Artificial intelligence-based method and system for evaluating the corrosion degree of ancient building exteriors

Abstract

The present invention discloses an artificial intelligence-based method and system for evaluating the degree of corrosion of the exterior of ancient buildings. The method includes image acquisition, image enhancement, establishing an ancient building exterior corrosion degree evaluation model, and evaluating the degree of corrosion of the exterior of ancient buildings. The present invention belongs to the field of image processing, and specifically refers to an artificial intelligence-based method and system for evaluating the degree of corrosion of the exterior of ancient buildings. This solution uses an atmospheric scattering model to estimate transmittance and restore scene radiation; utilizes cyclic channel compensation to ensure that the original texture is accurately restored; adopts local adaptive contrast enhancement and weighted fusion of IMF components at each layer of the pyramid; thereby improving the accuracy of corrosion degree evaluation; designs multi-branch exclusive feature extraction, focusing on color spot fading, microcrack pitting, and macro-peeling contours respectively; embeds SE-style coding attention in three-layer convolution; and adaptively adjusts noise intensity based on local variance by designing a perturbation activation function; thereby improving the effect of appearance corrosion degree evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to an artificial intelligence-based method and system for evaluating the degree of corrosion of the exterior of ancient buildings. Background Art

[0002] Methods for assessing the degree of corrosion on the exterior of ancient buildings utilize a combination of techniques and methods to quantitatively analyze and grade the corrosion caused by natural factors, human factors, and other factors. However, these methods often overlook the effects of sunlight angle and shadows, misjudge the extent of material damage due to color casts, and struggle to adjust contrast to accommodate differences in lighting or material, leading to poor accuracy in corrosion assessments. Furthermore, these methods are prone to feature extraction conflicts, lack detail mining, and inadequate processing of shooting noise and compression artifacts, resulting in poor corrosion assessments. Summary of the Invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an artificial intelligence-based method and system for evaluating the degree of corrosion of the exterior of ancient buildings. The general method for evaluating the degree of corrosion of the exterior of ancient buildings has the problem of ignoring the influence of sunlight angle and shadow, misjudging the degree of material damage due to color cast, and difficulty in adjusting contrast to adapt to differences in lighting or materials, which leads to poor accuracy in the final evaluation of the degree of corrosion. This solution estimates the transmittance through an atmospheric scattering model to restore scene radiation, remove the overall graying caused by haze and dust, and improve the details of the distant view and the authenticity of color; utilizes cyclic channel compensation to adaptively correct the color cast caused by sunlight, shadow and shooting angle under the goal of minimizing the sum of squares of channel differences, ensuring that the original texture of stone and brickwork is accurately restored; adopts local adaptive contrast enhancement to adaptively amplify the difference between light and dark, thereby highlighting tiny cracks and peeling edges; utilizes crack significance, texture frequency and illumination at each layer of the pyramid The weighted fusion of the IMF components is performed by balancing the three types of weights, which not only retains the optimal details but also avoids visual mutations between different scales; thereby improving the accuracy of corrosion degree assessment; in view of the fact that the general corrosion degree assessment method of ancient buildings is prone to feature extraction conflicts, insufficient detail mining, and insufficient processing capabilities for shooting noise and compression artifacts, which leads to poor appearance corrosion degree assessment results, this scheme designs multi-branch exclusive feature extraction, focusing on color spot fading, microcrack pitting and macro peeling contours respectively, avoiding redundant interference between different feature channels, and improving the richness and discriminability of overall feature expression; embedding SE-style coded attention in the three-layer convolution to realize dynamic weighting of each channel feature under different corrosion stages and lighting conditions; by designing a perturbation activation function, the noise intensity is adaptively adjusted based on the local variance, alleviating channel death and improving the fault tolerance to shooting noise and compression artifacts; thereby improving the appearance corrosion degree assessment effect.

[0004] The technical solution adopted by the present invention is as follows: The present invention provides an artificial intelligence-based method for evaluating the degree of corrosion of the exterior of ancient buildings, which comprises the following steps:

[0005] Step S1: image acquisition;

[0006] Step S2: image enhancement;

[0007] Step S3: establishing a model for evaluating the degree of corrosion of the ancient building exterior;

[0008] Step S4: Assess the degree of corrosion of the ancient building's exterior.

[0009] Furthermore, in step S1, the image acquisition is to acquire the image of the historical ancient building and the corrosion level; and the corrosion level is used as the image label.

[0010] Furthermore, in step S2, the image enhancement specifically includes the following steps:

[0011] Step S21: atmospheric scattering defogging; for the original image, calculate the transmittance , expressed as: ; where x is the image pixel position; is the experience weight; is the local window centered at x, and y is the pixel position index within the local window; is the global atmospheric illumination component; is the original image brightness; the recovered scene radiance is expressed as: ; Where c is the RGB color channel index; is the lower limit of transmittance; is the true color image after defogging; I(·) is the pixel value of the original image; A is the color value of the background atmosphere; the defogging image J is obtained;

[0012] Step S22: Circular channel compensation; the pixel values ​​of the RGB channels of the defogging image J are recorded as follows: 、 and ,but ; ; ;in, and are the minimum and maximum values ​​of the red channel in the entire image range; and are the maximum and minimum values ​​of the green channel respectively; 、 and are the values ​​of the red, green, and blue channels after compensation; the convergence goal is to minimize the sum of squares of channel differences, expressed as: ; Where Loss is the color loss; output the color-compensated image;

[0013] Step S23: Local adaptive contrast enhancement; for image processing after color compensation, including: local average brightness , ; Local brightness variance , ; Contrast magnification factor , ; Local brightness linear stretching, ; Where H and W are the height and width of the image respectively; i and j are image coordinate indices; B is the local block coordinate set; is the brightness value of the pixel in the local block; is the enhanced local brightness; is the maximum coefficient; is the global brightness standard deviation; the image with local contrast enhancement is obtained;

[0014] Step S24: iterative decomposition; the iterative decomposition method is used to obtain the intrinsic mode function { }, the convergence criterion is expressed as: ; Where SL is the convergence criterion; D is the pixel area; and They are the pixel values ​​of the IMF components at the t+1th iteration and the tth iteration; after all convergence, the IMF components are obtained ;in, is the kth IMF component; K is the total number of components;

[0015] Step S25: Multi-weighted image pyramid fusion; fuse the IMF components according to the three weights of crack significance, texture frequency and illumination balance. For each IMF component, ; Among them, for the first level of the pyramid, ; Finally, the enhanced image of the historical ancient building is reconstructed from the bottom up ;in, is the fusion result of the lth layer of the image pyramid; is the regularization constant; is the weighting coefficient of each level; is the image result of the pyramid level l; 、 and are the crack significance weight, texture frequency weight, and illumination balance weight under the kth IMF component; 、 and are the crack significance weight, texture frequency weight and illumination balance weight under the mth IMF component respectively.

[0016] Furthermore, in step S3, the establishment of an ancient building exterior corrosion degree assessment model is based on the historical ancient building images completed by deep learning and enhancement, and specifically includes the following steps:

[0017] Step S31: Overall model design; the model inputs the enhanced ancient building appearance image; after processing through the three-branch layer and the hybrid coding attention layer, it is processed through full connection + softmax to obtain the predicted distribution, and finally outputs the corrosion level predicted by the model; the cross entropy loss function is used, and the forward propagation is based on the gradient descent method;

[0018] Step S32: Three-branch layer design; construct color branch, texture branch and morphology branch respectively; color branch captures the color distribution changes of mottled and faded areas; texture branch extracts the detailed texture of microcracks and pits; morphology branch captures the peeling edge and macro contour changes; color branch uses 3×64 2D convolution kernels to extract color spots and fading features; and 3×128 1D convolution kernels to enhance the color difference distribution; texture branch uses 8 layers of densely connected 3×3 convolutions and inserts a maximum pooling layer in the middle; the morphology branch has the same structure as the color branch, but the number of kernels is reduced by S; the three branches are spliced ​​together to obtain ;

[0019] Step S33: Hybrid coding layer design; Perform three layers of convolution in sequence, perform SE processing on the output of each convolution layer, and use the SE processing output as the input of the next convolution layer; including: convolution processing; perturbation activation function processing; global average pooling; sigmoid processing; reweighting processing;

[0020] Step S34: Design of perturbation activation function; introduction of dynamic noise intensity , the perturbation activation function is expressed as: ; ; Where u is the perturbation activation function variable; , a set of noise is sampled and injected into the activation each time forward propagation; It is a noise bias, which is used to enhance the input image’s tolerance to shooting noise and artifacts; is the initial noise intensity; is the attenuation coefficient; is the standard deviation of the input pixel values ​​in the local window.

[0021] Furthermore, in step S4, the assessment of the degree of corrosion of the appearance of ancient buildings is based on the established model for assessing the degree of corrosion of the appearance of ancient buildings. The images of ancient buildings are collected in real time, and after image enhancement processing, they are input into the model for assessing the degree of corrosion of the appearance of ancient buildings. The corrosion level output by the model is used as the result of the assessment of the degree of corrosion of the appearance of ancient buildings.

[0022] The artificial intelligence-based ancient building appearance corrosion degree assessment system provided by the present invention includes an image acquisition module, an image enhancement module, an ancient building appearance corrosion degree assessment model establishment module and an ancient building appearance corrosion degree assessment module;

[0023] The image acquisition module acquires the appearance image of the historical building and marks the corrosion level;

[0024] The image enhancement module enhances the appearance image of historical buildings through atmospheric scattering defogging, cyclic channel compensation, local adaptive contrast enhancement, multi-scale intrinsic mode decomposition and multi-weighted pyramid fusion;

[0025] The module for establishing a model for assessing the degree of corrosion of the appearance of ancient buildings takes the enhanced image as input, extracts the color, texture, and morphology through three branches, and integrates them into the hybrid coding layer of SE attention, and establishes a model for assessing the degree of corrosion of the appearance of ancient buildings based on the perturbation activation function;

[0026] The ancient building appearance corrosion degree assessment module is based on the ancient building appearance corrosion degree assessment model, and implements appearance corrosion degree assessment on the ancient building appearance images collected in real time.

[0027] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0028] (1) In view of the fact that the general method for evaluating the degree of corrosion of ancient buildings ignores the influence of sunlight angle and shadow, the evaluation results misjudge the degree of material damage due to color cast, and the contrast adjustment is difficult to adapt to the difference in lighting or material, which leads to the problem of poor accuracy in the final corrosion degree evaluation, this scheme estimates the transmittance through the atmospheric scattering model to restore the scene radiation, remove the overall graying caused by haze and dust, and improve the details and color fidelity of the distant view; using cyclic channel compensation, under the goal of minimizing the sum of squares of channel differences, adaptively correct the color cast caused by sunlight, shadow and shooting angle, ensuring that the original texture of stone and brickwork is accurately restored; using local adaptive contrast enhancement, adaptively amplifying the difference between light and dark, thereby highlighting tiny cracks and peeling edges; using three types of weights, namely crack significance, texture frequency and illumination balance, to weightedly fuse the IMF components at each layer of the pyramid, which not only retains the optimal details but also avoids visual mutations between different scales, thereby improving the accuracy of corrosion degree evaluation.

[0029] (2) In view of the problems that the general method for evaluating the degree of corrosion of the exterior of ancient buildings is prone to feature extraction conflicts, insufficient detail mining, and insufficient ability to handle shooting noise and compression artifacts, which leads to poor evaluation of the degree of corrosion of the exterior, this scheme designs a multi-branch exclusive feature extraction, focusing on color fading, micro-crack pitting and macro-peeling contours respectively, avoiding redundant interference between different feature channels and improving the richness and discrimination of the overall feature expression; embedding SE-type coded attention in the three-layer convolution to achieve dynamic weighting of the features of each channel under different corrosion stages and lighting conditions; by designing a perturbation activation function, the noise intensity is adaptively adjusted based on the local variance, alleviating channel death and improving the fault tolerance to shooting noise and compression artifacts; thereby improving the evaluation effect of the degree of corrosion of the exterior. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A schematic diagram of the process of the method for evaluating the degree of corrosion of the exterior of ancient buildings based on artificial intelligence provided by the present invention;

[0031] Figure 2 A schematic diagram of the artificial intelligence-based ancient building exterior corrosion assessment system provided by the present invention;

[0032] Figure 3 Schematic diagram of the process of step S2.

[0033] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0035] In the description of the present invention, it should be understood that terms such as "up", "down", "front", "back", "left", "right", "top", "bottom", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the system or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they should not be understood as limiting the present invention.

[0036] Example 1, see Figure 1 The present invention provides an artificial intelligence-based method for evaluating the corrosion degree of the appearance of ancient buildings, which includes the following steps:

[0037] Step S1: Image acquisition: Acquire an image of the exterior of a historical building and mark the corrosion level;

[0038] Step S2: Image enhancement: The image of the exterior of the historical building is enhanced through atmospheric scattering defogging, cyclic channel compensation, local adaptive contrast enhancement, multi-scale intrinsic mode decomposition, and multi-weighted pyramid fusion.

[0039] Step S3: Establishing a model for evaluating the degree of corrosion of the exterior of ancient buildings; taking the enhanced image as input, extracting the color, texture, and morphology branches and integrating them into the hybrid coding layer of SE attention, and establishing a model for evaluating the degree of corrosion of the exterior of ancient buildings based on the perturbation activation function;

[0040] Step S4: evaluating the degree of corrosion of the ancient building's exterior; based on the ancient building's exterior corrosion degree evaluation model, evaluating the degree of corrosion of the ancient building's exterior image collected in real time.

[0041] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the corrosion level is used as an image label; the image label includes no corrosion, mild corrosion, moderate corrosion and severe corrosion.

[0042] Example 3, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S2, image enhancement specifically includes the following steps:

[0043] Step S21: atmospheric scattering defogging; the exterior of ancient buildings is often affected by dust and haze, which causes the overall image to be gray. Atmospheric scattering defogging is used to restore distant details and true colors; for the original image, the transmittance is calculated. , which represents the remaining proportion of light from the scene to the camera after passing through the atmosphere, is expressed as: ; where x is the image pixel position; is the experience weight; is the local window centered at x, and y is the pixel position index within the local window; is the global atmospheric illumination component; is the original image brightness; the recovered scene radiance is expressed as: ; Where c is the RGB color channel index; is the lower limit of transmittance; is the true color image after defogging; I(·) is the pixel value of the original image; A is the color value of the background atmosphere; the defogging image J is obtained;

[0044] Step S22: Circular channel compensation; used to remove the color cast caused by sunlight, shadows, and shooting angles, ensuring that the original color of stone and brickwork is accurately restored, and adaptively correcting the shadows of large-area walls and eaves; the RGB three-channel pixel values ​​of the dehazed image J are 、 and ,but ; ; ;in, and are the minimum and maximum values ​​of the red channel in the entire image range; and are the maximum and minimum values ​​of the green channel respectively; 、 and are the values ​​of the red, green, and blue channels after compensation; the convergence goal is to minimize the sum of squares of channel differences, expressed as: ; Where Loss is the color loss; output the color-compensated image;

[0045] Step S23: Local adaptive contrast enhancement; for the color-compensated image, amplify the brightness and darkness differences in the local block to highlight the peeling edges and cracks; including: local average brightness , ; Local brightness variance , ; Contrast magnification factor , ; Local brightness linear stretching, ; Where H and W are the height and width of the image respectively; i and j are image coordinate indices; B is the local block coordinate set; is the brightness value of the pixel in the local block; is the enhanced local brightness; is the maximum coefficient; is the global brightness standard deviation; the image with local contrast enhancement is obtained;

[0046] Step S24: iterative decomposition; in the evaluation of the appearance of ancient buildings, the crack width is small and the scale of the mottled is diverse, which requires multi-scale processing at the same time; the iterative decomposition method is used to obtain the intrinsic mode function { }, the convergence criterion is expressed as: ; The threshold is selected as 0.2; where SL is the convergence criterion of the image; D is the pixel area; and They are the pixel values ​​of the IMF components at the t+1th iteration and the tth iteration; after all convergence, the IMF components are obtained ;in, is the kth IMF component; K is the total number of components;

[0047] Step S25: Multi-weighted image pyramid fusion; fuse the IMF components according to the three weights of crack significance, texture frequency and illumination balance, retain the optimal details of each scale and smooth the transition, for each IMF component, ; Among them, for the first level of the pyramid, ; Finally, the enhanced image of the historical ancient building is reconstructed from the bottom up ;in, is the fusion result of the lth layer of the image pyramid; is the regularization constant; is the weighting coefficient of each level; is the image result of the pyramid level l; 、 and are the crack significance weight, texture frequency weight, and illumination balance weight under the kth IMF component; 、 and They are the crack significance weight, texture frequency weight and illumination balance weight under the mth IMF component respectively; the crack significance weight is obtained by extracting the gradient amplitude based on the Sobel operator, calculating the gradient and summing it over the entire image, and finally normalizing it; the texture frequency weight is obtained by using Laplacian high-pass filtering, averaging the absolute values, and finally normalizing it; the illumination balance weight is obtained by taking the inverse of the image variance and normalizing it.

[0048] By performing the above operations, we address the problems of general corrosion assessment methods for ancient buildings, which often ignore the effects of sunlight angle and shadows, misjudge the degree of material damage due to color cast, and make contrast adjustment difficult to adapt to differences in lighting or materials, resulting in poor accuracy in the final corrosion assessment. This solution uses an atmospheric scattering model to estimate transmittance and restore scene radiance, removing the overall graying caused by haze and dust, and improving distant detail and color fidelity. Circular channel compensation is used to adaptively correct color casts caused by sunlight, shadows, and shooting angles while minimizing the sum of squared channel differences, ensuring the accurate restoration of the original textures of stone and brickwork. Local adaptive contrast enhancement is used to adaptively amplify differences in brightness and darkness, thereby highlighting tiny cracks and peeling edges. At each pyramid level, the IMF components are weighted and fused using three weights: crack significance, texture frequency, and illumination balance. This method preserves optimal detail while avoiding visual abrupt changes between scales, thereby improving the accuracy of corrosion assessment.

[0049] Example 4, see Figure 1This embodiment is based on the above embodiment. In step S3, the establishment of the ancient building appearance corrosion degree assessment model is based on the historical ancient building image completed by deep learning and enhancement. The establishment of the ancient building appearance corrosion degree assessment model specifically includes the following steps:

[0050] Step S31: Overall model design; the model inputs the enhanced ancient building appearance image; after processing through the three-branch layer and the hybrid coding attention layer, it is processed through full connection + softmax to obtain the predicted distribution, and finally outputs the corrosion level predicted by the model; the cross entropy loss function is used, and the forward propagation is based on the gradient descent method;

[0051] Step S32: Three-branch layer design; construct color branch, texture branch and morphology branch respectively; the color branch captures the color distribution changes of mottled and faded areas, efficiently separates color information, and avoids repeated learning of the texture branch; the texture branch extracts the detailed texture of microcracks and pits, and enhances the detection of crack connectivity; the morphology branch captures the peeling edge and macro contour changes, and highlights the structural fracture and peeling areas; the color branch uses 3×64 2D convolution kernels to extract color spots and fading features; and 3×128 1D convolution kernels to enhance the color difference distribution; the texture branch uses 8 layers of densely connected 3×3 convolutions, and inserts a maximum pooling layer in the middle; the morphology branch has the same structure as the color branch, but the number of kernels is reduced by S; the three branches are spliced ​​and expressed as: ;in, It is the concatenated feature map tensor; It is a splicing operation; 、 and are the feature tensors extracted by the color branch, texture branch, and morphology branch respectively; X is the enhanced image of the ancient building appearance;

[0052] Step S33: Hybrid coding layer design; Perform three layers of convolution in sequence, perform SE processing on the output of each convolution layer, use the SE processing output as the input of the next convolution layer, and dynamically adjust the weights of each channel at different illumination and corrosion stages; including: convolution processing, expressed as: ;in, is the lth layer of convolution; is the result of the l-th layer convolution; is the input of the l-th layer convolution; the perturbation activation function is processed and expressed as: ;in, is the perturbation activation function; is the output of the perturbation activation function; Perform global average pooling to obtain ; sigmoid processing, expressed as: ;in, is the sigmoid function; and is a learnable weight matrix; the reweighted processing is expressed as: ;in, is the reweighted output of the lth layer, which serves as the convolution input of the l+1th layer;

[0053] Step S34: Design of perturbation activation function; Historical building images often contain shooting noise and compression artifacts. By designing the activation function, we can reduce channel death and enhance the robustness to shooting noise; introduce dynamic noise intensity , based on the local variance of the input features, adaptive adjustment is performed: the noise is increased in flat areas to improve the robustness of artifact removal, and the noise is reduced in high-texture areas to protect details; the perturbation activation function is expressed as: ; ; Where u is the perturbation activation function variable; , a set of noise is sampled and injected into the activation each time forward propagation; It is a noise bias, which is used to enhance the input image’s tolerance to shooting noise and artifacts; is the initial noise intensity; is the attenuation coefficient; is the standard deviation of the input pixel values ​​in the local window.

[0054] By performing the above operations, we can address the problems of general ancient building exterior corrosion degree assessment methods that are prone to feature extraction conflicts, insufficient detail mining, and insufficient processing capabilities for shooting noise and compression artifacts, which in turn lead to poor exterior corrosion degree assessment results. This solution designs multi-branch exclusive feature extraction, focusing on color spot fading, microcrack pitting, and macro-peeling contours respectively, to avoid redundant interference between different feature channels and improve the richness and discriminability of overall feature expression; embed SE-style coded attention in the three-layer convolution to achieve dynamic weighting of each channel feature under different corrosion stages and lighting conditions; by designing a perturbation activation function, the noise intensity is adaptively adjusted based on the local variance, alleviating channel death and improving the fault tolerance to shooting noise and compression artifacts, thereby improving the exterior corrosion degree assessment effect.

[0055] Example 5, see Figure 1 This embodiment is based on the above embodiment. In step S4, the evaluation of the degree of corrosion of the appearance of ancient buildings is based on the established model for evaluating the degree of corrosion of the appearance of ancient buildings. The image of the ancient building is collected in real time, and after image enhancement processing, it is input into the model for evaluating the degree of corrosion of the appearance of ancient buildings. The corrosion level output by the model is used as the result of the evaluation of the degree of corrosion of the appearance of ancient buildings.

[0056] Example 8, see Figure 2This embodiment is based on the above embodiment. The artificial intelligence-based ancient building appearance corrosion degree assessment system provided by the present invention includes an image acquisition module, an image enhancement module, an ancient building appearance corrosion degree assessment model establishment module and an ancient building appearance corrosion degree assessment module;

[0057] The image acquisition module acquires the appearance image of the historical building and marks the corrosion level;

[0058] The image enhancement module enhances the appearance image of historical buildings through atmospheric scattering defogging, cyclic channel compensation, local adaptive contrast enhancement, multi-scale intrinsic mode decomposition and multi-weighted pyramid fusion;

[0059] The module for establishing a model for assessing the degree of corrosion of the appearance of ancient buildings takes the enhanced image as input, extracts the color, texture, and morphology through three branches, and integrates them into the hybrid coding layer of SE attention, and establishes a model for assessing the degree of corrosion of the appearance of ancient buildings based on the perturbation activation function;

[0060] The ancient building appearance corrosion degree assessment module is based on the ancient building appearance corrosion degree assessment model, and implements appearance corrosion degree assessment on the ancient building appearance images collected in real time.

[0061] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0062] While the embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that various changes, modifications, substitutions, and alterations can be made to the embodiments without departing from the principles and spirit of the invention.

[0063] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. An artificial intelligence-based method for assessing the degree of corrosion of ancient building exteriors, characterized by: The method comprises the following steps: Step S1: Image acquisition: Acquire an image of the exterior of a historical building and mark the corrosion level; Step S2: Image enhancement: The image of the exterior of the historical building is enhanced through atmospheric scattering defogging, cyclic channel compensation, local adaptive contrast enhancement, multi-scale intrinsic mode decomposition, and multi-weighted pyramid fusion. Step S3: Establishing a model for evaluating the degree of corrosion of the exterior of ancient buildings; taking the enhanced image as input, extracting the color, texture, and morphology branches and integrating them into the hybrid coding layer of SE attention, and establishing a model for evaluating the degree of corrosion of the exterior of ancient buildings based on the perturbation activation function; Step S4: evaluating the degree of corrosion of the ancient building's exterior; based on the ancient building's exterior corrosion degree evaluation model, evaluating the degree of corrosion of the ancient building's exterior image collected in real time; In step S3, the establishment of an ancient building exterior corrosion degree assessment model is based on deep learning and enhanced historical ancient building images, and the establishment of an ancient building exterior corrosion degree assessment model specifically includes the following steps: Step S31: Overall model design; the model inputs the enhanced ancient building appearance image; after processing through the three-branch layer and the hybrid coding attention layer, it is processed through full connection + softmax to obtain the predicted distribution, and finally outputs the corrosion level predicted by the model; the cross entropy loss function is used, and the forward propagation is based on the gradient descent method; Step S32: Three-branch layer design; construct color branch, texture branch and morphology branch respectively; the color branch uses 3×64 2D convolution kernels to extract color spots and fading features; and 3×128 1D convolution kernels to enhance the color difference distribution; the texture branch uses 8 layers of densely connected 3×3 convolutions and inserts a maximum pooling layer in the middle; the morphology branch has the same structure as the color branch, but the number of kernels is reduced by S; the three branches are spliced ​​together to obtain ; Step S33: Hybrid coding layer design; Perform three layers of convolution in sequence, perform SE processing on the output of each convolution layer, and use the SE processing output as the input of the next convolution layer; including: convolution processing; perturbation activation function processing; global average pooling; sigmoid processing; reweighting processing; Step S34: Design of perturbation activation function; introduction of dynamic noise intensity , the perturbation activation function is expressed as: ; ; Where u is the perturbation activation function variable; ; is the noise bias; is the initial noise intensity; is the attenuation coefficient; is the standard deviation of the input pixel values ​​in the local window.

2. The method for evaluating the degree of corrosion of the exterior of ancient buildings based on artificial intelligence according to claim 1 is characterized in that: Furthermore, in step S2, the image enhancement specifically includes the following steps: Step S21: atmospheric scattering defogging; for the original image, calculate the transmittance , expressed as: ; where x is the image pixel position; is the experience weight; is the local window centered at x, and y is the pixel position index within the local window; is the global atmospheric illumination component; is the original image brightness; the recovered scene radiance is expressed as: ; Where c is the RGB color channel index; is the lower limit of transmittance; is the true color image after defogging; I(·) is the pixel value of the original image; A is the color value of the background atmosphere; the defogging image J is obtained; Step S22: Circular channel compensation; the pixel values ​​of the RGB channels of the defogging image J are recorded as follows: 、 and ,but ; ; ;in, and are the minimum and maximum values ​​of the red channel in the entire image range; and are the maximum and minimum values ​​of the green channel respectively; 、 and are the values ​​of the red, green, and blue channels after compensation; the convergence goal is to minimize the sum of squares of channel differences, expressed as: ; Where Loss is the color loss; output the color-compensated image; Step S23: Local adaptive contrast enhancement; for image processing after color compensation, including: local average brightness , ; Local brightness variance , ; Contrast magnification factor , ; Local brightness linear stretching, ; Where H and W are the height and width of the image respectively; i and j are image coordinate indices; B is the local block coordinate set; is the brightness value of the pixel in the local block; is the enhanced local brightness; is the maximum coefficient; is the global brightness standard deviation; the image with local contrast enhancement is obtained; Step S24: iterative decomposition; Step S25: Multi-weight image pyramid fusion.

3. The method for evaluating the degree of corrosion of the exterior of ancient buildings based on artificial intelligence according to claim 2 is characterized in that: In step S2, the iterative decomposition is to obtain the intrinsic mode function { }, the convergence criterion is expressed as: ; Where SL is the convergence criterion; D is the pixel area; and They are the pixel values ​​of the IMF components at the t+1th iteration and the tth iteration; after all convergence, the IMF components are obtained ;in, is the kth IMF component; K is the total number of components.

4. The method for evaluating the degree of corrosion of the exterior of ancient buildings based on artificial intelligence according to claim 3 is characterized by: In step S2, the multi-weighted image pyramid fusion is to fuse the IMF components according to the three weights of crack significance, texture frequency and illumination balance. For each IMF component, ; Among them, for the first level of the pyramid, ; Finally, the enhanced image of the historical ancient building is reconstructed from the bottom up ;in, is the fusion result of the lth layer of the image pyramid; is the regularization constant; is the weighting coefficient of each level; is the image result of the pyramid level l; 、 and are the crack significance weight, texture frequency weight, and illumination balance weight under the kth IMF component; 、 and are the crack significance weight, texture frequency weight and illumination balance weight under the mth IMF component respectively.

5. The method for evaluating the degree of corrosion of the exterior of ancient buildings based on artificial intelligence according to claim 4 is characterized in that: In step S4, the assessment of the degree of corrosion of the appearance of ancient buildings is based on the established model for assessing the degree of corrosion of the appearance of ancient buildings. The images of ancient buildings are collected in real time, and after image enhancement processing, they are input into the model for assessing the degree of corrosion of the appearance of ancient buildings. The corrosion level output by the model is used as the result of the assessment of the degree of corrosion of the appearance of ancient buildings.

6. The method for evaluating the degree of corrosion of the exterior of ancient buildings based on artificial intelligence according to claim 5 is characterized by: In step S1 , the corrosion level is used as an image label; the image label includes no corrosion, mild corrosion, moderate corrosion and severe corrosion.

7. An artificial intelligence-based ancient building exterior corrosion degree assessment system, used to implement the artificial intelligence-based ancient building exterior corrosion degree assessment method according to any one of claims 1 to 6, characterized in that: It includes an image acquisition module, an image enhancement module, an ancient building exterior corrosion degree assessment model establishment module, and an ancient building exterior corrosion degree assessment module; The image acquisition module acquires the appearance image of the historical building and marks the corrosion level; The image enhancement module enhances the appearance image of historical buildings through atmospheric scattering defogging, cyclic channel compensation, local adaptive contrast enhancement, multi-scale intrinsic mode decomposition and multi-weighted pyramid fusion; The module for establishing a model for assessing the degree of corrosion of the appearance of ancient buildings takes the enhanced image as input, extracts the color, texture, and morphology through three branches, and integrates them into the hybrid coding layer of SE attention, and establishes a model for assessing the degree of corrosion of the appearance of ancient buildings based on the perturbation activation function; The ancient building appearance corrosion degree assessment module is based on the ancient building appearance corrosion degree assessment model, and implements appearance corrosion degree assessment on the ancient building appearance images collected in real time.

Images