Biomedical Engineering Reference
In-Depth Information
Fig. 4.1
Schematic illustration of our partially coherent lensfree on-chip holography platform is
shown (
left
). The objects are placed directly on a digital sensor array with typically <5mm distance
to its active area. A partially coherent light source, such as an LED, is placed
4-10 cm away
from the sensor to illuminate the objects recording their digital in-line holograms with unit fringe
magnification over a large field of view. A typical full FOV .
24 mm
2
/ holographic image of a
heterogeneous sample recorded using a 5MP sensor with 2.2 m pixel size is shown (
right
). The
insets
show zoomed lensfree holograms for different micro-objects
where R.x;y;
z
/ is the reference wave, that is, the part of the wavefront that is not
scattered, s.x;y;
z
/ is the scattered field, and z
0
is the position of the digital detector
array.
Typically in digital in-line holography [
26
,
27
], the object is placed much closer to
the illumination source than the detector, and a laser source (after being filtered by,
e.g., a submicron aperture) is used for illumination to provide sufficient coherence
and power. As a result of this, the imaging field of view (FOV) is typically
much smaller than the detector size. In the lensfree on-chip imaging configuration
discussed in this chapter (see Fig.
4.1
a), the object to be imaged is placed much
closer to the detector array rather than to the illumination source. This permits using
a simple light source such as a light-emitting diode (LED) while maintaining a
sufficiently large spatial coherence diameter for hologram formation at the detector
array. Also, since the object is placed several centimeters away from the illumination
source, a large aperture of, for example,
0:1 mm diameter can be used, allowing
easier and more efficient light coupling compared to submicron apertures used in the
traditional in-line holography schemes. More importantly, from the perspective of
imaging applications that require high throughput, the presented configuration with
unit magnification has the entire active area of the sensor as the imaging FOV, which
can be up to several squared centimeters. A major trade-off that is made in return for
such advantages and simplicity is that the spatial resolution in our scheme is now
strongly affected by the pixel size at the detector array. This limitation, however,
can be addressed by pixel super-resolution techniques which will be discussed in
greater detail in Sects.
4.5
and
4.6
.
Once a lensfree hologram is recorded, it can be digitally propagated to the
object plane to undo the effect of diffraction that occurred between the object
and sensor planes. In our geometry, due to its long distance from the sensor, the
