ROI Image Splitting and Scaling for Hardware-Limited Model Input
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing image processing technologies face issues when the output limitation of the image scaling module prevents it from performing scaling processing on parts of the Region of Interest (ROI) image, leading to errors and the inability to execute subsequent processes normally.
Innovation Solution
The proposed solution involves an image processing method that splits a ROI image into multiple image blocks based on a supported image size by the image processing model and the hardware output size of the image scaling module. Each image block is then scaled to match the hardware output size, allowing all scaled image blocks to be normally output and input into the image processing model for sequential processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the image scaling module processes the ROI image directly, then the processing speed is fast, but the output size does not match the required image size supported by the image processing model
Solution Approach 1:
The patent divides the ROI image into multiple image blocks that can be processed separately by the image scaling module. Each block is scaled to match the required size, and then the blocks are concatenated to form the final processed image. This segmentation approach resolves the contradiction by making the processing compatible with hardware limitations while maintaining overall image quality.
Solution Approach 2:
The patent introduces a new dimension of processing by splitting the image not only spatially but also in terms of processing stages. The image is divided into blocks, each processed independently through scaling, and then reassembled. This multi-dimensional approach allows the system to work around the image scaling module's size limitations while achieving the desired output dimensions.
2Reliability
If the image scaling module outputs error due to size limitation, then the image size requirement is not met, but the processing time is wasted
Solution Approach 1:
The patent performs preliminary splitting of the ROI image into multiple blocks before processing. By pre-dividing the image according to the image scaling module's capabilities, the system avoids runtime errors and ensures that each block can be processed successfully. This preliminary action eliminates wasted processing time on incompatible images.
Solution Approach 2:
The patent preemptively prevents the error condition by splitting the image before it reaches the image scaling module. By anticipating the size limitation issue and addressing it in advance through image block division, the system avoids the error prompt and ensures reliable processing of all ROI images regardless of their original dimensions.
3Manufacturing precision
If the ROI image is split into multiple blocks, then the image size can be matched to hardware output size, but the device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the ROI image into multiple manageable blocks. This segmentation makes the processing compatible with the image scaling module's hardware output size limitations. The complexity is managed by processing each block independently and then concatenating them, which is a systematic approach that balances precision requirements with processing feasibility.
4Manufacturing precision
If the image is processed in blocks sequentially, then the image size matching is achieved, but the processing time increases
Solution Approach 1:
The patent segments the image processing into multiple block-level operations. Each block is processed to achieve accurate output size matching, and the blocks are then concatenated to form the final image. This segmentation enables precise size control while managing processing time through efficient block-level operations.
Data Source
AI summary
Embodiments of the present disclosure disclose an image processing method, method for generating instructions for image processing, and apparatuses therefor. The method includes: if an ROI image is obtained, splitting to obtain a plurality of image blocks based on a first image size supported by an image processing model, first split data, and the obtained ROI image, wherein each image size obtained by splitting the first image size based on the first split data matches a hardware output size of an image scaling module; performing image scaling on each image block to obtain scaled image blocks, wherein each image size of the scaled image blocks is consistent with a respective image size; and inputting all scaled image blocks to the image processing model sequentially. In the embodiments, although output of image scaling module is limited, subsequent processes involved in the visual image processing technology may be properly executed.