Graph-Based Object Separation for HDR Image Enhancement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image enhancement methods fail to effectively enhance high dynamic range (HDR) images for display on smaller screens, such as mobile devices, by maintaining the contrast and detail enhancement benefits seen on larger displays.
Innovation Solution
An iterative graph-based method that performs inter-object point cloud separation and spatial enhancement on individual objects within an image, using a hierarchical graph structure to prioritize and incrementally improve the contrast and detail of objects based on their importance, ensuring balanced global and local image enhancement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional image enhancement methods are applied to HDR images, then contrast and detail enhancement is achieved, but the enhancement benefits are not effectively maintained when displaying on smaller screens such as mobile devices
Solution Approach 1:
The image is segmented into multiple objects using graph-based segmentation, where each object is represented as a node in a graph. This segmentation allows the enhancement process to operate on individual objects rather than the entire image, enabling adaptive enhancement that works effectively on both large and small displays. The graph structure captures spatial relationships between objects, preserving the enhanced contrast and detail while adapting to different display sizes.
Solution Approach 2:
The patent applies local quality enhancement by processing different regions of the image (different objects) with different enhancement characteristics. Each object receives enhancement tailored to its specific properties and its relationship with neighboring objects, as captured by the graph structure. This local approach ensures that contrast and detail are enhanced appropriately for each region, maintaining effectiveness across different display sizes.
2Manufacturing precision
If graph-based inter-object separation is applied to enhance object distinction, then inter-object contrast is improved, but computational complexity increases
Solution Approach 1:
The graph-based segmentation and object representation is performed as a preliminary action before the enhancement process. By pre-establishing the graph structure and object identities, the subsequent enhancement computations can focus solely on adjusting contrast and detail for each object, reducing the overall computational burden during the enhancement phase.
Solution Approach 2:
The patent applies enhancement operations selectively to objects based on their importance and visual characteristics, rather than uniformly processing the entire image. This partial action approach concentrates computational resources on the most visually significant objects, achieving effective inter-object contrast enhancement while controlling overall computational complexity.
3Manufacturing precision
If iterative enhancement process is implemented to progressively improve image quality, then visual quality enhancement is achieved, but processing time increases
Solution Approach 1:
The iterative enhancement process incorporates feedback mechanisms that monitor image quality metrics and adjust the enhancement process accordingly. Based on feedback from quality assessment, the algorithm can determine when sufficient enhancement has been achieved and terminate early, or adjust the number of iterations needed, thereby balancing visual quality enhancement with processing time.
Solution Approach 2:
The patent implements a stopping criterion that allows the iterative process to terminate when sufficient enhancement is achieved, avoiding unnecessary additional iterations. This partial action approach ensures that the enhancement process stops at the optimal point, balancing improved visual quality with acceptable processing time.
Data Source
AI summary
Systems and methods for enhancing images using graph-based inter- and intra-object separation. One method includes receiving an object within the image frame, the object including a plurality of pixels, performing an inter-object point cloud separation operation on the image, and expanding the plurality of pixels of the object. The method includes performing a spatial enhancement operation on the plurality of pixels of the object and generating an output image based on the inter-object point cloud separation operation, the expansion of the plurality of pixels, and the spatial enhancement operation.