A method for correcting fixed-pattern noise of an infrared detector

Abstract

The application relates to a fixed-pattern noise correction method of an infrared detector and belongs to the technical field of infrared imaging. The application can eliminate the fixed-pattern noise of the infrared detector in a real-time image by taking the fixed-pattern noise phenomenon of the infrared detector as a template and calculating the fixed-pattern noise intensity in the real-time image through correlation.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of infrared imaging technology, and particularly relates to a method for correcting fixed pattern noise of an infrared detector. BACKGROUND

[0002] A staring infrared detector is a large array semiconductor device. Due to the discreteness of the material parameters of the device, the complexity of the processing technology and the multiple processes, the performance parameters of each detection unit of the infrared focal plane array will be different, resulting in the inconsistency of the infrared response between the detection units, which is the non-uniformity of the detector, also known as fixed pattern noise or fixed pattern noise. The fixed pattern is generated during the preparation process of the infrared detector assembly, and each detector has its own fixed pattern noise phenomenon. The noise will change in intensity with different working conditions.

[0003] The non-uniformity of the infrared detector needs to be corrected to eliminate the fixed pattern noise, so as to ensure the normal use of the detector.

[0004] At present, most of the infrared imaging systems are based on the calibration method of black body, that is, a black body is placed in front of the infrared detector or infrared imager, the black body is used as a uniform radiation source, the responses of each pixel of the detector are measured, and the correction parameters are obtained according to the response values. This process is called calibration. After calibration, these data are read and corresponding operations are performed in the working process, which is called correction. The specific correction algorithms include single temperature point correction and two temperature point correction, as shown in equations (1)-(4).

[0005] 1) Single temperature point correction algorithm

[0006] Calibration formula:

[0007]

[0008] Correction formula:

[0009] Y ij (T)=X ij (T)+O ij (2)

[0010] 2) Two temperature point correction algorithm

[0011] Calibration formula:

[0012]

[0013] Correction formula:

[0014] Y ij (T)=G ij X ij (T)+O ij (4)

[0015] where X ij (T) is the response output of the pixel (i,j) to the blackbody of radiation temperature T, is the average value of the response output of all pixels to the blackbody of radiation temperature T, T1, T2 are the temperatures of the calibration blackbody, G ij and O ij is the correction parameter of each pixel, Y ij (T) is the corrected result.

[0016] The non-uniformity correction method based on blackbody calibration takes the blackbody as a reference source and has the characteristics of scene dependence. When the scene used is inconsistent with the radiation characteristics of the calibration blackbody, the blackbody calibration method cannot achieve good correction effect, and there are two limiting conditions:

[0017] 1) The radiation brightness of the scene during correction is equivalent to the radiation brightness of the blackbody during calibration;

[0018] 2) The spectral distribution of the scene during correction is equivalent to the spectral distribution of the blackbody radiation during calibration.

[0019] Therefore, when the use environment is inconsistent with the calibration environment, the phenomenon of degradation of the detector fixed pattern will occur. SUMMARY

[0020] (I) Technical problems to be solved

[0021] The technical problem to be solved by the present application is to design a detector fixed pattern correction method independent of the environment.

[0022] (II) Technical solutions

[0023] In order to solve the above technical problems, the present application provides an infrared detector fixed pattern noise correction method, comprising the following steps:

[0024] 1) Step 1, calibration:

[0025] After the output image of the infrared detector without non-uniformity correction is processed by median filtering, an image with clean background and only the phenomenon of infrared detector fixed pattern noise is obtained as a template image Z0(x,y);

[0026] 2) Step 2, correlation calculation:

[0027] Through the correlation operation of the real-time image generated by the infrared detector and the template image Z0(x,y), the feature parameters are extracted in the image, and the correlation coefficient of the real-time image and the template image is obtained as the correction coefficient;

[0028] Let the real-time image generated by the infrared detector when working be I(x, y), and the correlation coefficient of the real-time image and the template image is calculated according to the following formula:

[0029] The autocorrelation coefficient is:

[0030] The cross-correlation coefficient is:

[0031] The correlation coefficient is: α = r2 / r1;

[0032] In the formula, I is the infrared image in the current state, is the average value of the gray scale of all pixels of the infrared image; is the average value of the gray scale of all pixels of the template;

[0033] 3) Step 3, correction:

[0034] In the real-time image, the product of the template image and the correlation coefficient is subtracted to obtain the corrected infrared image: I = I - α × Z0.

[0035] The application also provides a system for implementing the method.

[0036] The application also provides an application of the method in the field of infrared imaging technology.

[0037] The application also provides an application of the system in the field of infrared imaging technology.

[0038] (III) Beneficial effects

[0039] The effect of the application is that for a general infrared imaging system, the application can eliminate the infrared detector fixed pattern noise in the real-time image by taking the fixed pattern noise phenomenon of the infrared detector as a template and calculating the fixed pattern noise intensity in the real-time image through correlation calculation. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 It is a schematic diagram of an infrared imaging system composition;

[0041] Figure 2 It is a schematic diagram of the working process of an infrared imaging system;

[0042] Figure 3 It is an infrared fixed pattern noise template;

[0043] Figure 4 It is a degraded image of the pattern noise of the infrared detector;

[0044] Figure 5 It is an image corrected by the fixed pattern noise of the application. DETAILED DESCRIPTION

[0045] In order to make the objects, contents and advantages of the present application clearer, the specific embodiments of the present application are described in further detail below in combination with the drawings and examples.

[0046] The present application provides a correction method for fixed pattern noise of infrared detector, and particularly relates to a correction method based on image similarity calibration.

[0047] A general infrared imaging system comprises optical system, infrared detector assembly, image pre-processing circuit and other components, as shown in the figure. Figure 1 Radiation of the scene is converged by the lens of the optical system to the focal plane of the infrared detector, the infrared detector converts the scene information into electrical signal, and the electrical signal is read out and processed and displayed by the back-end information processing to obtain the image of the scene.

[0048] In the present application, for the infrared detector or the imaging device with infrared detector as the core, the output image of the infrared detector is excited by the change of the environment after non-uniformity correction, and the fixed pattern noise of the infrared detector appears in the output image, the image is stored as a template; or in the original image without non-uniformity correction, the image with clean background and only the fixed pattern noise phenomenon of the detector is made by filtering method as a template. The above template is called fixed noise image template, and is simply called template. In use, the real-time infrared image is correlated with the template to obtain the correlation coefficient, the template is multiplied by the correlation coefficient as the real-time fixed pattern noise, which is subtracted in the real-time image to obtain the corrected image.

[0049] The infrared detector fixed pattern noise correction method of the present application is described further below:

[0050] 1) Step 1, calibration:

[0051] The output image of the infrared detector without non-uniformity correction is processed by median filtering to obtain an image with clean background and only the fixed pattern noise phenomenon of the infrared detector, which is used as a template image Z0(x, y) and stored in the memory of the infrared imaging system.

[0052] 2) Step 2, correlation calculation:

[0053] The correlation operation between the real-time image generated by the infrared detector and the template image Z0(x, y) extracts the characteristic parameters in the image to obtain the correlation coefficient (correlation coefficient) of the real-time image and the template image as the correction coefficient.

[0054] Let the real-time image generated by the infrared detector in operation be I(x, y), and the correlation coefficient of the real-time image and the template image is calculated according to the following formula:

[0055] Autocorrelation coefficient:

[0056] Cross-correlation coefficient:

[0057] Correlation coefficient: α = r2 / r1.

[0058] In the formula, I is the infrared image in the current state, is the average value of the gray scale of all pixels of the infrared image; is the average value of the gray scale of all pixels of the template.

[0059] 3) Step 3, correction:

[0060] In the real-time image, the product of the template image and the correlation coefficient is subtracted to obtain the corrected infrared image.

[0061] Image correction: I = I - α × Z0

[0062] Figure 3 Z0 is the infrared detector fixed pattern noise template made using the original image without non-uniformity correction; Figure 4 Z is the fixed pattern degradation phenomenon of the infrared imaging system after non-uniformity correction, which occurs in use; Figure 5 I is the infrared image after correction using the method. It can be seen that for the infrared imaging system, the present application can eliminate the infrared detector fixed pattern noise in the real-time image.

[0063] The above only describes the preferred embodiments of the present application, and it should be noted that for ordinary skilled in the art, without departing from the technical principles of the present application, a number of improvements and modifications can be made, and these improvements and modifications should be considered as the protection scope of the present application.

Claims

1. An infrared detector fixed pattern noise correction method, characterized by, The method comprises the following steps: 1) Step 1, calibration: The output image of the infrared detector without non-uniformity correction is processed by median filtering to obtain an image with clean background and only fixed-pattern noise phenomenon of the infrared detector, which is used as a template image Z0(x, y); 2) Step 2, correlation calculation: The correlation between the real-time image generated by the infrared detector and the template image Z0(x, y) is calculated to extract the feature parameters in the image and obtain the correlation coefficient of the real-time image and the template image as the correction coefficient; Let the real-time image generated by the infrared detector during operation be I(x, y), and the correlation coefficient of the real-time image and the template image is calculated according to the following formula: Autocorrelation coefficient: Cross-correlation coefficient: Correlation coefficient: α = r2 / r1; where I is the infrared image in the current state, is the average value of the gray scale of all pixels of the infrared image; is the average value of the gray scale of all pixels of the template; 3) Step 3, correction: In the real-time image, the product of the template image and the correlation coefficient is subtracted to obtain the corrected infrared image: I = I - α × Z0.

2. The method of claim 1, wherein, In step 1, the template image Z0(x, y) is also stored in the memory of the infrared imaging system.

3. The method of claim 2, wherein, In the infrared imaging system, the radiation of the scene is focused by the lens of the optical system to the focal plane of the infrared detector, the infrared detector converts the scene information into an electrical signal, and the rear-end information processing reads and processes the signal to obtain the image of the scene.

4. The method of claim 2, wherein, The infrared imaging system further comprises a power supply circuit.

5. The method of claim 2, wherein, The infrared imaging system further comprises an analog-to-digital conversion circuit.

6. The method according to claim 1, wherein The infrared detector is a staring infrared detector.

7. The method of claim 6, wherein, The staring infrared detector is a large-area semiconductor device.

8. A system for implementing the method according to any one of claims 1 to 7.

9. The use of the method according to any one of claims 1 to 7 in the field of infrared imaging technology.

10. The use of the system according to claim 8 in the field of infrared imaging technology.

Images