Automated Defect Detection Using Prior Inspection Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated inspection techniques for images captured by borescopes, such as those used in aircraft engine blade inspection, are prone to errors due to limited defect classification and fail to detect defects outside predefined categories, relying heavily on human interpretation and low-level feature extraction.
Innovation Solution
An automated defect detection method and system that utilizes prior inspection data by capturing and comparing current images with stored prior images, employing image registration and analysis to determine transformations and update the database, allowing for the identification of defects and tracking their progression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automated inspection techniques are used with predefined defect classes, then human error is reduced, but defects outside predefined categories are not detected
Solution Approach 1:
The system performs preliminary actions by capturing and storing reference images of defect-free components before inspection. These reference images are used to automatically identify defects by comparing current inspection images against the stored references, enabling detection of any defect type without requiring predefined categories.
Solution Approach 2:
The system creates copies of defect-free component images from previous inspections and stores them in a database. These copied reference images serve as the basis for automatic defect detection, allowing the system to identify any deviations from the known good state without human intervention or predefined defect classes.
2Adaptability or versatility
If human inspectors manually examine images, then all defect types can be detected, but human inattention and errors occur
Solution Approach 1:
The inspection system performs self-service by automatically comparing current images against stored reference images and generating defect reports without human intervention. The system autonomously identifies defects, determines their locations, and creates inspection records, eliminating human inattention and errors while maintaining comprehensive defect detection capability.
Solution Approach 2:
The system replaces the mechanical human inspection process with an automated computer-based system that uses image processing algorithms. The mechanical action of human inspectors examining images is substituted by electronic image comparison algorithms that automatically detect deviations from reference images, improving both reliability and versatility.
3Productivity
If low-level feature extraction is used, then processing speed is improved, but complex defects are not accurately identified
Solution Approach 1:
The system uses copied reference images from previous inspections as the basis for comparison. By comparing current images against these stored references, the system can identify complex defects through direct visual comparison rather than relying solely on low-level feature extraction, thereby improving detection precision while maintaining processing efficiency.
Solution Approach 2:
The system extracts and stores reference images of defect-free components from previous inspections. These extracted reference images are then used for automatic comparison with current inspection images, allowing the system to identify defects including complex ones, without requiring complex real-time analysis algorithms.
Data Source
AI summary
A system and method for performing automated defect detection by utilizing data from prior inspections is disclosed. The system and method may include providing a image capture device for capturing and transmitting at least one current image of an object and providing a database for storing at least one prior image from prior inspections. The system and method may further include registering the at least one current image with the at least one prior image, comparing the registered at least one current image with the at least one prior image to determine a transformation therebetween and updating the database with the at least one current image.