Search for: All records

Coded aperture imaging has emerged as a solution to enhance light sensitivity and enable imaging in challenging conditions. However, the computational expense of image reconstruction poses limitations in processing efficiency. To address this, we propose a direct classification method using convolutional neural networks. By leveraging raw coded measurements, our approach eliminates the need for explicit image reconstruction, reducing computational overhead. We evaluate the effectiveness of this approach compared to traditional methods on the MNIST and CIFAR10 datasets. Our results demonstrate that direct image classification using raw coded measurements achieves comparable performance to traditional methods while reducing computational overhead and enabling real-time processing. These findings highlight the potential of machine learning in enhancing the decoding process and improving the overall performance of coded aperture imaging systems. 

We investigate methods for learning partial differential equation (PDE) models from spatio-temporal data under biologically realistic levels and forms of noise. Recent progress in learning PDEs from data have used sparse regression to select candidate terms from a denoised set of data, including approximated partial derivatives. We analyse the performance in using previous methods to denoise data for the task of discovering the governing system of PDEs. We also develop a novel methodology that uses artificial neural networks (ANNs) to denoise data and approximate partial derivatives. We test the methodology on three PDE models for biological transport, i.e. the advection–diffusion, classical Fisher–Kolmogorov–Petrovsky–Piskunov (Fisher–KPP) and nonlinear Fisher–KPP equations. We show that the ANN methodology outperforms previous denoising methods, including finite differences and both local and global polynomial regression splines, in the ability to accurately approximate partial derivatives and learn the correct PDE model.