Contour Coordinate Reduction for Accurate Whole Slide Target Images

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing technologies struggle with generating target images from whole slide images (WSIs) due to size limitations, leading to overlapping contours and reduced accuracy when reducing location information, causing connected contours in the generated images.

Innovation Solution

Define a target region and decrease the coordinates of contour points uniformly to generate a target image within this region, using the minimum coordinates as the origin to avoid drawing outside the target region, ensuring accurate representation of contours.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If the location information of contour points is reduced to decrease image size, then the image size is reduced to fit processing limits, but the contours may overlap and connect to form wrong target images

Engineering Contradiction:
Improveimage sizeVSAvoidcontour accuracy
Core Design Contradiction:
Area of stationary objectVSManufacturing precision

Solution Approach 1:

The patent divides the WSI into multiple smaller images by segmenting the contour points into different groups based on their spatial relationships. Each group forms a separate contour in the reduced image, preventing overlap and connection errors while maintaining manageable image size for processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different reduction strategies to different regions of the WSI. By identifying clusters of contour points and applying localized coordinate transformations to each cluster, the method preserves local contour accuracy while achieving overall size reduction, preventing the uniform reduction that causes contour adhesion.

Inventive Principle:
Principle #3Local quality

2Area of stationary object

If uniform coordinate reduction is applied to all contour points, then the image size is reduced, but the relative distances between contours are distorted causing adhesion

Engineering Contradiction:
Improveimage sizeVSAvoidcontour distance preservation
Core Design Contradiction:
Area of stationary objectVSStability of the object's composition

Solution Approach 1:

The patent implements a dynamic coordinate transformation system where the reduction factor is not fixed but adapts based on the spatial distribution of contour points. By calculating local scaling factors that preserve relative distances within contour groups, the method dynamically adjusts the reduction to prevent adhesion while achieving size reduction.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the reduction parameter from a uniform scalar to a variable transformation matrix that depends on the local geometry of contour point clusters. This parameter change allows the coordinate reduction to adapt to different spatial configurations, maintaining contour distances where needed while reducing overall image size.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4009232B1Image generation method, device, and storage medium
Publication Date: 2026.02.18 TENCENT TECHNOLOGY (SHENZHEN) CO LTD
  • EP4009232B1 patent drawingFigure 1
  • EP4009232B1 patent drawingFigure 2
  • EP4009232B1 patent drawingFigure 3~5

AI summary

An image generation method and apparatus, a device, and a storage medium, relating to the technical field of computers. The method comprises: obtaining contour information and target region information; determining at least one target contour according to the contour information and the target region information, all or part of each target contour being located in a target region; reducing the first coordinates of multiple contour points in each target contour to obtain the second position information of each target contour, the second position information of the target contour comprising the second coordinates of the multiple contour points in the target contour, and the magnitude of reduction of the coordinates of each contour point being equal; and according to the second position information of each target contour and the target region information, generating a target image corresponding to the target region, the target image comprising contours located in the target region. The method reduces the size of the target image, and further ensures the accuracy of the generated target image.