Bayes-DIC Net: Estimating Digital Image Correlation Uncertainty with Bayesian Neural Networks

TL;DR

This work tackles the realism and reliability challenges in Digital Image Correlation by introducing a high-quality synthetic dataset generated from non-uniform B-spline displacement fields and a novel Bayes-DIC Net architecture that jointly extracts detail and global information. The network employs a bilateral encoder–decoder design with a single skip-based fusion path and lightweight blocks to expand the receptive field efficiently. By integrating dropout during inference, Bayes-DIC Net becomes Bayesian, delivering both displacement predictions and uncertainty estimates on unlabeled data, which enhances practical applicability. Experimental results on a large synthetic dataset show state-of-the-art accuracy and meaningful uncertainty maps, underscoring the method's potential for robust real-world DIC tasks.

Abstract

This paper introduces a novel method for generating high-quality Digital Image Correlation (DIC) dataset based on non-uniform B-spline surfaces. By randomly generating control point coordinates, we construct displacement fields that encompass a variety of realistic displacement scenarios, which are subsequently used to generate speckle pattern datasets. This approach enables the generation of a large-scale dataset that capture real-world displacement field situations, thereby enhancing the training and generalization capabilities of deep learning-based DIC algorithms. Additionally, we propose a novel network architecture, termed Bayes-DIC Net, which extracts information at multiple levels during the down-sampling phase and facilitates the aggregation of information across various levels through a single skip connection during the up-sampling phase. Bayes-DIC Net incorporates a series of lightweight convolutional blocks designed to expand the receptive field and capture rich contextual information while minimizing computational costs. Furthermore, by integrating appropriate dropout modules into Bayes-DIC Net and activating them during the network inference stage, Bayes-DIC Net is transformed into a Bayesian neural network. This transformation allows the network to provide not only predictive results but also confidence levels in these predictions when processing real unlabeled datasets. This feature significantly enhances the practicality and reliability of our network in real-world displacement field prediction tasks. Through these innovations, this paper offers new perspectives and methods for dataset generation and algorithm performance enhancement in the field of DIC.

Figures (8)

• Figure 1: Schematic of the dataset generation workflow.
• Figure 2: The architecture of the proposed network. During the Feature Extraction stage, the feature maps with darker colors correspond to higher levels of information.
• Figure 3: The architecture of the Down Block and Up Block.
• Figure 4: The architecture of the Small Block and Wide Block.
• Figure 5: The architecture of the Fusion Block and Prediction Head.