prosi3d.sensors package

Submodules

prosi3d.sensors.acousticair module

Subclass from Abstract Base Class featureExtractor that outputs features of the raw data that are required for machine learning models

class prosi3d.sensors.acousticair.Accousticair

Bases: FeatureExtractor

Attribute:

peaks_x (numpy.ndarray): frequency of the peaks.

peaks_y (numpy.ndarray): spectral power density of the peaks.

get_data(hdf)

Extract the measurements of the accousticair sensor from the hdf5 file.

Parameters:

hdf (str) – path of the hdf5 file.

Raises:

IOError – File can not found.

get_feature()

Determine the sensor specific features as array [variance, peaks over a boundary xxx in the frequency domain, main components of the PCA].

Call get_data and process before using this method otherwise this method throws a error.

Returns:

Array with the sensor specific features.

Return type:

features (numpy.ndarray)

plot_test()

Plot the diagramms of the time domain and freqency domain with the identified peaks to check the result

process()

Convert the signal of the time domain to the representation in the frequency domain using the rFFT. Identify particularly conspicuous peaks in the frequency domain.

write()

Print the conspicuous peaks which are detected in the method Accousticair.process(). The frequency and the power spectral density are saved in numpy arrays.

prosi3d.sensors.acousticplatform module

Subclass from Abstract Base Class featureExtractor that outputs features of the raw data that are required for machine learning models

class prosi3d.sensors.acousticplatform.Accousticplatform

Bases: FeatureExtractor

Attribute:

peaks_x (numpy.ndarray): frequency of the peaks.

peaks_y (numpy.ndarray): spectral power density of the peaks.

get_data(hdf)

Extract the measurements of the accousticplatform sensor from the hdf5 file.

Parameters:

hdf (str) – path of the hdf5 file.

Raises:

IOError – File can not found.

get_feature()

Determine the sensor specific features as array [variance, peaks over a boundary xxx in the frequency domain, main components of the PCA].

Call get_data and process before using this method otherwise this method throws a error.

Returns:

Array with the sensor specific features.

Return type:

features (numpy.ndarray)

plot_test()

Plot the diagramms of the time domain and freqency domain with the identified peaks to check the result

process()

Convert the signal of the time domain to the representation in the frequency domain using the rFFT. Identify particularly conspicuous peaks in the frequency domain.

write()

Print the x-values and the y-values of the peaks which are saved in to numpy arrays.

prosi3d.sensors.acousticplatform.lab2hdf_4ch_simple(path)

Process 4-channel layerwise acoustic measurement txt files written by labview and saves as hdf-files with pandas dataframe key=’df’. Timestamp is changed to float total seconds starting with zero in each layer. If day changes, then 24*60*60 seconds are added to each float value of the new day. By this, the maximum processable layertime is 24 hours. Calls folder2Files to get ttldir in project path and collects all txt files by calling filesInfolder() from ttldir.

Args: path Returns: <saved h5 files>

prosi3d.sensors.nircamera module

Subclass from Abstract Base Class featureExtractor that outputs features of the raw data that are required for machine learning models

class prosi3d.sensors.nircamera.Nircamera

Bases: FeatureExtractor

Attribute:

xxx: xxx. xxx: xxx.

get_data()

Beschreibung

get_feature()

Beschreibung

process()

Beschreibung

write()

Beschreibung

prosi3d.sensors.recoater module

Subclass from Abstract Base Class featureExtractor that outputs features of the raw data that are required for machine learning models

class prosi3d.sensors.recoater.Recoater

Bases: FeatureExtractor

Attribute:

peaks_x (numpy.ndarray): frequency of the peaks.

peaks_y (numpy.ndarray): spectral power density of the peaks.

get_data(hdf)

Extract the measurements of the recoater sensor from the hdf5 file.

Parameters:

hdf (str) – path of the hdf5 file.

Raises:

IOError – File can not found.

get_feature()

Determine the sensor specific features as array [variance, peaks over a boundary xxx in the frequency domain, peaks over a boundary xxx in the time domain].

Call get_data and process before using this method otherwise this method throws a error.

Returns:

Array with the sensor specific features.

Return type:

features (numpy.ndarray)

plot_test()

Plot the diagramms of the time domain and freqency domain with the identified peaks to check the result

process()

Convert the signal of the time domain to the representation in the frequency domain using the rFFT. Identify particularly conspicuous peaks in the frequency domain.

write()

Print the conspicuous peaks which are detected in the method Accousticair.process(). The frequency and the power spectral density are saved in numpy arrays.

prosi3d.sensors.viscamera module

Subclass from Abstract Base Class featureExtractor that outputs features of the raw data that are required for machine learning models

class prosi3d.sensors.viscamera.Viscamera

Bases: FeatureExtractor

Attribute:

xxx: xxx. xxx: xxx.

get_data()

Beschreibung

get_feature()

Beschreibung

process()

Beschreibung

write()

Beschreibung

Module contents