3D Source Localization Imaging System Using 2D Angular Buffering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing imaging systems for source localization face significant computational burdens when reconstructing radiation events in three-dimensional space, with the burden increasing linearly with the number of image elements.

Innovation Solution

The proposed imaging system employs a combined two-dimensional to three-dimensional approach, where events are first imaged in a common two-dimensional angular space, corrected for detector pose, and then projected into three-dimensional space, reducing the computational load by reconstructing multiple events simultaneously rather than on an event-by-event basis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If events are reconstructed one-by-one in full three-dimensional space, then accurate source localization is achieved, but computational burden increases significantly

Engineering Contradiction:
Improvesource localization accuracyVSAvoidcomputational burden
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the reconstruction process into two distinct stages: first projecting all events onto a two-dimensional angular space buffer, then performing a single three-dimensional reconstruction operation. This segmentation separates the frequent angular projection operations from the computationally intensive 3D reconstruction, reducing overall computational burden while preserving localization accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary projection of radiation events onto a two-dimensional angular space buffer before the final three-dimensional reconstruction. This preliminary action organizes and pre-processes the event data in a computationally efficient manner, so that when 3D reconstruction is performed, the computational workload is significantly reduced compared to performing full 3D reconstruction on each individual event.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If the number of image elements is increased, then source localization precision is improved, but computational burden increases linearly

Engineering Contradiction:
Improvesource localization precisionVSAvoidcomputational efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent introduces a two-dimensional angular space as an intermediate dimension between the detected radiation events and the final three-dimensional source localization. By projecting events onto this 2D angular buffer first, the system reduces the dimensionality of the reconstruction problem, allowing for higher precision with reduced computational burden compared to direct 3D reconstruction of all events.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP4063919B1Imaging system for three-dimensional source localization
Publication Date: 2025.04.30 H3D INC
  • EP4063919B1 patent drawingFigure 1~2
  • EP4063919B1 patent drawingFigure 3
  • EP4063919B1 patent drawingFigure 4

AI summary

An imaging system includes a detector configured to obtain radiation data from one or more sources and a controller. The controller is configured to define plurality of buffers based on at least one initial condition. The radiation data includes a plurality of events. The controller is configured to receive an individual event of the plurality of events and determine if the individual event falls within a designated current buffer. Each of the plurality of events in the current buffer is corrected for pose and aligned in a common two-dimensional space. The plurality of events in the current buffer are reconstructed into a three-dimensional space, the reconstruction being done once for each of the plurality of buffers. The controller is configured to create a three-dimensional image based in part on the reconstruction in the three-dimensional space.