Point spread function
865348
224677172
2008-07-09T22:46:12Z
Sakurambo
1070795
/* Theory */ replace image
[[Image:Convolution_Illustrated_eng.png|thumb|right|250px|Image formation in a [[confocal microscope]]: central longitudinal (XZ) slice. The 3D acquired distribution arises from the [[convolution]] of the real light sources with the PSF.]]
[[image:spherical-aberration-disk.jpg|thumb|300 px|A [[point source]] as imaged by a system with negative (top), zero (center), and positive (bottom) [[spherical aberration]]. Images to the left are defocused toward the inside, images on the right toward the outside.]]
The '''point spread function''' ('''PSF''') describes the response of an imaging system to a [[point source]] or point object. A related but more general term for the PSF is a system's [[impulse response]]. The PSF in many contexts can be thought of as the extended blob in an image that represents an unresolved object. In functional terms it is the [[spatial domain]] version of the [[modulation transfer function]]. It is a useful concept in [[Fourier optics]], [[astronomy|astronomical imaging]], electron microscopy and other imaging techniques such as [[dimension|3D]] [[microscopy]] (like in [[Confocal laser scanning microscopy]]) and [[fluorescence microscopy]]. The degree of spreading (blurring) of the point object is a measure for the quality of an imaging system. In [[coherence (physics)|incoherent]] imaging systems such as [[fluorescent]] [[microscopes]], [[telescopes]] or optical microscopes, the image formation process is linear in power and described by [[linear system]] theory. When the light is [[coherence (physics)|coherent]], image formation is linear in complex field.{{vague}} This means that when two objects A and B are imaged simultaneously, the result is equal to the sum of the independently imaged objects. In other words: the imaging of A is unaffected by the imaging of B and ''vice versa'', owing to the non-interacting property of photons. (The sum is of the light waves which may result in destructive and constructive interference at non-image planes.)
==Introduction==
By virtue of the linearity property of optical imaging systems, i.e.,
: ''Image''(''Object''<sub>1</sub> + ''Object''<sub>2</sub>) = ''Image''(''Object''<sub>1</sub>) + ''Image''(''Object''<sub>2</sub>)
the image of an object in a microscope or telescope can be computed by expressing the object-plane field as a weighted sum over 2D impulse functions, and then expressing the image plane field as the weighted sum over the ''images'' of these impulse functions. This is known as the ''superposition principle'', valid for [[linear systems]]. The images of the individual object-plane impulse functions are called point spread functions, reflecting the fact that a mathematical ''point'' of light in the object plane is ''spread'' out to form a finite area in the image plane (in some branches of mathematics and physics, these might be referred to as [[Green's functions]] or [[impulse response]] functions).
When the object is divided into discrete point objects of varying intensity, the image is computed as a sum of the PSF of each point. As the PSF is typically determined entirely by the imaging system (that is, microscope or telescope), the entire image can be described by knowing the optical properties of the system. This process is usually formulated by a [[convolution]] equation. In [[microscope image processing]] and [[astronomy]], knowing the PSF of the measuring device is very important for restoring the (original) image with [[deconvolution]].
==Theory==
The point spread function may be independent of position in the object plane, in which case it is called ''shift invariant''. In addition, if there is no distortion in the system, the image plane coordinates are linearly related to the object plane coordinates via the [[magnification]] ''M'' as:
:<math>(x_i, y_i) = (M x_o, M y_o)</math>.
If the imaging system produces an inverted image, we may simply regard the image plane coordinate axes as being reversed from the object plane axes. With these two assumptions, i.e., that the PSF is shift-invariant ''and'' that there is no distortion, calculating the image plane convolution integral is a straightforward process.
Mathematically, we may represent the object plane field as:
:<math> O(x_o,y_o) = \int\!\!\int O(u,v) ~ \delta(x_o-u,y_o-v) ~ du\, dv</math>
i.e., as a sum over weighted impulse functions, although this is also really just stating the sifting property of 2D delta functions (discussed further below). Rewriting the object transmittance function in the form above allows us to calculate the image plane field as the superposition of the images of each of the individual impulse functions, i.e., as a superposition over weighted point spread functions in the image plane using the ''same'' weighting function as in the object plane, i.e., <math>O(x_o,y_o)</math>. Mathematically, the image is expressed as:
:<math>I(x_i,y_i) = \int\!\!\int O(u,v) ~ \mathrm{PSF}(x_i - M u,y_i - M v) \, du\, dv</math>
in which PSF(''x<sub>i</sub>'' − ''Mu'',''y<sub>i</sub>'' − ''Mv'') is the image of the impulse function δ(''x<sub>o</sub>'' − ''u'',''y<sub>o</sub>'' − ''v'').
The 2D impulse function may be regarded as the limit (as side dimension ''w'' tends to zero) of the "square post" function, shown in the figure below (click to enlarge).
[[Image:SquarePost.jpg|Square Post Function|left|thumb|220px]]
We imagine the object plane as being decomposed into square areas such as this, with each having its own associated square post function. If the height, ''h'', of the post is maintained at 1/w<sup>2</sup>, then as the side dimension ''w'' tends to zero, the height, ''h'', tends to infinity in such a way that the volume (integral) remains constant at 1. This gives the 2D impulse the sifting property (which is implied in the equation above), which says that when the 2D impulse function, δ(''x'' − ''u'',''y'' − ''v''), is integrated against any other continuous function, ''f''(''u'',''v''), it "sifts out" the value of ''f'' at the location of the impulse, ''i''.''e''., at the point (''x'',''y'').
Since the concept of a perfect point source object is so central to the idea of PSF, it's worth spending some time on that before proceeding further. First of all, there is no such thing in nature as a perfect mathematical point source radiator; the concept is completely non-physical and is nothing more that a mathematical construct used to model and understand optical imaging systems. The utility of the point source concept comes from the fact that a point source in the 2D object plane can only radiate a perfect uniform-amplitude, spherical wave - a wave having perfectly spherical, outward travelling phase fronts with uniform intensity everywhere on the spheres (see [[Huygens-Fresnel principle]]). Such a source of uniform spherical waves is shown in the figure below (click to enlarge). We also note that a perfect point source radiator will not only radiate a uniform spectrum of propagating plane waves, but a uniform spectrum of exponentially decaying ([[Evanescent wave|evanescent]]) waves as well, and it is these which are responsible for resolution finer than one wavelength (see [[Fourier optics]]). This follows from the following [[Fourier transform]] expression for a 2D impulse function,
:<math>\delta (x,y) \propto \int\!\!\int e^{j(k_x x + k_y y)} \, d k_x\, d k_y</math>
[[Image:PSF.jpg|Truncation of Spherical Wave by Lens|left|thumb|400px]]
The quadratic [[lens (optics)|lens]] intercepts a ''portion'' of this spherical wave (and the key phrase here is ''a portion''), and refocuses it onto a blurred point in the image plane. For a single [[lens (optics)|lens]], an on-axis point source in the object plane produces an [[Airy disc]] PSF in the image plane. This comes about in the following way. ''It can be shown'' (see [[Fourier optics]], [[Huygens-Fresnel principle]], [[Fraunhofer diffraction]]) that the field radiated by a planar object (or, by reciprocity, the field converging onto a planar image) is related to its corresponding source (or image) plane distribution via a [[Fourier transform]] (FT) relation. In addition, a uniform function over a circular area (in one FT domain) corresponds to the Airy function, ''J''<sub>1</sub>(''x'')/''x'' in the other FT domain, where ''J''<sub>1</sub>(''x'') is the first-order Bessel function of the first kind. That is, a uniformly-illuminated circular aperture radiates an Airy function pattern plus a converging uniform spherical wave (over a circular angular sector) and yields an Airy function image. A graph of a sample 2D Airy function is shown in the adjoining figure (click to enlarge).
[[Image:Airy-3d.svg|Airy Function|left|thumb|300px]]
Therefore, the converging (''partial'') spherical wave shown in the figure above produces an [[Airy disc]] in the image plane. The argument of the Airy function is important, because this determines the ''scaling'' of the Airy disc (in other words, how big the disc is in the image plane). If Θ<sub>max</sub> is the maximum angle that the converging waves make with the lens axis, ''r'' is radial distance in the image plane, and [[wavenumber]] ''k'' = 2π/λ where λ = wavelength, then the argument of the Airy function is: kr sin(Θ<sub>max</sub>). If Θ<sub>max</sub> is small (only a small portion of the converging spherical wave is available to form the image), then radial distance, r, has to be very large before the total argument of the Airy function moves away from the central spot. In other words, if Θ<sub>max</sub> is small, the Airy disc is large (which is just another statement of Heisenberg's [[uncertainty principle]] for FT pairs, namely that small extent in one domain corresponds to wide extent in the other domain, and the two are related via the ''space-bandwidth product''. By virtue of this, high [[magnification]] systems, which typically have small values of Θ<sub>max</sub> (by the [[Abbe sine condition]]), can have more blur in the image, owing to the broader PSF. The size of the PSF is proportional to the [[magnification]], so that the blur is no worse in a relative sense, but it is definitely worse in an absolute sense.
==History and methods==
The diffraction theory of point-spread functions was first studied by [[George Biddell Airy|Airy]] in the nineteenth century. He developed an expression for the point-spread function amplitude and intensity of a perfect instrument, free of aberrations (the so-called [[Airy disc]]). The theory of aberrated point-spread functions close to the optimum focal plane was studied by the Dutch physicists [[Fritz Zernike]] and Nijboer in the 1930–40s. A central role in their analysis is played by [[Fritz Zernike|Zernike’s]] [[Zernike polynomials|circle polynomials]] that allow an efficient representation of the aberrations of any optical system with rotational symmetry. Recent analytic results have made it possible to extend Nijboer and Zernike’s approach for point-spread function evaluation to a large volume around the optimum focal point. This [http://www.nijboerzernike.nl Extended Nijboer-Zernike (ENZ) theory] is instrumental in studying the imperfect imaging of three-dimensional objects in [[confocal microscopy]] or astronomy under non-ideal imaging conditions. The ENZ-theory has also been applied to the characterization of optical instruments with respect to their aberration by measuring the through-focus intensity distribution and solving an appropriate [[inverse problem]].
==PSF in microscopy==
In microscopy, experimental determination of a PSF is usually tricky, due to the difficulty of finding sub-resolution (point-like) radiating sources. [[Quantum dot]]s and [[fluorescent]] [[bead]]s are usually considered for this purpose.
==The PSF in astronomy==
In [[observational astronomy]] the experimental determination of a PSF is often very straightforward due to the ample supply of point sources ([[star]]s or [[quasars]]). The form and source of the PSF may vary widely depending on the instrument and the context in which it is used.
For [[radio telescopes]] and diffraction-limited space [[telescopes]] the dominant terms in the PSF may be inferred from the configuration of the aperture in the [[Fourier domain]]. In practice there may be multiple terms contributed by the various components in a complex optical system. A complete description of the PSF will also include diffusion of light (or photo-electrons) in the detector, as well as tracking errors in the spacecraft.
For ground based optical telescopes, atmospheric turbulence (known as [[astronomical seeing]]) dominates the contribution to the PSF. In high-resolution ground-based imaging, the PSF is often found to vary with position in the image (an effect called anisoplanatism). In ground based [[adaptive optics]] systems the PSF is a combination of the aperture of the system with residual uncorrected atmospheric terms.
== Point spread functions in ophthalmology==
<!-- Unsourced image removed: [[Image:PSF mild myopia.jpg|frame|A white-light PSF of a human eye with moderate optical aberrations.]] -->
PSFs have recently become a useful diagnostic tool in clinical [[ophthalmology]]. Patients are measured with a [[wavefront]] sensor, and special software calculates the PSF for that patient’s eye. In this manner a physician can "see" what the patient sees. This method also allows a physician to simulate potential treatments on a patient, and see how those treatments would alter the patient’s PSF. Additionally, once measured the PSF can be minimized using an adaptive optics system. This, in conjunction with a CCD, can be used to visualize anatomical structures not otherwise visible ''in vivo'', such as cone photoreceptors.
==See also==
* [[Circle of confusion]], for the closely related topic in general photography.
* [[Airy disc]]
==External links==
* [http://support.svi.nl/wiki/NyquistCalculator PSF calculator] for fluorescence microscopes.
* [http://www.nijboerzernike.nl Extended Nijboer-Zernike (ENZ) theory], calculating the PSF for a general Optical system.
[[Category:Optics]]
[[Category:Ophthalmology]]
[[de:Punktspreizfunktion]]
[[fr:Fonction d'étalement du point]]