mne.preprocessing.compute_proj_eog(raw, raw_event=None, tmin=-0.2, tmax=0.2, n_grad=2, n_mag=2, n_eeg=2, l_freq=1.0, h_freq=35.0, average=True, filter_length='10s', n_jobs=None, reject={'eeg': 0.0005, 'eog': inf, 'grad': 2e-10, 'mag': 3e-12}, flat=None, bads=(), avg_ref=False, no_proj=False, event_id=998, eog_l_freq=1, eog_h_freq=10, tstart=0.0, filter_method='fir', iir_params=None, ch_name=None, copy=True, return_drop_log=False, meg='separate', verbose=None)[source]#

Compute SSP (signal-space projection) vectors for EOG artifacts.

This function will:

  1. Filter the EOG data channel.

  2. Find the peaks of eyeblinks in the EOG data using mne.preprocessing.find_eog_events().

  3. Filter the raw data.

  4. Create Epochs around the eyeblinks.

  5. Optionally average the Epochs to produce an Evoked if average=True was passed (default).

  6. Calculate SSP projection vectors on that data to capture the artifacts.


Raw data must be preloaded.


Raw input file. or None

Raw file to use for event detection (if None, raw is used).


Time before event in seconds.


Time after event in seconds.


Number of SSP vectors for gradiometers.


Number of SSP vectors for magnetometers.


Number of SSP vectors for EEG.

l_freqfloat | None

Filter low cut-off frequency for the data channels in Hz.

h_freqfloat | None

Filter high cut-off frequency for the data channels in Hz.


Compute SSP after averaging. Default is True.

filter_lengthstr | int | None

Number of taps to use for filtering.

n_jobsint | None

The number of jobs to run in parallel. If -1, it is set to the number of CPU cores. Requires the joblib package. None (default) is a marker for ‘unset’ that will be interpreted as n_jobs=1 (sequential execution) unless the call is performed under a joblib.parallel_config context manager that sets another value for n_jobs.

rejectdict | None

Epoch rejection configuration (see Epochs).

flatdict | None

Epoch flat configuration (see Epochs).


List with (additional) bad channels.


Add EEG average reference proj.


Exclude the SSP projectors currently in the fiff file.


ID to use for events.


Low pass frequency applied to the E0G channel for event detection.


High pass frequency applied to the EOG channel for event detection.


Start artifact detection after tstart seconds.


Method for filtering (‘iir’ or ‘fir’).

iir_paramsdict | None

Dictionary of parameters to use for IIR filtering. See mne.filter.construct_iir_filter for details. If iir_params is None and method=”iir”, 4th order Butterworth will be used.

ch_namestr | None

If not None, specify EOG channel name.


If False, filtering raw data is done in place. Defaults to True.


If True, return the drop log.

New in v0.15.


Can be ‘separate’ (default) or ‘combined’ to compute projectors for magnetometers and gradiometers separately or jointly. If ‘combined’, n_mag == n_grad is required and the number of projectors computed for MEG will be n_mag.

New in v0.18.

verbosebool | str | int | None

Control verbosity of the logging output. If None, use the default verbosity level. See the logging documentation and mne.verbose() for details. Should only be passed as a keyword argument.

projslist of Projection

List of computed projection vectors.

eog_events: ndarray

Detected EOG events.


The drop log, if requested.


Filtering is applied to the EOG channel while finding events using eog_l_freq and eog_h_freq, and then to the raw instance using l_freq and h_freq before creation of the epochs used to create the projectors.

Examples using mne.preprocessing.compute_proj_eog#

Working with CTF data: the Brainstorm auditory dataset

Working with CTF data: the Brainstorm auditory dataset

Repairing artifacts with SSP

Repairing artifacts with SSP