Image Inspection Apparatus Mode Switching for Rule and Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image inspection systems face challenges in accurately distinguishing between non-defective and defective products, especially when the inspection target's features change with environmental conditions, and they require complex setting flows when switching between rule-based and learning-based inspection modes.

Innovation Solution

An image inspection apparatus with a mode switching unit that allows for seamless transition between rule-based and learning-based inspection modes, featuring an imaging setting unit, master image registering unit, inspection type selecting unit, image processing tool selecting unit, and a distinguishing device generating unit, which enables users to set imaging conditions, select image processing tools, and generate distinguishing devices for accurate pass/fail determinations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If rule-based inspection is used to allow user control over pass/fail determination, then ease of operation is improved, but measurement precision deteriorates when feature values change with imaging conditions

Engineering Contradiction:

Improveuser control capabilityVSAvoidpass/fail determination accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The system dynamically switches between rule-based inspection mode (for user control) and learning-based inspection mode (for high precision). The mode switching unit allows the system to adapt its operation mode based on the inspection requirements, combining the advantages of both approaches.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

A mode switching unit acts as an intermediary between rule-based and learning-based inspection systems. This mediator allows the user to select the appropriate inspection mode, enabling the system to leverage both user-controlled rule-based methods and AI-driven learning-based methods for optimal performance.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If learning-based inspection is used to automatically determine feature values, then ease of operation is improved, but reliability deteriorates when encountering unknown defective states

Engineering Contradiction:

Improveinspection setting simplicityVSAvoiddetermination stability

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The system dynamically switches between rule-based inspection mode (for reliability with known defects) and learning-based inspection mode (for ease of operation). The mode switching unit allows the user to select the appropriate inspection mode based on the inspection requirements and the nature of the defects being sought.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If both rule-based and learning-based inspection modes are implemented, then adaptability is improved, but device complexity increases

Engineering Contradiction:

Improveinspection mode flexibilityVSAvoidsetting flow complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The image inspection apparatus is designed with multi-functionality, supporting both rule-based and learning-based inspection modes within a single system. The mode switching unit enables the system to perform different types of inspections based on user selection, making the apparatus versatile while managing complexity through unified architecture.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Measurement precision

If feature values are set based on master image to improve measurement precision, then measurement precision is improved, but loss of time increases due to repeated adjustments

Engineering Contradiction:

Improvefeature value accuracyVSAvoidsetting adjustment time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-registering master images and pre-determining feature values and thresholds before actual inspection. This preparation work is done once during system setup, eliminating the need for repeated adjustments during production inspections, thus saving time while maintaining precision.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system creates a copy of the ideal inspection target in the form of a master image. This master image serves as a reference template that is registered once and then used repeatedly for comparisons during inspection, eliminating the need to重新 adjust feature values for each inspection batch.

Inventive Principle:

Principle #26Copying

Data Source

PatentUS11042977B2Image inspection apparatus and setting method for image inspection apparatus

Publication Date: 2021.06.22 KEYENCE CORP

AI summary

To enable inspection setting flows appropriate for a rule-based inspection mode and a learning-based inspection mode to be easily created when these modes are implemented in an image inspection apparatus. The rule-based inspection mode and the learning-based inspection mode are implemented in the image inspection apparatus. In the setting mode in a rule-based inspection, the user can set an imaging condition, select an image processing tool, set the application range of an image processing tool, and adjust the parameters of the image processing tool. In the setting mode in a learning-based inspection, a non-defective product image and a defective product image are input to generate a distinguishing device.