READ ME File For 'Dataset supporting the publciation "Diatom lensless imaging using laser scattering and deep learning"'


ReadMe Author: James A Grant-Jacob  University of Southampton https://orcid.org/0000-0002-4270-4247

This dataset supports the publication:
AUTHORS: Ben Mills, Michalis N. Zervas and James A. Grant-Jacob
TITLE: Diatom lensless imaging using laser scattering and deep learning
JOURNAL: ACS ES&T Water
PAPER DOI IF KNOWN: 


This dataset contains:

Figure_1. a) Diagram of the experimental setup used to simultaneously image the diatoms and capture their scattering pattern from laser illumination. Three neural networks employed in this work were b) image generation, c) speed classification, and d) angular direction classification.

Figure_2. Experimental diatom scattering patterns (column 1) and the associated neural network predicted images (column 2). Also included are the corresponding experimental images (column 3) and the difference between the two images (column 4) (one minus the other, such that white pixels indicate less difference).

Figure_3. a) Predicted diatom velocity compared with experimental velocity. b) Predicted diatom angle compared with experimental angle when moving at a velocity of 0.2 mm/s. Test data 1 is data from angles used during training of the neural network, whilst test data 2 is data from angles not present in the training data.

Table_1 PSNR, SSIM, RMSE and PIQE for the predicted and experimental pollen images shown in Figure 2.

TOC_Graphic Table of contents graphic.

Licence: CC-BY

Related projects:
EPSRC grant EP/T026197/1
EPSRC grant EP/W028786/1
EPSRC grant  EP/Z002567/1
Date that the file was 03, 2025