Recognizer for the powerful owl #370
Comments
towsey
added a commit
that referenced
this issue
Aug 27, 2020
Issue #370: Reconizer is detecting syllables.
towsey
added a commit
that referenced
this issue
Aug 28, 2020
Issue #370: Shift event bandwidth filter and component count filter to Generic class. Rework filter for acoustic activity in neighbourhood - i.e. event sidebands.
towsey
added a commit
that referenced
this issue
Aug 28, 2020
Issue #370 Rearrange the order of post-processing steps and where they are done. I.e. shift from species specific to generic class. This will make the generic recognizers more powerful and give user more control.
towsey
added a commit
that referenced
this issue
Aug 28, 2020
Issue #370 Use different formula to calculate maximum length of a syllable sequence.
towsey
added a commit
that referenced
this issue
Sep 3, 2020
towsey
added a commit
that referenced
this issue
Sep 3, 2020
Issue #370 The SyllableSequenceConfig contains all the info to combine a potential event sequence into a single event and then to filter those combined events if the sequence does not have the correct properties.
towsey
added a commit
that referenced
this issue
Sep 3, 2020
towsey
added a commit
that referenced
this issue
Sep 3, 2020
towsey
added a commit
that referenced
this issue
Sep 3, 2020
towsey
added a commit
that referenced
this issue
Sep 4, 2020
Issue #370 Changes to accomodate change in calculation of the neighbourhood acoustic activity.
towsey
added a commit
that referenced
this issue
Sep 7, 2020
Issue #370 Changed the config file to optimise the parameters on a large data set provided by Elizabeth Znidersic. Then reworked the unit test.
towsey
added a commit
that referenced
this issue
Sep 8, 2020
Issue #370 Conditions inserted to provent potential out of range index exceptions.
towsey
added a commit
that referenced
this issue
Sep 9, 2020
Issue #370 Refactoring and preparing way forsome methods involving dynamic gain.
towsey
added a commit
that referenced
this issue
Sep 10, 2020
Issue #370 Exploring possibility of doing amplitude normalisation on detected events.
towsey
added a commit
that referenced
this issue
Sep 15, 2020
towsey
added a commit
that referenced
this issue
Sep 15, 2020
towsey
added a commit
that referenced
this issue
Sep 16, 2020
Issue #370 Change the decibel intensity array that is displayed with the spectrogram. Helps in debugging results.
towsey
added a commit
that referenced
this issue
Sep 17, 2020
Issue #370 Change Log.Debug to Log.Info in order to get information included in log file.
towsey
added a commit
that referenced
this issue
Sep 17, 2020
Issue #370 Do not add very short tracks to list of potential events
towsey
added a commit
that referenced
this issue
Sep 17, 2020
Issue #370 Add more information with regard to events before and after post-processing
towsey
added a commit
that referenced
this issue
Sep 17, 2020
Issue #370 Add canonical recording to Fixtures
towsey
added a commit
that referenced
this issue
Sep 18, 2020
Issue #370 Some unit tests broken because changes to one of the Generic Algorithms have compromised others.
towsey
added a commit
that referenced
this issue
Sep 23, 2020
Issue #370: shift PreparePlot() method from GenericRecognizer class to Plot class.
towsey
added a commit
that referenced
this issue
Sep 23, 2020
Issue #370 Change double to array of double. THis is with a view to running the call recognizers over a set of decibel thresholds rather than just one threshold. This is to accomodate calls that are close and distant to the microphone, i.e. to accomodate a range of call SNR.
towsey
added a commit
that referenced
this issue
Sep 23, 2020
Issue #370 These changes allow track recognizers to be derived from an array of different decibel thresholds.
towsey
added a commit
that referenced
this issue
Sep 23, 2020
Issue #370 Start process of getting the Test class for Generic recognizers to accept the new array of decibel thresholds. Next step is to ensure those tests pass.
towsey
added a commit
that referenced
this issue
Sep 24, 2020
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Do not add very short tracks to list of potential events
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Add more information with regard to events before and after post-processing
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Add canonical recording to Fixtures
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Some unit tests broken because changes to one of the Generic Algorithms have compromised others.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370: shift PreparePlot() method from GenericRecognizer class to Plot class.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Change double to array of double. THis is with a view to running the call recognizers over a set of decibel thresholds rather than just one threshold. This is to accomodate calls that are close and distant to the microphone, i.e. to accomodate a range of call SNR.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 These changes allow track recognizers to be derived from an array of different decibel thresholds.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Start process of getting the Test class for Generic recognizers to accept the new array of decibel thresholds. Next step is to ensure those tests pass.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Part of ongoing refactoring to allow all recognizers to work with an array of decibel thresholds.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Fix positioning of text at extreme right end of plot.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 efactor POwerful Owl test so that it accepts an array of decibel thresholds. The test results are no different after extensive refactoring.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Have yet to check whether unit tests are passing after this change.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Main reason for refactoring is so that the generic component recognizers are able to do multiple passes ofr an array of different decibel threshold values.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Implemented changes requested by Anthony.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Refactor test methods to accomodate recent changes. All tests pass.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Fix bug where window was longer than the matrix.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Update the config.yml for the Australasian Bittern.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Separate the component count filter from the periodicity filter.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Previously the filter was based on component events but now that we have multiple decibel thresholds, composite events are accumulating more component events, so now need to calculate the number of probable syllables within a composite event.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Unit tests need to be revised because of accumulating more acoustic events as a result of having multiple decibel thresholds.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Adjust test for changes to generic algorith, The recognizer is now more accurate.
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 To filter on event duration during generic post-processing
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Incorporate all latest event combiners and filters into this recognizer
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Issue #370 Incorporate recent changes into this recognizer
atruskie
pushed a commit
that referenced
this issue
Oct 14, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Is your feature request related to a problem? Please describe.
Creating a new species recognizer for the Powerful Owl, Ninnox strenua. This owl is in same genus as the boobook.
Describe the solution you'd like
Will use same approach as with the boobook..
Describe alternatives you've considered
This owl should be clearly visible in LDFC spectrograms as is the boobook - but users want a species recognizer.
Additional context
THis recognizer requested by Brad Law.
The text was updated successfully, but these errors were encountered: