Quickstart

Installation

Install from PyPI:

pip install pylorenzmie

For GPU-accelerated hologram computation (requires CUDA 12):

pip install pylorenzmie[gpu]

The interactive GUI tool (LMTool) is included in the standard install and can be launched immediately:

lmtool

Computing a hologram

from pylorenzmie.theory import LorenzMie, Sphere, Instrument
import matplotlib.pyplot as plt

# Describe the microscope
instrument = Instrument()
instrument.wavelength = 0.447      # vacuum wavelength [um]
instrument.magnification = 0.048   # image scale [um/pixel]
instrument.n_m = 1.340             # medium refractive index

# Describe the particle
particle = Sphere()
particle.a_p = 0.75    # radius [um]
particle.n_p = 1.45    # refractive index
particle.r_p = [100, 100, 200]   # position [pixels]

# Compute hologram on a 201x201 grid
model = LorenzMie(instrument=instrument)
model.coordinates = model.meshgrid((201, 201))
model.particle = particle
hologram = model.hologram().reshape(201, 201)

plt.imshow(hologram, cmap='gray')
plt.show()

Fitting a hologram

import cv2
from pylorenzmie.analysis import Feature
from pylorenzmie.theory import LorenzMie

# Load and normalize a hologram (divide by background level)
data = cv2.imread('crop.png', cv2.IMREAD_GRAYSCALE).astype(float)
data /= data.mean()

# Set up the model
model = LorenzMie()
model.instrument.wavelength = 0.447
model.instrument.magnification = 0.048
model.instrument.n_m = 1.340
model.particle.r_p = [data.shape[1]//2, data.shape[0]//2, 330.]
model.particle.a_p = 1.0
model.particle.n_p = 1.45

# Optimize
feature = Feature(data=data,
                  coordinates=model.meshgrid(data.shape),
                  model=model)
result = feature.optimize()
print(result)

Analyzing a full frame

import cv2
from pylorenzmie.analysis import Frame

frame = Frame()
data = cv2.imread('image.png', cv2.IMREAD_GRAYSCALE).astype(float)
data /= data.mean()
results = frame.analyze(data)
print(results)   # pandas.DataFrame with one row per particle