X-ray Container Load Detection via Background Subtraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting internal loads in X-ray images of containers face challenges due to overlapping images of internal loads and background, which reduces detection performance, especially in cases where internal loads are distributed in multiple layers and the container itself is imaged.
Innovation Solution
A system that includes an X-ray imaging unit, a database of reference images, a condition detection unit, a comparison image selection unit, a matching unit, and an internal load detection unit to remove the container background by generating a difference image between the target X-ray image and a reference image, using calibration and registration phantoms to correct geometric and brightness conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If X-ray imaging is performed on containers with internal loads distributed in multiple layers, then the imaging capability is improved, but the detection performance deteriorates due to overlapping images of internal loads and background
Solution Approach 1:
The patent segments the X-ray image into multiple layers corresponding to different depth positions within the container. By dividing the three-dimensional imaging space into discrete layers, the system can separately process and analyze internal loads at different depths, preventing the overlapping and confusion that occurs in traditional single-layer imaging approaches.
Solution Approach 2:
The patent introduces calibration phantoms as intermediary objects with known geometric and radiographic properties. These phantoms are placed within the container to serve as reference markers that help distinguish between the container background and internal loads. The known characteristics of the phantoms allow the system to identify and subtract background structures, thereby improving the detection of actual internal loads.
2Measurement precision
If the container background is included in the X-ray image, then the complete imaging is achieved, but the internal load detection accuracy deteriorates due to background interference
Solution Approach 1:
The patent extracts the container background from the X-ray image by identifying and removing the phantom structures that represent the container walls and supporting frameworks. By taking out the background components through phantom-based identification and subtraction, the system isolates the internal loads for more accurate detection without the interference of container structures.
Solution Approach 2:
The patent performs preliminary imaging and processing of the container background using calibration phantoms before the actual internal load detection. The phantoms are imaged first to establish the background template, which is then subtracted from subsequent images. This preliminary action prepares the imaging system to focus specifically on internal loads, reducing the complexity of real-time background separation.
3Measurement precision
If calibration and registration phantoms are used to correct geometric and brightness conditions, then the detection accuracy is improved, but the system complexity increases
Solution Approach 1:
The patent implements self-service functionality where the calibration and registration phantoms automatically provide their own reference information for geometric and brightness correction. The phantoms contain embedded markers and known structures that enable the system to self-calibrate without requiring external reference measurements or manual calibration procedures, thereby reducing operational complexity while maintaining high detection accuracy.
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
This system effectively detects internal loads by removing the container background from X-ray images, improving detection performance and enabling accurate identification of internal loads under various imaging conditions.
Implementation Method 1
an X-ray imaging unit including a source and a detector and configured to image a target with X-rays
Data Source
AI summary
Provided is a system for detecting an internal load by using an X-ray image of a container. The system includes an X-ray imaging unit and configured to image a target with X-rays, a database that stores an X-ray reference image obtained by imaging a target having an empty interior according to X-ray imaging information including an X-ray imaging condition, a condition detection unit that acquires X-ray imaging information when the target is imaged, a comparison image selection unit that selects an X-ray reference image from the database, a matching unit that matches a difference between the X-ray image of the target and the selected X-ray reference image, and an internal load detection unit that obtains a difference image from the X-ray image and the X-ray reference image, and detects an internal loading image having no background of the X-ray image of the target through the obtained difference image.