Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Fast and accurate evaluation of a generalized incomplete gamma function

Abstract : We present a computational procedure to evaluate the integral ∫xy sp-1 e-μs ds, for 0 ≤ x < y ≤+∞, μ = ±1, p > 0, which generalizes the lower (x=0) and upper (y=+∞) incomplete gamma functions. To allow for large values of x, y, and p while avoiding under/overflow issues in the standard double precision floating point arithmetic, we use an explicit normalization that is much more efficient than the classical ratio with the complete gamma function. The generalized incomplete gamma function is estimated with continued fractions, integrations by parts, or, when x ≈ y, with the Romberg numerical integration algorithm. We show that the accuracy reached by our algorithm improves a recent state-of-the-art method by two orders of magnitude, and is essentially optimal considering the limitations imposed by the floating point arithmetic. Moreover, the admissible parameter range of our algorithm (0 ≤ p,x,y ≤ 1015) is much larger than competing algorithms and its robustness is assessed through massive usage in an image processing application.
Complete list of metadatas

Cited literature [39 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01329669
Contributor : Lionel Moisan <>
Submitted on : Thursday, November 7, 2019 - 8:38:31 AM
Last modification on : Friday, April 10, 2020 - 5:24:16 PM

File

gammainc_revised_preprint.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01329669, version 2

Collections

Citation

Rémy Abergel, Lionel Moisan. Fast and accurate evaluation of a generalized incomplete gamma function. 2019. ⟨hal-01329669v2⟩

Share

Metrics

Record views

77

Files downloads

1384