Sector Integration Example

Loads the example data set, sets diffrent Sectors for intergration. Shows Acoular’s Sector und Sound Pressure level Integration functionality.

from pathlib import Path

import acoular as ac
import matplotlib.pyplot as plt
import numpy as np
from acoular.tools.helpers import get_data_file
from matplotlib.patches import Polygon, Rectangle

Obtain necessary data

time_data_file = get_data_file('example_data.h5')

Define the necessary objects

micgeofile = Path(ac.__file__).parent / 'xml' / 'array_56.xml'

mg = ac.MicGeom(file=micgeofile)
ts = ac.TimeSamples(file=time_data_file)
ps = ac.PowerSpectra(source=ts, block_size=128, window='Hanning')
rg = ac.RectGrid(x_min=-0.6, x_max=-0.0, y_min=-0.3, y_max=0.3, z=-0.68, increment=0.02)
st = ac.SteeringVector(grid=rg, mics=mg)
f = ac.PowerSpectra(source=ts, block_size=128)
bf = ac.BeamformerBase(freq_data=f, steer=st)

Integrate function can deal with multiple methods for integration:

1. a circle containing of three values: x-center, y-center and radius

circle = np.array([-0.3, -0.1, 0.05])

2. a rectangle containing of 4 values: lower corner(x1, y1) and upper corner(x2, y2).

rect = np.array([-0.5, -0.15, -0.4, 0.15])

3. a polygon containing of vector tuples: x1,y1,x2,y2,…,xi,yi

poly = np.array([-0.25, -0.1, -0.1, -0.1, -0.1, -0.2, -0.2, -0.25, -0.3, -0.2])

4th alternative: define those sectors as Classes

circle_sector = ac.CircSector(x=-0.3, y=-0.1, r=0.05)
rect_sector = ac.RectSector(x_min=-0.5, x_max=-0.4, y_min=-0.15, y_max=0.15)

PolySector is a class that takes a list of points as input list of points containing x1,y1,x2,y2,…,xi,yi

poly_sector = ac.PolySector(edges=[-0.25, -0.1, -0.1, -0.1, -0.1, -0.2, -0.2, -0.25, -0.3, -0.2])

The MultiSector class allows to sum over multiple different sectors

multi_sector = ac.MultiSector(sectors=[circle_sector, rect_sector, poly_sector])

Two integration variants exist (with same outcome): 1. use Acoular’s integrate function. Integrate SPL values from beamforming results using the shapes

levels_circ = ac.integrate(bf.result[:], rg, circle)
levels_rect = ac.integrate(bf.result[:], rg, rect)
levels_poly = ac.integrate(bf.result[:], rg, poly)

[('example_data_cache.h5', 18)]
[('example_data_cache.h5', 19)]

integrate SPL values from beamforming results using sector classes

levels_circ_sector = ac.integrate(bf.result[:], rg, circle_sector)
levels_rect_sector = ac.integrate(bf.result[:], rg, rect_sector)
levels_poly_sector = ac.integrate(bf.result[:], rg, poly_sector)
levels_multi_sector = ac.integrate(bf.result[:], rg, multi_sector)

2. use beamformers integrate function (does not require explicit assignment of grid object). Integrate SPL values from beamforming results using the shapes

levels_circ = bf.integrate(circle)
levels_rect = bf.integrate(rect)
levels_poly = bf.integrate(poly)

integrate SPL values from beamforming results using sector classes

levels_circ_sector = bf.integrate(circle_sector)
levels_rect_sector = bf.integrate(rect_sector)
levels_poly_sector = bf.integrate(poly_sector)
levels_multi_sector = bf.integrate(multi_sector)

Plot map and sectors

plt.figure()
map = bf.synthetic(2000, 1)
mx = ac.L_p(map.max())
plt.imshow(ac.L_p(map.T), origin='lower', vmin=mx - 15, interpolation='nearest', extent=rg.extent, cmap=plt.cm.hot_r)
plt.colorbar()
circle1 = plt.Circle((-0.3, 0.1), 0.05, color='k', fill=False)
plt.gcf().gca().add_artist(circle1)
polygon = Polygon(poly.reshape(-1, 2), color='k', fill=False)
plt.gcf().gca().add_artist(polygon)
rect = Rectangle((-0.5, -0.15), 0.1, 0.3, linewidth=1, edgecolor='k', facecolor='none')
plt.gcf().gca().add_artist(rect)

# calculate the discrete frequencies for the integration
fftfreqs = np.arange(128 / 2 + 1) * (51200 / 128)

# plot from shapes
plt.figure()
plt.plot(fftfreqs, ac.L_p(levels_circ))
plt.plot(fftfreqs, ac.L_p(levels_rect))
plt.plot(fftfreqs, ac.L_p(levels_poly))
plt.xlim([2000, 20000])
plt.ylim([10, 80])
plt.legend(['Circle', 'Rectangle', 'Polygon'])

# plot from sector classes
plt.figure()
plt.plot(fftfreqs, ac.L_p(levels_circ_sector))
plt.plot(fftfreqs, ac.L_p(levels_rect_sector))
plt.plot(fftfreqs, ac.L_p(levels_poly_sector))
plt.plot(fftfreqs, ac.L_p(levels_multi_sector))
plt.xlim([2000, 20000])
plt.ylim([10, 80])
plt.legend(['Circle Sector', 'Rectangle Sector', 'Polygon Sector', 'Multisector'])

plt.show()

• 
• 
•