Validation of non-invasive measures for life stage in wild octopuses and AI-Assisted Behavioral Analysis

Grantee: Michaella Andrade

 

Institution: Federal University of ABC, Brazil

Grant amount: $10,000

 

Grant type: Seed grants

Focal species: Octopuses (Octopus insularis)

 

Conservation status: Not evaluated

Disciplines: Animal behavior

 

Research locations: Brazil


Project summary

This project aims to develop non-invasive methods to assess the life stage and welfare of wild octopuses. The project will develop an innovative AI approach that automatically learns to recognize complex behaviors, such as body patterns and ventilation rates, directly from pre-recorded videos in the field. In addition, the project will develop a methodology to automatically measure the life stage of octopuses in pre-recorded videos based on our database and non-invasive in situ measurements of the distance between the octopuses’ eyes and eyeballs. The main goal is to correlate these metrics with different behavioral contexts to determine whether they can be used as reliable indicators of the animal’s affective state. In the long term, the project aims to create an automated tool that accelerates and improves behavioral data analysis, enabling large-scale studies of the welfare of wild octopuses.

Why we funded this project

This project extends a previous WAI-funded project and uses innovative approaches to analyzing behavioral data for welfare assessment. 


Previous
Previous

Assessment of welfare impacts of parasites and pathogens on migratory flamingos among Salt Lakes of East Africa

Next
Next

Effect of an invasive competitor on the welfare of a threatened fish in a soft-release program