Adaptive Boundary Line Detection for Candidate Region Estimation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for object candidate region estimation in images face challenges such as parameter inefficiency, incomplete boundary line detection, and high computational costs, particularly in accurately detecting target objects across varying image conditions.

Innovation Solution

A candidate region estimation apparatus that determines parameters based on the ratio of boundary line densities within an image and a specified region, using these parameters to detect boundary lines and estimate candidate regions for multiple target objects with high precision.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If common parameters are used for boundary line detection, then the device complexity is reduced, but the measurement precision of boundary lines deteriorates

Engineering Contradiction:

Improveparameter setting complexityVSAvoidboundary line detection precision

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent automatically determines optimal parameters (standard deviation of Gaussian function, threshold values) based on image characteristics such as average gradient magnitude and variance. This transforms the manual parameter setting process into an automated adaptation process, resolving the contradiction between device complexity and measurement precision by dynamically adjusting parameters to match specific image conditions

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system performs self-calibration by automatically analyzing image characteristics and determining appropriate detection parameters without external intervention. The boundary line detection apparatus calculates optimal parameters from the input image itself, enabling the system to adapt to different image conditions autonomously while maintaining high detection precision

Inventive Principle:

Principle #25Self-service

2Device complexity

If the standard deviation of Gaussian function is increased, then the device complexity is reduced, but the measurement precision of boundary lines deteriorates

Engineering Contradiction:

Improveparameter configuration simplicityVSAvoidboundary line detection accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent calculates the standard deviation of the Gaussian function based on the variance of gradient magnitudes in the image. By establishing a relationship between image characteristics and optimal parameter values, the system automatically determines the appropriate standard deviation for each image, eliminating the need for manual configuration while ensuring high detection accuracy

Inventive Principle:

Principle #35Parameter changes

3Productivity

If threshold values are increased, then the productivity is improved, but the measurement precision of boundary lines deteriorates

Engineering Contradiction:

Improveboundary line detection speedVSAvoidboundary line detection accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent determines threshold values based on the average gradient magnitude and its variance in the image. By establishing adaptive thresholding that considers image-specific characteristics, the system achieves both high detection speed and high accuracy, as the thresholds are optimized for each image rather than using fixed high values that would compromise precision

Inventive Principle:

Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables precise estimation of candidate regions for multiple target objects by adaptively determining detection parameters, reducing erroneous detections and computational costs, and improving the accuracy of boundary line detection across different image conditions.

Implementation Method 1

a two-dimensional Gaussian function obtained by finding a first derivative is superimposed in the x direction and the y direction of an image

Methodology Applied

Scientific EffectGaussian function:

Implementation Method 2

the magnitude of the gradient and the gradient direction are obtained based on the derivative values of the pixels

Methodology Applied

Scientific EffectGradient calculation:

Data Source

PatentUS12008765B2Candidate region estimation device, candidate region estimation method, and program

Publication Date: 2024.06.11 NIPPON TELEGRAPH & TELEPHONE CORP

AI summary

The present invention makes it possible to estimate, with high precision, a candidate region indicating each of multiple target objects included in an image. A parameter determination unit 11 determines parameters to be used when detecting a boundary line of an image 101 based on a ratio between a density of boundary lines included in an image 101 and a density of boundary lines in a region indicated by region information 102 indicating the region including at least one of the multiple target objects included in the image 101. A boundary line detection unit 12 detects the boundary line in the image 101 using the parameter. For each of the multiple target objects included in the image 101, the region estimation unit 13 estimates the candidate region of the target object based on the detected boundary line.