Infrared image flat-field noise correction method and device based on noise model

Abstract

The application discloses an infrared image flat-field noise correction method and device based on a noise model. The application firstly acquires infrared original images I i at different temperatures, extracts Gaussian noise thereof as initial Gaussian noise templates at the temperatures, then performs difference to obtain initial Gaussian noise templates at all temperatures within a working environment temperature range, then adjusts parameters of the initial Gaussian noise template at the working environment temperature according to a working environment temperature of a real-time infrared image, extracts a Gaussian noise template after each adjustment parameter, and takes a fine-tuned Gaussian noise template corresponding to the minimum distance between the real-time infrared image and the fine-tuned Gaussian noise template as an updated Gaussian noise template at the working environment temperature, and finally subtracts the updated Gaussian noise template of the Gaussian noise template updating module from the real-time infrared image to realize noise reduction of the real-time infrared image. The application can be updated in real time to ensure real-time correction of the infrared image and stability of the image.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of non-cooled infrared imaging, and particularly relates to an infrared image flat-field noise correction method and device based on a noise model. BACKGROUND

[0002] Non-cooled infrared imaging sensors cause non-uniform brightness of images due to non-uniform response of the sensors, illumination differences or other reasons, which is manifested as inconsistent response between the center and the edge of the image. At present, the most common method for solving the problem of non-uniform response is to rely on calibration correction or noise estimation. The method relying on calibration correction measures the radiation response of a black body under uniform radiation at high and low temperature conditions, and corrects the image using the response of a single point or two points. This method is based on the premise that the response of the detector is time-stable. However, in real use scenarios, rapid changes in environmental temperature and temperature differences of the detector can easily cause inaccurate calibration correction or inadaptation to environmental changes. The method relying on noise estimation relies on the estimation ability of the algorithm, and the estimation process is prone to overestimation, estimation failure and other problems, resulting in missing image details or introducing new noise and other problems. How to update and correct the noise model in real time is very important for maintaining the stability of infrared for a long time. SUMMARY

[0003] Therefore, the present application provides an infrared image flat-field noise correction method and device based on a noise model, which determines the parameters of the noise model through initial calibration, and updates the parameters in real time to correct the flat field of the infrared image.

[0004] The infrared image flat-field noise correction method based on a noise model of the present application comprises the following steps:

[0005] Step one, acquiring infrared original images I i under different temperatures (assuming N are collected) , i = 1, 2, …, N. The acquisition method of each image is as follows:

[0006] Suppose the temperatures of the N temperature points are T i , i = 1, 2, …, N, then the original image I i can be acquired by placing the imaging assembly in a constant-temperature box with a temperature of T i and facing the photosensitive surface of the imaging assembly to the black body with a uniform temperature field.

[0007] Step two, extracting an initial Gaussian noise template under the temperature T i condition from the acquired infrared original image I i and storing it, wherein the Gaussian model template is:

[0008]

[0009] a i ,b i ,c i ,d i ,e i ,f i are the Gaussian model fitting parameters to be solved, which can be solved by minimizing the fitting error function E(a i ,b i ,c i ,d i ,e i ,f i ) that measures the difference between the fitted surface and the actual data points, and the specific fitting error function formula is as follows:

[0010]

[0011] where W and H are the width and height of the image.

[0012] Step three, obtain the initial Gaussian noise template parameters under a specific temperature condition T according to interpolation, a T ,b T ,c T ,d T ,e T ,f T The calculation method is:

[0013] a T =L(a i )

[0014] b T =L(b i )

[0015] c T =L(c i )

[0016] d T =L(d i )

[0017] e T =L(e i )

[0018] f T =L(f i )

[0019] where L is the interpolation function, which can be linear interpolation, quadratic function interpolation, or polynomial interpolation.

[0020] Step four, according to the real-time infrared image and the initial Gaussian noise template parameters a T ,b T ,cT ,d T ,e T ,f T The Gaussian noise template is updated in real time within a certain threshold range, and the updated Gaussian noise template is:

[0021]

[0022] The updated Gaussian noise template parameter a T ′ ,b T ′ ,c T ′ ,d T ′ ,e T ′ ,f T ′ The minimum Euclidean distance D in the range of Delta a, Delta b, Delta c, Delta d, Delta e and Delta f is obtained by calculation:

[0023]

[0024] Step five, obtain the real-time image after noise reduction, and the calculation method is:

[0025] I out = I T -G

[0026] I out is the image I T after noise reduction.

[0027] The application also provides an infrared image flat-field noise correction device based on a noise model, which comprises a memory, a Gaussian noise template extraction module, a Gaussian noise template updating module and a noise reduction module; the noise reduction is performed by using the above method; wherein,

[0028] The memory stores infrared original images I i under different sampling point temperatures T i in a working environment temperature range of a non-cooled infrared imaging sensor.

[0029] The Gaussian noise template is used to extract the Gaussian noise of the infrared original image I i under each sampling point temperature T i as the initial Gaussian noise template under the temperature, and the initial Gaussian noise templates under all temperatures in the working environment temperature range are obtained through difference.

[0030] The Gaussian noise template updating module adjusts parameters of an initial Gaussian noise template under the working environment temperature of the real-time infrared image, calculates distances between the Gaussian noise templates after the adjustment of the parameters and the real-time infrared image, and extracts a Gaussian noise template corresponding to a minimum distance as an updated Gaussian noise template under the working environment temperature.

[0031] The denoising module subtracts the updated Gaussian noise template of the Gaussian noise template updating module from the real-time infrared image, so as to realize the denoising of the real-time infrared image.

[0032] The application further provides an infrared image flat-field noise correction computer program product based on a noise model, which comprises a non-transient readable storage medium and a computer program.

[0033] Advantages:

[0034] 1) The parameters of the noise model are determined through initial calibration, and the parameters are updated in real time when the real-time image is obtained, so that the real-time correction of the infrared image is ensured, and the stability of the image is ensured.

[0035] 2) The technical algorithm has low complexity and is convenient to realize on different platforms such as FPGA, DSP, ARM and PC. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 The method flowchart of the application.

[0037] Figure 2 The effect comparison chart before and after the method of the application is processed, a) is a chart before processing, and b) is an effect chart after processing. Figure 2 It can be seen from the chart that the image processed by the method of the application has obvious correction effect on flat-field noise compared with the original image. DETAILED DESCRIPTION

[0038] The application will be described in detail below with reference to the drawings and examples.

[0039] The application provides an infrared image flat-field noise correction method based on a noise model.

[0040] Step one, acquire infrared original images I i under different temperatures (suppose six are acquired), i.e., I i=1, 2, …, 6. The acquisition method of each image is as follows:

[0041] The temperature range of the collected temperature is -40℃ to 60℃, and the temperatures of the six temperature points are -40℃, -20℃, 0℃, 20℃, 40℃ and 60℃, respectively, and the temperature collection step is 20℃. By sequentially placing the imaging assembly in the temperature environment of the six temperature points of the thermostat and facing the photosensitive surface of the imaging assembly to the black body with a uniform temperature field, the original image I i .

[0042] Step two, by acquiring the infrared original image I i Extract the initial Gaussian noise template under six temperature conditions and store them, wherein the Gaussian model template is:

[0043]

[0044] a i ,b i ,c i ,d i ,e i ,f i are the Gaussian model fitting parameters to be solved, which can be solved by minimizing the fitting error function E(a i ,b i ,c i ,d i ,e i ,f i ) that measures the difference between the fitted surface and the actual data points. The specific fitting error function formula is as follows:

[0045]

[0046] Step three, according to the interpolation, the initial Gaussian noise template parameters under any temperature T in the working temperature range are obtained. Taking linear interpolation as an example, a T ,b T ,c T ,d T ,e T ,f T The calculation method is:

[0047]

[0048] Where T belongs to the range of T i ~T i+1 .

[0049] Step four, extract the initial Gaussian noise template parameters a T ,b T ,c T ,d T ,e T ,f T, fine-tune the template parameters within a certain threshold range and update the Gaussian noise template; the updated Gaussian noise template is:

[0050]

[0051] Updated Gaussian noise template parameter a T ′ ,b T ′ ,c T ′ ,d T ′ ,e T ′ ,f T ′ By calculating the minimum Euclidean distance D within the set thresholds Δa, Δb, Δc, Δd, Δe, and Δf, we can get:

[0052]

[0053] Step 5: Obtain the real-time image after noise reduction. The calculation method is:

[0054] I out =I T -G

[0055] I out For image I T Final image after denoising.

[0056] The present invention also provides an infrared image flat field noise correction device based on a noise model, comprising: a memory, a Gaussian noise template extraction module, a Gaussian noise template update module and a denoising module; denoising is performed using the above method; wherein,

[0057] The memory stores the temperature T of different sampling points within the working environment temperature range of the uncooled infrared imaging sensor. i The original infrared image I i ;

[0058] The Gaussian noise template is used to extract the temperature T of each sampling point i The original infrared image I i The Gaussian noise of is used as the initial Gaussian noise template at the temperature, and the initial Gaussian noise templates at all temperatures within the working environment temperature range are obtained by difference;

[0059] The Gaussian noise template updating module adjusts parameters of an initial Gaussian noise template under a working environment temperature of a real-time infrared image, calculates distances between the Gaussian noise template after adjustment of each parameter and the real-time infrared image, and extracts a Gaussian noise template corresponding to a minimum distance value as an updated Gaussian noise template under the working environment temperature.

[0060] The denoising module subtracts the updated Gaussian noise template of the Gaussian noise template updating module from the real-time infrared image, so as to realize denoising of the real-time infrared image.

[0061] The application further provides an infrared image flat-field noise correction computer program product based on a noise model, which comprises a non-transient readable storage medium and a computer program, the computer program is tangibly stored in the non-transient readable storage medium, and the computer program is executed by a processor in an FPGA, a DSP, an ARM or other computers to realize steps of the above correction method.

[0062] To sum up, the above is only a preferred embodiment of the application, and is not used to limit the protection scope of the application. Any modification, equivalent replacement, improvement and the like within the spirit and principle of the application should be included in the protection scope of the application.

Claims

1. A method for correcting flat-field noise in infrared images based on a noise model, characterized in that: include: Step 1: Obtain the temperature of different sampling points within the working environment temperature range of the uncooled infrared imaging sensor Infrared original image , i=1, 2, …, N; Step 2: Assuming the noise is Gaussian noise, extract the temperature of each sampling point Lower infrared original image Gaussian noise, as the temperature The initial Gaussian noise template; Step 3: interpolate the parameters of the initial Gaussian noise template at each sampling point temperature obtained in step 2 to obtain the initial Gaussian noise template at all temperatures within the working environment temperature range; Step 4: Update the initial Gaussian noise template at the working environment temperature according to the working environment temperature of the real-time infrared image, specifically: Fine-tune the parameters of the initial Gaussian noise template at the working environment temperature, calculate the Euclidean distance between the Gaussian noise template after each fine-tuning parameter and the real-time infrared image, and the Gaussian noise template corresponding to the minimum value of the Euclidean distance is the updated Gaussian noise template; Step 5: Subtract the Gaussian noise template updated in step 4 from the real-time infrared image to achieve noise reduction of the real-time infrared image.

2. The method according to claim 1, wherein In the step 1, the imaging components are placed in a temperature environment of Place the photosensitive surface of the imaging component in a constant temperature box and face the black body with a uniform temperature field to obtain the original image. .

3. The method according to claim 1, wherein In step 2, the expression of Gaussian noise is: Where (x, y) is the pixel coordinate of the original infrared image; The Gaussian model fitting parameters to be solved are obtained by minimizing the fitting error function that measures the difference between the Gaussian fitting surface and the actual data points. To obtain; the fitting error function is: in, W and H are the width and height of the image respectively.

4. The method according to claim 3, wherein In step 4, the updated Gaussian noise template is: Updated Gaussian noise template parameters Set by calculation Euclidean distance within range The minimum is: 。 5. The method according to any one of claims 1 to 4, characterized in that: In the step three, the interpolation is linear interpolation, quadratic function interpolation or polynomial interpolation.

6. An infrared image flat field noise correction device based on a noise model, characterized in that: include: Memory, Gaussian noise template extraction module, Gaussian noise template update module and denoising module; denoising is performed using the method according to any one of claims 1 to 5; wherein, The memory stores the temperature of different sampling points within the working environment temperature range of the uncooled infrared imaging sensor. Infrared original image ; The Gaussian noise template extraction module is used to extract the temperature of each sampling point Infrared original image The Gaussian noise of is used as the initial Gaussian noise template at the temperature, and the initial Gaussian noise templates at all temperatures within the working environment temperature range are obtained by interpolation; The Gaussian noise template updating module is used to fine-tune the parameters of the initial Gaussian noise template at the working environment temperature of the real-time infrared image, calculate the Euclidean distance between the Gaussian noise template after each fine-tuning parameter and the real-time infrared image, and extract the Gaussian noise template corresponding to the minimum value of the Euclidean distance as the updated Gaussian noise template at the working environment temperature; The denoising module is used to subtract the updated Gaussian noise template of the Gaussian noise template updating module from the real-time infrared image to achieve denoising of the real-time infrared image.

7. A computer program product for infrared image flat-field noise correction based on a noise model, characterized in that: The method comprises a non-transitory readable storage medium and a computer program, wherein the computer program is tangibly stored on the non-transitory readable storage medium, and the computer program is executed by a processor in an FPGA, DSP, ARM or other computer to implement the steps of the correction method according to any one of claims 1 to 5.

Images