Advancing Biodiversity Research on Barro Colorado Island
Panama
The Problem
In 2017, Rainforest Connection (RFCx) partnered with the Smithsonian Tropical Research Institute (STRI) to enhance biodiversity monitoring on Barro Colorado Island, one of the world’s most intensively studied tropical forests. The project leveraged acoustic technology to support long-term ecological research and conservation efforts.
Partners
Microsoft AI for Good Research Lab
Objectives
• Develop a convolutional neural network (CNN) for identifying ~100 species of birds, ~30 species of amphibians, and three species of monkeys in field audio recordings from Barro Colorado Island (BCI), Panama.
• Research will focus on creating a generic pipeline for species identification of known species and to discover new species.
• Goal is that the final product will be user-friendly, making it accessible to a large range of users that will vary greatly in terms of their AI knowledge and programming skills.
• In addition, to creating the CNN, the pipeline will include the ability to identify and remove known species, apply autoencoder analyses to cluster similar unknown calls, data augmentation for increasing the samples size of rare species, and annotation tools for assisting the identification of new species. To make this complex pipeline accessible to all potential users, we will concurrently build the front-end so that we can receive feedback from the user community.
• Improve long-term biodiversity monitoring of 1000s of species around the world by creating a user-friendly data analyses platform that will provide scientists, wildlife managers, conservationists, and public and private environmental organizations the information they need to make informed conservation and management decision. To ensure that the pipeline is used by the community, we will publish scientific articles, share the code, implement the pipeline in the ARBIMON platform, provide workshops at scientific meetings, and partner with organization that have a global reach to promote the use of the pipeline.
Implementation
• In phase 1, we will create a convolutional neural network (CNN) for identifying ~100 known species of birds, ~30 species of amphibians, and three species of monkeys in >100,000 1-minute field audio recordings from Barro Colorado Island (BCI), Panama.To complete this task, we will use tools developed to generate training and testing data sets for each species.
• In phase 2, we will create a pipeline that will identify known species, mark unknown sounds, cluster the unknown sounds based on feature similarities, and provide the user with annotation tools for classifying the new species.
• In phase 3, we will exclude subsets of known species from the BCI-CNN and test the ability of the pipeline to discover these excluded species. In these tests we will use recordings that have been thoroughly annotated, so we will know what species are present and what species should be “discovered” when they are left out of the CNN.
• In phase 4, we will make the generic pipeline available to the community, through publications, sharing the code in GitHub, implementing the pipeline in the ARBIMON platform, and providing workshops at scientific meetings.
