Vectorial PSF Generator Debye-Wolf + Gibson-Lanni

Research-grade point-spread function simulation for quantitative microscopy

Parameters

Emission wavelength (e.g., 525 for GFP)

Objective NA (e.g., 1.4 for oil immersion)

1.0 (air), 1.33 (water), 1.518 (oil)

Optional Parameters (intelligent defaults applied)

Leave empty for optimal sampling (shows Airy rings)

Output image dimensions (64-512)

Refractive Index Mismatch (Gibson-Lanni Model)

Accounts for cover glass / sample RI differences

Initializing...

PSF Output

Configure parameters and click "Compute PSF"

z = 0.000 µm

Intensity Scaling Note

PSF intensities are normalized to peak = 1.0. TIFF exports preserve full 32-bit floating-point values. PNG uses linear mapping to 8-bit grayscale (0-255).

📖 Physical Model & Scientific Documentation

Vectorial Diffraction Model (Debye-Wolf Integral)

This PSF generator implements the vectorial diffraction theory based on the Debye-Wolf integral, which provides an exact solution for high-NA optical systems. Unlike scalar (Airy/Gaussian) approximations, this model correctly accounts for:

  • Polarization effects at high numerical apertures
  • Apodization from the Abbe sine condition
  • Non-paraxial propagation (valid for all NA values)
  • Asymmetric axial PSF structure

Mathematical Formulation

For circularly polarized or unpolarized illumination, the total intensity PSF is computed from three diffraction integrals:

I₀(ρ,z) = ∫₀^θmax √cos(θ) · sin(θ) · (1+cos(θ)) · J₀(kρ·sin(θ)) · exp(ikΦ) dθ
I₁(ρ,z) = ∫₀^θmax √cos(θ) · sin²(θ) · J₁(kρ·sin(θ)) · exp(ikΦ) dθ
I₂(ρ,z) = ∫₀^θmax √cos(θ) · sin(θ) · (1-cos(θ)) · J₂(kρ·sin(θ)) · exp(ikΦ) dθ

PSF(ρ,z) = |I₀|² + 2|I₁|² + |I₂|²
                

Where θmax = arcsin(NA/nᵢ), k = 2πnᵢ/λ, and Jₙ are Bessel functions of the first kind.

Gibson-Lanni Aberration Model

When RI mismatch is enabled, the optical path difference (OPD) includes contributions from:

Φ(θ) = z·cos(θ) + OPD_mismatch(θ)

OPD_mismatch = Δt_g·(n_g·cos(θ_g) - n_i·cos(θ_i))  [cover glass term]
             + d·(n_s·cos(θ_s) - n_i·cos(θ_i))      [sample depth term]
                

where Snell's law relates angles in each medium: nᵢ·sin(θ) = n_g·sin(θ_g) = n_s·sin(θ_s)

Physical Assumptions

Assumption Value/Condition Notes
Illumination Circularly polarized / unpolarized Standard fluorescence microscopy
Objective Airy pupil, no aberrations Perfect lens (aberration-free)
Detection Incoherent intensity Sum of polarization components
Numerical integration 128-point Gauss-Legendre quadrature High accuracy (~10⁻¹⁰ relative error)

Comparison to Established Tools

Feature This Tool PSFGenerator (MATLAB) Python-microscPSF
Diffraction model Vectorial Debye-Wolf Vectorial Debye-Wolf Scalar/Vectorial
RI mismatch Gibson-Lanni Gibson-Lanni Gibson-Lanni
Aberrations Spherical only (from RI) Zernike polynomials Varies
Precision 64-bit float (JS) 64-bit float 64-bit float

Browser Limitations

Limitation Impact Mitigation
JavaScript precision IEEE 754 64-bit (~15-17 digits) Sufficient for PSF accuracy
Memory constraints Large 3D stacks may fail Grid size limited to 512×512
Single-threaded UI may freeze during computation Web Worker (async computation)
Bessel functions Custom implementation Polynomial approximation (error < 10⁻⁸)

References

  • Wolf, E. (1959). Electromagnetic diffraction in optical systems. Proc. Roy. Soc. A 253, 349-357.
  • Richards, B. & Wolf, E. (1959). Electromagnetic diffraction in optical systems II. Proc. Roy. Soc. A 253, 358-379.
  • Gibson, S.F. & Lanni, F. (1992). Experimental test of an analytical model of aberration in an oil-immersion objective. J. Opt. Soc. Am. A 9, 154-166.
  • Haeberle, O. et al. (2003). The point spread function of optical microscopes imaging through stratified media. Opt. Express 11, 2249-2259.