3D Surface Projection for Angle-Independent Graphic Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing technologies struggle with accurate recognition of targets due to variations in image capture angles and target colors, leading to inconsistent target identification.

Innovation Solution

An image processing method and apparatus that projects surface points from a three-dimensional image to a two-dimensional coordinate system to determine a minimum circumscribed graphic, allowing for precise recognition of targets by comparing the area ratio of the projected polygon to a predefined threshold.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional image processing methods are used for target recognition, then the processing is simple and fast, but the recognition accuracy is low due to angle and color variations

Engineering Contradiction:
Improvetarget recognition accuracyVSAvoidimage processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transforms the target recognition problem from two-dimensional image space to three-dimensional spatial coordinates. By extracting 3D coordinates of target points and calculating spatial geometric features (circumcenter, circumradius, area), the method achieves angle-independent recognition that overcomes the limitations of traditional 2D image processing affected by capture angles and colors.

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

2Measurement precision

If 3D coordinate transformation and geometric feature calculation are implemented, then target recognition accuracy is improved, but the computational complexity increases

Engineering Contradiction:
Improvetarget recognition accuracyVSAvoidcomputational power consumption
Core Design Contradiction:
Measurement precisionVSPower

Solution Approach 1:

The patent extracts only the essential geometric features (3D coordinates of target points, circumcenter, circumradius, and area) needed for recognition, rather than processing the entire 3D point cloud or performing complex geometric transformations. This selective extraction reduces computational overhead while maintaining high recognition accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If multiple geometric features are calculated for each surface, then the recognition precision is enhanced, but the processing time increases

Engineering Contradiction:
Improvegraphic recognition precisionVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent pre-calculates and stores the geometric features (circumcenter, circumradius, area) for each extracted graphic during the image processing stage. These pre-computed features are then directly compared with threshold values for rapid recognition, avoiding the need to perform complex geometric calculations during real-time recognition operations.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4198776B1Image processing method and apparatus, electronic device, and computer readable storage medium
Publication Date: 2026.04.22 TENCENT TECHNOLOGY (SHENZHEN) CO LTD
  • EP4198776B1 patent drawingFigure 1
  • EP4198776B1 patent drawingFigure 2
  • EP4198776B1 patent drawingFigure 3

AI summary

The present disclosure provides an image processing method and apparatus, an electronic device, and a computer-readable storage medium. The method includes: obtaining a first three-dimensional image in a three-dimensional coordinate system (S01); determining multiple surfaces in the first three-dimensional image, the multiple surfaces each comprising surface points respectively represented by three-dimensional coordinates (S02); projecting the surface points of each of the multiple surfaces to a two-dimensional coordinate system, in which a plane extracted from the respective surface is located, to obtain two-dimensional coordinates respectively representing the surface points (S03); determining, for the plane extracted from the respective surface, a polygon and a minimum circumscribed graphic of the polygon according to the two-dimensional coordinates respectively representing the surface points, the minimum circumscribed graphic having a same shape as a first graphic to be recognized (S04); and recognizing a surface whose area ratio obtained by dividing its corresponding polygon by its corresponding minimum circumscribed graphic is greater than a pre-defined threshold as the first graphic in the first three-dimensional image (S05).