Image Quality Enhancement via ROI and RONI Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image quality enhancement methods either apply the same model to entire images or videos, failing to emphasize key regions, or only enhance regions of interest without improving non-interest areas, leading to limited subjective quality and user experience.
Innovation Solution
The method involves determining regions-of-interest (ROI) and regions-of-non-interest (RONI) in images, using separate image quality enhancement models for each to produce more targeted enhancements, blending the outputs for improved overall image or video quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the same image quality enhancement model is applied to entire images or videos, then the processing is simple and consistent, but key regions are not emphasized and subjective quality is limited
Solution Approach 1:
The image is segmented into multiple regions based on importance: regions of interest (ROI) and regions of non-interest (RONI). Different enhancement models are applied to different regions, with more complex processing for ROI and simpler processing for RONI, thus resolving the contradiction between processing simplicity and enhancement precision
Solution Approach 2:
Different quality enhancement standards and models are applied to different regions of the image. ROI regions receive high-quality enhancement with detailed processing, while RONI regions receive standard-quality enhancement, achieving local optimization of image quality that resolves the contradiction between consistent processing and targeted quality improvement
2Manufacturing precision
If only regions of interest are enhanced, then key regions are emphasized, but non-interest areas are not improved and overall quality is limited
Solution Approach 1:
The patent applies partial enhancement action by using different enhancement intensities for different regions. ROI regions receive excessive enhancement (higher quality model) to ensure key areas are emphasized, while RONI regions receive standard enhancement, achieving both key region emphasis and overall quality improvement
3Manufacturing precision
If different enhancement models are applied to different regions, then targeted enhancement is achieved, but processing complexity increases
Solution Approach 1:
The image is segmented into ROI and RONI regions, allowing different enhancement models to be applied. This segmentation enables targeted enhancement while managing complexity by clearly defining region boundaries and using appropriate models for each region
Solution Approach 2:
Different quality levels and models are applied locally to different regions. The system manages complexity by establishing clear local quality standards for ROI versus RONI, making the multi-model approach systematic and controllable rather than chaotic
Data Source
AI summary
The present disclosure provides a method and an apparatus for enhancing image quality, a device, and a medium, relates to the field of artificial intelligence and specifically to computer vision and deep learning technologies, and can be applied to an image processing scenario. The method includes: determining an ROI and an RONI in an image to be processed; inputting the ROI to an ROI image quality enhancement model, to obtain first image data output from the ROI image quality enhancement model; inputting the RONI to an RONI image quality enhancement model, to obtain second image data output from the RONI image quality enhancement model; and blending the first image data and the second image data.