Corner Point Detection Using Region Proportion Filtering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing contour-based corner point detection methods in image processing have high computational complexity, making them inefficient for high-efficiency detection scenarios.

Innovation Solution

A method that determines corner points based on the region proportion of a predetermined region centered on the corner point, reducing the need for additional comparisons and traversals, and utilizing region of interest and polygon approximation to filter out interference points.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If contour-based corner point detection method is used, then detection accuracy is improved, but computational complexity increases

Engineering Contradiction:
Improvedetection accuracyVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the necessary information (region proportion) from the contour data to determine corner points, eliminating the need for complex multi-point traversals and comparisons while maintaining detection accuracy

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the detection parameter from traditional multi-point geometric comparisons to a single region proportion metric, simplifying the computational process while preserving corner point identification accuracy

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS12505556B2Corner point detection method and apparatus
Publication Date: 2025.12.23 CONTEMPORARY AMPEREX TECHNOLOGY (HONG KONG) LIMITED
  • US12505556B2 patent drawing
  • US12505556B2 patent drawing
  • US12505556B2 patent drawing

AI summary

A corner point detection method and apparatus, which may improve the efficiency of corner point detection is disclosed. The corner point detection method includes: obtaining a plurality of corner points of a target object in an image to be detected, where the plurality of corner points are corner points on a contour of a first object in the image to be detected, and the contour of the first object includes a contour of the target object; obtaining a region proportion corresponding to each of the plurality of corner points, where the region proportion is a proportion of a region of the first object in a predetermined region centered on the corner point; and determining a target corner point of the target object from the plurality of corner points based on the region proportion.