ROI-Based Image Enhancement in Portable Terminals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image enhancement methods for portable terminals apply uniform algorithms to entire images, failing to account for varying characteristics across different regions, leading to suboptimal results such as increased noise in high-frequency regions like grass or trees while decreasing noise in human faces.
Innovation Solution
An apparatus and method that utilize a Region of Interest (ROI) processor to extract and enhance specific image regions based on their characteristics, applying tailored Image Signal Processing (ISP) techniques and synthesizing the enhanced ROIs with the original image, while controlling boundary areas to minimize artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If an identical image enhancement algorithm is applied to the whole image, then the processing is simple and fast, but the image cannot be enhanced according to the characteristic of each region
Solution Approach 1:
The image is divided into multiple regions based on their characteristics. The image processing unit divides the input image into first regions and second regions, allowing different enhancement algorithms to be applied to different regions. This segmentation enables region-specific optimization while maintaining overall processing efficiency.
Solution Approach 2:
Different image enhancement algorithms are applied to different regions according to their specific characteristics. First regions receive a first image enhancement algorithm optimized for their characteristics, while second regions receive a second image enhancement algorithm suited to their needs, achieving local quality optimization.
2Measurement precision
If sharpening increases and denoise decreases to increase resolution, then the resolution of high-frequency regions increases, but the noise of face regions increases
Solution Approach 1:
The image is segmented into different regions (first regions and second regions) with distinct characteristics. High-frequency regions like grass and trees are separated from face regions, allowing independent processing of each region to optimize resolution enhancement while controlling noise in sensitive areas.
Solution Approach 2:
Different denoise and sharpening parameters are applied locally to different regions. High-frequency regions receive aggressive sharpening for resolution enhancement, while face regions receive milder processing to preserve skin quality and minimize noise, achieving local quality optimization.
Data Source
AI summary
Disclosed is an apparatus and a method for enhancing an image in a portable terminal. The apparatus includes an image processor for performing an Image Signal Processing (ISP) for an image received in a preview mode, an ROI processor for extracting an ROI from the image received from the image processor, enhancing an image of the extracted ROI, and synthesizing the image of the enhanced ROI with the image received from the image processor, and a controller for controlling the ROI processor to extract an ROI from an image received from the image processor, enhancing an image of the extracted ROI when a photographing is selected, synthesizing an image of the enhanced ROI with the image received from the image processor, and outputting an synthesized image.