Three-dimensional image security processing system and method based on diffusion model

Abstract

The invention relates to the field of image data processing, in particular to a three-dimensional image safety processing system and method based on a diffusion model. The system comprises an image format conversion module, an image feature training module, an image three-dimensional structure analysis module, an image light effect simulation module, an image rendering module and an artificial intelligence security and image anti-counterfeiting module. Through the convolutional neural network, deep semantic analysis of image content can be realized by processing a large amount of three-dimensional image data, so that the model not only identifies key structure elements in an image, but also can predict positions and interrelationships of the elements in a three-dimensional space, thereby remarkably improving identification accuracy and application efficiency of the model, and improving the application range of the model. A light tracing technology and a neural renderer of a neural network are combined, and the sense of reality of the rendering process is enhanced by simulating the physical propagation process of light. And meanwhile, the accuracy, authenticity and safety of image processing are effectively integrated, and the reliability and the expression effect in image generation and use are comprehensively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and in particular to a three-dimensional image security processing system and method based on a diffusion model. Background Art

[0002] 3D image generation systems involve the use of 3D image processing technology to create highly realistic 3D visual effects. They are often used in areas such as film production, video games, virtual reality, and simulation training. In addition, the purpose of 3D image generation is to provide a visual experience that is almost indistinguishable from reality, allowing users to interact more naturally and deeply in the virtual environment.

[0003] Traditional 3D image processing image-based rendering or neural rendering methods have difficulty in achieving satisfactory rendering effects in reconstructing fine geometry in larger views. Secondly, due to the limited information provided by a single perspective, inferring 3D information from a single image is a complex and error-prone process. In addition, many existing 3D generation networks mainly model objects of a specific category and perform poorly on generalization to general or uncommon 3D objects. In addition, even with advanced techniques such as neural radiance fields (NeRF), there are still deviations in the fidelity of texture and color, making the generated 3D models look too smooth or inaccurate in color saturation. These defects emphasize the need for more innovative methods to improve the efficiency and quality of conversion from a single image to high-fidelity 3D content. In terms of security and anti-counterfeiting, it is easy to cause the risk of information leakage or tampering in the generated images, or the presence of harmful information, and the security and credibility of the image content cannot be guaranteed. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a three-dimensional image security processing system and method based on a diffusion model.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is as follows: A first aspect provides a three-dimensional image security processing system based on a diffusion model, the system comprising: An image format conversion module is used to perform coordinate alignment, resolution calibration and format conversion on the image data obtained by three-dimensional scanning to generate standardized three-dimensional data; An image feature training module is used to construct a deep learning model based on the diffusion model and the normalized three-dimensional data, and to train the image data to learn the features and structural information of the image, and obtain a trained deep model; An image three-dimensional structure analysis module is used to analyze the shape, size and relative position of three-dimensional objects in the image using the trained depth model, reconstruct the three-dimensional space, and obtain three-dimensional space image data; An image light effect simulation module is used to simulate the influence of the light source on the image based on the three-dimensional space image data, calculate the light scattering, reflection and shadow effects after the light interacts with the three-dimensional object, and obtain light simulation data; An image rendering module, used to adjust rendering parameters based on the light simulation data, perform image rendering processing, and output an optimized rendered image; The artificial intelligence security and image anti-counterfeiting module is used to use artificial intelligence to perform security analysis and content review on the optimized rendered image, and generate anti-counterfeiting image data by embedding digital watermarks and performing image content monitoring.

[0006] In one implementation, the image format conversion module includes: A coordinate alignment submodule is used to perform consistency check and alignment of spatial coordinates based on the image data obtained by three-dimensional scanning, and generate aligned image data; A resolution calibration submodule, used to detect and adjust the image pixel resolution of the aligned image data to generate calibrated image data; The data formatting submodule performs a data format conversion operation based on the calibrated image data, converts the image data into a unified point cloud format, and generates standardized three-dimensional data.

[0007] In one embodiment, the image feature training module includes: A model building submodule, for selecting a network layer structure and an activation function, defining input and output nodes and the number of network layers, and generating an initialized deep learning model framework based on the normalized three-dimensional data; A feature learning submodule is used to use the initialized deep learning model framework, input three-dimensional image data for feature extraction, analyze the structural information and texture features of the image, optimize model parameters, and generate a feature optimization model; The model training submodule performs multiple rounds of training based on the feature optimization model, adjusts the learning rate and the loss function, uses a batch training method to optimize the generalization ability of the model, and generates a trained deep model. In each round of training, a diffusion model is used to calculate the loss based on the difference between the current noise prediction and the actual noise, so as to guide the model to learn how to process noise terms at different time steps, wherein the noise term is used to reflect the noise accumulation effect based on the time step.

[0008] In one embodiment, the image three-dimensional structure analysis module includes: An image shape analysis submodule is used to draw the outline of a three-dimensional object based on the trained depth model, perform shape analysis by aligning the drawn outline with the image edge, record shape features, and generate image shape data; An image size analysis submodule is used to measure the size of the three-dimensional object based on the image shape data, by measuring the height, width and depth of each three-dimensional object, comparing it with the standard size reference in the three-dimensional model, and summarizing the size data to obtain image size data; The image position analysis submodule is used to locate the position of the three-dimensional object in space through the image size data, record the coordinate information of each three-dimensional object according to the spatial relationship and depth information between the three-dimensional objects, and update the data in the three-dimensional model to obtain three-dimensional space image data.

[0009] In one embodiment, the image light effect simulation module includes: A light source setting submodule is used to configure the light source based on the three-dimensional space image data through the BRDF model to simulate the effect of natural light or indoor lighting, adjust the light source position and intensity, and set a variety of lighting angles and color temperatures to generate light source configuration data. The BRDF model is a bidirectional reflectance distribution function model; A light interaction submodule, for analyzing the interaction between light and the surface of a three-dimensional object, calculating reflection and scattering effects, adjusting the interaction behavior of light according to the material and surface characteristics of the object, simulating the performance of light in various environments, and generating light environment interaction data using the light source configuration data; The shadow generation submodule is used to locate the shadow area caused by the obstruction between the light source and the object through the light environment interaction data, calculate the intensity and extension of the shadow, adjust the blurriness and boundary of the shadow, optimize the light-dark contrast in the image, and obtain light simulation data.

[0010] In one embodiment, the BRDF model is according to the formula:

[0011] Calculate the relationship between incident light and reflected light from various angles to simulate the effects of natural light or indoor lighting; Among them, the function Represents the bidirectional reflectance distribution function, which is used to calculate the incident light direction of a given material surface. and reflected light direction The light reflectivity under is the incident light direction, is the direction of reflected light, is the incident angle, is the observation angle, For material properties, is the diffuse reflection coefficient, which indicates the proportion of the diffuse reflection part of the material surface. is the cosine of the angle, which is used to calculate the effective component of the light incident on the surface. is the specular reflection coefficient, which is used to indicate the proportion of the specular reflection part of the material surface. is the specular highlight attenuation factor, according to the observation angle Adjust the intensity of specular reflection so that the highlight effect weakens as the viewing angle increases. is the light radiation intensity in the unit reflection direction, representing the reflection direction The brightness per unit area, is the light intensity per unit incident direction, representing the light intensity in the incident direction The amount of light energy received per unit area.

[0012] In one embodiment, the image rendering module includes: A parameter adjustment submodule, for analyzing the light intensity and color distribution in the image based on the light simulation data, adjusting the parameters one by one to match the predetermined visual effect, and adjusting the color saturation and brightness level to generate image parameter optimization data; A detail enhancement submodule, for using the image parameter optimization data to perform hierarchical separation, contrast and sharpness adjustment on the details in the image, and optimize key visual elements in the image, including texture and edges, to obtain a detail-optimized image; The image output submodule is used to perform color correction and adjust the resolution of the detail-optimized image, and output an optimized rendered image.

[0013] Based on the same inventive concept, the second aspect of the present invention provides a three-dimensional image security processing method based on a diffusion model, which is implemented based on the three-dimensional image security processing system based on the diffusion model in the first aspect, and the method includes the following steps: S1: coordinate alignment, resolution calibration and format conversion are performed on the image data obtained by three-dimensional scanning to generate standardized three-dimensional data; S2: constructing a deep learning model based on the diffusion model and the normalized three-dimensional data, and training the image data to learn the features and structural information of the image, and obtaining a trained deep model; S3: using the trained depth model, analyzing the shape, size and relative position of the three-dimensional object in the image, reconstructing the three-dimensional space, and obtaining three-dimensional space image data; S4: Based on the three-dimensional space image data, simulate the influence of the light source on the image, calculate the light scattering, reflection and shadow effects after the light interacts with the three-dimensional object, and obtain light simulation data; S5: Based on the light simulation data, adjusting rendering parameters, performing image rendering processing, and outputting an optimized rendered image; S6: Using artificial intelligence to perform security analysis and content review on the optimized rendered image, and generating anti-counterfeiting image data by embedding a digital watermark and performing image content monitoring.

[0014] Based on the same inventive concept, the third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the three-dimensional image security processing method based on the diffusion model described in the second aspect.

[0015] Based on the same inventive concept, the fourth aspect of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the three-dimensional image security processing method based on the diffusion model described in the second aspect is implemented.

[0016] Compared with the prior art, the advantages and beneficial technical effects of the present invention are as follows: The system or method proposed in the present invention constructs a deep model based on a convolutional neural network. The convolutional neural network can realize deep semantic analysis of image content by processing a large amount of three-dimensional image data, so that the model can not only identify key structural elements in the image, but also predict the position and relationship of these elements in three-dimensional space, thereby significantly improving the recognition accuracy and application efficiency of the model. The neural renderer combining ray tracing technology and neural network enhances the realism of the rendering process by simulating the physical propagation process of light, and creates a single image of high-fidelity three-dimensional content from a single image. Using diffusion prior as 3D perception supervision, the generated 3D model shows faithful geometry and real texture with diffusion clipping loss and texture point cloud enhancement. Make 3D suitable for general objects and give multi-purpose applications. Provide users with a brand new 3D creation experience. Through artificial intelligence technology, digital watermarks are embedded and image monitoring is performed to ensure the security and traceability of image content. It effectively integrates the accuracy, authenticity and security of image processing, and comprehensively improves the reliability and performance of image generation and use. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A module diagram of a three-dimensional image security processing system based on a diffusion model proposed in an embodiment of the present invention; Figure 2A framework diagram of a three-dimensional image security processing system based on a diffusion model proposed in an embodiment of the present invention; Figure 3 A schematic diagram of the processing flow of an image format conversion module in a three-dimensional image security processing system based on a diffusion model proposed in an embodiment of the present invention; Figure 4 A schematic diagram of the processing flow of an image feature training module in a three-dimensional image security processing system based on a diffusion model proposed in an embodiment of the present invention; Figure 5 A schematic diagram of the processing flow of an image three-dimensional structure analysis module in a three-dimensional image security processing system based on a diffusion model proposed in an embodiment of the present invention; Figure 6 A schematic diagram of the processing flow of an image light effect simulation module in a three-dimensional image security processing system based on a diffusion model proposed in an embodiment of the present invention; Figure 7 A schematic diagram of the processing flow of an image rendering module in a three-dimensional image security processing system based on a diffusion model proposed in an embodiment of the present invention; Figure 8 Schematic diagram of the processing flow of the artificial intelligence security and image anti-counterfeiting module in the three-dimensional image security processing system based on the diffusion model proposed in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The present invention proposes a three-dimensional image security processing system based on a diffusion model, including: an image format conversion module that performs coordinate alignment and resolution calibration on the image data obtained by three-dimensional scanning, converts the image into VOX voxel or point cloud format, and generates standardized three-dimensional data. The present invention can realize deep semantic analysis of image content by processing a large amount of three-dimensional image data through a convolutional neural network, so that the model can not only identify the key structural elements in the image, but also predict the position and relationship of these elements in three-dimensional space, thereby significantly improving the recognition accuracy and application efficiency of the model, and combining ray tracing technology and neural network neural renderer, by simulating the physical propagation process of light, enhances the realism of the rendering process. At the same time, it effectively integrates the accuracy, authenticity and security of image processing, and comprehensively improves the reliability and performance of image generation and use.

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Embodiment 1 See also Figure 1 The present invention provides a three-dimensional image security processing system based on a diffusion model, the system comprising: An image format conversion module is used to perform coordinate alignment, resolution calibration and format conversion on the image data obtained by three-dimensional scanning to generate standardized three-dimensional data; An image feature training module is used to construct a deep learning model based on the diffusion model and the normalized three-dimensional data, and to train the image data to learn the features and structural information of the image, and obtain a trained deep model; An image three-dimensional structure analysis module is used to analyze the shape, size and relative position of three-dimensional objects in the image using the trained depth model, reconstruct the three-dimensional space, and obtain three-dimensional space image data; An image light effect simulation module is used to simulate the influence of the light source on the image based on the three-dimensional space image data, calculate the light scattering, reflection and shadow effects after the light interacts with the three-dimensional object, and obtain light simulation data; An image rendering module, used to adjust rendering parameters based on the light simulation data, perform image rendering processing, and output an optimized rendered image; The artificial intelligence security and image anti-counterfeiting module is used to perform security analysis and content review through artificial intelligence. It embeds digital watermarks and performs image content monitoring to ensure that the image does not contain harmful information and the source is traceable, thereby generating anti-counterfeiting image data.

[0022] Specifically, the normalized three-dimensional data includes the spatial resolution of voxel data, the grayscale level of voxels, and the point cloud density parameters; the trained deep model includes the model's weight distribution, activation function type, and number of training cycles; the three-dimensional spatial image data includes the reconstructed three-dimensional grid density, the grid's topological structure, and the spatial coordinates of key feature points; the light simulation data includes the light source position parameters, light intensity level, and the angular distribution of reflected light and scattered light; the optimized rendered image includes the image's resolution, dynamic range, and color correction parameters.

[0023] See also Figure 2 and Figure 3 ,The processing flow of the image format conversion module includes: The coordinate alignment submodule performs consistency check and alignment of spatial coordinates based on the image data obtained by 3D scanning. The specific process of generating aligned image data is as follows: Based on the image data obtained by 3D scanning, the coordinate alignment submodule first analyzes the position information of multiple key points in the 3D space, extracts key feature points, and compares them with the reference points in the standard model library. The position of each point is optimized by the least squares method to ensure the accurate alignment of each point in the 3D space. During the alignment process, the deviation between each data point and the reference point is calculated, and the points that exceed the error range are corrected or eliminated to ensure that the generated aligned image data has high accuracy and consistency. In addition, the alignment submodule maintains an error log to record the accuracy and correction measures of each alignment operation for subsequent analysis and optimization to generate aligned image data.

[0024] The resolution calibration submodule uses the aligned image data to detect and adjust the image pixel resolution. In the specific implementation process, the specific process of generating the calibrated image data is as follows: The resolution calibration submodule uses the aligned image data to analyze each pixel of the image and calculate the color and grayscale differences between it and the surrounding pixels to determine whether its resolution needs to be adjusted. Using a set of predefined standard resolution profiles, the actual resolution of the image pixels is compared with these standards and the closest configuration is selected for adjustment. The image resolution is adjusted through interpolation or downsampling techniques to meet the needs of subsequent processing. The calibration process is repeated until the average resolution error of the image is minimized. After the calibration is completed, the resolution comparison data before and after the adjustment of each pixel is output to generate the calibrated image data.

[0025] The data formatting submodule performs data format conversion based on the calibrated image data, converting the image data into a unified point cloud format. In the specific implementation process, the specific process of generating standardized 3D data is as follows: Based on the calibrated image data, the data formatting submodule first evaluates the structural density and spatial distribution of the data, and selects VOX voxel or point cloud format according to the needs. In the case of selecting VOX voxel, the module will calculate the spatial position and corresponding grayscale of each voxel, and map the two-dimensional pixels to three-dimensional voxels through volume rendering technology. For point cloud format, the module converts the position and color information of each pixel into a point in the point cloud, retaining its three-dimensional coordinates and color attributes. During the conversion process, the data will be compressed to optimize storage and processing speed, and finally generate standardized three-dimensional data, which is suitable for subsequent deep processing and analysis.

[0026] See also Figure 2 and Figure 4 ,The processing flow of the image feature training module includes: The model building submodule selects the network layer structure and activation function based on the normalized three-dimensional data, defines the input and output nodes and the number of network layers, and generates the initialization deep learning model framework. The specific process is as follows: The model building submodule generates an initialized deep learning model framework based on the normalized three-dimensional data. This process includes data feature analysis, network structure selection, and network parameter initialization. First, by statistically analyzing the feature density and distribution uniformity in the three-dimensional data, the appropriate network layer structure is selected, such as deciding how many convolutional layers to use and the number of filters in each layer. For the activation function, ReLU is selected according to the data characteristics to increase the ability to calculate nonlinear processing, or Sigmoid to achieve the probability distribution of the output layer. The input and output nodes are determined by the number of voxels in the three-dimensional data and the number of expected classification categories. The depth and width of the network are set to achieve the expected recognition accuracy and processing speed, and the initialized deep learning model framework is generated. The diffusion model calculates the loss based on the difference between the current noise prediction and the actual noise at each iteration, thereby guiding the model to learn how to better handle the noise term at different time steps. As the training progresses, the NeRF parameters are updated, during which the 3D object gradually displays its texture and geometry.

[0027] The feature learning submodule uses the initialization deep learning model framework, inputs the 3D image data for feature extraction, analyzes the image's structural information and texture features, optimizes the model parameters, and generates the feature optimization model in the following specific process: The feature learning submodule uses the initialization deep learning model framework, inputs the processed three-dimensional image data, and extracts features layer by layer through the network layer. First, the convolution layer extracts the local features of the image through filters, and then the pooling layer reduces the spatial dimension of the features to reduce the amount of calculation. At this stage, each convolution layer may be followed by a normalization layer to stabilize the learning process. By iteratively adjusting the network parameters, the model parameters including weights and biases are optimized. In this process, the loss function is used to evaluate the difference between the predicted results and the actual labels. The mean square error (MSE) is usually used as the loss function. The loss function expression is: ,in is the sample size, is the actual value, is the predicted value. The loss is optimized by gradient descent method to improve the model's fitting accuracy to the data and generate a feature optimization model.

[0028] The model training submodule performs multiple rounds of training based on the feature optimization model, adjusts the learning rate and loss function, and uses batch training methods to optimize the generalization ability of the model. The specific process of generating a trained deep model is as follows: The model training submodule sets up multiple rounds of training based on the feature optimization model to meet the learning needs at different stages. The learning rate and loss function are adjusted to improve the model training process and improve the generalization ability. For example, a higher learning rate may be used for rapid convergence in the early stage, and the learning rate may be gradually reduced as the training progresses to refine the learning effect. The batch training method is used, that is, a batch of data is used in each iteration. This can reduce memory usage and stabilize the gradient estimation by using the statistical characteristics of batch data. In addition, to prevent overfitting, the Dropout technology may be introduced to randomly shut down some neurons in the network in each training iteration. Through repeated training, the network structure and parameters are adjusted, and finally a trained deep model is generated, which can effectively process and interpret the complex structure of three-dimensional image data.

[0029] Before formal training of the explanation, some preparation is needed. A text-to-image diffusion model can be used to guide the optimization of the 3D representation. For a given viewpoint , where G is a differentiable rendering function represented by a parameterized 3D representation. Specifically, the diffusion model renders the image x by introducing a random amount of noise , at different time steps t, that is ,in, represents the noise term introduced during the rendering process, Represents a rendered image, which is usually used to describe an image representation with noise perturbations, reflecting the noise accumulation effect based on the time step t. is the attenuation coefficient or noise scaling factor in the diffusion model, which is used to control the change of noise with time step t during the image generation process. Usually, as the time step t increases, the attenuation coefficient Gradually decrease to balance the effect of noise and original image data. is the standard deviation of the noise in the diffusion model, which is used to describe the noise intensity added in the t-th step. It determines the noise term The size of the current step. As t increases, the standard deviation Usually gets bigger. is the introduced noise term, which is used to simulate the random perturbations added during the rendering process. By adding different noises, diversified image samples can be generated. ; and A noise table is defined with a logarithmic signal-to-noise ratio , which decreases linearly with time step t. In order to optimize the 3D representation parameters so that the image is as close as possible to the better generated samples, a fractional distillation loss L pushes the rendered image to a higher density region based on the text embedding. Specifically, L is the symbol of the loss function, which is used to measure the difference between the noise predicted by the model and the actual noise added in this image generation model or diffusion model. L calculates the difference between the predicted noise and the added noise by pixel gradient and is used to update the scene parameters. It can be shown that this loss is essentially a measure of the similarity between the image and the text prompt. The loss function L in the diffusion model serves as the main objective function during the training process. The diffusion model calculates the loss based on the difference between the current noise prediction and the actual noise at each iteration, thereby guiding the model to learn how to better handle the noise term at different time steps. As the training progresses, the NeRF parameters are updated, during which the 3D objects gradually reveal their textures and geometry.

[0030] See also Figure 2 and Figure 5 ,The processing flow of the image three-dimensional structure analysis module includes: The image shape analysis submodule draws the contour of the three-dimensional object based on the trained deep model, performs shape analysis by aligning the drawn contour line with the image edge, records the shape features, and generates image shape data. In the specific implementation process, the specific process of generating image shape data is as follows; The image shape parsing submodule is based on the trained deep model. It first performs high-precision edge detection on the 3D scanned image and uses the edge detection algorithm to accurately draw the outline of each 3D object. The algorithm calculates the gradient strength and direction of each pixel based on the gradient information of the image and extracts clear edge lines. Subsequently, the module compares these contour lines with the actual edges of the image, uses the Euclidean distance to quantify the deviation between the contour lines and the actual edges, and adjusts the model parameters to minimize this deviation. During this process, shape features such as angles and curvatures are recorded in detail for further shape analysis. This ensures a high degree of consistency between the drawn contour lines and the actual 3D objects, and generates image shape data.

[0031] The image size analysis submodule uses image shape data to measure the size of three-dimensional objects. By measuring the height, width and depth of each object, comparing it with the standard size reference in the three-dimensional model, and summarizing the size data, the image size data is obtained. The specific process is as follows: The Image Size Analysis submodule uses the image shape data to perform detailed measurements of the dimensions of each 3D object. The measurement reference points for each object are first defined, and then the distances between these reference points are calculated to determine the height, width, and depth of the object. The measurement uses the straight-line distance formula in 3D space: Distance = , where (x1, y1, z1) and (x2, y2, z2) are the 3D coordinates of the reference points. Compare these measured dimensions with the standard dimension references in the 3D model to verify the accuracy of the dimensions and make adjustments. All dimension data are counted and summarized to form complete image dimension data, providing a basis for subsequent model verification and adjustment.

[0032] The image position analysis submodule locates the position of the three-dimensional object in space through the image size data, records the coordinate information of each object according to the spatial relationship and depth information between the objects, and updates the data in the three-dimensional model. The specific process of obtaining the three-dimensional space image data is as follows: The image position analysis submodule accurately locates the position of three-dimensional objects in space through image size data. Vector calculation is used to describe the spatial position of each object and calculate its vector coordinates relative to the model origin. The specific calculation method involves the vector representation from the reference point of each three-dimensional object to the model origin. The vector is represented as v=(x, y, z), where x, y and z represent the coordinates of the point in three-dimensional space respectively. This vector expression provides accurate position information of each object in three-dimensional space, allowing the module to accurately locate each object and analyze the relative relationship between them. The position information of each object and its spatial relationship with other objects are recorded in detail, and these data are updated in the three-dimensional model using a specific update algorithm. These position data provide important spatial geometric information for further analysis and optimization of the model, and ultimately form detailed three-dimensional spatial image data.

[0033] See also Figure 2 and Figure 6 ,The processing flow of the image light effect simulation module includes: The light source setting submodule is based on the three-dimensional space image data. It configures the light source through the BRDF model to simulate the effect of natural light or indoor lighting, adjusts the light source position and intensity, and sets a variety of lighting angles and color temperatures to generate light source configuration data. The specific process is as follows; The light source setting submodule is based on 3D spatial image data. It first uses the BRDF (bidirectional reflectance distribution function) model to simulate different types of light source effects, such as natural light and indoor lighting. According to the needs of the model, the position of the light source is adjusted to match the actual environment of the scene, such as the position of the skylight or the position of the indoor lamps. In addition, the intensity of the light source is adjusted to match the changes in natural light at different times or the needs of indoor lighting. The light effect is further refined by changing the lighting angle and color temperature, such as the sunset color temperature of sunlight or the warm tones of indoor lights. The precise setting of these parameters ensures that the simulated light can realistically reflect the lighting conditions in the real world and generate detailed light source configuration data.

[0034] The BRDF model follows the formula:

[0035] Calculate the relationship between incident and reflected light from various angles, simulating the effects of natural light or indoor lighting; Among them, the function Represents the bidirectional reflectance distribution function, which is used to calculate the reflection distribution function of a given material surface in a specific incident light direction. and reflected light direction The light reflectivity under . is the incident light direction, which indicates the direction from which the light enters the surface of an object. It is the direction of reflected light, which indicates the direction in which the light is reflected from the surface of an object. is the angle of incidence, which is the angle between the incident light and the surface normal of the object. is the observation angle, which is the angle between the reflected light seen by the observer and the normal of the object's surface. Material properties that represent the physical and optical characteristics of the surface hit by light. It is the diffuse reflection coefficient, which is used to indicate the proportion of the diffuse reflection part of the material surface. is the cosine of the angle and is used to calculate the effective component of the light incident on the surface. It is the specular reflection coefficient, which is used to indicate the proportion of the specular reflection part of the material surface. is the specular highlight attenuation factor, according to the observation angle Adjusts the intensity of specular reflections so that the highlight effect decreases as the viewing angle increases. is the light radiation intensity in the unit reflection direction, that is, The brightness per unit area. is the light intensity per unit incident direction, that is, The amount of light energy received per unit area.

[0036] During the specific implementation process, the execution process is as follows: Determine the angle parameters and : (Angle of incidence) is determined by the angle between the incident light and the normal line of the object surface. In actual calculations, It can be obtained by taking the arc cosine of the dot product of the light source direction vector and the surface normal vector: in, is the unit normal vector, is the unit incident ray vector.

[0037] (Observation angle) is determined by the angle between the direction of the reflected light and the direction of the observer. , It can be obtained by taking the arc cosine of the dot product of the reflected light direction vector and the viewing direction vector: in, is the reflection vector, is the observation vector.

[0038] Diffuse reflectance and the specular reflection coefficient Determination of: These coefficients are usually preset according to the physical and optical properties of the material, which can be obtained from the material database or determined by experimental measurement. They reflect the ability of different materials to absorb and reflect light.

[0039] Light radiation intensity and light intensity Calculation: In a given reflection direction The intensity of light radiation reflected from a unit area of ​​a material surface. This is usually simulated in rendering software using a ray tracing algorithm. At a given incident direction In general, the light intensity received per unit area on the surface of an object is also measured through ray tracing or actual lighting environment.

[0040] Formula integration calculation: The final BRDF value is calculated by plugging all these calculated parameter values ​​into the formula to take into account the effects of material properties, light angle and light intensity on reflective properties.

[0041] The calculated results Used to accurately simulate and render the visual appearance of objects in 3D scenes under different lighting conditions. This includes how light is absorbed, scattered, and reflected by the surface of objects, enabling the generation of realistic images for filmmaking, video games, virtual reality, and any application that requires accurate lighting and material representation.

[0042] The light interaction submodule uses light source configuration data to analyze the interaction between light and the surface of three-dimensional objects, calculates reflection and scattering effects, adjusts the interactive behavior of light according to the material and surface characteristics of the object, simulates the performance of light in various environments, and generates the specific process of light environment interaction data as follows: The Light Interaction submodule uses the light source configuration data to analyze the interaction between light and the surface of a 3D object. The module first uses the Bidirectional Reflection Distribution Function (BRDF) model to calculate the reflection and scattering effects of light emitted from the light source when it encounters the surface of a 3D object. Specifically, the module considers the incident angle θ and the reflection angle (i.e. observation angle). The formula for calculating reflection intensity is: reflection intensity = light source intensity × BRDF (θ, )×cos(θ), where the light source intensity represents the initial intensity of the light emitted by the light source, BRDF(θ, ) is the reflection coefficient, and cos(θ) represents the effect of the incident angle on the light intensity. The scattering effect depends on the roughness of the object surface and the scattering coefficient, and its calculation formula is: scattered light intensity = light source intensity × scattering coefficient × (1 / 2π), where the scattering coefficient describes the scattering ability of the object surface, and (1 / 2π) is the normalized coefficient of omnidirectional scattering. Through these calculations, the module simulates the performance of light in a variety of environments, and the generated light environment interaction data not only accurately reflects the physical interaction between light and objects, but also adapts to the needs of different visual scenes.

[0043] The shadow generation submodule uses the light environment interaction data to locate the shadow area caused by the obstruction between the light source and the object, calculate the intensity and extension of the shadow, adjust the blur and boundary of the shadow, optimize the light-dark contrast in the image, and obtain the specific process of light simulation data as follows: The shadow generation submodule uses the light environment interaction data to locate the shadow area caused by the obstruction between the light source and the object. Ray tracing technology is used to calculate the straight line path from the light source to each part of the model, and detect whether these paths are blocked by other objects, so as to determine the shadow formation area. The intensity and extension of the shadow are calculated based on the distance and angle of the light source. The "shadow attenuation factor" is used to adjust the blurriness of the shadow edge. This factor is adjusted as a function of the distance from the light source to the shadow area. The specific calculation formula is: , where k is the attenuation coefficient and d is the distance from the light source to the point where the shadow is generated. This optimizes the light-dark contrast in the image and ultimately obtains accurate light simulation data.

[0044] See also Figure 2 and Figure 7 , the image rendering module includes: The parameter adjustment submodule analyzes the light intensity and color distribution in the image based on the light simulation data, adjusts the parameters one by one to match the predetermined visual effect, and adjusts the color saturation and brightness level. The specific process of generating image parameter optimization data is as follows: The parameter adjustment submodule first performs a detailed analysis of the illumination intensity and color distribution in the image based on the light simulation data. This process involves evaluating the overall brightness of the image and the intensity distribution of each color channel. The color saturation and brightness level of the image are adjusted based on these data. L essentially measures the similarity between the image and the given text prompt. While L can generate 3D models corresponding to the text prompt, they cannot be fully aligned with the reference image because the text prompt cannot capture all the object details. Through a diffusion CLIP loss, it is expressed as , which additionally forces the generated model to match the reference image: ; here is a CLIP image encoder, rather than directly measuring the CLIP loss on X , which encodes the new view rendering , and then denoise it into a clean image , where X represents the image generated by the model or the data sample. By applying a similarity loss to the denoised image sampled from the diffusion model, the rendering is encouraged to align with the reference image while resembling a high-quality sample from frozen diffusion. Incorporating the diffusion prior ensures that the generated 3D model is visually appealing and plausible while also conforming to the given image. Nevertheless, even if the rendered image seems to make sense to the diffusion model, there are still shape ambiguities, leading to problems such as concave surfaces, through-plane geometry, or depth ambiguity. These problems are alleviated by leveraging the depth previously learned from rich external images and performing supervision directly in 3D. Specifically, an off-the-shelf single-view depth estimator is used to estimate the depth d of the input image. Although the estimated depth may not accurately describe the geometric details, it is sufficient to ensure reasonable geometry and resolve most of the ambiguities. To account for inaccuracies and scale mismatches in d, a negative Pearson correlation between the estimated depth and the depth d is regularized. Modeled by NeRF on the reference viewpoint, the texture point cloud construction diagram. The goal is to build dense points and texture visible points using the reference image, and invisible points from NeRF. Through this regularization, NeRF depth estimation is encouraged to be linearly related to the depth prior. Multiple loss functions are used during model training, which are used for different objectives and optimization processes. By combining these loss functions, the model is able to optimize in terms of overall performance, similarity to the reference image, stability of the diffusion process, and consistency of the depth information, thereby improving the quality of the generated image and ensuring the stability of the model during the optimization process. A progressive training strategy is adopted here, starting from a narrow view range near the reference view and gradually expanding the range during training. Through progressive training, 360-degree reconstruction of an object can be achieved.

[0045] The detail enhancement submodule uses image parameter optimization data to perform hierarchical separation of details in the image, adjust contrast and sharpness, and optimize key visual elements in the image, including texture and edges. The specific process of obtaining a detail-optimized image is as follows: The detail enhancement submodule constructs a point cloud according to and NeRF, existing points ,in are the extrinsic and intrinsic matrices of the camera, P represents point-to-point projection, Indicates the depth of rendering, Indicates that according to the reference view Rendering generated alpha mask, M is a mask generation function, the input is the reference view Parameters are used to distinguish different areas in the image (such as foreground and background), thereby providing a basis for subsequent image optimization and adjustment. Represents a reference view Generated point cloud. In the image or 3D rendering process, the point cloud is a set of points in the 3D space of the image, used to describe the depth information and geometry in the 3D space. These points are visible under the reference view and are therefore colored with the ground truth texture. For the projection of the remaining views, the new view is used , it is important to avoid introducing points that overlap with existing points but have conflicting colors. To do this, replace the existing points Project to New View , generate a mask indicating the presence of existing points. Use this as a guide to only promote points that have not yet been observed , which has not been observed. Then, NeRF is used to render , and integrated into a dense point cloud, Indicates that in a specific view NeRF rendering operation performed under , G is the rendering function used to So far, a set of texture point clouds V = {V(β ref ), V(β1), ..., V(β N )}, although V(β ref ) already has a high-fidelity texture projected from the reference image, and other points occluded in the reference view still suffer from the smooth texture from the rough NeRF. To enhance the texture, the textures of other points are optimized and the new view rendering is constrained using a diffusion prior. Specifically, a 19-dimensional descriptor F is optimized for each point, and its first three dimensions are initialized with the initial RGB colors. In order to avoid noisy colors and artifacts, a multi-scale deferred rendering scheme is adopted. In particular, given a new view β, the point cloud V is rasterized K times to obtain K feature maps, which are then concatenated and rendered into an image.

[0046] The image output submodule performs color correction and adjusts the resolution through the detail-optimized image, and the specific process of outputting the optimized rendered image is as follows: The image output submodule performs final color correction and resolution adjustment on the detail-optimized image to suit different display devices and output requirements. Color correction is based on the standard configuration in the color management system, adjusting each color channel to achieve real-world color performance. Resolution adjustment is based on the needs of the output media, adjusting the image resolution through sampling and resampling techniques to ensure that the image maintains the best quality at various sizes. The output optimized rendered image therefore has high-quality color accuracy and clarity.

[0047] In addition, GPU costs can be reduced by combining the diffusion model and the InstructBLIP model, by optimizing the image processing flow in the generation task. The diffusion model is mainly used in high-fidelity 3D image generation to gradually build and refine the details of the image, gradually transforming from a disordered noise state to a detailed image output through an iterative process. This model is particularly suitable for generating complex 3D textures and details because its iterative process allows precise control of the generated details at each step, thereby generating extremely realistic images, because its generation process simulates the physical diffusion process, so that while maintaining the naturalness of the image, it can handle visual effects such as lighting, reflections and textures well. The InstructBLIP model can generate or improve 3D images based on language instructions by combining natural language processing and visual data processing techniques. The model is suitable for applications that require detailed language descriptions to be converted into accurate 3D models, such as virtual reality, game development, and industrial design, and can quickly generate or modify 3D images based on the user's specific instructions. Compared with traditional large-scale language-image pre-training models, the InstructBLIP model is more efficient in resource usage. By optimizing image and text processing algorithms, the model reduces dependence on high-performance computing resources, especially in GPU-intensive operations, which can significantly reduce costs and energy consumption.

[0048] See also Figure 2 and Figure 8 , the AI ​​security and image anti-counterfeiting modules include: The security analysis submodule detects the image content based on the optimized rendered image, screens the sensitive content in the image, including privacy data and illegal information, and identifies the content that needs to be reviewed. The specific process of generating the security analysis results is as follows: The optimized image becomes the starting point for system processing. The system analyzes each pixel one by one through multi-level image scanning technology. In this way, the system can fully cover and scan the entire image data and detect sensitive content in each area. In this process, the data of each area is accurately analyzed and recorded, and the system uses advanced algorithms to identify and classify suspicious areas. This sophisticated scanning process ensures the comprehensive identification of sensitive content. Subsequently, according to the preset content classification standards, the system further subdivides and prepares for review of these sensitive contents, and finally forms a comprehensive security analysis result.

[0049] The content review submodule checks the sensitive content identified in the image based on the security analysis results, reviews it based on pre-trained data or preset artificial intelligence models, and deletes image content that does not meet the standards. The specific process for obtaining the review confirmation results is as follows: After obtaining the preliminary security analysis results, the review system begins to perform a comprehensive inspection of the image content. During this process, the system compares each identified sensitive content in the image with the organization's internal release standards, and through detailed comparison and verification work, ensures that every element in the image meets the specified requirements. This series of precise comparison and review operations allows the content that does not meet the regulations to be identified and removed one by one, thereby ensuring the purity of the image content. Through this meticulous inspection process, the accuracy and reliability of the image review and confirmation results are ultimately ensured.

[0050] The anti-counterfeiting coding submodule verifies the result, inserts a digital watermark into the image, sets up a monitoring program to track the distribution and use of the image, and ensures that the source and distribution records of the image can be queried. The specific process of obtaining the anti-counterfeiting image data is as follows; According to the results of the review and confirmation, the anti-counterfeiting coding system immediately begins to embed a digital watermark in the image. This operation is completed through a specially designed coding technology to ensure that the embedding position and pattern of the digital watermark are accurately calculated, which not only ensures that the image quality is not affected, but also ensures the security of the image. At the same time, the system sets up an image monitoring program to monitor the distribution and use of the image in real time. The program can record the details of each access and use of the image to ensure that the source and purpose of the image can be traced and verified. Through this series of operations, a complete anti-counterfeiting image data system is formed.

[0051] Embodiment 2 Based on the same inventive concept, this embodiment discloses a 3D image security processing method based on a diffusion model, which is implemented based on the 3D image security processing system based on the diffusion model described in Embodiment 1. The method includes the following steps: S1: coordinate alignment, resolution calibration and format conversion are performed on the image data obtained by three-dimensional scanning to generate standardized three-dimensional data; S2: constructing a deep learning model based on the diffusion model and the normalized three-dimensional data, and training the image data to learn the features and structural information of the image, and obtaining a trained deep model; S3: using the trained depth model, analyzing the shape, size and relative position of the three-dimensional object in the image, reconstructing the three-dimensional space, and obtaining three-dimensional space image data; S4: Based on the three-dimensional space image data, simulate the influence of the light source on the image, calculate the light scattering, reflection and shadow effects after the light interacts with the three-dimensional object, and obtain light simulation data; S5: Based on the light simulation data, adjusting rendering parameters, performing image rendering processing, and outputting an optimized rendered image; S6: Using artificial intelligence to perform security analysis and content review on the optimized rendered image, and generating anti-counterfeiting image data by embedding a digital watermark and performing image content monitoring.

[0052] Since the method introduced in the second embodiment of the present invention is a method implemented by the three-dimensional image security processing system based on the diffusion model in the first embodiment of the present invention, the specific implementation steps of the method can be understood by those skilled in the art based on the system introduced in the first embodiment of the present invention, so they are not described here in detail. All methods implemented by the system in the first embodiment of the present invention belong to the scope of protection of the present invention.

[0053] Embodiment 3 Based on the same inventive concept, the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described in the first embodiment is implemented.

[0054] Since the computer-readable storage medium introduced in the third embodiment of the present invention is the computer-readable storage medium used to implement the three-dimensional image security processing method based on the diffusion model in the second embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, the person skilled in the art can understand the specific structure and deformation of the computer-readable storage medium, so it is not repeated here. All computer-readable storage media used in the method of the first embodiment of the present invention belong to the scope of protection of the present invention.

[0055] Embodiment 4 The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in Embodiment 1 when executing the program.

[0056] Since the computer device introduced in the fourth embodiment of the present invention is a computer device used to implement the three-dimensional image security processing method based on the diffusion model in the second embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, the person skilled in the art can understand the specific structure and deformation of the computer device, so it is not repeated here. All computer devices used in the method of the first embodiment of the present invention belong to the scope of protection of the present invention.

[0057] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0058] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0059] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic creative concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the embodiments of the present invention without departing from the spirit and scope of the embodiments of the present invention. Thus, if these modifications and variations of the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A three-dimensional image security processing system based on a diffusion model, characterized in that: The system comprises: An image format conversion module is used to perform coordinate alignment, resolution calibration and format conversion on the image data obtained by three-dimensional scanning to generate standardized three-dimensional data; An image feature training module is used to construct a deep learning model based on the diffusion model and the normalized three-dimensional data, and to train the image data to learn the features and structural information of the image, and obtain a trained deep model; An image three-dimensional structure analysis module is used to analyze the shape, size and relative position of three-dimensional objects in the image using the trained depth model, reconstruct the three-dimensional space, and obtain three-dimensional space image data; An image light effect simulation module is used to simulate the influence of the light source on the image based on the three-dimensional space image data, calculate the light scattering, reflection and shadow effects after the light interacts with the three-dimensional object, and obtain light simulation data; An image rendering module, used to adjust rendering parameters based on the light simulation data, perform image rendering processing, and output an optimized rendered image; The artificial intelligence security and image anti-counterfeiting module is used to use artificial intelligence to perform security analysis and content review on the optimized rendered image, and generate anti-counterfeiting image data by embedding digital watermarks and performing image content monitoring.

2. The three-dimensional image security processing system based on the diffusion model as claimed in claim 1, characterized in that: The image format conversion module comprises: A coordinate alignment submodule is used to perform consistency check and alignment of spatial coordinates based on the image data obtained by three-dimensional scanning, and generate aligned image data; A resolution calibration submodule, used to detect and adjust the image pixel resolution of the aligned image data to generate calibrated image data; The data formatting submodule performs a data format conversion operation based on the calibrated image data, converts the image data into a unified point cloud format, and generates standardized three-dimensional data.

3. The three-dimensional image security processing system based on the diffusion model as claimed in claim 1, characterized in that: The image feature training module includes: A model building submodule, for selecting a network layer structure and an activation function, defining input and output nodes and the number of network layers, and generating an initialized deep learning model framework based on the normalized three-dimensional data; A feature learning submodule is used to use the initialized deep learning model framework, input three-dimensional image data for feature extraction, analyze the structural information and texture features of the image, optimize model parameters, and generate a feature optimization model; The model training submodule performs multiple rounds of training based on the feature optimization model, adjusts the learning rate and the loss function, uses a batch training method to optimize the generalization ability of the model, and generates a trained deep model. In each round of training, a diffusion model is used to calculate the loss based on the difference between the current noise prediction and the actual noise, so as to guide the model to learn how to process noise terms at different time steps, wherein the noise term is used to reflect the noise accumulation effect based on the time step.

4. The three-dimensional image security processing system based on the diffusion model as claimed in claim 1, characterized in that: The image three-dimensional structure analysis module includes: An image shape analysis submodule is used to draw the outline of a three-dimensional object based on the trained depth model, perform shape analysis by aligning the drawn outline with the image edge, record shape features, and generate image shape data; An image size analysis submodule is used to measure the size of the three-dimensional object based on the image shape data, by measuring the height, width and depth of each three-dimensional object, comparing it with the standard size reference in the three-dimensional model, and summarizing the size data to obtain image size data; The image position analysis submodule is used to locate the position of the three-dimensional object in space through the image size data, record the coordinate information of each three-dimensional object according to the spatial relationship and depth information between the three-dimensional objects, and update the data in the three-dimensional model to obtain three-dimensional space image data.

5. The three-dimensional image security processing system based on the diffusion model as claimed in claim 1, characterized in that: The image light effect simulation module includes: A light source setting submodule is used to configure the light source based on the three-dimensional space image data through the BRDF model to simulate the effect of natural light or indoor lighting, adjust the light source position and intensity, and set a variety of lighting angles and color temperatures to generate light source configuration data. The BRDF model is a bidirectional reflectance distribution function model; A light interaction submodule, for analyzing the interaction between light and the surface of a three-dimensional object, calculating reflection and scattering effects, adjusting the interaction behavior of light according to the material and surface characteristics of the object, simulating the performance of light in various environments, and generating light environment interaction data using the light source configuration data; The shadow generation submodule is used to locate the shadow area caused by the obstruction between the light source and the object through the light environment interaction data, calculate the intensity and extension of the shadow, adjust the blurriness and boundary of the shadow, optimize the light-dark contrast in the image, and obtain light simulation data.

6. The three-dimensional image security processing system based on the diffusion model as claimed in claim 5, characterized in that: The BRDF model follows the formula: Calculate the relationship between incident light and reflected light from various angles to simulate the effects of natural light or indoor lighting; Among them, the function Represents the bidirectional reflectance distribution function, which is used to calculate the incident light direction of a given material surface. and reflected light direction The light reflectivity under is the incident light direction, is the direction of reflected light, is the incident angle, is the observation angle, For material properties, is the diffuse reflection coefficient, which indicates the proportion of the diffuse reflection part of the material surface. is the cosine of the angle, which is used to calculate the effective component of the light incident on the surface. is the specular reflection coefficient, which is used to indicate the proportion of the specular reflection part of the material surface. is the specular highlight attenuation factor, according to the observation angle Adjust the intensity of specular reflection so that the highlight effect weakens as the viewing angle increases. is the light radiation intensity in the unit reflection direction, representing the reflection direction The brightness per unit area, is the light intensity per unit incident direction, representing the light intensity in the incident direction The amount of light energy received per unit area.

7. The three-dimensional image security processing system based on the diffusion model as claimed in claim 1, characterized in that: The image rendering module comprises: A parameter adjustment submodule, for analyzing the light intensity and color distribution in the image based on the light simulation data, adjusting the parameters one by one to match the predetermined visual effect, and adjusting the color saturation and brightness level to generate image parameter optimization data; A detail enhancement submodule, for using the image parameter optimization data to perform hierarchical separation, contrast and sharpness adjustment on the details in the image, and optimize key visual elements in the image, including texture and edges, to obtain a detail-optimized image; The image output submodule is used to perform color correction and adjust the resolution of the detail-optimized image, and output an optimized rendered image.

8. A three-dimensional image security processing method based on a diffusion model, characterized in that: Based on the implementation of the three-dimensional image security processing system based on the diffusion model according to any one of claims 1 to 7, the method comprises the following steps: S1: coordinate alignment, resolution calibration and format conversion are performed on the image data obtained by three-dimensional scanning to generate standardized three-dimensional data; S2: constructing a deep learning model based on the diffusion model and the normalized three-dimensional data, and training the image data to learn the features and structural information of the image, and obtaining a trained deep model; S3: using the trained depth model, analyzing the shape, size and relative position of the three-dimensional object in the image, reconstructing the three-dimensional space, and obtaining three-dimensional space image data; S4: Based on the three-dimensional space image data, simulate the influence of the light source on the image, calculate the light scattering, reflection and shadow effects after the light interacts with the three-dimensional object, and obtain light simulation data; S5: Based on the light simulation data, adjusting rendering parameters, performing image rendering processing, and outputting an optimized rendered image; S6: Using artificial intelligence to perform security analysis and content review on the optimized rendered image, and generating anti-counterfeiting image data by embedding a digital watermark and performing image content monitoring.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the three-dimensional image security processing method based on the diffusion model as claimed in claim 8 is implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the three-dimensional image security processing method based on the diffusion model as claimed in claim 8 is implemented.

Images