Infrared image dynamic compression and detail enhancement method, storage medium and device

Abstract

The present invention discloses a method, storage medium and device for dynamic compression and detail enhancement of infrared images. The method comprises: using a parameter-adaptive guided filter to divide the original image into a base layer and a detail layer; the adaptive parameters of the guided filter are calculated based on the local variance histogram of the original image to be processed, and the local variance histogram of the original image is obtained by the local variance statistics of the original image to be processed; the base layer is dynamically compressed; the detail layer is enhanced based on a gain coefficient representing the overall clarity of the image; the gain coefficient is obtained based on the adaptive parameters; the base layer image after dynamic range compression and the detail enhanced image are combined to obtain the final dynamic range compressed image. The present invention introduces adaptive parameters to make the filter suitable for different infrared images. In the enhancement of the detail layer, the gain coefficient representing the overall clarity of the image is calculated based on the aforementioned adaptive parameters to improve the detail performance in the image.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to an infrared image dynamic compression and detail enhancement method, storage medium and device, in particular to an infrared image dynamic range compression and detail enhancement method, storage medium and device based on parameter adaptation. Background Art

[0002] Since the actual temperature range of objects is wide and the infrared band is much wider than the visible light band, the information that infrared detectors can receive is much richer than visible light imaging. The infrared image data they output usually has a higher bit width (10bit-16bit), which is much higher than the currently commonly used standard display (8bit) and far exceeds the human eye's ability to resolve grayscale images. Therefore, when transmitting the infrared image from the detector end to the display end, dynamic range compression is usually required to reduce the bit width of the infrared image.

[0003] In the process of infrared image compression, information loss is inevitable. Infrared images generally have low contrast, and noise and redundant information occupy most of the grayscale, with low signal-to-noise ratio. If the difference between the target and the background in the image is not considered during the compression process, and the method is not appropriate, it will cause problems such as reduced contrast, loss of details, and noise amplification, reducing image quality. Further detail enhancement and noise reduction of compressed infrared images can effectively alleviate these problems.

[0004] At present, image compression and enhancement methods based on the layered concept are a hot topic. The infrared image is divided into a basic layer and a detail layer through a filter, and dynamic range compression and detail enhancement are performed respectively. The merged infrared image can compress the bit width while retaining the details. However, the performance of one type of method depends on the layered filter, and the generally applicable fixed parameter values ​​are not well applicable to infrared images in different scenes. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method, device and computer-readable storage medium for dynamic compression and detail enhancement of infrared images. Taking into account the influence of filtering parameters on the dynamic range compression effect in infrared images under different conditions, adaptive parameters are introduced, and image filtering layering and detail enhancement are performed on this basis, thereby improving the dynamic range compression quality of infrared images under different conditions.

[0006] Based on the first aspect, the present invention provides a method for dynamic compression and detail enhancement of infrared images, the method comprising:

[0007] The original image to be processed is divided into a base layer and a detail layer using a parameter-adaptive guided filter; the adaptive parameter of the guided filter is calculated according to a local variance histogram of the original image to be processed, and the local variance histogram of the original image is obtained by statistics of local variance of the original image to be processed;

[0008] Performing dynamic range compression on the base layer;

[0009] Performing image enhancement on the detail layer according to a gain coefficient representing the overall clarity of the image; the gain coefficient is obtained according to the adaptive parameter;

[0010] The base layer image after dynamic range compression is merged with the detail enhanced image to obtain the final dynamic range compressed image.

[0011] Optionally, the step of using a parameter-adaptive guided filter to divide the original image to be processed into a base layer and a detail layer comprises:

[0012] Calculating the local variance of the original image to be processed, and obtaining a local variance histogram of the original image by counting the local variance of the entire image;

[0013] Obtaining adaptive parameters using the mean of the local variance histogram of the original image;

[0014] Substituting the adaptive parameters into the guided filter, filtering the original image to be processed, and dividing the original image to be processed into a base layer and a detail layer;

[0015] Optionally, obtaining the adaptive parameter by using the mean of the local variance histogram of the original image includes:

[0016] According to the local variance histogram of the original image, the adaptive parameter ε is calculated. The calculation formula of the adaptive parameter ε is as follows:

[0017]

[0018] In the formula, k represents the adjustment coefficient of ε, represents the local variance of the original image to be processed, Represents the quantity of each local variance, and n represents the range of the local variance mean.

[0019] Optionally, substituting the adaptive parameters into the guided filter, filtering the original image to be processed, and dividing the original image to be processed into a base layer and a detail layer comprises:

[0020] The original image to be processed is used as the guide image, the adaptive parameter ε is substituted, and the guide filter is used for filtering. The filter is as follows:

[0021]

[0022]

[0023] Where q(i,j) represents the output filtered image as the base layer, I(i,j) represents the original image to be processed, and They are respectively k and b k The mean of Represents the mean value of the original image to be processed within the window.

[0024] The filtering result is used as the base layer, and the detail layer is obtained by subtracting the base layer from the original image to be processed.

[0025] Optionally, the performing dynamic range compression on the base layer includes:

[0026] A histogram-based mapping is used on the base layer to achieve dynamic range compression.

[0027] Optionally, the performing image enhancement on the detail layer according to the gain coefficient representing the overall clarity of the image includes:

[0028] The adaptive parameters ε and local variance calculated during the image layering process are Mapped to the gain coefficient that characterizes the overall clarity of the image, as a mask for suppressing image noise, the mapping relationship is as follows:

[0029]

[0030] In the formula, α represents the normalized adjustment factor of ε, and its value is close to the ε obtained from the image in the standard clear scene, g max and g min They are the high and low gain value coefficients of the image, representing the gain range of the image.

[0031] Optionally, merging the base layer image after dynamic range compression with the detail enhanced image to obtain a final dynamic range compressed image includes:

[0032] The base layer image after dynamic range compression and the detail enhanced image are weightedly added to obtain a final dynamic range compressed image.

[0033] Based on the second aspect, the present invention also provides a computer-readable storage medium, on which is stored an image dynamic range compression and detail enhancement program, which, when read and executed, implements the steps of the infrared image dynamic range compression and detail enhancement method based on parameter adaptation as described in any of the above items.

[0034] Based on the third aspect, the present invention provides an infrared image dynamic range compression and detail enhancement device based on parameter adaptation, including a processor, wherein the processor is used to execute a computer program stored in a memory to implement the steps of the infrared image dynamic range compression and detail enhancement method based on parameter adaptation as described in any one of the above items.

[0035] Based on the third aspect, the present invention further provides an infrared image dynamic range compression and detail enhancement device based on parameter adaptation, comprising:

[0036] An image layering program module is used to use a parameter-adaptive guided filter to divide the original image to be processed into a base layer and a detail layer; the adaptive parameter of the guided filter is calculated according to a local variance histogram of the original image to be processed, and the local variance histogram of the original image is obtained by local variance statistics of the original image to be processed;

[0037] A base layer dynamic range compression program module, used for performing dynamic range compression on the base layer;

[0038] A detail layer detail enhancement program module is used to perform image enhancement on the detail layer according to a gain coefficient representing the overall clarity of the image; the gain coefficient is obtained according to the adaptive parameter;

[0039] The image layer merging program module is used to merge the base layer image after dynamic range compression with the detail enhanced image to obtain the final dynamic range compressed image.

[0040] Beneficial effects of the present invention:

[0041] The present invention uses a guided filter to decompose the original high dynamic range image into an image base layer containing large dynamic temperature information and an image detail layer containing small dynamic temperature detail information. It has the advantages of low computational complexity, better edge retention ability, and the ability to effectively suppress phenomena such as gradient reversal. After adding adaptive parameters, the guided filter can achieve good filtering layering effects in different infrared scenes, laying a good foundation for subsequent processing. The method of the present invention can compress the dynamic range of the image while maintaining the image detail information. The detail enhancement method based on adaptive parameters introduces a gain coefficient that characterizes the overall clarity level of the image, thereby reducing the image background noise while highlighting the target details of the image. On the basis of considering both the local clarity of the image and the overall clarity level of the image, the image detail layer is enhanced.

[0042] In addition, the present invention also provides a corresponding implementation device and a computer-readable storage medium for the method, which improves the practicability of the infrared image dynamic range compression and detail enhancement method, and the device and computer-readable storage medium have corresponding advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart of the method of the present invention.

[0044] Figure 2 A schematic flow chart of a method for dynamic compression and detail enhancement of infrared images provided by an embodiment of the present invention.

[0045] Figure 3 An original infrared image is provided as an exemplary application scenario of an embodiment of the present invention.

[0046] Figure 4 for Figure 3 Schematic diagram of the local variance histogram corresponding to the original image in .

[0047] Figure 5 for Figure 3 The result image obtained by processing the original image in the embodiment of the present invention.

[0048] Figure 6 A schematic diagram of the program module composition of the infrared image dynamic compression and detail enhancement device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to enable technicians to better understand the content, technical solutions and advantages of the present invention, the present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. The embodiments are only some embodiments, not all embodiments. Based on the embodiments in the present invention, all other embodiments obtained without creative work are within the scope of protection of the present invention.

[0050] The embodiment of the present invention provides a method for dynamic range compression and detail enhancement of infrared images based on adaptive parameter filtering, such as Figure 1 As shown, the specific method includes the following steps:

[0051] Step 101: Use a parameter-adaptive guided filter to divide the original image to be processed into a base layer and a detail layer.

[0052] The original image to be processed is the original image data collected by the image acquisition device and needs to be compressed in the dynamic range. This example can be applied to the infrared photoelectric imaging system. The image data source can be the infrared image generated by the infrared detector imaging system, or it can be the data read from other devices such as the memory. There is no clear requirement for the dynamic range and size of the original image to be processed. For example, the present invention is applicable to the original image to be processed with a bit width of 14 bit, 16 bit, and a size of 256×256, 1024×768, etc.

[0053] In the embodiment of the present invention, the local variance value of the original image to be processed is first counted. Variance can characterize the edge and detail richness in the neighborhood of the pixel point in the image. The local variance value of the whole image is counted to obtain the local variance histogram of the original image. Similar to the image grayscale histogram, the local variance histogram is a statistical feature of the image. The distribution of the local variance in the image is counted. The horizontal axis represents the local variance value, and the vertical axis represents the number of pixels corresponding to the local variance value. See Figure 4 , the local variance statistics of the image are widely distributed and the parts with large values ​​are few. Since the local variance value reflects the details and edge texture of the image, the statistical information of the local variance statistics histogram of the image with rich detail texture is more widely distributed, and the number of distributions in the interval with large local variance is more. Therefore, introducing the local variance statistics into the filter parameters can better adapt to the infrared image layering processing in different scenes.

[0054] Step 102: Perform dynamic range compression on the base layer.

[0055] The base layer is obtained after filtering the original image to be processed, removing most of the detail texture information, representing the general information of the image, and reducing the bit width while retaining the basic information of the image through dynamic range compression.

[0056] Step 103: performing image enhancement on the detail layer according to a gain coefficient representing the overall clarity of the image.

[0057] The gain coefficient characterizing the overall clarity of the image is obtained by the adaptive parameters of the local variance statistical histogram and is applied to the mask for suppressing image noise. The gain is small in the image with clear overall detail texture, avoiding excessive enhancement of image edge information and noise, while the gain is large in the blurred image, which can effectively enhance the image detail information. The image noise mask is based on the principle of noise masking in the detail-rich area, and the gain is small in the flat area. In this way, the detail layer can be enhanced on the basis of considering both the local and overall clarity of the image.

[0058] Step 104: merging the base layer image after dynamic range compression and the detail enhanced image to obtain a final dynamic range compressed image.

[0059] In the technical solution provided by the embodiment of the present invention, the infrared image dynamic range compression method based on the layered concept is improved and optimized, and the original image to be processed is divided into a base layer and a detail layer using a parameter-adaptive guided filter, wherein the adaptive parameters are derived from the local variance statistical histogram of the original image to be processed, and different parameters are given for infrared images in different scenes, so that the filter can be adjusted according to the image features, avoiding the performance fluctuation of fixed parameters in different scenes. Applying the parameters to detail enhancement can suppress the interference of noise in the detail layer on image enhancement, and combined with the noise mask, effectively improves the image detail enhancement effect, and also improves the result of the final image dynamic range compression.

[0060] In the above embodiment, for using a parameter-adaptive guided filter to divide the original image to be processed into a base layer and a detail layer, a specific implementation method is given in this embodiment:

[0061] A1: Calculate the local variance of the original image to be processed, and count the local variance of the whole image to obtain a local variance histogram of the original image.

[0062] A2: Calculate the adaptive parameters using the mean of the local variance histogram of the original image.

[0063] A3: Substitute the adaptive parameters into the guided filter, filter the original image to be processed, and divide the original image to be processed into a base layer and a detail layer.

[0064] The method of calculating the adaptive parameters according to the local variance histogram may include the following steps:

[0065] The sliding window is used to calculate the variance of each pixel in the image within the neighborhood one by one as the local variance of the pixel.

[0066] The local variance value of the entire image is counted to obtain the local variance histogram.

[0067] The adaptive parameter is obtained by using the local variance histogram mean. The relationship can be expressed as:

[0068]

[0069] In the formula, k represents the adjustment coefficient of ε, represents the local variance of the original image to be processed, Indicates the number of local variances, n indicates the range of local variance means, such as n = 5000, which means that the local variance value is less than 5000. The mean of .

[0070] In the above embodiment, for layering the image using the guided filter based on the adaptive parameter, an implementation method is provided in this embodiment, which may include the following steps:

[0071] The original image to be processed is used as the guide image, the adaptive parameter ε is substituted, and the guided filter is used for filtering:

[0072]

[0073]

[0074] Where q(i,j) represents the output filtered image as the base layer, I(i,j) represents the original image to be processed, and They are respectively k and b k The mean of Represents the mean of the original image to be processed in the window. The parameter ε can be adaptively adjusted according to the local variance of the image. In images with rich details, the local variance distribution is wider and a larger ε value can be calculated.

[0075] The result obtained by guided filtering is used as the base layer. After subtracting the base layer from the original image to be processed, the detail layer is obtained. The relationship can be expressed as:

[0076] I detail =I ori -I base

[0077] In the formula, I ori represents the original image to be processed, I detail represents the detail layer, I base The parameter-adaptive filter introduces image local variance statistics when determining parameters, which can make the filter more suitable for various scenarios.

[0078] In the embodiment of the present invention, there is no limitation on the method of performing dynamic range compression on the base layer. In this example, histogram-based mapping is used for the base layer to achieve dynamic range compression. Specifically, the embodiment of the present invention further provides an implementation method of step 102, which uses contrast-limited histogram equalization and may include the following steps:

[0079] B1: for the base layer image I base Divide the data into blocks and perform platform histogram restriction on the local histogram of each sub-block.

[0080] B2: The clipped part of the histogram is evenly distributed to each gray level in turn according to the proportion of each gray level pixel in the whole image to obtain an improved histogram.

[0081] B3: Perform dynamic range compression on each sub-block image according to the improved histogram.

[0082] B4: Use the bilinear interpolation method to recalculate the mapping values ​​of pixels in the non-central area.

[0083] In the above embodiment, an implementation of the step of "performing image enhancement on the detail layer according to the gain coefficient representing the overall clarity of the image" is as follows:

[0084] The adaptive parameters ε and local variance calculated during the image layering process are Mapped to the gain coefficient that characterizes the overall clarity of the image, as a mask for suppressing image noise, the mapping relationship is as follows:

[0085]

[0086] In the formula, g(i,j) represents the detail gain coefficient corresponding to the pixel point (i,j), α represents the normalized adjustment factor of ε, and its value is close to the ε obtained in the image under the standard clear scene, g max and g min They are the high and low gain value coefficients of the image, representing the gain range of the image. Represents the overall gain adjustment parameter of the image, It indicates the richness of detail texture in each area of ​​the image. The gain can take into account the flatness of the area in the image and give different degrees of detail enhancement, thereby suppressing noise to a certain extent.

[0087] The final detail layer enhanced image is determined by the gain coefficient, and the relationship can be expressed as:

[0088] I′ detail =I detail ×g

[0089] In the formula, I′ detail shows the detail layer after the detail enhancement processing, I detail It represents the detail layer obtained after filtering the original image, and g represents the detail gain coefficient.

[0090] In the above embodiment, the method for merging the base layer image and the detail enhanced image after dynamic range compression is not limited. This embodiment takes weighted addition as an example to provide a merging method, which may include the following steps:

[0091] The base layer image after dynamic range compression and the detail enhancement image are weightedly added in a certain ratio to obtain a final dynamic range compression image. The weighted addition relationship is as follows:

[0092] Iout =γ*I′ base +(1-γ)*I′ detail

[0093] In the formula, I′ base represents the base layer after dynamic range compression, I′ detail represents the detail layer after detail enhancement processing, and γ represents a weighting coefficient, which can be adjusted by oneself in practical applications without limitation.

[0094] In order to verify the effectiveness of the method in this application, a verification experiment was also conducted to apply this method to the original infrared image to be processed. Figure 3 The processing results are as follows Figure 5 As shown, by comparison, the method of the present application can compress the dynamic range of the image while maintaining the image detail information. The detail enhancement method based on adaptive parameters introduces a gain coefficient that characterizes the overall clarity level of the image, thereby reducing the image background noise while highlighting the target details of the image. The image detail layer is enhanced while considering both the local clarity and the overall clarity level of the image.

[0095] The embodiment of the present invention provides an infrared image dynamic range compression and detail enhancement device based on parameter adaptation, which further improves the practicality of the method. Figure 6 As shown, based on the program module perspective, the device of the present invention includes:

[0096] The image layering program module 601 is used to use a parameter-adaptive guided filter to separate the original image to be processed into a base layer and a detail layer; the adaptive parameter of the guided filter is calculated according to the local variance histogram of the original image to be processed, and the local variance histogram of the original image is obtained by local variance statistics of the original image to be processed;

[0097] A base layer dynamic range compression program module 602 is used to perform dynamic range compression on the base layer;

[0098] A detail layer detail enhancement program module 603 is used to enhance the image of the detail layer according to a gain coefficient representing the overall clarity of the image; the gain coefficient is obtained according to the adaptive parameter;

[0099] The image layer merging program module 604 is used to merge the base layer image after dynamic range compression with the detail enhanced image to obtain a final dynamic range compressed image.

[0100] Optionally, in some implementation methods of this embodiment, the image layering program module 601 may include:

[0101] Image local variance calculation submodule: calculates the local variance of the original image to be processed, and counts the local variance of the whole image to obtain the local variance histogram of the original image;

[0102] Adaptive parameter calculation submodule: The adaptive parameter is obtained by using the mean of the local variance histogram of the original image. The adaptive parameter calculation formula is as follows:

[0103]

[0104] In the formula, k represents the adjustment coefficient of ε, represents the local variance of the original image to be processed, Indicates the number of local variances, n indicates the range of local variance means, such as n = 5000, which means that the local variance value is less than 5000. The mean of .

[0105] Filtering layer submodule: Substitute the adaptive parameters into the guided filter, filter the original image to be processed, and divide the original image to be processed into a basic layer and a detail layer.

[0106] In some implementations of the embodiments of the present invention, the filtering layer submodule may include:

[0107] Guided filtering unit: The original image to be processed is used as a guided image, the adaptive parameters are substituted, and the guided filter is used for filtering. The filter is as follows:

[0108]

[0109]

[0110] Where q(i,j) represents the output filtered image as the base layer, I(i,j) represents the original image to be processed, and They are respectively k and b k The mean of Represents the mean of the original image to be processed in the window. The parameter ε can be adaptively adjusted according to the local variance of the image. In images with rich details, the local variance distribution is wider and a larger ε value can be calculated.

[0111] Layering unit: The filtering result is used as the base layer, and the detail layer is obtained by subtracting the base layer from the original image to be processed.

[0112] Optionally, in some implementation methods of this embodiment, the base layer dynamic range compression module 602 may include:

[0113] The basic dynamic range compression submodule uses contrast-limited histogram equalization to compress the dynamic range of the image base layer.

[0114] Optionally, in some implementation methods of this embodiment, the detail layer detail enhancement module 603 may include:

[0115] Gain coefficient calculation submodule: maps the adaptive parameters and local variance calculated in the image layering process to the gain coefficient that characterizes the overall clarity of the image as a mask to suppress image noise. The mapping relationship is as follows:

[0116]

[0117] In the formula, g(i,j) represents the detail gain coefficient corresponding to the pixel point (i,j), α represents the normalized adjustment factor of ε, and its value is close to the ε obtained in the image under the standard clear scene, g max and g min They are the high and low gain value coefficients of the image, representing the gain range of the image. Represents the overall gain adjustment parameter of the image, It indicates the richness of detail texture in each area of ​​the image. The gain can take into account the flatness of the area in the image and give different degrees of detail enhancement, thereby suppressing noise to a certain extent.

[0118] Coefficient mapping submodule: using the detail gain coefficient to map each pixel in the detail layer image to obtain an enhanced detail layer image.

[0119] The functions of the functional modules of the infrared image dynamic range compression and detail enhancement device described in the embodiment of the present invention can be specifically implemented according to the above method, and the specific process can refer to the relevant description of the above method embodiment.

[0120] From the hardware perspective, the present invention also provides an infrared image dynamic range compression and detail enhancement device based on parameter adaptation, the device comprising:

[0121] A processor is used to execute a computer program stored in a memory to implement the steps of the infrared image dynamic range compression and detail enhancement method based on parameter adaptation as described in any of the above items. The specific form, model, and performance are not limited.

[0122] The memory is used to store computer programs, and the stored contents include but are not limited to computer programs, operating systems, input / output and operation data, etc., used to implement the steps of any of the above-mentioned embodiments. The present invention does not limit the specific type of memory.

[0123] In addition, the device may also include an input and output interface, a communication interface, a communication bus, a power supply and a display screen. The structure is not limited to the image dynamic range compression and detail enhancement method described in the embodiment, and components can be increased or decreased according to specific circumstances.

[0124] The image dynamic range compression and detail enhancement method described in the above embodiment can be stored in a computer-readable medium for separate exchange and use. Therefore, the present invention also provides a computer-readable storage medium, on which an image dynamic range compression and detail enhancement program is stored. When the image dynamic range compression and detail enhancement program is read and executed, the steps of the infrared image dynamic range compression and detail enhancement method based on parameter adaptation as described in any of the above items are implemented.

[0125] The method of the present invention can compress the dynamic range of an image while maintaining image detail information. The detail enhancement method based on adaptive parameters introduces a gain coefficient that characterizes the overall clarity level of the image, thereby reducing the image background noise while highlighting the target details of the image. The image detail layer is enhanced while considering both the local clarity and the overall clarity level of the image.

[0126] The principles and implementation methods of the present invention are described in detail using specific embodiments. The above embodiments are only used to help technicians understand and use the methods and ideas of the present invention. Modifications made without departing from the principles of the present invention also fall within the scope of protection of the claims of this application.

Claims

1. A method for dynamic compression and detail enhancement of infrared images, characterized in that: The following steps are involved: Step S1, using a parameter-adaptive guided filter to divide the original image to be processed into a base layer and a detail layer; the adaptive parameters of the guided filter are calculated according to the local variance histogram of the original image to be processed, and the adaptive parameters are calculated according to the local variance histogram of the original image. : In the formula, express The adjustment factor, represents the local variance of the original image to be processed, Represents the quantity of each local variance, and n represents the range of the local variance mean; The local variance histogram of the original image is obtained by statistics of the local variance of the original image to be processed; Step S2, performing dynamic range compression on the base layer; Step S3, performing image enhancement on the detail layer according to a gain coefficient representing the overall clarity of the image; The gain coefficient is obtained according to the adaptive parameter; The image enhancement of the detail layer according to the gain coefficient representing the overall clarity of the image comprises: The adaptive parameters calculated during the image layering process and local variance Mapped to the gain coefficient that characterizes the overall clarity of the image, as a mask for suppressing image noise, the mapping relationship is shown in the following formula: In the formula, Represents pixel The corresponding detail gain coefficient, a represents The normalization adjustment factor is and They are the high and low gain value coefficients of the image, representing the gain range of the image; Represents the overall gain adjustment parameter of the image, Indicates the richness of detail texture in each area of ​​the image. The gain can take into account the flatness of the area in the image and give different degrees of detail enhancement, thereby suppressing noise to a certain extent; Step S4: The base layer image after dynamic range compression is combined with the detail enhanced image to obtain a final dynamic range compressed image.

2. The method according to claim 1, characterized in that In step S1, the use of a parameter-adaptive guided filter to divide the original image to be processed into a base layer and a detail layer includes: Calculating the local variance of the original image to be processed, and obtaining a local variance histogram of the original image by counting the local variance of the entire image; Obtaining adaptive parameters using the mean of the local variance histogram of the original image; The adaptive parameters are substituted into the guided filter to filter the original image to be processed, and the original image to be processed is divided into a basic layer and a detail layer.

3. The method according to claim 2, characterized in that Substituting the adaptive parameters into the guided filter, filtering the original image to be processed, and dividing the original image to be processed into a base layer and a detail layer comprises: The original image to be processed is used as the guide image and the adaptive parameters are substituted into the , using a guided filter for filtering, the filter is shown in the following formula: In the formula, represents the output filtered image, as the base layer, Represents the original image to be processed. and They are and The mean of Represents the mean value of the original image to be processed within the window; The detail layer is obtained by subtracting the base layer from the original image to be processed.

4. The method according to claim 1, characterized in that In step S2, the dynamic range compression of the base layer includes: A histogram-based mapping is used on the base layer to achieve dynamic range compression.

5. The method according to any one of claims 1 to 4, characterized in that: In step S4, the base layer image after dynamic range compression is combined with the detail enhanced image to obtain a final dynamic range compressed image, including: The base layer image after dynamic range compression and the detail enhanced image are weightedly added to obtain a final dynamic range compressed image.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the steps of the infrared image dynamic compression and detail enhancement method as described in any one of claims 1 to 5.

7. An infrared image dynamic compression and detail enhancement device, characterized in that: Includes a processor and the computer-readable storage medium as claimed in claim 6.

8. An infrared image dynamic compression and detail enhancement device, characterized in that: include: An image layering program module, used in step S1 of the infrared image dynamic compression and detail enhancement method according to any one of claims 1 to 5; A base layer dynamic range compression program module, used in step S2 of the infrared image dynamic compression and detail enhancement method according to any one of claims 1 to 5; A detail layer detail enhancement program module, used in step S3 of the infrared image dynamic compression and detail enhancement method according to any one of claims 1 to 5; The image layer merging program module is used in step S4 of the infrared image dynamic compression and detail enhancement method as described in any one of claims 1 to 5.

Images