3D Millimeter Wave Parcel Imaging for High-Throughput Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing high-frequency electromagnetic imaging technologies for detecting contraband in mail and packages face challenges in achieving high throughput with improved accuracy.
Innovation Solution
A system utilizing millimeter wave transmitting and receiving antennas, image reconstruction, and a logistic regression classifier to create a 3D millimeter wave image and estimate the probability of anomalies in parcels, triggering diversion when the probability exceeds a threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If higher-frequency electromagnetic imaging technologies are used to detect contraband, then detection accuracy is improved, but throughput is reduced
Solution Approach 1:
The imaging system is divided into multiple sensor modules that scan different angular positions independently. Each module captures a portion of the complete image, allowing parallel processing and faster acquisition while maintaining high-frequency electromagnetic imaging capabilities for accurate contraband detection
Solution Approach 2:
The system performs preliminary scanning at multiple angular positions simultaneously using multiple sensor modules before complete image reconstruction. This preliminary action at different angles enables faster data collection while the final image reconstruction maintains high detection accuracy for contraband items
2Productivity
If multiple sensor modules are used to increase throughput, then productivity is improved, but device complexity increases
Solution Approach 1:
Multiple sensor modules perform identical scanning functions at different angular positions, with each module being a standardized, interchangeable unit. This universal design allows the system to increase throughput by simply adding more identical modules rather than designing complex different components, thereby managing system complexity while improving productivity
3Measurement precision
If complete 3D image reconstruction is performed, then detection accuracy is improved, but processing time increases
Solution Approach 1:
The system performs complete 3D image reconstruction from multiple angular views to achieve high detection accuracy, but uses classification algorithms that process only the most informative features and voxels rather than analyzing every pixel in the complete reconstructed image. This partial processing approach maintains high anomaly detection accuracy while significantly reducing the time required for image analysis
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 high-throughput, non-destructive imaging with accurate detection of non-paper inclusions like metalized plastic and metal items, achieving resolution better than 1 cm and reducing the need for human labor.
Implementation Method 1
an array of millimeter wave transmitting and receiving antennas disposed proximate to the conveyor and operative to produce millimeter wave data
Implementation Method 2
collecting reflection measurements from the sensor modules
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
Systems and methods for detecting anomalies in an envelope including scanning an envelope moving along a conveyor with multiple millimeter wave imaging sensor modules and collecting reflection measurements from the sensor modules. The systems and methods can further include constructing a one or more voxel thick horizontal 3D image of the envelope and then creating an image of the envelope comprising the magnitude of the mean of the one or more voxel layers. The image of the envelope can be fed into a binary logistic regression classifier to provide a probability that the envelope contains an anomaly from the classifier. The systems and methods can include causing the envelope to be diverted from the conveyor for further inspection when the probability is greater than a selected threshold.