Multi-scale Fusion Method, Device and Equipment for Infrared Images and Visible Light Images

Abstract

The present invention discloses a multi-scale fusion method and device for infrared images and visible light images. The method includes: obtaining a visible light grayscale image and an infrared image for the same scene as the first and second images; performing image processing operations on the first and second images according to a filtering function and a layer processing function to obtain a first processed image corresponding to the first image, first layer parameters, a second processed image corresponding to the second image, and second layer parameters; determining a first weight matrix corresponding to the first layer parameters and a second weight matrix corresponding to the second layer parameters according to the first and second layer parameters; performing a color coding operation on the second processed image to obtain a color coding result; and performing a preset dual-light fusion operation on the first image and the color coding result according to the first and second weight matrices to obtain a target output image. It can be seen that implementing the present invention can fuse infrared images and visible light images and improve the fusion effect of the fused images.

Description

Technical Field

[0001] The present invention relates to the technical field of image fusion, and in particular to a multi-scale fusion method and device for infrared images and visible light images. Background Art

[0002] Infrared thermal imaging technology uses optoelectronic technology to detect the infrared signal in a specific band of the thermal radiation on the surface of an object, and converts this signal into the temperature information of the object surface, and then into image information that can be distinguished by humans. However, the infrared images generated by the system are grayscale images. Since the black-and-white interval that can be distinguished by the human eye is narrow, the information of the original thermal infrared image cannot be well observed; in addition, the thermal infrared images generated by the system have defects such as low image grayscale, low contrast, and visual blurriness, and cannot provide texture details.

[0003] Color visible light images have a huge amount of information and can provide the most intuitive details for computer vision tasks. However, due to the influence of the data collection environment, visible light images generally cannot highlight important targets. Different from visible light images, infrared images can distinguish targets from the background according to the thermal radiation difference, and are not affected by lighting and weather conditions.

[0004] In view of the complementary characteristics of infrared images and visible light images, their fused images will be a great help in promoting the application of computer vision tasks. However, the fusion effect achieved by the existing dual-light fusion technology is far from satisfactory. It can be seen that it is particularly important to provide a method for improving the fusion effect of infrared images and visible light images. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a multi-scale fusion method and device for infrared images and visible light images, which can fuse infrared images and visible light images and improve the fusion effect of infrared images and visible light images.

[0006] To solve the above technical problem, a first aspect of the present invention discloses a multi-scale fusion method for infrared images and visible light images, the method comprising:

[0007] Obtain a first image and a second image, where the first image is an image obtained after the original image undergoes a primary image processing operation, the second image is an image obtained after performing a shooting operation on the scene corresponding to the original image, the first image is a visible light grayscale image and the second image is an infrared image, and the original image is a visible light image obtained after performing a shooting operation on the scene;

[0008] Perform a preset secondary image processing operation on the first image and the second image according to a preset filtering function and a layer processing function, to obtain a first processed image corresponding to the first image, first layer parameters, a second processed image corresponding to the second image, and second layer parameters;

[0009] Determine a first weight matrix corresponding to the first layer parameters and a second weight matrix corresponding to the second layer parameters according to the first layer parameters and the second layer parameters;

[0010] Perform a preset color encoding operation on the second processed image to obtain a color encoding result corresponding to the second processed image;

[0011] Perform a preset dual - light fusion operation on the first image and the color encoding result according to the first weight matrix and the second weight matrix to obtain a target output image.

[0012] As an optional implementation manner, in the first aspect of the present invention, the performing a preset secondary image processing operation on the first image and the second image according to a preset filtering function and a layer processing function, to obtain a first processed image corresponding to the first image, first layer parameters, a second processed image corresponding to the second image, and second layer parameters includes:

[0013] Perform a normalization processing operation on the first image and the second image according to a preset normalization processing function, to obtain a first processed image corresponding to the first image and a second processed image corresponding to the second image;

[0014] Determine target layer parameters for calculating the first layer parameters of the first image and the second layer parameters of the second image, where the target layer parameters include the number of decomposition levels, a window coefficient queue, and an offset coefficient queue;

[0015] Perform a parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, a maximum value function, and the target layer parameters, to obtain the first layer parameters corresponding to the first processed image and the second layer parameters corresponding to the second processed image.

[0016] As an optional implementation manner, in the first aspect of the present invention, performing a parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, a maximum value function, and the target layer parameters, to obtain the first layer parameters corresponding to the first processed image and the second layer parameters corresponding to the second processed image includes:

[0017] Perform a first-level parameter decomposition operation on the first processed image and the second processed image according to a preset guiding filtering function, a maximum value function, the target layer parameters, and the determined guiding matrix queue to obtain a first base layer queue corresponding to the first processed image and a second base layer queue corresponding to the second processed image;

[0018] Perform a second-level parameter decomposition operation on the first base layer queue and the second base layer queue according to the maximum value function to obtain a first detail layer queue, a first detail layer weight queue corresponding to the first base layer queue, a second detail layer queue, and a second detail layer weight queue corresponding to the second base layer queue;

[0019] Determine the first base layer queue, the first detail layer queue, and the first detail layer weight queue as the first layer parameters corresponding to the first processed image, and determine the second base layer queue, the second detail layer queue, and the second detail layer weight queue as the second layer parameters corresponding to the second processed image.

[0020] As an optional implementation manner, in the first aspect of the present invention, the determining the first weight matrix corresponding to the first layer parameters and the second weight matrix corresponding to the second layer parameters according to the first layer parameters and the second layer parameters includes:

[0021] Calculate the first weight matrix corresponding to the first layer parameters according to the first base layer queue, the first detail layer queue, and the first detail layer weight queue;

[0022] Calculate the second weight matrix corresponding to the second layer parameters according to the second base layer queue, the second detail layer queue, and the second detail layer weight queue.

[0023] As an optional implementation manner, in the first aspect of the present invention, the performing a preset color coding operation on the second processed image to obtain a color coding result corresponding to the second processed image includes:

[0024] Determine a preset number of coding parameters within a preset numerical range;

[0025] Calculate the channel components corresponding to the second processed image according to all the coding parameters to obtain a channel component set corresponding to the second processed image, and the total number of the channel components in the channel component set is the preset number;

[0026] Determine the channel component set corresponding to the second processed image as the color coding result corresponding to the second processed image.

[0027] As an alternative implementation manner, in the first aspect of the present invention, performing a preset dual - light fusion operation on the first image and the color - coding result according to the first weight matrix and the second weight matrix to obtain a target output image includes:

[0028] Obtain the original image before the first - level image - processing operation of the first image;

[0029] According to the number of channel components in the color - coding result and the original image, determine all fusion parameters corresponding to the original image, where the number of fusion parameters is the preset number;

[0030] For each channel component in the color - coding result, determine the fusion parameter corresponding to each channel component, and calculate a fusion component corresponding to each channel component according to the first weight matrix, the second weight matrix, the channel component, and the fusion parameter corresponding to the channel component;

[0031] Determine the target output image according to all the fusion components.

[0032] As an alternative implementation manner, in the first aspect of the present invention, the method further includes:

[0033] Obtain the target output image and a reference image, where the reference image is used to calculate the fusion value of the target output image;

[0034] Analyze the target output image and the reference image according to the original image to obtain the fusion value corresponding to the target output image;

[0035] Judge whether the fusion value is greater than a preset fusion threshold. When the judgment result is yes, determine that the target output image is an image that meets the preset dual - light fusion standard;

[0036] When the judgment result is no, generate a failure message indicating that the target output image does not meet the preset dual - light fusion standard according to the fusion value, the target output image, the first weight matrix, and the second weight matrix;

[0037] According to the failure message, adjust the target layer parameters, and perform the operation of decomposing parameters on the first processed image and the second processed image according to the preset guided filter function, maximum function, and the adjusted target layer parameters to obtain the first layer parameters corresponding to the first processed image and the second layer parameters corresponding to the second processed image.

[0038] As an optional implementation manner, in the first aspect of the present invention, the calculation formulas of the first weight matrix and the second weight matrix are specifically:

[0039]

[0040]

[0041] Wherein, n is the number of decomposition layers, imgC_W is the first weight matrix corresponding to the first layer parameters, C n is the first base layer queue, W_c k is the first detail layer weight queue, C_d k is the first detail layer queue; similarly, imgT_W is the second weight matrix corresponding to the second layer parameters, T n is the second base layer queue, W_t k is the second detail layer weight queue, T_d k The second detail layer queue;

[0042] Among them, the first detail layer weight queue W_c k And the second detail layer weight queue W_t k The corresponding calculation formula is:

[0043]

[0044]

[0045] in, , in order to find L for A p norm.

[0046] A second aspect of the present invention discloses a multi-scale fusion device for infrared images and visible light images, the device comprising:

[0047] an acquisition module, configured to acquire a first image and a second image, wherein the first image is an image obtained by performing a primary image processing operation on an original image, and the second image is an image obtained by performing a photographing operation on a scene corresponding to the original image, wherein the first image is a visible light gray image and the second image is an infrared image, and the original image is a visible light image obtained by performing the photographing operation on the scene;

[0048] An image processing module, configured to perform a preset secondary image processing operation on the first image and the second image according to a preset filtering function and a layer processing function, to obtain a first processed image corresponding to the first image, first layer parameters, a second processed image corresponding to the second image, and second layer parameters;

[0049] A determination module, configured to determine a first weight matrix corresponding to the first layer parameters and a second weight matrix corresponding to the second layer parameters according to the first layer parameters and the second layer parameters;

[0050] An encoding module, configured to perform a preset color encoding operation on the second processed image to obtain a color encoding result corresponding to the second processed image;

[0051] A fusion module, configured to perform a preset dual-light fusion operation on the first image and the color encoding result according to the first weight matrix and the second weight matrix to obtain a target output image.

[0052] As an optional implementation manner, in the second aspect of the present invention, the manner in which the image processing module performs a preset secondary image processing operation on the first image and the second image according to a preset filtering function and a layer processing function, to obtain a first processed image corresponding to the first image, first layer parameters, a second processed image corresponding to the second image, and second layer parameters specifically includes:

[0053] Performing a normalization processing operation on the first image and the second image according to a preset normalization processing function to obtain a first processed image corresponding to the first image and a second processed image corresponding to the second image;

[0054] Determining target layer parameters for calculating the first layer parameters of the first image and the second layer parameters of the second image, where the target layer parameters include the number of decomposition levels, a window coefficient queue, and an offset coefficient queue;

[0055] Performing a parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, a maximum value function, and the target layer parameters to obtain a first layer parameter corresponding to the first processed image and a second layer parameter corresponding to the second processed image.

[0056] As an alternative implementation manner, in the second aspect of the present invention, the manner in which the image processing module performs parameter decomposition operations on the first processed image and the second processed image according to a preset guided filtering function, maximum value function, and the target layer parameters to obtain the first layer parameters corresponding to the first processed image and the second layer parameters corresponding to the second processed image specifically includes:

[0057] Perform a first-level parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, maximum value function, the target layer parameters, and a determined guided matrix queue to obtain a first base layer queue corresponding to the first processed image and a second base layer queue corresponding to the second processed image;

[0058] Perform a second-level parameter decomposition operation on the first base layer queue and the second base layer queue according to the maximum value function to obtain a first detail layer queue, a first detail layer weight queue corresponding to the first base layer queue, a second detail layer queue, and a second detail layer weight queue corresponding to the second base layer queue;

[0059] Determine the first base layer queue, the first detail layer queue, and the first detail layer weight queue as the first layer parameters corresponding to the first processed image, and determine the second base layer queue, the second detail layer queue, and the second detail layer weight queue as the second layer parameters corresponding to the second processed image.

[0060] As an alternative implementation manner, in the second aspect of the present invention, the manner in which the determination module determines a first weight matrix corresponding to the first layer parameters and a second weight matrix corresponding to the second layer parameters according to the first layer parameters and the second layer parameters includes:

[0061] Calculate a first weight matrix corresponding to the first layer parameters according to the first base layer queue, the first detail layer queue, and the first detail layer weight queue;

[0062] Calculate a second weight matrix corresponding to the second layer parameters according to the second base layer queue, the second detail layer queue, and the second detail layer weight queue.

[0063] As an alternative implementation manner, in the second aspect of the present invention, the manner in which the encoding module performs a preset color encoding operation on the second processed image to obtain a color encoding result corresponding to the second processed image specifically includes:

[0064] Determine a preset number of encoding parameters within a preset numerical range;

[0065] Calculate the channel components corresponding to the second processed image according to all the encoding parameters, and obtain a set of channel components corresponding to the second processed image, where the total number of channel components in the set of channel components is the preset number;

[0066] Determine the set of channel components corresponding to the second processed image as the color encoding result corresponding to the second processed image.

[0067] As an alternative implementation, in the second aspect of the present invention, the way for the fusion module to perform a preset dual-light fusion operation on the first image and the color encoding result according to the first weight matrix and the second weight matrix to obtain a target output image specifically includes:

[0068] Obtain the original image of the first image before the primary image processing operation;

[0069] Determine all fusion parameters corresponding to the original image according to the number of channel components in the color encoding result and the original image, where the number of fusion parameters is the preset number;

[0070] For each channel component in the color encoding result, determine the fusion parameter corresponding to each channel component, and calculate the fusion component corresponding to each channel component according to the first weight matrix, the second weight matrix, the channel component, and the fusion parameter corresponding to the channel component;

[0071] Determine the target output image according to all the fusion components.

[0072] As an alternative implementation, in the second aspect of the present invention, the acquisition module is further configured to acquire the target output image and a reference image, and the reference image is used to calculate the fusion value of the target output image;

[0073] The device further includes:

[0074] An analysis module, configured to analyze the target output image and the reference image according to the original image to obtain the fusion value corresponding to the target output image;

[0075] A judgment module, configured to judge whether the fusion value is greater than a preset fusion threshold, and when the judgment result is yes, determine that the target output image is an image that meets the preset dual-light fusion standard;

[0076] A generation module, configured to, when the judgment result of the judgment module is no, generate a failure message indicating that the target output image does not meet the preset dual-light fusion standard according to the fusion value, the target output image, the first weight matrix, and the second weight matrix;

[0077] An adjustment module, configured to adjust the parameters of the target layer according to the failure information, and trigger the image processing module to execute the operation of performing parameter decomposition on the first processed image and the second processed image according to a preset guided filtering function, a maximum value function, and the parameters of the target layer, so as to obtain the first layer parameters corresponding to the first processed image and the second layer parameters corresponding to the second processed image.

[0078] As an optional implementation manner, in the second aspect of the present invention, the calculation formulas of the first weight matrix and the second weight matrix are specifically:

[0079]

[0080]

[0081] where n is the number of decomposition layers, imgC_W is the first weight matrix corresponding to the first layer parameters, C n is the first base layer queue, W_c k is the first detail layer weight queue, C_d k is the first detail layer queue; similarly, imgT_W is the second weight matrix corresponding to the second layer parameters, T n is the second base layer queue, W_t k is the second detail layer weight queue, T_d k is the second detail layer queue;

[0082] where the first detail layer weight queue W_c k and the second detail layer weight queue W_t k corresponding calculation formulas are specifically:

[0083]

[0084]

[0085] where , is to find the L p norm of A.

[0086] The third aspect of the present invention discloses another multi-scale fusion device for infrared images and visible light images, and the device includes:

[0087] A memory storing executable program code;

[0088] A processor coupled to the memory;

[0089] The processor calls the executable program code stored in the memory and executes the multi-scale fusion method of infrared images and visible light images disclosed in the first aspect of the present invention.

[0090] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0091] In the embodiments of the present invention, a multi-scale fusion method of infrared images and visible light images is provided. The method includes: obtaining a first image and a second image, where the first image is an image obtained after the original image undergoes a primary image processing operation, and the second image is an image obtained after performing a shooting operation on the scene corresponding to the original image. The first image is a visible light grayscale image and the second image is an infrared image, and the original image is a visible light image obtained after performing a shooting operation on the scene; according to a preset filtering function and layer processing function, performing a preset secondary image processing operation on the first image and the second image to obtain a first processed image corresponding to the first image, a first layer parameter, a second processed image corresponding to the second image, and a second layer parameter; determining a first weight matrix corresponding to the first layer parameter and a second weight matrix corresponding to the second layer parameter according to the first layer parameter and the second layer parameter; performing a preset color encoding operation on the second processed image to obtain a color encoding result corresponding to the second processed image; and performing a preset dual-light fusion operation on the first image and the color encoding result according to the first weight matrix and the second weight matrix to obtain a target output image. It can be seen that after obtaining the first and second images in the embodiments of the present invention, the preset filtering function and layer processing function can be automatically used to perform a secondary image processing operation on the first and second images to obtain the first and second processed images and the first and second layer parameters, realizing the intelligent image parameter decomposition of the dual-light images; and then determining the first and second weight matrices, and finally intelligently fusing the first image and the color encoding result through the first and second weight matrices. Through the parameter decomposition of the light source image and the generation of the weight matrix, the fusion of the visible light image and the infrared image is realized, and the fusion effect of the dual-light fusion result is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0093] Figure 1It is a schematic flowchart of a multi-scale fusion method for infrared images and visible light images disclosed in an embodiment of the present invention;

[0094] Figure 2 It is a schematic flowchart of another multi-scale fusion method for infrared images and visible light images disclosed in an embodiment of the present invention;

[0095] Figure 3 It is a schematic structural diagram of a multi-scale fusion device for infrared images and visible light images disclosed in an embodiment of the present invention;

[0096] Figure 4 It is a schematic structural diagram of another multi-scale fusion device for infrared images and visible light images disclosed in an embodiment of the present invention;

[0097] Figure 5 It is a schematic structural diagram of yet another multi-scale fusion device for infrared images and visible light images disclosed in an embodiment of the present invention;

[0098] Figure 6 It is an example diagram corresponding to the original image, the second image, and the color image of the second image in an embodiment of the present invention;

[0099] Figure 7 It is a comparison diagram of the fusion effect of the output image obtained after executing the multi-scale fusion method for infrared images and visible light images disclosed in an embodiment of the present invention. Detailed implementation manners

[0100] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0101] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product, or terminal that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or terminals.

[0102] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments can be included in at least one embodiment of the invention. The phrase occurs in various places in the specification and is not necessarily referring to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive of other embodiments. Those skilled in the art will understand explicitly and implicitly that the embodiments described herein can be combined with other embodiments.

[0103] The present invention discloses a multi-scale fusion method and apparatus for infrared images and visible light images. After obtaining the first and second images, it can automatically perform a two-level image processing operation on the first and second images through a preset filtering function and layer processing function to obtain the first and second processed images and the first and second layer parameters, realizing the intelligent decomposition of image parameters for dual-light images. Furthermore, the first and second weight matrices are determined, and finally, the first image and the color coding result are intelligently fused through the first and second weight matrices. Through the parameter decomposition of the light source image and the generation of the weight matrix, the fusion of visible light images and infrared images is achieved, improving the fusion effect of the dual-light fusion result. The following will be described in detail respectively.

[0104] Embodiment 1

[0105] Please refer to Figure 1 and Figure 6 , Figure 1 is a flowchart showing a multi-scale fusion method for infrared images and visible light images disclosed in an embodiment of the present invention. Figure 6 is an example diagram corresponding to the original image, the second image, and the color image of the second image in an embodiment of the present invention. Among them, Figure 1 The described multi-scale fusion method for infrared images and visible light images can be applied to a multi-scale fusion apparatus for infrared images and visible light images, which is not limited in the embodiments of the present invention. As Figure 1 shown, the multi-scale fusion method for infrared images and visible light images may include the following operations:

[0106] 101. Obtain a first image and a second image. The first image is an image obtained after the original image undergoes a first-level image processing operation, and the second image is an image obtained after performing a shooting operation on the scene corresponding to the original image.

[0107] In an embodiment of the present invention, the first image is a visible light grayscale image and the second image is an infrared image, and the original image is a visible light image obtained after performing a shooting operation on the scene.

[0108] In the embodiments of the present invention, for the visible light image corresponding to the original image, if it is in the YUV space, the Y space is taken as a single channel; if it is in the RGB space, it remains as a three-channel image; if the infrared image corresponding to the second image is a pseudo-color image, it is converted into a temperature value matrix and becomes a single channel.

[0109] 102. According to a preset filtering function and layer processing function, perform a preset secondary image processing operation on the first image and the second image to obtain a first processed image corresponding to the first image, a first layer parameter, a second processed image corresponding to the second image, and a second layer parameter.

[0110] In the embodiments of the present invention, in step 102, according to a preset filtering function and layer processing function, performing a preset secondary image processing operation on the first image and the second image to obtain a first processed image corresponding to the first image, a first layer parameter, a second processed image corresponding to the second image, and a second layer parameter may specifically include the following operations:

[0111] Perform a normalization processing operation on the first image and the second image according to a preset normalization processing function to obtain a first processed image corresponding to the first image and a second processed image corresponding to the second image;

[0112] Determine target layer parameters for calculating the first layer parameter of the first image and the second layer parameter of the second image. The target layer parameters include the number of decomposition levels, a window coefficient queue, and an offset coefficient queue;

[0113] Perform a parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, maximum value function, and target layer parameters to obtain a first layer parameter corresponding to the first processed image and a second layer parameter corresponding to the second processed image.

[0114] In the embodiments of the present invention, the number of decomposition levels is specifically denoted as n for the number of multi-scale decompositions; the window coefficient queue for each layer , where is an odd integer; the offset coefficient queue for each layer , where is a positive integer.

[0115] 103. Determine a first weight matrix corresponding to the first layer parameter and a second weight matrix corresponding to the second layer parameter according to the first layer parameter and the second layer parameter.

[0116] In the embodiment of the present invention, in step 103, according to a preset guiding filter function, maximum value function, and target layer parameters, the parameter decomposition operation is performed on the first processed image and the second processed image to obtain the first layer parameters corresponding to the first processed image and the second layer parameters corresponding to the second processed image. The specific operations may include the following:

[0117] According to the preset guiding filter function, maximum value function, target layer parameters, and the determined guiding matrix queue, perform a first-level parameter decomposition operation on the first processed image and the second processed image to obtain a first base layer queue corresponding to the first processed image and a second base layer queue corresponding to the second processed image;

[0118] According to the maximum value function, perform a second-level parameter decomposition operation on the first base layer queue and the second base layer queue to obtain a first detail layer queue corresponding to the first base layer queue, a first detail layer weight queue, a second detail layer queue corresponding to the second base layer queue, and a second detail layer weight queue;

[0119] Determine the first base layer queue, the first detail layer queue, and the first detail layer weight queue as the first layer parameters corresponding to the first processed image, and determine the second base layer queue, the second detail layer queue, and the second detail layer weight queue as the second layer parameters corresponding to the second processed image.

[0120] In the embodiment of the present invention, it should be noted that the above-mentioned first base layer queue The specific calculation formula is:

[0121]

[0122] The second base layer queue

[0123]

[0124] Among them, For the guiding filter function, imgC is a single-channel grayscale image obtained by processing the gray value and normalizing the original image (visible light image), is the normalized infrared image, S k is a preset guiding matrix, r k is the window coefficient queue, p k is the offset coefficient queue;

[0125] The guiding matrix queue The calculation method is as follows:

[0126]

[0127] And, is the maximum value function.

[0128] Furthermore, the visible light image base layer queue , and the infrared image base layer queue After that, the visible light image detail layer queue , and the infrared image detail layer queue The calculation formulas are as follows:

[0129]

[0130]

[0131] In an embodiment of the present invention, furthermore, the method for determining the first weight matrix corresponding to the first layer parameter and the second weight matrix corresponding to the second layer parameter according to the first layer parameter and the second layer parameter is specifically as follows:

[0132] Calculate the first weight matrix corresponding to the first layer parameter according to the first base layer queue, the first detail layer queue, and the first detail layer weight queue;

[0133] Calculate the second weight matrix corresponding to the second layer parameter according to the second base layer queue, the second detail layer queue, and the second detail layer weight queue.

[0134] Among them, the calculation formulas of the first weight matrix and the second weight matrix are specifically as follows:

[0135]

[0136]

[0137] Among them, n is the number of decomposition layers, imgC_W is the first weight matrix corresponding to the first layer parameter, C n is the first base layer queue, W_c k is the first detail layer weight queue, C_d k is the first detail layer queue; similarly, imgT_W is the second weight matrix corresponding to the second layer parameter, T n is the second base layer queue, W_t k is the second detail layer weight queue, T_d k is the second detail layer queue;

[0138] Among them, the first detail layer weight queue W_c k and the second detail layer weight queue W_t k The corresponding calculation formulas are specifically as follows:

[0139]

[0140]

[0141] Among them, , it is to find the L p norm of A.

[0142] 104. Perform a preset color coding operation on the second processed image to obtain a color coding result corresponding to the second processed image.

[0143] 105. According to the first weight matrix and the second weight matrix, perform a preset dual-light fusion operation on the first image and the color coding result to obtain a target output image.

[0144] It can be seen that when implementing Figure 1 the described multi-scale fusion method of infrared images and visible light images, after obtaining the first and second images, it can automatically perform secondary image processing operations on the first and second images through preset filtering functions and layer processing functions to obtain the first and second processed images and the first and second layer parameters, realizing the intelligent image parameter decomposition of dual-light images; and then determining the first and second weight matrices, and finally intelligently fusing the first image and the color coding result through the first and second weight matrices. Through the parameter decomposition of the light source image and the generation of the weight matrix, the fusion of visible light images and infrared images is realized, and the fusion effect of the dual-light fusion result is improved.

[0145] Embodiment 2

[0146] Please refer to Figure 2 and Figure 7 , Figure 2 which is a schematic flowchart of another multi-scale fusion method of infrared images and visible light images disclosed in the embodiments of the present invention, Figure 7 and Figure 2 is a comparison diagram of the fusion effect of the output image obtained after performing the multi-scale fusion method of infrared images and visible light images disclosed in the embodiments of the present invention. Among them, Figure 2 the described multi-scale fusion method of infrared images and visible light images can be applied to a multi-scale fusion device for infrared images and visible light images, which is not limited in the embodiments of the present invention. As

[0147] 201. Obtain a first image and a second image. The first image is an image obtained after the original image undergoes a primary image processing operation, and the second image is an image obtained after performing a shooting operation on the scene corresponding to the original image.

[0148] 202. Perform a preset secondary image processing operation on the first image and the second image according to a preset filtering function and a layer processing function, to obtain a first processed image corresponding to the first image, first layer parameters, a second processed image corresponding to the second image, and second layer parameters.

[0149] 203. Determine a first weight matrix corresponding to the first layer parameters and a second weight matrix corresponding to the second layer parameters according to the first layer parameters and the second layer parameters.

[0150] 204. Perform a preset color encoding operation on the second processed image to obtain a color encoding result corresponding to the second processed image.

[0151] In an embodiment of the present invention, optionally, the manner of performing a preset color encoding operation on the second processed image in step 204 to obtain a color encoding result corresponding to the second processed image may specifically include the following operations:

[0152] Determine a preset number of encoding parameters within a preset numerical range;

[0153] Calculate channel components corresponding to the second processed image according to all the encoding parameters to obtain a set of channel components corresponding to the second processed image, and the total number of channel components in the set of channel components is a preset number;

[0154] Determine the set of channel components corresponding to the second processed image as the color encoding result corresponding to the second processed image.

[0155] In an embodiment of the present invention, it should be noted that, denote , that is, select three encoding parameters l1, l2, l3, and select the specific value calculated from the temperature distribution statistical result of the actual scene data ; the encoded infrared color image is C, and the calculation formula is as follows:

[0156]

[0157]

[0158]

[0159] Wherein: , , are respectively the three channel components of the infrared color image C.

[0160] 205. Perform a preset dual - light fusion operation on the first image and the color encoding result according to the first weight matrix and the second weight matrix to obtain a target output image.

[0161] In an embodiment of the present invention, in step 205, according to the first weight matrix and the second weight matrix, a preset dual-light fusion operation is performed on the first image and the color coding result to obtain the target output image, and the specific method may include the following operations:

[0162] Obtain the original image before the first image undergoes the primary image processing operation;

[0163] According to the number of channel components in the color coding result and the original image, determine all fusion parameters corresponding to the original image, and the number of fusion parameters is a preset number;

[0164] For each channel component in the color coding result, determine the fusion parameter corresponding to each channel component, and according to the first weight matrix, the second weight matrix, combine the channel component and the fusion parameter corresponding to the channel component, and calculate to obtain the fusion component corresponding to each channel component;

[0165] Determine the target output image according to all the fusion components.

[0166] In an embodiment of the present invention, further, according to the obtained original image (visible light image) , the first weight matrix , the second weight matrix and the color coding result, calculate the three channels of the target output image in the following manner:

[0167]

[0168]

[0169]

[0170] In an embodiment of the present invention, optionally, after the three-channel data calculation is successful, output the image imgO, and according to the data call requirements, perform subsequent data conversion and encapsulation operations on the output image to encapsulate the data into the required data format for storage and call, which is not limited in the embodiment of the present invention.

[0171] In an embodiment of the present invention, for other descriptions of steps 201 - 205, please refer to the other specific descriptions of steps 101 - 105 in Embodiment 1, which will not be elaborated in the embodiment of the present invention.

[0172] 206. Obtain the target output image and a reference image, where the reference image is used to calculate the fusion value of the target output image.

[0173] 207. Analyze the target output image and the reference image according to the original image to obtain the fusion value corresponding to the target output image.

[0174] 208. Determine whether the fusion value is greater than a preset fusion threshold.

[0175] In an embodiment of the present invention, when the determination result of step 208 is yes, step 209 is executed; when the determination result of step 208 is no, step 210 is executed.

[0176] 209. Determine that the target output image is an image that meets the preset dual - light fusion standard.

[0177] 210. Generate a failure message indicating that the target output image does not meet the preset dual - light fusion standard based on the fusion value, the target output image, the first weight matrix, and the second weight matrix.

[0178] 211. According to the failure message, adjust the target layer parameters and, based on the adjusted target layer parameters, perform the above - mentioned operation of performing parameter decomposition on the first processed image and the second processed image according to the preset guided filtering function, maximum function, and target layer parameters to obtain the first layer parameters corresponding to the first processed image and the second layer parameters corresponding to the second processed image.

[0179] In an embodiment of the present invention, for a comparison diagram of the fusion effect of the multi - scale fusion method of infrared images and visible - light images described in the embodiment of the present invention, please refer to Figure 7 , where Figure 7 the left figure is the fusion effect diagram when the decomposition layer number n = 1, and the right figure is the fusion effect diagram when the decomposition layer number n = 5.

[0180] It can be seen that implementing Figure 2 the multi - scale fusion method of infrared images and visible - light images described can, after outputting the target output image, automatically determine the fusion value of the target output image according to the reference image, and then determine the target output image with a fusion value greater than the preset fusion threshold as an image that meets the preset dual - light fusion standard, and for the target output image with a fusion value less than or equal to the preset fusion threshold, automatically generate a failure message and adjust the target layer parameters according to the failure message, improving the success rate and accuracy of finally determining that the target output image meets the dual - light fusion standard.

[0181] In an optional embodiment, before obtaining the first image and the second image, the method may further include:

[0182] Determine the spatial type for processing the original image, where the spatial type includes the yuv spatial type or the RGB spatial type;

[0183] Determine the number of channels for processing the original image according to the spatial type. According to the spatial type and the number of channels corresponding to the spatial type, perform a primary image processing operation on the original image to obtain a first image and trigger the operation of obtaining the first image and the second image as described above.

[0184] It can be seen that in the embodiments of the present invention, a processing solution for processing the original image is provided, which improves the accuracy of processing the original image and the reliability of obtaining the first image.

[0185] Embodiment III

[0186] Please refer to Figure 3 , Figure 3 which is a schematic structural diagram of a multi-scale fusion device for infrared images and visible light images disclosed in the embodiments of the present invention. Among them, the multi-scale fusion device for infrared images and visible light images can be a multi-scale fusion terminal for infrared images and visible light images, a multi-scale fusion device for infrared images and visible light images, a multi-scale fusion system for infrared images and visible light images, or a multi-scale fusion server for infrared images and visible light images. The multi-scale fusion server for infrared images and visible light images can be a local server, a remote server, or a cloud server (also known as a cloud server). When the multi-scale fusion server for infrared images and visible light images is a non-cloud server, the non-cloud server can communicate with the cloud server. The embodiments of the present invention do not make any limitations. As Figure 3 shown, the multi-scale fusion device for infrared images and visible light images may include an acquisition module 301, an image processing module 302, a determination module 303, an encoding module 304, and a fusion module 305, where:

[0187] The acquisition module 301 is configured to acquire a first image and a second image. The first image is an image obtained after a primary image processing operation on the original image, and the second image is an image obtained after performing a shooting operation on the scene corresponding to the original image. The first image is a visible light grayscale image and the second image is an infrared image. The original image is a visible light image obtained after performing a shooting operation on the scene.

[0188] The image processing module 302 is configured to perform a preset secondary image processing operation on the first image and the second image according to a preset filtering function and a layer processing function to obtain a first processed image corresponding to the first image, a first layer parameter, a second processed image corresponding to the second image, and a second layer parameter.

[0189] The determination module 303 is configured to determine a first weight matrix corresponding to the first layer parameter and a second weight matrix corresponding to the second layer parameter according to the first layer parameter and the second layer parameter.

[0190] The encoding module 304 is configured to perform a preset color encoding operation on the second processed image to obtain a color encoding result corresponding to the second processed image.

[0191] The fusion module 305 is configured to perform a preset dual-light fusion operation on the first image and the color encoding result according to the first weight matrix and the second weight matrix to obtain a target output image.

[0192] In an embodiment of the present invention, optionally, the image processing module 302 performs a preset secondary image processing operation on the first image and the second image according to a preset filtering function and a layer processing function to obtain a first processed image corresponding to the first image, a first layer parameter, a second processed image corresponding to the second image, and a second layer parameter, and the specific manner includes:

[0193] Perform a normalization processing operation on the first image and the second image according to a preset normalization processing function to obtain a first processed image corresponding to the first image and a second processed image corresponding to the second image;

[0194] Determine target layer parameters for calculating a first layer parameter of the first image and a second layer parameter of the second image. The target layer parameters include the number of decomposition levels, a window coefficient queue, and an offset coefficient queue;

[0195] Perform a parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, a maximum value function, and the target layer parameters to obtain a first layer parameter corresponding to the first processed image and a second layer parameter corresponding to the second processed image.

[0196] Further, the manner in which the image processing module 302 performs a parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, a maximum value function, and the target layer parameters to obtain a first layer parameter corresponding to the first processed image and a second layer parameter corresponding to the second processed image specifically includes:

[0197] Perform a primary parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, a maximum value function, the target layer parameters, and a determined guided matrix queue to obtain a first base layer queue corresponding to the first processed image and a second base layer queue corresponding to the second processed image;

[0198] Perform a secondary parameter decomposition operation on the first base layer queue and the second base layer queue according to the maximum value function to obtain a first detail layer queue corresponding to the first base layer queue, a first detail layer weight queue, a second detail layer queue corresponding to the second base layer queue, and a second detail layer weight queue;

[0199] Determine the first layer parameters corresponding to the first processed image as the first base layer queue, the first detail layer queue, and the first detail layer weight queue, and determine the second layer parameters corresponding to the second processed image as the second base layer queue, the second detail layer queue, and the second detail layer weight queue.

[0200] In an embodiment of the present invention, optionally, the encoding module 304 performs a preset color encoding operation on the second processed image to obtain the color encoding result corresponding to the second processed image, and the specific method includes:

[0201] Determine a preset number of encoding parameters within a preset numerical range;

[0202] According to all the encoding parameters, calculate the channel components corresponding to the second processed image to obtain a set of channel components corresponding to the second processed image, and the total number of channel components in the set of channel components is a preset number;

[0203] Determine the set of channel components corresponding to the second processed image as the color encoding result corresponding to the second processed image.

[0204] In an embodiment of the present invention, optionally, the fusion module 305 performs a preset dual-light fusion operation on the first image and the color encoding result according to the first weight matrix and the second weight matrix to obtain the target output image, and the specific method includes:

[0205] Obtain the original image of the first image before the first-level image processing operation;

[0206] According to the number of channel components in the color encoding result and the original image, determine all the fusion parameters corresponding to the original image, and the number of fusion parameters is a preset number;

[0207] For each channel component in the color encoding result, determine the fusion parameter corresponding to each channel component, and calculate the fusion component corresponding to each channel component according to the first weight matrix, the second weight matrix, the channel component, and the fusion parameter corresponding to the channel component;

[0208] Determine the target output image according to all the fusion components.

[0209] In an embodiment of the present invention, optionally, the determination module 303 determines the first weight matrix corresponding to the first layer parameters and the second weight matrix corresponding to the second layer parameters according to the first layer parameters and the second layer parameters, including:

[0210] Calculate the first weight matrix corresponding to the first layer parameters according to the first base layer queue, the first detail layer queue, and the first detail layer weight queue;

[0211] Calculate the second weight matrix corresponding to the second layer parameters according to the second base layer queue, the second detail layer queue, and the second detail layer weight queue.

[0212] Among them, the calculation formulas for the first weight matrix and the second weight matrix are specifically as follows:

[0213]

[0214]

[0215] Among them, n is the number of decomposition layers, imgC_W is the first weight matrix corresponding to the first layer parameters, C n is the first base layer queue, W_c k is the first detail layer weight queue, C_d k is the first detail layer queue; similarly, imgT_W is the second weight matrix corresponding to the second layer parameters, T n is the second base layer queue, W_t k is the second detail layer weight queue, T_d k is the second detail layer queue;

[0216] Among them, the first detail layer weight queue W_c k and the second detail layer weight queue W_t k The corresponding calculation formulas are specifically as follows:

[0217]

[0218]

[0219] Among them, , is to find the L p norm of A.

[0220] It can be seen that implementing Figure 3 The described multi-scale fusion device for infrared images and visible light images can automatically perform secondary image processing operations on the first and second images through preset filtering functions and layer processing functions after obtaining the first and second images, obtain the first and second processed images and the first and second layer parameters, realizing the intelligent image parameter decomposition of dual-light images; furthermore, determine the first and second weight matrices, and finally intelligently fuse the first image and the color coding result through the first and second weight matrices. Through the parameter decomposition of the light source image and the generation of the weight matrix, the fusion of visible light images and infrared images is realized, and the fusion effect of the dual-light fusion result is improved.

[0221] In an alternative embodiment, the obtaining module 301 is further configured to obtain a target output image and a reference image, where the reference image is used to calculate a fusion value of the target output image.

[0222] As Figure 4 shown, the apparatus may further include an analysis module 306, a judgment module 307, a generation module 308, and an adjustment module 309, where:

[0223] The analysis module 306 is configured to analyze the target output image and the reference image according to the original image, and obtain a fusion value corresponding to the target output image.

[0224] The judgment module 307 is configured to judge whether the fusion value is greater than a preset fusion threshold. When the judgment result is yes, it is determined that the target output image is an image that meets the preset dual - light fusion standard.

[0225] The generation module 308 is configured to, when the judgment result of the judgment module 307 is no, generate a failure message indicating that the target output image does not meet the preset dual - light fusion standard according to the fusion value obtained by the analysis module 306, the target output image obtained by the fusion module 305, the first weight matrix determined by the determination module 303, and the second weight matrix.

[0226] The adjustment module 309 is configured to adjust the target layer parameters according to the failure message, and trigger the image processing module 302 to perform the above - mentioned operation of performing parameter decomposition on the first processed image and the second processed image according to the preset guided filtering function, maximum value function, and target layer parameters to obtain the first layer parameters corresponding to the first processed image and the second layer parameters corresponding to the second processed image according to the adjusted target layer parameters.

[0227] It can be seen that the multi - scale fusion apparatus for infrared images and visible - light images described in Figure 4 can, after outputting the target output image, automatically determine the fusion value of the target output image according to the reference image, and then determine the target output image with a fusion value greater than the preset fusion threshold as an image that meets the preset dual - light fusion standard, and for the target output image with a fusion value less than or equal to the preset fusion threshold, automatically generate a failure message and adjust the target layer parameters according to the failure message, improving the success rate and accuracy of finally determining that the target output image meets the dual - light fusion standard.

[0228] Embodiment 4

[0229] Please refer to Figure 5 , Figure 5 which is a schematic structural diagram of another multi - scale fusion apparatus for infrared images and visible - light images disclosed in the embodiments of the present invention. As Figure 5As shown, the multi-scale fusion device for infrared images and visible light images may include:

[0230] A memory 401 storing executable program code;

[0231] A processor 402 coupled to the memory 401;

[0232] The processor 402 calls the executable program code stored in the memory 401 and executes the steps in the multi-scale fusion method for infrared images and visible light images described in Embodiment 1 or Embodiment 2 of the present invention.

[0233] Embodiment 5

[0234] Embodiment of the present invention discloses a computer storage medium. The computer storage medium stores computer instructions, which, when called, are used to execute the steps in the multi-scale fusion method for infrared images and visible light images described in Embodiment 1 or Embodiment 2 of the present invention.

[0235] Embodiment 6

[0236] Embodiment of the present invention discloses a computer program product. The computer program product includes a non-transitory computer storage medium storing a computer program, and the computer program is operable to cause a computer to execute the steps in the multi-scale fusion method for infrared images and visible light images described in Embodiment 1 or Embodiment 2.

[0237] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0238] Through the specific descriptions of the above embodiments, those skilled in the art can clearly understand that each implementation manner can be realized by means of software plus a necessary general hardware platform. Of course, it can also be realized by hardware. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer storage medium, and the storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc memories, magnetic disc memories, tape memories, or any other computer-readable medium that can be used to carry or store data.

[0239] Finally, it should be noted that: what is disclosed in a multi-scale fusion method and device for infrared images and visible light images disclosed in the embodiments of the present invention is only the preferred embodiments of the present invention, and is only used to illustrate the technical solutions of the present invention, rather than limiting it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-scale fusion method for infrared images and visible light images, characterized in that The method includes: Obtaining a first image and a second image, where the first image is an image obtained after a first-level image processing operation on an original image, the second image is an image obtained after performing a shooting operation on the scene corresponding to the original image, the first image is a visible light grayscale image and the second image is an infrared image, and the original image is a visible light image obtained after performing a shooting operation on the scene; Performing a preset second-level image processing operation on the first image and the second image according to a preset filtering function and a layer processing function to obtain a first processed image corresponding to the first image, a first layer parameter, a second processed image corresponding to the second image, and a second layer parameter; Determining a first weight matrix corresponding to the first layer parameter and a second weight matrix corresponding to the second layer parameter according to the first layer parameter and the second layer parameter; Performing a preset color encoding operation on the second processed image to obtain a color encoding result corresponding to the second processed image; Performing a preset dual-light fusion operation on the first image and the color encoding result according to the first weight matrix and the second weight matrix to obtain a target output image.

2. The multi-scale fusion method of infrared images and visible light images according to claim 1, characterized in that The performing a preset second-level image processing operation on the first image and the second image according to a preset filtering function and a layer processing function to obtain a first processed image corresponding to the first image, a first layer parameter, a second processed image corresponding to the second image, and a second layer parameter includes: Performing a normalization processing operation on the first image and the second image according to a preset normalization processing function to obtain a first processed image corresponding to the first image and a second processed image corresponding to the second image; Determining target layer parameters for calculating a first layer parameter of the first image and a second layer parameter of the second image, where the target layer parameters include the number of decomposition levels, a window coefficient queue, and an offset coefficient queue; Performing a parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, a maximum value function, and the target layer parameters to obtain a first layer parameter corresponding to the first processed image and a second layer parameter corresponding to the second processed image.

3. The multi-scale fusion method of infrared images and visible light images according to claim 2, characterized in that The performing a parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, a maximum value function, and the target layer parameters to obtain a first layer parameter corresponding to the first processed image and a second layer parameter corresponding to the second processed image includes: Performing a first-level parameter decomposition operation on the first processed image and the second processed image according to a preset guided filtering function, a maximum value function, the target layer parameters, and a determined guided matrix queue to obtain a first base layer queue corresponding to the first processed image and a second base layer queue corresponding to the second processed image; Perform a secondary parameter decomposition operation on the first basic layer queue and the second basic layer queue according to the maximum value function to obtain a first detail layer queue corresponding to the first basic layer queue, a first detail layer weight queue, a second detail layer queue corresponding to the second basic layer queue, and a second detail layer weight queue; Determine the first basic layer queue, the first detail layer queue, and the first detail layer weight queue as the first layer parameters corresponding to the first processed image, and determine the second basic layer queue, the second detail layer queue, and the second detail layer weight queue as the second layer parameters corresponding to the second processed image.

4. The multi-scale fusion method of infrared images and visible light images according to claim 3, characterized in that, The determining the first weight matrix corresponding to the first layer parameters and the second weight matrix corresponding to the second layer parameters according to the first layer parameters and the second layer parameters includes: Calculate the first weight matrix corresponding to the first layer parameters according to the first basic layer queue, the first detail layer queue, and the first detail layer weight queue; Calculate the second weight matrix corresponding to the second layer parameters according to the second basic layer queue, the second detail layer queue, and the second detail layer weight queue.

5. The multi-scale fusion method of infrared images and visible light images according to any one of claims 2-4, characterized in that The performing a preset color coding operation on the second processed image to obtain a color coding result corresponding to the second processed image includes: Determine a preset number of coding parameters within a preset numerical range; Calculate the channel components corresponding to the second processed image according to all the coding parameters to obtain a set of channel components corresponding to the second processed image, and the total number of the channel components in the set of channel components is the preset number; Determine the set of channel components corresponding to the second processed image as the color coding result corresponding to the second processed image.

6. The multi-scale fusion method for infrared images and visible light images according to claim 5, wherein, The performing a preset dual-light fusion operation on the first image and the color coding result according to the first weight matrix and the second weight matrix to obtain a target output image includes: Obtain the original image before the first image undergoes the primary image processing operation; Determine all the fusion parameters corresponding to the original image according to the number of the channel components in the color coding result and the original image, and the number of the fusion parameters is the preset number; For each of the channel components in the color coding result, determine the fusion parameter corresponding to each of the channel components, and calculate the fusion component corresponding to each of the channel components according to the first weight matrix, the second weight matrix, the channel component, and the fusion parameter corresponding to the channel component; Determine the target output image according to all the fusion components.

7. The multi-scale fusion method of infrared images and visible light images according to claim 6, wherein The method further includes: Obtain the target output image and a reference image, where the reference image is used to calculate the fusion value of the target output image; Analyze the target output image and the reference image according to the original image to obtain the fusion value corresponding to the target output image; Judge whether the fusion value is greater than a preset fusion threshold, and when the judgment result is yes, determine that the target output image is an image that meets the preset dual-light fusion standard; When the judgment result is negative, generate a failure message indicating that the target output image does not meet the preset dual - light fusion standard according to the fusion value, the target output image, the first weight matrix, and the second weight matrix; According to the failure message, adjust the target layer parameters, and according to the adjusted target layer parameters, perform the operation of performing parameter decomposition on the first processed image and the second processed image according to the preset guided filtering function, maximum function, and the target layer parameters, to obtain the first layer parameters corresponding to the first processed image and the second layer parameters corresponding to the second processed image.

8. The multi-scale fusion method of infrared images and visible light images according to claim 4, characterized in that The specific calculation formulas of the first weight matrix and the second weight matrix are as follows: where n is the number of decomposition layers, imgC_W is the first weight matrix corresponding to the first layer parameter, C n is the first base layer queue, W_c k is the first detail layer weight queue, C_d k is the first detail layer queue; similarly, imgT_W is the second weight matrix corresponding to the second layer parameter, T n is the second base layer queue, W_t k is the second detail layer weight queue, T_d k is the second detail layer queue; Among them, the first detail layer weight queue W_c k and the second detail layer weight queue W_t k The corresponding calculation formula is specifically as follows: Among them, , it is to calculate the L p norm of A.

9. A multi-scale fusion device for infrared images and visible light images, characterized in that, The device includes: An acquisition module, configured to acquire a first image and a second image. The first image is an image obtained after a primary image processing operation on the original image, and the second image is an image obtained after performing a shooting operation on the scene corresponding to the original image. The first image is a visible - light grayscale image and the second image is an infrared image, and the original image is a visible - light image obtained after performing a shooting operation on the scene; An image - processing module, configured to perform a preset secondary image - processing operation on the first image and the second image according to a preset filtering function and layer - processing function, to obtain a first processed image corresponding to the first image, first layer parameters, a second processed image corresponding to the second image, and second layer parameters; A determination module, configured to determine a first weight matrix corresponding to the first layer parameters and a second weight matrix corresponding to the second layer parameters according to the first layer parameters and the second layer parameters; An encoding module, configured to perform a preset color - encoding operation on the second processed image to obtain a color - encoding result corresponding to the second processed image; A fusion module, configured to perform a preset dual - light fusion operation on the first image and the color - encoding result according to the first weight matrix and the second weight matrix to obtain a target output image.

10. A multi-scale fusion device for infrared images and visible light images, characterized in that, The device includes: A memory storing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory and executes the multi - scale fusion method of infrared images and visible - light images according to any one of claims 1 - 8.

Images