X-Ray Inspection Imaging for Multi-Angle Foreign Matter Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional X-ray inspection methods face challenges in accurately detecting foreign matters and maintaining posture changes or shape variations during transportation due to fixed observation directions, leading to potential false detections and the need for multiple X-ray sources and detectors, resulting in a large and costly apparatus.
Innovation Solution
An X-ray inspection apparatus and method that generates pseudo inspection images from multiple observation directions using a pseudo three-dimensional information generation model and a multi-layer neural network to accurately detect foreign matters and deformities by creating pseudo inspection images with different observation angles, allowing for a simple and cost-effective setup.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple X-ray sources and detectors are used to obtain X-ray images from multiple directions, then the inspection accuracy for foreign matter detection is improved, but the device size and cost increase
Solution Approach 1:
The patent creates pseudo three-dimensional information and pseudo inspection images by copying and processing the original X-ray image data through a learning device. This allows simulation of multiple observation directions without physically implementing multiple X-ray sources and detectors, thereby maintaining high inspection accuracy while reducing device complexity and size
2Device complexity
If X-ray images are obtained from a fixed observation direction, then the device structure is simplified, but the inspection reliability decreases when the inspection object changes posture or shape
Solution Approach 1:
The patent dynamically generates pseudo inspection images with different observation directions based on the original X-ray image data. The learning device adapts the inspection images to various possible postures and shapes of the inspection object, allowing the system to maintain high inspection reliability regardless of the actual posture of the object being inspected, while using a simple fixed device structure
3Measurement precision
If independent component analysis is used to separate foreign matter components, then foreign matter detection accuracy is improved, but the system cannot handle shape changes and posture variations of the inspection object
Solution Approach 1:
The patent changes the parameter of observation direction by generating pseudo inspection images from multiple angles using the learning device. This allows the system to handle various shapes and postures of inspection objects by selecting appropriate observation directions, while maintaining the foreign matter detection accuracy achieved through independent component analysis. The system can now adapt to shape changes and posture variations without sacrificing detection precision
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 accurate detection of foreign matters and deformities regardless of posture or shape changes during transportation, preventing false detections and reducing the size and cost of the X-ray inspection apparatus.
Implementation Method 1
an X-ray generator (21) that generates an X-ray of a predetermined energy band which passes through the inspection object W
Implementation Method 2
a learning device that generates pseudo three-dimensional information about the inspection object W, based on the X-ray image data (D1)
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An X-ray inspection apparatus, comprising: an X-ray generator (21); an X-ray detector (24); and a determination unit (33) that determines a quality state of an inspection object (W) based on an X-ray detection signal, wherein the X-ray inspection apparatus is provided with: an X-ray image storing unit (31) that stores a first inspection image (Pe), corresponding to the X-ray detection signal outputted from the X-ray detector, whose observation direction is the direction in which the X-rays transmits the inspection object; a pseudo three-dimensional information generation model (41) that generates a pseudo three-dimensional information regarding a type of object to be learned; and an inspection image generation unit (42) that creates a second inspection image (Ps1, Ps2) regarding the type of object to be learned having the observation direction different from the first inspection image (Pe), based on the first inspection image (Pe) regarding the type of object to be learned, the determination unit (33) performs the determination, based on at least the second inspection image (Ps1, Ps2) created by the inspection image generation unit (42).