X-ray transform

From HandWiki
Short description: Integral transform

In mathematics, the X-ray transform (also called ray transform[1] or John transform) is an integral transform introduced by Fritz John in 1938[2] that is one of the cornerstones of modern integral geometry. It is very closely related to the Radon transform, and coincides with it in two dimensions. In higher dimensions, the X-ray transform of a function is defined by integrating over lines rather than over hyperplanes as in the Radon transform. The X-ray transform derives its name from X-ray tomography (used in CT scans) because the X-ray transform of a function ƒ represents the attenuation data of a tomographic scan through an inhomogeneous medium whose density is represented by the function ƒ. Inversion of the X-ray transform is therefore of practical importance because it allows one to reconstruct an unknown density ƒ from its known attenuation data.

In detail, if ƒ is a compactly supported continuous function on the Euclidean space Rn, then the X-ray transform of ƒ is the function defined on the set of all lines in Rn by

[math]\displaystyle{ Xf(L) = \int_L f = \int_{\mathbf{R}} f(x_0+t\theta)dt }[/math]

where x0 is an initial point on the line and θ is a unit vector in Rn giving the direction of the line L. The latter integral is not regarded in the oriented sense: it is the integral with respect to the 1-dimensional Lebesgue measure on the Euclidean line L.

The X-ray transform satisfies an ultrahyperbolic wave equation called John's equation.

The Gauss hypergeometric function can be written as an X-ray transform (Gelfand Gindikin).

References

  1. Natterer, Frank; Wübbeling, Frank (2001). Mathematical Methods in Image Reconstruction. Philadelphia: SIAM. doi:10.1137/1.9780898718324.fm. 
  2. Fritz, John (1938). "The ultrahyperbolic differential equation with four independent variables". Duke Mathematical Journal 4: 300–322. doi:10.1215/S0012-7094-38-00423-5. http://projecteuclid.org/euclid.dmj/1077490637. Retrieved 23 January 2013.