It is a schematic diagram of the shaped image information of removing optical non-uniformity;

[0179] Figure 11 It is a schematic diagram of the processing device provided for this application. Detailed implementation manners

[0180] The present application will be further described in detail below with reference to the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related invention, rather than limiting the invention. In addition, it should be noted that, for the convenience of description, only the parts related to the invention are shown in the drawings.

[0181] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the drawings and embodiments.

[0182] Embodiment 1

[0183] Please refer to Figure 1 A method for correcting optical non-uniformity of an infrared image based on multi-scale frequency domain filtering as shown.

[0184] Figure 1 In [reference], a method for correcting optical non-uniformity of an infrared image based on multi-scale frequency domain filtering is specifically given, including the following steps:

[0185] S1: Obtain first infrared image information, which is obtained by an infrared thermal imager; specifically, the infrared thermal imager can be: a cooled infrared thermal imager with a resolution of 640*512 is used to obtain it, and the first infrared image information can be referred to Figure 2 as shown.

[0186] S2: Convert the first infrared image information into second image information of a first format type; wherein: the first format type is double-precision floating-point type; specifically, the specific steps for converting the first infrared image information into second image information of double-precision floating-point type are:

[0187] F' = double(F)

[0188] wherein: F is the first infrared image information; F' is the second image information.

[0189] Please refer toFigure 3 The second image information of double-precision floating-point type obtained by format conversion from the first infrared image information as shown.

[0190] S3: Obtain a spectrogram corresponding to the second image information to get first spectrogram image information; specifically, perform a two-dimensional Fourier transform on the second image information to obtain the first spectrogram image information

[0191] G = FFT2(F')

[0192] Where: G is the first spectrogram image information; F' is the second image information.

[0193] S4: Perform offset processing on the first spectrogram image information so that the low-frequency part of the first spectrogram image information is moved to the center to obtain second spectrogram image information;

[0194] Specifically, the specific steps for performing offset processing on the first spectrogram image information so that the low-frequency part of the first spectrogram image information is moved to the center to obtain second spectrogram image information are as follows:

[0195] G' = FFTSHIFT(G)

[0196] Where: G' is the second spectrogram image information, and G is the first spectrogram image information.

[0197] Perform image four-quadrant exchange on the first spectrogram image information to move the low-frequency part to the center of the image, with the high-frequency part around the perimeter, to obtain the second spectrogram image information, please refer to Figure 4 as shown.

[0198] S5: Use at least two Gaussian low-pass filter radii to perform filtering processing on the second spectrogram image information respectively to obtain a set of first low-pass filtered image information;

[0199] Specifically:

[0200] Calculate the first distance between each pixel in the second spectrogram image information and the center of the second spectrogram image information, that is:

[0201]

[0202] Where:

[0203] i represents the horizontal position of the current pixel on the second spectrogram image information;

[0204] j represents the vertical position of the current pixel on the second spectrogram image information;

[0205] m represents the total number of pixels in the horizontal direction of the second spectrogram image information;

[0206] n represents the total number of pixels in the vertical direction of the second spectral image information;

[0207] Calculate the Gaussian low-pass filter kernel at each Gaussian low-pass filter radius at the first distance, that is:

[0208]

[0209] Where: r k represents the k-th Gaussian low-pass filter radius, 1 ≤ k ≤ q;

[0210] Calculate the first low-pass filtered image information using the Gaussian low-pass filter kernel at each Gaussian low-pass filter radius and the second spectral image information, that is:

[0211] Gauss k = GaussKernel k * G'

[0212] Specifically, let q be 5, then k can take values of 1, 2, 3, 4, 5, and we get: r1 = 3, r2 = 4, r3 = 5, r4 = 6, r5 = 7.

[0213] Calculate in sequence:

[0214] The Gaussian low-pass filter kernel corresponding to r1 = 3 to obtain the first Gaussian low-pass filter kernel GaussKernel1;

[0215] The Gaussian low-pass filter kernel corresponding to r2 = 4 to obtain the second Gaussian low-pass filter kernel GaussKernel2;

[0216] The Gaussian low-pass filter kernel corresponding to r3 = 5 to obtain the third Gaussian low-pass filter kernel GaussKernel3;

[0217] The Gaussian low-pass filter kernel corresponding to r4 = 6 to obtain the fourth Gaussian low-pass filter kernel GaussKernel4;

[0218] The Gaussian low-pass filter kernel corresponding to r5 = 7 to obtain the fifth Gaussian low-pass filter kernel GaussKernel5.

[0219] Then calculate in sequence:

[0220] Using the first Gaussian low-pass filter kernel and the second spectral image information, the first low-pass filtered image information Gauss1 corresponding to the first Gaussian low-pass filter kernel is obtained;

[0221] Using the second Gaussian low-pass filter kernel and the second spectral image information, the first low-pass filtered image information Gauss2 corresponding to the second Gaussian low-pass filter kernel is obtained;

[0222] Using the third Gaussian low-pass filter kernel and the second spectral image information, the first low-pass filtered image information Gauss3 is obtained corresponding to the third Gaussian low-pass filter kernel;

[0223] Using the fourth Gaussian low-pass filter kernel and the second spectral image information, the first low-pass filtered image information Gauss4 is obtained corresponding to the fourth Gaussian low-pass filter kernel;

[0224] Using the fifth Gaussian low-pass filter kernel and the second spectral image information, the first low-pass filtered image information Gauss5 is obtained corresponding to the fifth Gaussian low-pass filter kernel.

[0225] Taking the first low-pass filtered image information Gauss1 as an example, the obtained image information can be referred to Figure 5 as shown; taking the first low-pass filtered image information Gauss2 as an example, the obtained image information can be referred to Figure 6 as shown.

[0226] Summarize all the first low-pass filtered image information to obtain the first low-pass filtered image information set {Gauss1, Gauss2, Gauss3, Gauss4, Gauss5}.

[0227] S6: Convert each piece of the first low-pass filtered image information in the first low-pass filtered image information set to the third image information to obtain the third image information set;

[0228] Specifically, GaussLow k = IFFT(IFFTSHIFT(Gauss k ))

[0229] where: GaussLow k is the third image information;

[0230] Referring to {Gauss1, Gauss2, Gauss3, Gauss4, Gauss5} in S5, the third image information set {GaussLow1, GaussLow2, GaussLow3, GaussLow4, GaussLow5} is obtained.

[0231] S7: Convert the second image information to the fourth image information of the second format type;

[0232] The second format type is the logarithmic type;

[0233] The specific steps to convert the second image information to the fourth image information of the second format type are as follows:

[0234] L = lnF'

[0235] where: L is the fourth image information.

[0236] Specifically, in this step, the second image information is converted into the fourth image information in logarithmic format type.

[0237] S8: Obtain the spectrogram corresponding to the fourth image information to obtain the third spectrogram image information; specifically, the specific steps for obtaining the spectrogram corresponding to the fourth image information to obtain the third spectrogram image information are as follows:

[0238] LF = FFT(L)

[0239] where: LF is the third spectrogram image information.

[0240] S9: Use different homomorphic filtering radii to perform filtering processing on the third spectrogram image information respectively to obtain the second low-pass filtered image information set;

[0241] Specifically:

[0242] Calculate the second distance between each pixel in the third spectrogram image information and the center of the third spectrogram image information, that is:

[0243]

[0244] where:

[0245] i represents the horizontal position of the current pixel on the third spectrogram image information;

[0246] j represents the vertical position of the current pixel on the third spectrogram image information;

[0247] m represents the total number of pixels in the horizontal direction of the third spectrogram image information;

[0248] n represents the total number of pixels in the vertical direction of the third spectrogram image information;

[0249] Calculate the homomorphic low-pass filter kernel at each homomorphic filtering radius with the second distance, that is:

[0250]

[0251] where:

[0252] D s represents the s-th homomorphic filtering radius, 1 ≤ s ≤ p;

[0253] r H is the homomorphic filtering parameter, r H ≥ 1;

[0254] r L is the homomorphic filtering parameter, r L <1;

[0255] The second low-pass filtered image information is calculated using the homomorphic low-pass filter kernel and the third spectral image information at each homomorphic filtering radius, i.e.:

[0256] Homo s = H s * LF

[0257] Specifically, let p be 4, then s can take values of 1, 2, 3, 4, and we get: D1 = 4, D1 = 6, D1 = 8, D1 = 10.

[0258] Calculate in sequence:

[0259] The homomorphic low-pass filter kernel corresponding to D1 = 4 is obtained, getting the first homomorphic low-pass filter kernel H1;

[0260] The homomorphic low-pass filter kernel corresponding to D1 = 6 is obtained, getting the second homomorphic low-pass filter kernel H2;

[0261] The homomorphic low-pass filter kernel corresponding to D1 = 8 is obtained, getting the third homomorphic low-pass filter kernel H3;

[0262] The homomorphic low-pass filter kernel corresponding to D1 = 10 is obtained, getting the fourth homomorphic low-pass filter kernel H4.

[0263] Calculate in sequence:

[0264] Using the first homomorphic low-pass filter kernel and the third spectral image information, the second low-pass filtered image information Homo1 corresponding to the first homomorphic low-pass filter kernel is obtained;

[0265] Using the second homomorphic low-pass filter kernel and the third spectral image information, the second low-pass filtered image information Homo2 corresponding to the second homomorphic low-pass filter kernel is obtained;

[0266] Using the third homomorphic low-pass filter kernel and the third spectral image information, the second low-pass filtered image information Homo3 corresponding to the third homomorphic low-pass filter kernel is obtained;

[0267] Using the fourth homomorphic low-pass filter kernel and the third spectral image information, the second low-pass filtered image information Homo4 corresponding to the fourth homomorphic low-pass filter kernel is obtained.

[0268] Taking the second low-pass filtered image information Homo2 as an example, the obtained image information can be referred to Figure 7 as shown; taking the second low-pass filtered image information Homo3 as an example, the obtained image information can be referred to Figure 8 as shown.

[0269] All the second low-pass filtered image information is aggregated to obtain the second low-pass filtered image information set {Homo1, Homo2, Homo3, Homo4}.

[0270] S10: Convert the second low-pass filtered image information in the second low-pass filtered image information set to the fifth image information one by one to obtain the fifth image information set;

[0271] The specific steps for converting the second low-pass filtered image information in the second low-pass filtered image information set to the fifth image information are as follows:

[0272] HomoLow s = exp[IFFT(IFFTSHIFT(Homo s ))]

[0273] where: HomoLow s the fifth image information;

[0274] Referring to {Homo1, Homo2, Homo3, Homo4} in S9, obtain the fifth image information set {HomoLow1, HomoLow2, HomoLow3, HomoLow4}

[0275] S11: Perform a weighted transformation process on the third image information set and the fifth image information set together to obtain the optical non-uniformity image information;

[0276] Specifically, the specific steps for performing a weighted transformation process on the third image information set and the fifth image information set together to obtain the optical non-uniformity image information are as follows:

[0277]

[0278] Specifically, calculate NULow using the third image information set {GaussLow1, GaussLow2, GaussLow3, GaussLow4, GaussLow5} in S6 and the fifth image information set {HomoLow1, HomoLow2, HomoLow3, HomoLow4} in S10.

[0279] S12: Correct the second image information with the optical non-uniformity image information to obtain the de-optically non-uniformity image information, for reference Figure 9 as shown.

[0280] Specifically, the specific steps for correcting the second image information with the optical non-uniformity image information to obtain the de-optically non-uniformity image information are as follows:

[0281] ImgTemp = F' - NULow

[0282] S13: Calculate the de-optically non-uniformity shaped image information using the de-optically non-uniformity image information, as well as the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information.

[0283] Specifically, the specific steps for calculating the de-optically non-uniformity shaped image information from the de-optically non-uniformity image information and the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information are as follows:

[0284] Obtain the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information;

[0285] Calculate the de-optically non-uniformity shaped image information from the de-optically non-uniformity image information, the maximum and minimum values of the image pixel values, that is:

[0286]

[0287] Where: min(ImgTemp) is the minimum value of the image pixel values in the de-optically non-uniformity image information; max(ImgTemp) is the maximum value of the image pixel values in the de-optically non-uniformity image information.

[0288] Combining the above steps, the final de-optically non-uniformity shaped image information can be referred to Figure 10 as shown.

[0289] The optical non-uniformity of the infrared imaging system usually appears as low-frequency noise in the image, with the characteristic of slow change of neighborhood gray values. After the above correction process of S1-S13, the technical solution of the present application performs low-pass filtering processing on the second image information at different scales respectively to obtain the third image information set and the fifth image information set, and performs weighted transformation processing on the third image information set and the fifth image information set together to obtain the optical non-uniformity image information. Correct the second image information with the optical non-uniformity image information to obtain the de-optically non-uniformity image information. Finally, calculate the de-optically non-uniformity shaped image information from the de-optically non-uniformity image information and the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information.

[0290] Compare Figure 10 and Figure 2 It can be seen that the infrared image optical non-uniformity correction method provided by this embodiment can effectively improve the display effect of the infrared image, suppress the influence of optical non-uniformity, improve the contrast of details, and enhance the expressiveness of the infrared image under multi-scale frequency domain filtering processing.

[0291] Embodiment 2

[0292] Specifically, this embodiment provides an infrared image optical non-uniformity correction device based on multi-scale frequency domain filtering, including:

[0293] The first acquisition module, which is configured to obtain first infrared image information captured by an infrared thermal imager.

[0294] The first conversion module, which is configured to convert the first infrared image information into second image information of a first format type; specifically, the first format type is double-precision floating-point type. The first conversion module is configured to convert the first infrared image information into second image information of double-precision floating-point type. Specifically:

[0295] F' = double(F)

[0296] Where: F is the first infrared image information; F' is the second image information.

[0297] The first operation module, which is configured to obtain a spectrogram corresponding to the second image information to obtain first spectrogram image information. The first operation module is configured to perform a two-dimensional Fourier transform on the second image information to obtain first spectrogram image information.

[0298] G = FFT2(F')

[0299] Where: G is the first spectrogram image information; F' is the second image information.

[0300] The second operation module, which is configured to perform offset processing on the first spectrogram image information so that the low-frequency part of the first spectrogram image information is moved to the center to obtain second spectrogram image information. Specifically, the second operation module is configured to perform offset processing on the first spectrogram image information so that the low-frequency part of the first spectrogram image information is moved to the center to obtain second spectrogram image information. The specific steps are:

[0301] G' = FFTSHIFT(G)

[0302] Where: G' is the second spectrogram image information, and G is the first spectrogram image information.

[0303] The first filtering module, which is configured to perform filtering processing on the second spectrogram image information respectively using at least two Gaussian low-pass filter radii to obtain a set of first low-pass filtered image information. Specifically, the first filtering module includes:

[0304] The first operation unit, which is configured to calculate a first distance between each pixel in the second spectrogram image information and the center of the second spectrogram image information, that is:

[0305]

[0306] Where:

[0307] i represents the horizontal position of the current pixel in the second spectral image information;

[0308] j represents the vertical position of the current pixel in the second spectral image information;

[0309] m represents the total number of pixels in the horizontal direction of the second spectral image information;

[0310] n represents the total number of pixels in the vertical direction of the second spectral image information;

[0311] A second arithmetic unit, the second arithmetic unit is configured to calculate a Gaussian low-pass filter kernel at each Gaussian low-pass filter radius at a first distance, that is:

[0312]

[0313] where: r k represents the k-th Gaussian low-pass filter radius, 1 ≤ k ≤ q;

[0314] A third arithmetic unit, the third arithmetic unit is configured to calculate first low-pass filtered image information with the Gaussian low-pass filter kernel at each Gaussian low-pass filter radius and the second spectral image information, that is:

[0315] Gauss k = GaussKernel k * G'

[0316] A fourth arithmetic unit, the fourth arithmetic unit is configured to aggregate all the first low-pass filtered image information to obtain a set of first low-pass filtered image information.

[0317] A second conversion module, the second conversion module is configured to convert the first low-pass filtered image information in the set of first low-pass filtered image information to third image information one by one to obtain a set of third image information; specifically, the specific steps for the second conversion module to convert the first low-pass filtered image information in the set of first low-pass filtered image information to third image information are:

[0318] GaussLow k = IFFT(IFFTSHIFT(Gauss k ))

[0319] where: GaussLow k is the third image information.

[0320] The third conversion module, which is configured to convert the second image information into the fourth image information of the second format type; specifically, the second format type is a logarithmic type; the third conversion module is configured to convert the second image information into the fourth image information of the second format type; specifically:

[0321] L = ln F'

[0322] where: L is the fourth image information.

[0323] The third operation module, which is configured to obtain the spectrogram corresponding to the fourth image information to obtain the third spectrogram image information; specifically, the third operation module, the specific steps for the third operation module to obtain the spectrogram corresponding to the fourth image information to obtain the third spectrogram image information are:

[0324] LF = FFT(L)

[0325] where: LF is the third spectrogram image information.

[0326] The second filtering module, which is configured to perform filtering processing on the third spectrogram image information respectively by using different homomorphic filtering radii to obtain the second low-pass filtered image information set; specifically, the second filtering module includes:

[0327] The fifth operation unit, which is configured to calculate the second distance between each pixel in the third spectrogram image information and the center of the third spectrogram image information, that is:

[0328]

[0329] where:

[0330] i represents the horizontal position of the current pixel on the third spectrogram image information;

[0331] j represents the vertical position of the current pixel on the third spectrogram image information;

[0332] m represents the total number of pixels in the horizontal direction of the third spectrogram image information;

[0333] n represents the total number of pixels in the vertical direction of the third spectrogram image information;

[0334] The sixth operation unit, which is configured to calculate the homomorphic low-pass filter kernel at each homomorphic filtering radius with the second distance, that is:

[0335]

[0336] where:

[0337] Ds Denote the s-th homomorphic filtering radius, where 1 ≤ s ≤ p;

[0338] r H is a homomorphic filtering parameter, and r H ≥ 1;

[0339] r L is a homomorphic filtering parameter, and r L < 1;

[0340] The seventh operation unit, which is configured to calculate the second low-pass filtered image information with the homomorphic low-pass filter kernel and the third spectral image information at each homomorphic filtering radius, i.e.:

[0341] Homo s = H s * LF

[0342] The eighth operation unit, which is configured to aggregate all the second low-pass filtered image information to obtain a set of second low-pass filtered image information.

[0343] The fourth conversion module, which is configured to convert the second low-pass filtered image information in the set of second low-pass filtered image information into the fifth image information one by one to obtain a set of fifth image information; specifically, the fourth conversion module, the specific steps for the fourth conversion module to convert the second low-pass filtered image information in the set of second low-pass filtered image information into the fifth image information are:

[0344] HomoLow s = exp[IFFT(IFFTSHIFT(Homo s ))]

[0345] where: HomoLow s is the fifth image information.

[0346] The fourth operation module, which is configured to perform weighted transformation processing on the third image information set and the fifth image information set to obtain optical non-uniformity image information; specifically, the fourth operation module, the specific steps for the fourth operation module to perform weighted transformation processing on the third image information set and the fifth image information set to obtain optical non-uniformity image information are:

[0347]

[0348] The first correction module, which is configured to correct the second image information with the optical non-uniformity image information to obtain the de-optically non-uniformity image information. Specifically, the specific steps for the first correction module to correct the second image information with the optical non-uniformity image information to obtain the de-optically non-uniformity image information are as follows:

[0349] ImgTemp = F' - NULow.

[0350] The fifth operation module, which is configured to calculate the de-optically non-uniformity shaped image information with the de-optically non-uniformity image information, as well as the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information. Specifically, the fifth operation module includes:

[0351] The ninth operation unit, which is configured to obtain the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information;

[0352] The tenth operation unit, which is configured to calculate the de-optically non-uniformity shaped image information with the de-optically non-uniformity image information, the maximum and minimum values of the image pixel values, that is:

[0353]

[0354] Where:

[0355] min(ImgTemp) is the minimum value of the image pixel values in the de-optically non-uniformity image information;

[0356] max(ImgTemp) is the maximum value of the image pixel values in the de-optically non-uniformity image information.

[0357] Embodiment 3

[0358] This embodiment provides a processing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering as described above are implemented.

[0359] As Figure 11As shown, the computer system 800 includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 802 or the program loaded from the storage section into the random access memory (RAM) 803. In the RAM 803, various programs and data required for system operation are also stored. The CPU 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0360] The following components are connected to the I / O interface 805: an input section 806 including a keyboard, a mouse, etc.; an output section including a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN card, a modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive is also connected to the I / O interface 805 as required. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 810 as required so that a computer program read from it can be installed into the storage section 808 as required.

[0361] Specifically, according to an embodiment of the present invention, the process described above with reference to the flow Figure 1 can be implemented as a computer software program. For example, Embodiment 1 of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes program codes for performing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication section, and / or installed from the removable medium. When the computer program is executed by the central processing unit (CPU) 501, the above functions defined in the system of the present application are executed.

[0362] Embodiment 4

[0363] This embodiment provides a computer-readable storage medium, which has a computer program, and when the computer program is executed by a processor, it implements the steps of the infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering as described above.

[0364] It should be noted that the computer-readable medium shown in the present invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. In the present invention, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and this computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0365] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and the above module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the block can occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0366] The units involved in the embodiments of the present invention can be implemented in software or in hardware, and the described units can also be provided in a processor. Among them, the names of these units do not, in some cases, constitute a limitation on the units themselves. The described units or modules can also be provided in a processor. For example, it can be described as: a processor includes a processing module, an initialization module, and a determination module.

[0367] Among them, the names of these units or modules do not, in some cases, constitute a limitation on the units or modules themselves. For example, the initialization module can be described as "the initialization module for initializing the timer structure".

[0368] As another aspect, the present embodiment also provides a computer-readable medium, which can be included in the electronic device described in the above embodiments; or it can exist alone without being assembled into the electronic device. The above computer-readable medium carries one or more programs. When the above one or more programs are executed by an electronic device, the electronic device is caused to implement the method steps of timer extension as described in the above embodiments.

[0369] It should be noted that although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0370] In addition, although the steps of the methods in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be executed in that specific order, or that all the steps shown must be executed to achieve the desired result. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step for execution, and / or one step can be decomposed into multiple steps for execution, etc.

[0371] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described here can be implemented by software or by a combination of software and necessary hardware.

[0372] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages. For example, object-oriented programming languages such as Java and interpreted scripting languages such as JavaScript, etc.

[0373] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0374] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0375] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are performed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0376] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present application.

[0377] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalent technologies, this application is also intended to cover these modifications and variations.

Claims

1. An infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering, characterized in that, It includes the following steps: Obtain the first infrared image information, which is captured by an infrared thermal imager; Convert the first infrared image information into second image information of the first format type; the first format type is double-precision floating-point type; Obtain the spectrogram corresponding to the second image information to get the first spectrogram image information; Perform offset processing on the first spectrogram image information so that the low-frequency part of the first spectrogram image information is moved to the center to obtain the second spectrogram image information; Use at least two Gaussian low-pass filter radii to perform filtering processing on the second spectrogram image information respectively to obtain the first set of low-pass filtered image information; Convert the first low-pass filtered image information in the first set of low-pass filtered image information into third image information one by one to obtain the third set of image information; Convert the second image information into fourth image information of the second format type; the second format type is logarithmic type; Obtain the spectrogram corresponding to the fourth image information to get the third spectrogram image information; Use different homomorphic filter radii to perform filtering processing on the third spectrogram image information respectively to obtain the second set of low-pass filtered image information; Convert the second low-pass filtered image information in the second set of low-pass filtered image information into fifth image information one by one to obtain the fifth set of image information; Perform weighted transformation processing on the third set of image information and the fifth set of image information together to obtain the optical non-uniformity image information; Correct the second image information with the optical non-uniformity image information to obtain the de-optically non-uniformity image information; Calculate the de-optically non-uniformity shaped image information based on the de-optically non-uniformity image information and the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information.

2. The method for correcting the optical non-uniformity of an infrared image based on multi-scale frequency domain filtering according to claim 1, wherein: The first format type is double-precision floating-point type; The specific steps for converting the first infrared image information into second image information of double-precision floating-point type are: Wherein: F is the first infrared image information; F' is the second image information.

3. The method for correcting the optical non-uniformity of an infrared image based on multi-scale frequency domain filtering according to claim 2, wherein: The specific steps for obtaining the spectrogram corresponding to the second image information to get the first spectrogram image information are: Perform two-dimensional Fourier transform on the second image information to obtain the first spectrogram image information Wherein: G is the first spectral image information; F' is the second image information.

4. The method for correcting the optical non-uniformity of an infrared image based on multi-scale frequency domain filtering according to claim 3, wherein: The specific steps for performing offset processing on the first spectrogram image information so that the low-frequency part of the first spectrogram image information is moved to the center to obtain the second spectrogram image information are: Wherein: G' is the second spectral image information, G is the first spectral image information.

5. The method for correcting the optical non-uniformity of an infrared image based on multi-scale frequency domain filtering according to claim 4, wherein: The specific steps for using at least two Gaussian low-pass filter radii to perform filtering processing on the second spectrogram image information respectively to obtain the first set of low-pass filtered image information are: Calculate the first distance between each pixel in the second spectrogram image information and the center of the second spectrogram image information, that is: Where: i Indicates the horizontal position of the current pixel on the second spectral image information; j Indicates the vertical position of the current pixel in the second spectral image information; m Indicates the total number of pixels in the horizontal direction on the second spectral image information; n Indicates the total number of pixels in the vertical direction on the second spectral image information; Calculate the Gaussian low-pass filter kernel at each Gaussian low-pass filter radius at the first distance, that is Wherein: r k represents the k th Gaussian low-pass filtering radius, 1 ≤ k ≤ q ; Using the Gaussian low-pass filter kernel at each Gaussian low-pass filter radius and the second spectral image information Calculate the first low-pass filtered image information, that is Aggregate all the first low-pass filtered image information to obtain the first low-pass filtered image information set 6. A method for correcting optical non-uniformity of an infrared image based on multi-scale frequency domain filtering according to any one of claims 1 to 5, characterized in that The second format type is logarithmic The specific steps for converting the second image information to the fourth image information of the second format type are Wherein: L is the fourth image information.

7. A method for correcting optical non-uniformity of an infrared image based on multi-scale frequency domain filtering according to claim 6, characterized in that The specific steps for obtaining the spectral diagram corresponding to the fourth image information to obtain the third spectral image information are Wherein: LF is the third spectral image information.

8. A method for correcting optical non-uniformity of an infrared image based on multi-scale frequency domain filtering according to claim 7, characterized in that The specific steps for filtering the third spectral image information with different homomorphic filter radii respectively to obtain the second low-pass filtered image information set are Calculate the second distance between each pixel in the third spectral image information and the center of the third spectral image information, that is Where i Indicates the horizontal position of the current pixel in the third spectral image information; j Indicates the vertical position of the current pixel on the third spectral image information; m Indicates the total number of pixels in the horizontal direction on the third spectral image information; n Indicates the total number of pixels in the vertical direction on the third spectral image information; Calculate the homomorphic low-pass filter kernel at each homomorphic filter radius at the second distance, that is Where D S Indicates the S th homomorphic filtering radius, 1 ≤ s ≤ p ; r H is the homomorphic filtering parameter, r H ≥1; r L is the homomorphic filtering parameter, r L <1; Using the homomorphic low-pass filter kernel at each homomorphic filter radius and the third spectral image information, calculate the second low-pass filtered image information, that is Aggregate all the second low-pass filtered image information to obtain the second low-pass filtered image information set 9. A method for correcting optical non-uniformity of an infrared image based on multi-scale frequency domain filtering according to claim 8, characterized in that The specific steps for converting the first low-pass filtered image information in the first low-pass filtered image information set to the third image information are Wherein: GaussLow k is the third image information; The specific steps for converting the second low-pass filtered image information in the second low-pass filtered image information set to the fifth image information are Wherein: HomoLow s The fifth image information; The specific steps for jointly performing weighted transformation processing on the third image information set and the fifth image information set to obtain the optical non-uniformity image information are 。 10. A method for correcting optical non-uniformity of an infrared image based on multi-scale frequency domain filtering according to claim 9, characterized in that The specific steps for correcting the second image information with the optical non-uniformity image information to obtain the de-optically non-uniformity image information are The specific steps for calculating the de-optically non-uniformity shaped image information with the de-optically non-uniformity image information and the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information are Obtain the maximum and minimum values of the image pixel values in the de-optically non-uniformity image information Calculate the de-optically non-uniformity shaped image information with the de-optically non-uniformity image information, the maximum and minimum values of the image pixel values, that is Where min(ImgTemp) is the minimum value of the image pixel values in the de-optically non-uniformity image information max(ImgTemp) is the maximum value of the image pixel values in the image information after removing optical non-uniformity.

11. An infrared image optical non-uniformity correction device based on multi-scale frequency domain filtering, characterized in that, Including: A first acquisition module configured to acquire first infrared image information captured by an infrared thermal imager. A first conversion module configured to convert the first infrared image information into second image information of a first format type, where the first format type is double-precision floating-point type. A first operation module configured to obtain a spectrogram corresponding to the second image information to obtain first spectral image information. A second operation module configured to perform offset processing on the first spectral image information so that the low-frequency part of the first spectral image information is moved to the center to obtain second spectral image information. A first filtering module configured to perform filtering processing on the second spectral image information using at least two Gaussian low-pass filter radii respectively to obtain a set of first low-pass filtered image information. A second conversion module configured to convert the first low-pass filtered image information in the set of first low-pass filtered image information into third image information one by one to obtain a set of third image information. A third conversion module configured to convert the second image information into fourth image information of a second format type, where the second format type is logarithmic type. A third operation module configured to obtain a spectrogram corresponding to the fourth image information to obtain third spectral image information. A second filtering module configured to perform filtering processing on the third spectral image information using different homomorphic filter radii respectively to obtain a set of second low-pass filtered image information. A fourth conversion module configured to convert the second low-pass filtered image information in the set of second low-pass filtered image information into fifth image information one by one to obtain a set of fifth image information. A fourth operation module configured to perform weighted transformation processing on the set of third image information and the set of fifth image information together to obtain optical non-uniformity image information. A first correction module configured to correct the second image information with the optical non-uniformity image information to obtain image information after removing optical non-uniformity. A fifth operation module configured to calculate and obtain image information after removing optical non-uniformity and shaping based on the image information after removing optical non-uniformity, as well as the maximum and minimum values of the image pixel values in the image information after removing optical non-uniformity.

12. An electronic processing device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering according to any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the infrared image optical non-uniformity correction method based on multi-scale frequency domain filtering according to any one of claims 1 to 10.

Images