Math 2.0

Integral transforms, operational calculus Definition

Filtered back-projection

Builds on Radon transformDual Radon transformConvolution

Definition

Let ff be in the Schwartz space 𝒮(ℝ2)\mathcal{S}(\R^2), with Radon transform RfR f, and write Rθf(s)=Rf(θ,s)R_\theta f(s) = R f(\theta, s). Let Φ ⁣:ℝ→ℝ\Phi \colon \R \to \R be an even bounded function with Φ(σ)=0\Phi(\sigma) = 0 for ∣σ∣>1\abs{\sigma} > 1, continuous at 00, with Φ(0)=1\Phi(0) = 1. For b>0b > 0, let wbw_b be the filter with Fourier transform w^b(σ)=∣σ∣ Φ(σ/b)\hat w_b(\sigma) = \abs{\sigma}\, \Phi(\sigma/b). Its filtered back-projection reconstruction with filter wbw_b is fb(x)=14πR∗(wb∗Rθf)(x)=14π∫S1(wb∗Rθf)(x⋅θ) dθ,f_b(x) = \frac{1}{4\pi} R^* \left( w_b * R_\theta f \right)(x) = \frac{1}{4\pi} \int_{S^1} (w_b * R_\theta f)(x \cdot \theta)\, d\theta, where ∗* is convolution in the variable ss and R∗R^* is the dual Radon transform.

Remarks

The Fourier convention is g^(σ)=∫g(s) e−isσ ds\hat g(\sigma) = \int g(s)\, e^{-is\sigma}\, ds. The projection-slice theorem gives f^b(ξ)=Φ(∣ξ∣/b) f^(ξ)\hat f_b(\xi) = \Phi(\abs{\xi}/b)\, \hat f(\xi), so fb→ff_b \to f as b→∞b \to \infty, uniformly on ℝ2\R^2. With Φ\Phi the indicator of [−1,1][-1, 1] the filter is the Ram–Lak filter, a band-limited ramp.

Example: if Φ\Phi is the indicator of [−1,1][-1, 1] and f^\hat f vanishes for ∣ξ∣>b\abs{\xi} > b, then f^b=f^\hat f_b = \hat f and the reconstruction is exact: fb=ff_b = f.

Non-example: back-projection without filter, 14πR∗Rf\frac{1}{4\pi} R^* R f, is not a reconstruction of ff. It corresponds to w^(σ)=1\hat w(\sigma) = 1 instead of ∣σ∣\abs{\sigma}, and its Fourier transform is f^(ξ)/∣ξ∣\hat f(\xi)/\abs{\xi}, a blurred image.

In practice RfR f is known only at finitely many angles and offsets, and the integral and the convolution are replaced by sums.

Reading path

To understand this page from the ground up, read these 7 pages in order. The foundations come first.

Show the path as a graph
Each page uses the pages above it. The colour of the bar on a box shows how strong the evidence is, and the box also gives it as text. The list has the same pages.
  • no evidence level
  1. Dual Radon transform concept
  2. Haar measure concept
  3. Locally compact group concept
  4. Lp space concept
  5. Radon transform concept
  6. Convolution definition
  7. Filtered back-projection definition

Further reading

  1. Web https://en.wikipedia.org/wiki/Tomographic_reconstruction
About this page
TagMHC7X73NG9
URIm2://MHC7X73NG9
Kinddefinition
WritingW0 stub
MSC44A12, 92C55, 65R10
DefinesFiltered back-projection
UsesRadon transform, Dual Radon transform, Convolution

Markdown · JSON · Edit on GitHub

What this page needs for W1

  • an inventory row: the row of the inventory that the node accounts for (0 of 1)
  • a source with a locator (0 of 1)

The writing levels say how complete the text of a page is. They do not measure its truth.

What links here

No page links to this page yet.