X-Ray CT Reconstruction of Additively Manufactured Parts using 2.5D Deep Learning MBIR

TL;DR

This paper introduces a fast, deep learning-based CT reconstruction method for additive manufacturing parts that leverages CAD models and simulated measurements to produce high-quality 3D images in real-time.

Contribution

It presents a novel 2.5D deep learning approach trained on simulated data to enhance CT reconstruction speed and quality for AM parts.

Findings

Achieves high-quality 3D reconstructions rapidly using GPU acceleration.

Utilizes CAD models as priors to improve reconstruction accuracy.

Enables real-time CT imaging suitable for manufacturing inspection.

Abstract

In this paper, we present a deep learning algorithm to rapidly obtain high quality CT reconstructions for AM parts. In particular, we propose to use CAD models of the parts that are to be manufactured, introduce typical defects and simulate XCT measurements. These simulated measurements were processed using FBP (computationally simple but result in noisy images) and the MBIR technique. We then train a 2.5D deep convolutional neural network [4], deemed 2.5D Deep Learning MBIR (2.5D DL-MBIR), on these pairs of noisy and high-quality 3D volumes to learn a fast, non-linear mapping function. The 2.5D DL-MBIR reconstructs a 3D volume in a 2.5D scheme where each slice is reconstructed from multiple inputs slices of the FBP input. Given this trained system, we can take a small set of measurements on an actual part, process it using a combination of FBP followed by 2.5D DL-MBIR. Both steps can be rapidly performed using GPUs, resulting in a real-time algorithm that achieves the high-quality of MBIR as fast as standard techniques. Intuitively, since CAD models are typically available for parts to be manufactured, this provides a strong constraint "prior" which can be leveraged to improve the reconstruction.