Figure 6
From: Computational On-Chip Imaging of Nanoparticles and Biomolecules using Ultraviolet Light
(a) Phase reconstruction of a holographic raw image after pixel SR and circular averaging. (b) The same FOV after background compensation and averaging over different reconstruction heights (see Fig. 5). Note that the two larger particles (>150 nm), appearing red in the bottom-right corner, are digitally peeled29 to highlight the smaller particles in (c), as detailed in the Methods section. (c) Particle candidates are automatically determined by thresholding, and each candidate is checked for focusing criteria. The insets show the peak phase vs. focusing depth values (plotted in green) of 3 physical particles (color-coded arrows: red, orange and green) and 3 noise grains (purple arrows), all verified by SEM images, as illustrated in (d), following the same color coding as the arrows in (c). The best fit parabolas (plotted in yellow) are used to assess focusing criteria (see the Methods sections): the coefficient of the quadratic term for the bottom-right noise grain is significantly larger than the threshold and the R-squared goodness of fit values for the top and bottom-left noise grains do not satisfy the threshold unlike the physical/real particles, which satisfy these criteria. (d) We show the remaining physical particles after checking the focusing criteria for each particle candidate in (c). The insets show the SEM images of the detected particles as well as the empty regions co-localized with the noise grains (numbered and outlined in purple). The smallest particle detected is 26.4 nm in diameter, denoted with the red arrow and red image frame in (c) and (d), respectively. The color coded arrows point to these particles across (a),(b) and (c).
