Object-Aware Image Scaling for Multi-Part Substrates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computer-implemented methods for applying images to substrates face challenges in maintaining image quality and realism when there are significant size differences between the image and substrate, especially with multi-part substrates, leading to unrealistic and degraded image representation.
Innovation Solution
A computer-implemented method that utilizes artificial intelligence (AI) to automatically adjust images by detecting objects, analyzing their properties and context, and applying scaling, mirroring, and expansion techniques to ensure high-quality and realistic image application on substrates of any size, including multi-part substrates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If the image is automatically scaled to the size of the substrate, then the image covers the substrate, but image quality and realism are lost
Solution Approach 1:
The patent segments the substrate into multiple parts and applies different scaling factors to different segments. This allows each segment to be scaled appropriately without compromising overall image quality, resolving the contradiction between full substrate coverage and maintaining image quality.
Solution Approach 2:
The patent applies local quality by using context-aware scaling that adjusts scaling factors based on the specific region of the substrate and image being processed. Different areas receive different scaling treatments to preserve image quality while ensuring complete substrate coverage.
2Area of stationary object
If different scaling factors are applied to multi-part substrates, then each part fits its region, but uniform overall image cannot be created
Solution Approach 1:
The patent employs feedback mechanisms where the system analyzes the relationships between different substrate parts and their corresponding image regions, then adjusts scaling factors to ensure uniformity across the entire multi-part substrate while maintaining proper coverage of each individual part.
Solution Approach 2:
The patent uses dynamic scaling factors that are adjusted based on the specific configuration of multi-part substrates. The system dynamically determines appropriate scaling for each part while maintaining overall image uniformity, rather than applying static uniform scaling.
3Area of stationary object
If the image is enlarged significantly, then it covers larger substrates, but quality loss and unrealistic representation occur
Solution Approach 1:
The patent changes parameters by using context-aware scaling factors that are determined based on the specific image content, substrate characteristics, and regional requirements. This allows significant enlargement when necessary while maintaining image realism through intelligent parameter adjustment.
Solution Approach 2:
The patent performs preliminary analysis of the image and substrate characteristics before applying scaling. This preliminary action allows the system to plan the scaling strategy in advance, ensuring that enlargement is done in a way that preserves realism and avoids quality loss.
Data Source
AI summary
The invention relates to a computer-implemented method for applying an image to a substrate, the method comprising the following steps: selecting the image, in particular from an image database; determining a substrate width and a substrate height of the substrate; automatically adjusting the image so that the adjusted image at least completely covers the substrate; and automatically applying the adjusted image to the substrate.The automatic adjustment comprises the following steps: automatically detecting at least one object in the image; automatically deciding, depending on properties of the at least one object, whether the at least one object is enlarged by scaling the at least one object, by arranging the at least one object multiple times in succession and mirroring them at the edges of adjacent objects, and/or by AI-assisted expansion of the at least one object; and automatically enlarging the image at least to the size of the substrate, wherein the at least one object is enlarged according to the result of the decision.