Circle Detection via Perpendicular Bisector Intersection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for identifying circular shapes in images, such as those used in safe driving support systems, face high computational costs due to the need for extensive edge detection and center calculation using all points on the circle's circumference, which increases processing time and memory requirements.

Innovation Solution

An image identification apparatus that extracts contour images from input frames, detects pixels in regions separated by predetermined distances, calculates the intersection points of perpendicular bisectors of line segments formed by detected pixels, and identifies circles based on these intersections, reducing computational load and memory usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If Hough transformation is used to identify circular shapes, then circle detection capability is improved, but computational cost and processing time increase

Engineering Contradiction:

Improvecircle detection capabilityVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the continuous circle detection problem into discrete sampling points. Instead of processing all pixels or edge points, the method selects specific sampling points along the circle circumference at predetermined intervals, reducing the computational burden while maintaining detection accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by using only a subset of points (sampling points) rather than all available points for circle detection. By selecting 3 or more sampling points at specific intervals along the circumference, the method achieves sufficient detection accuracy with reduced computational effort compared to using all edge pixels.

Inventive Principle:

Principle #16Partial or excessive action

2Measurement precision

If all points on the circle's circumference are used for center calculation, then identification accuracy is improved, but memory requirements and computational load increase

Engineering Contradiction:

Improveidentification accuracyVSAvoidmemory requirements

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent extracts only the necessary information for circle detection by identifying and using specific sampling points along the circumference. Instead of storing and processing all edge point coordinates, the method extracts coordinates of 3 or more sampling points at predetermined intervals, significantly reducing memory requirements while maintaining sufficient accuracy for identifying circular shapes and their centers.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS11275963B2Image identification apparatus, image identification method, and non-transitory computer-readable storage medium for storing image identification program

Publication Date: 2022.03.15 FUJITSU LTD

AI summary

An image identification apparatus includes: a memory; and a processor coupled to the memory, the processor being configured to perform processing, the processing including: acquiring a contour image where pixels indicating a contour of a subject are extracted from an input image; detecting pixels of 3 points or more over the contour from each of regions in the contour image, which are separated by a predetermined distance in a vertical direction or a horizontal direction; calculating, for each combination of pixels of every 2 points among the detected pixels, an intersection point of a perpendicular bisector of a line segment having 2 points as end points and a distance between the intersection point and one of the detected pixels; and identifying, from the input image, a circle having the intersection point as a center of the circle and the distance as a radius.