Image Inference Model Auxiliary Area Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image classification systems only identify primary areas of interest in images, overlooking auxiliary areas that contributed to the inference generation, which can limit the completeness of analysis and storage efficiency.
Innovation Solution
A system that uses an inference model to calculate pixel gradients, identify both primary and auxiliary areas of interest, and segment images for storage in varying performance tiers based on likelihood of future access, ensuring that more relevant image segments are stored in higher performance storage while less relevant ones are stored in lower performance storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If only primary areas of interest are identified in images, then the system complexity is reduced and processing speed is improved, but the completeness of analysis is degraded and storage efficiency is worsened
Solution Approach 1:
The patent segments the image into multiple regions of interest (ROIs) based on pixel contribution levels to inference outcomes. Primary ROIs contain pixels with high contribution levels, while secondary ROIs contain pixels with moderate contribution levels. This segmentation allows the system to process and store different regions with different priorities, improving both processing efficiency and analysis completeness.
2Speed
If all image segments are stored in high performance storage, then access speed is improved, but storage cost increases
Solution Approach 1:
The patent applies local quality by storing different image segments in different storage tiers based on their importance. Primary ROIs with high pixel contribution levels are stored in high performance storage for fast access, while secondary ROIs with moderate contribution levels are stored in lower performance, cost-effective storage. This differential storage strategy optimizes both access speed and storage cost.
3Reliability
If auxiliary areas of interest are identified and stored, then the completeness of inference understanding is improved, but the device complexity increases
Solution Approach 1:
The patent performs preliminary action by pre-calculating pixel contribution levels to inference outcomes and pre-segmenting the image into primary and secondary ROIs before storage. This preliminary segmentation and classification simplifies subsequent retrieval and analysis operations, reducing system complexity while maintaining completeness of inference understanding.
Data Source
AI summary
Methods and systems for identifying areas of interest in an image are disclosed. To manage identification of areas of interest in an image, a trained inference model may generate inferences based on the pixels of the image. The inferences may include areas of interest that contributed to the generation of the inferences. Some areas of interest may be highly relevant to the inferences and may be classified as primary areas of interest. Auxiliary areas of interest may also be identified using a trained inference model. Auxiliary areas of interest may be obtained by calculating gradients for each pixel that contributed to the identification of the primary areas of interest. By rank ordering the pixels, pixels with the highest contribution to the identification of the primary areas of interest may be identified. Proximate groupings of these pixels may be classified as auxiliary areas of interest in the image.