Dynamic Threshold Inspection for Image Edge Defect Detection

Solution Overview

Problem

Existing image inspection devices face challenges in accurately detecting defects, particularly at image edges, due to the limitations of simply decreasing the threshold value for defect detection when comparing correct and target images.

Innovation Solution

An inspection device that acquires image information from both correct and target images, extracts edge information, and adjusts the threshold value using brightness and color information to improve defect detection accuracy by dynamically setting the threshold for each pixel based on edge and brightness characteristics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a threshold value for detecting a defect of an edge portion is decreased to improve defect detection sensitivity, then defect detection sensitivity is improved, but erroneous detection of normal edge variations as defects increases

Engineering Contradiction:

Improvedefect detection sensitivityVSAvoiderroneous detection rate

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent applies local quality by setting different threshold values for different regions of the image. Specifically, edge portions are assigned different threshold criteria compared to non-edge portions. This allows the inspection device to use a lower threshold for detecting defects in non-edge areas while maintaining a higher threshold for edge areas, thereby reducing erroneous detection of normal edge variations while still detecting actual defects.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent implements dynamics by dynamically adjusting the threshold value based on the local characteristics of each pixel. The inspection device determines whether each pixel belongs to an edge portion or non-edge portion and automatically selects appropriate threshold values accordingly. This dynamic adaptation allows the system to optimize defect detection sensitivity for each region without manual intervention.

Inventive Principle:

Principle #15Dynamics

2Reliability

If a threshold value for detecting a defect of an edge portion is increased to reduce erroneous detection, then erroneous detection rate is reduced, but defect detection sensitivity deteriorates

Engineering Contradiction:

Improveerroneous detection rateVSAvoiddefect detection sensitivity

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent applies local quality by setting different threshold values for different regions of the image. Specifically, edge portions are assigned different threshold criteria compared to non-edge portions. This allows the inspection device to use a lower threshold for detecting defects in non-edge areas while maintaining a higher threshold for edge areas, thereby reducing erroneous detection of normal edge variations while still detecting actual defects.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent implements dynamics by dynamically adjusting the threshold value based on the local characteristics of each pixel. The inspection device determines whether each pixel belongs to an edge portion or non-edge portion and automatically selects appropriate threshold values accordingly. This dynamic adaptation allows the system to optimize defect detection sensitivity for each region without manual intervention.

Inventive Principle:

Principle #15Dynamics

3Device complexity

If a single threshold value is used for the entire image to simplify the inspection process, then device complexity is reduced, but defect detection accuracy deteriorates

Engineering Contradiction:

Improvethreshold setting complexityVSAvoiddefect detection accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies self-service by enabling the inspection device to automatically determine edge portions and select appropriate threshold values without requiring manual configuration. The device performs edge detection, identifies edge portions, and autonomously adjusts threshold values based on the detected edge characteristics. This self-service capability maintains high defect detection accuracy while keeping the operation simple for users.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent implements dynamics by dynamically adjusting the threshold value based on the local characteristics of each pixel. The inspection device determines whether each pixel belongs to an edge portion or non-edge portion and automatically selects appropriate threshold values accordingly. This dynamic adaptation allows the system to optimize defect detection sensitivity for each region without manual intervention.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS12125243B2Inspection device, image forming apparatus, and non-transitory computer readable medium storing inspection program

Publication Date: 2024.10.22 FUJIFILM BUSINESS INNOVATION CORP

AI summary

An inspection device includes a processor configured to acquire image information of each of a correct image and a target image as an inspection target, extract edge information of each of the correct image and the target image by using the acquired image information, obtain a difference image between the correct image and the target image, and change a threshold value for detecting a defect by using brightness information or color information of the correct image along with the edge information and detect a defect of the target image by using the difference image and the threshold value.