Edge Extraction Apparatus Using Length-Enhanced Thresholding

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional edge detection methods struggle to accurately extract the contour of an article from an image due to noise detection and operational inefficiencies, particularly when using differential filters that rely solely on brightness information and fail to consider edge length and curvature.

Innovation Solution

An edge extraction device that calculates edge strength and length, performs edge enhancement processing by multiplying these values, and applies adjustable thresholding to binarize edges, reducing noise and improving operability through a single parameter adjustment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Difficulty of detecting and measuring

If differential filters are used for edge detection based on brightness information, then edge detection can be performed, but noise such as pattern and color changes inside articles are also detected

Engineering Contradiction:

Improveedge detection capabilityVSAvoidcontour detection accuracy

Core Design Contradiction:

Difficulty of detecting and measuringVSMeasurement precision

Solution Approach 1:

The invention changes the parameters used for edge detection from only brightness information to a combination of edge strength, edge length, and curvature degree. By introducing length and curvature as additional parameters, the system can distinguish between true contour edges and internal patterns, thereby improving measurement precision while maintaining edge detection capability

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The invention replaces the traditional differential filter method with a new approach that calculates edge strength from gradient information and combines it with geometric properties (length and curvature). This substitution allows for more accurate contour detection by considering multiple factors rather than relying solely on brightness gradients

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If cubic spline curve is used to estimate edge direction and calculate edge strength, then directional characteristic of edge can be obtained, but the process is time-consuming and difficult to automate

Engineering Contradiction:

Improveedge direction estimation accuracyVSAvoidprocessing speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The invention replaces the complex cubic spline curve method with a simpler calculation approach that uses gradient information directly to estimate edge direction and calculate edge strength. This simpler method achieves comparable accuracy without the computational burden of spline curve fitting, thereby improving processing speed and enabling automation

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The invention performs preliminary calculation of gradient information that can be directly used for edge direction estimation and edge strength calculation. By preparing this gradient data in advance, the system avoids time-consuming iterative curve fitting processes while maintaining accurate edge detection

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If threshold process is performed using length and curvature degree as independent parameters, then edge detection can be performed, but operability is deteriorated and intuitive adjustment is difficult

Engineering Contradiction:

Improveedge detection accuracyVSAvoidparameter adjustment ease

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The invention merges the multiple parameters (edge strength, edge length, curvature degree) into a single integrated threshold value. This consolidation allows operators to adjust edge detection sensitivity through one parameter rather than managing multiple independent parameters, significantly improving ease of operation while maintaining detection accuracy

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The single threshold value serves multiple functions by simultaneously controlling the acceptance criteria for edge strength, length, and curvature. This universal parameter enables intuitive adjustment of edge detection behavior across different scenarios without requiring specialized knowledge of each individual parameter's optimal value

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentEP2296114B1Edge extraction apparatus, surveying equipment, and program

Publication Date: 2016.09.28 TOPCON CORPORATION

Figure 1~2
Figure 3~4
Figure 5A~5C

AI summary

An edge extraction device can reduce detected noise other than a contour of an article, and can improve the operability. The edge extraction device includes: an edge detection section which calculates edge strength from an image and detects an edge; a labeling processing section which performs labeling processing on the edge detected by the edge detection section and calculates a length of the edge; an edge enhancement processing section which performs edge enhancement processing by using a value corresponding to the length of the edge, which is calculated by the labeling processing section, and the edge strength, which is calculated by the edge detection; and an edge extraction section which performs binarization processing on a value of the image, which is enhanced by the edge enhancement processing section, by using an adjustable threshold value, and extracts a predetermined edge.