A method and system for pre-distortion compensation of curved surface transfer images based on artificial intelligence
By using an AI-based parameter set simulation and automatic generation method, the problem of image distortion in curved surface transfer was solved, achieving high-precision automated distortion compensation and improving transfer effect and efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 张立晖
- Filing Date
- 2026-04-29
- Publication Date
- 2026-07-03
AI Technical Summary
In existing curved surface transfer processes, images are prone to geometric distortion, parameter adjustment methods are limited, correction accuracy is low, manual adjustment is inefficient, and the process cannot adapt to various types of curved surfaces and transfer processes, resulting in poor transfer effects.
By employing an artificial intelligence-based approach, the first and second parameter sets are independently set to simulate image deformation patterns and generate inverse compensation distortion. Combined with the automatic parameter generation of the artificial intelligence model, high-precision automated distortion compensation of images is achieved.
It improves distortion correction accuracy and processing efficiency, adapts to various types of curved surfaces, is compatible with multiple transfer processes, and ensures image integrity and accuracy.
Smart Images
Figure CN122335631A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of curved surface transfer image processing technology, specifically relating to an artificial intelligence-based method and system for pre-distortion compensation of curved surface transfer images, applicable to image distortion correction and preprocessing in the process of transferring planar two-dimensional images to three-dimensional curved surfaces. Background Technology
[0002] Surface transfer printing is a common process for transferring planar patterns, text, and images onto the surface of three-dimensional curved workpieces. It is widely used in vacuum heat transfer printing, water transfer printing, curved surface coating, and curved surface inkjet printing. In existing surface transfer printing processes, when a planar image is applied to a curved surface, geometric distortions such as radial stretching, radial compression, and local twisting occur, leading to problems such as distortion and disproportion in the transferred image. Current technologies typically involve not adjusting the original image or manual adjustment, which is inefficient and produces poor results, failing to meet customer needs. Summary of the Invention
[0003] Technical problem solved by the present invention To address the problems of geometric distortion, limited parameter adjustment methods, low correction accuracy, low efficiency of manual adjustment, and inability to adapt to various types of curved surfaces and transfer processes in existing curved surface transfer processes, this invention provides an artificial intelligence-based method and system for pre-distortion compensation of curved surface transfer images, achieving high-precision and automated distortion compensation for curved surface transfer images.
[0004] Technical solution of the present invention An AI-based pre-distortion compensation method for surface transfer images acquires the original 2D image to be transferred and the structural parameters of the target 3D surface. It configures two independent parameter sets: a first parameter set to simulate the deformation pattern of the image during surface transfer, and a second parameter set to generate compensation distortion that is the opposite of the deformation pattern. Based on the first parameter set, the actual deformation effect of the original image after fitting to the target 3D surface is simulated. Combining the actual deformation effect, the second parameter set is used to perform global pre-distortion processing on the original image, generating a compensated target image. This compensated target image then cancels out the deformation generated during the surface transfer process. The system automatically generates the first and second parameter sets based on the structural parameters of the target 3D surface using an AI model, while also supporting manual parameter adjustment and real-time preview of the compensation effect. An AI-based pre-distortion compensation system for surface transfer images includes a data acquisition module, a parameter configuration module, a deformation simulation module, a pre-distortion processing module, an AI parameter generation module, and a manual adjustment and preview module. These modules work together to complete image acquisition, parameter configuration, deformation simulation, pre-distortion processing, and parameter calibration.
[0005] Beneficial effects of the present invention This invention separates the parameters for deformation simulation and pre-distortion compensation by independently setting a first parameter set and a second parameter set, thereby improving the accuracy of distortion correction. It uses artificial intelligence to automatically generate parameters, replacing manual adjustment and significantly improving processing efficiency and parameter consistency. The pre-distortion processing covers geometric distortion, color distortion, and sharpness compensation, adapting to various types of curved surfaces such as external curved surfaces, internal curved surfaces, and irregular free-form surfaces. It is compatible with various transfer processes such as vacuum heat transfer, water transfer, and curved surface inkjet printing, offering a wide range of applications and stable correction effects, effectively ensuring the integrity and accuracy of the image after curved surface transfer. Attached Figure Description
[0006] Figure 1. Document titled "A Method and System for Pre-distortion Compensation of Curved Surface Transfer Images Based on Artificial Intelligence" Detailed Implementation
[0007] The present invention will be further described in detail below with reference to specific embodiments. An original two-dimensional image to be transferred is acquired. This original two-dimensional image includes patterns, text, trademarks, QR codes, graphics, etc. Simultaneously, the structural parameters of the target three-dimensional surface are collected. The target three-dimensional surface includes outer surfaces, inner surfaces, concave-convex composite surfaces, and irregular free-form surfaces. Two independent parameter sets are configured. The first parameter set is used to simulate the deformation law of the image during the surface transfer process, and the second parameter set is used to generate inverse compensation distortion. An artificial intelligence model automatically generates and optimizes the two sets of parameters based on the three-dimensional surface structural parameters. The original two-dimensional image and the first parameter set are input into a deformation simulation unit to simulate the actual deformation effect after the image is fitted to the surface, determining the deformation type and degree. The simulation results and the second parameter set are input into a pre-distortion processing unit to perform global pre-distortion processing on the original image, including radial stretching, radial compression, tangential deformation correction, and local distortion correction, generating a compensated target image. The compensated target image is applied to transfer processes such as vacuum heat transfer, water transfer, curved surface lamination, and curved surface inkjet printing. The transferred image can compensate for distortions caused during the process, maintaining the integrity of image proportions, outlines, and clarity. The system supports fine-tuning of the first and second parameter sets through a manual adjustment interface, and previewing the adjusted pre-distortion effect to meet personalized processing needs.
Claims
1. A method for pre-distortion compensation of curved surface transfer images, characterized in that, Includes the following steps: Obtain the original 2D image to be transferred and the structural parameters of the target 3D surface. Configure a first parameter set and a second parameter set that are independent of each other. The first parameter set is used to simulate the deformation law of the image during the surface transfer process. The second parameter set is used to generate compensation distortion that is opposite to the deformation law. Based on the first parameter set, simulate the actual deformation effect of the original image after it is attached to the target 3D surface. Combine the actual deformation effect, call the second parameter set to perform global pre-distortion processing on the original image to generate a compensated target image. The compensated target image cancels out all the deformations generated during the transfer process after the surface transfer.
2. The method for pre-distortion compensation of curved surface transfer images according to claim 1, characterized in that, The pre-distortion processing includes geometric distortion compensation and sharpness compensation.
3. The method for pre-distortion compensation of curved surface transfer images according to claim 2, characterized in that, The geometric distortion compensation includes radial stretching, radial compression, tangential stretching, tangential compression, local torsion correction, and rotation correction.
4. The method for pre-distortion compensation of curved surface transfer images according to claim 1, characterized in that, The target three-dimensional surface includes outer surface, inner surface, concave-convex composite surface, irregular free surface, rotationally symmetric surface, and non-rotationally symmetric surface.
5. The method for pre-distortion compensation of curved surface transfer images according to claim 1, characterized in that, The curved surface transfer includes vacuum heat transfer, water transfer, water-based coating transfer, and curved surface lamination.
6. The method for pre-distortion compensation of curved surface transfer images according to claim 1, characterized in that, Based on an artificial intelligence model, the first parameter set and the second parameter set are automatically generated according to the structural parameters of the target three-dimensional surface.
7. The method for pre-distortion compensation of curved surface transfer images according to claim 1, characterized in that, A manual adjustment interface is provided, allowing users to manually modify the values of the first and second parameter sets and preview the modified pre-distortion effect in real time.
8. The method for pre-distortion compensation of curved surface transfer images according to claim 1, characterized in that, The original two-dimensional image includes patterns, text, trademarks, QR codes, barcodes, photographs, graphics, gradient backgrounds, and solid color blocks.
9. A pre-distortion compensation system for curved surface transfer images, characterized in that, include: The data acquisition module is used to acquire the original two-dimensional image to be transferred and the structural parameters of the target three-dimensional curved surface. The parameter configuration module is used to configure a first parameter set and a second parameter set that are independent of each other. The first parameter set is used to simulate the deformation law of the image during the surface transfer process, and the second parameter set is used to generate compensation distortion that is opposite to the deformation law. The deformation simulation module is used to simulate the actual deformation effect of the original image after it is fitted to the target three-dimensional curved surface based on the first parameter set. The pre-distortion processing module is used to combine the actual deformation effect and call the second parameter set to perform global pre-distortion processing on the original image to generate the compensated target image.
10. The curved surface transfer image pre-distortion compensation system according to claim 9, characterized in that, It also includes an artificial intelligence parameter generation module, which is used to automatically generate the first parameter set and the second parameter set based on the structural parameters of the target three-dimensional surface.
11. The curved surface transfer image pre-distortion compensation system according to claim 9, characterized in that, It also includes a manual adjustment and preview module, which allows users to manually modify the values of the first and second parameter sets and preview the modified pre-distortion effect in real time.