ROI-Aware Image Splitting for Continuous Multi-Display Reading

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current image display technologies often disconnect important information, such as text, when splitting images across multiple displays, leading to reading difficulties due to improper splitting.

Innovation Solution

An image display method that analyzes the original image to identify regions of interest, calculates pixel gradients, and determines optimal splitting paths to prioritize important information, while performing point deletion or insertion to maintain readability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If the original image is split into multiple sub-images based on display arrangement, then the image can be displayed across multiple displays, but important information may be disconnected and reading becomes difficult

Engineering Contradiction:
Improvedisplay areaVSAvoidinformation continuity
Core Design Contradiction:
Area of stationary objectVSLoss of information

Solution Approach 1:

The patent applies segmentation by dividing the image into multiple sub-images for display across multiple screens. However, it enhances this basic segmentation by introducing intelligent detection of text and important regions, ensuring that segmentation occurs only in appropriate areas rather than uniformly across the entire image.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements local quality by treating different regions of the image differently. Text regions and important areas are identified and protected from splitting, while other regions can be divided as needed. This creates locally optimized quality where critical information maintains integrity while non-critical areas are efficiently distributed.

Inventive Principle:
Principle #3Local quality

2Loss of information

If text-based splitting is performed to avoid disconnecting information, then reading difficulty is reduced, but the splitting may still not align with actual important regions in the image

Engineering Contradiction:
Improveinformation integrityVSAvoidsplitting adaptability
Core Design Contradiction:
Loss of informationVSAdaptability or versatility

Solution Approach 1:

The patent applies preliminary action by performing gradient calculations and text detection on the entire image before actual splitting occurs. This preprocessing identifies all critical regions in advance, allowing the splitting algorithm to make informed decisions about where to divide the image without disrupting important content.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces gradient calculation as an additional dimensional analysis beyond simple text detection. By computing gradients in multiple directions and analyzing regional characteristics, the system gains a more comprehensive understanding of image structure, enabling more accurate and adaptable splitting decisions.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Loss of information

If multiple algorithms are used to analyze image priority and determine splitting paths, then the priority of important information is increased, but the processing complexity increases

Engineering Contradiction:
Improveinformation priority preservationVSAvoidprocessing complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent segments the complex processing task into distinct algorithmic components: gradient calculation, text detection, regional analysis, and priority assignment. Each component handles a specific aspect of image analysis, making the overall complex process more manageable and systematic while preserving information priority.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces gradient matrices and priority matrices as intermediary data structures that bridge the gap between raw image data and splitting decisions. These intermediary representations organize complex information in a structured format, facilitating easier processing and decision-making in the splitting algorithm.

Inventive Principle:
Principle #24Intermediary (Mediator)

4Area of stationary object

If the image is split to fit multiple displays, then the display coverage is improved, but the reading experience deteriorates due to framing effects

Engineering Contradiction:
Improvedisplay coverageVSAvoidreading experience
Core Design Contradiction:
Area of stationary objectVSEase of operation

Solution Approach 1:

The patent ensures local quality by protecting specific regions (text and important areas) from splitting while allowing other regions to be divided for optimal display coverage. This creates a localized reading experience that maintains integrity where needed while achieving good coverage overall.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent converts the potential harm of display framing into a benefit by using the frame structure to guide intelligent splitting decisions. Rather than fighting against the framing effects, the algorithm uses knowledge of frame positions and display geometry to make splitting decisions that work with the display structure while protecting content integrity.

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Data Source

PatentEP4597423B1Image display method and control system
Publication Date: 2025.11.05 WISTRON CORP
  • EP4597423B1 patent drawingFigure 1
  • EP4597423B1 patent drawingFigure 2
  • EP4597423B1 patent drawingFigure 3~4

AI summary

An image display method and a control system (100) are provided. A processor (110) is configured to split an original image (RI, I) into multiple sub-images (720, 810, ML) and then output the sub-images (720, 810, ML) to multiple corresponding displays to respectively display (130, 910, 920). Process of splitting the original image (RI, I) includes: finding multiple regions of interest (ROIs) (310, 320) in the original image (RI, I) to generate a ROI matrix (M1, IROI); using a mask (50) to perform a specified operation on each pixel of the original image (RI, I) to obtain a priority matrix (M2); performing gradient calculation on each pixel of the original image (RI, I) to generate a gradient matrix; generating an integration matrix (MT) based on the ROI matrix (M1, IROI), the priority matrix (M2) and the gradient matrix; determining a splitting path based on the integration matrix (MT); and splitting the original image (RI, I) into multiple sub-images (720, 810, ML) based on the splitting path.