Nkhotakota camera traps

Wildlife monitoring in Nkhotakota Wildlife Reserve, Malawi

I am extremely fortunate to get to work with collaborators from African Parks and the U.S. Forest Service on a large monitoring program in Nkhotakota Wildlife Reserve, Malawi, which is the site of one of the largest-ever translocation of African elephants in 2016 and 2017. To date, over 2000 animals of 9 species have been released into the reserve as part of the restoration effort.

We are using remote camera traps to monitor mammal populations in the reserve, along with computer vision models to process the data. We published a paper in Ecological Applications presenting the development of our model and a user-friendly software tool, Njobvu-AI. We also published our dataset of annotated images on LILA BC (Nkhotakota Camera Traps).

Njobvu-AI was developed with collaborators at Oregon State University and is a user-friendly, collaborative software tool to label images and train custom computer vision models. (Github) (Preprint) (Paper)

References

2025

  1. EcolApps
    Developing custom computer vision models with Njobvu-AI: A collaborative, user-friendly platform for ecological research
    Cara L Appel, Ashwin Subramanian, Jonathan L Koning, Marnet Ngosi, Christopher M Sullivan, Taal Levi, and Damon L Lesmeister
    Ecological Applications, 2025
    HTML PDF

2023

  1. arXiv
    Njobvu-AI: An open-source tool for collaborative image labeling and implementation of computer vision models
    Jonathan S Koning, Ashwin Subramanian, Mazen Alotaibi, Cara L Appel, Christopher M Sullivan, Thon Chao, Lisa Truong, Robyn L Tanguay, Pankaj Jaiswal, Taal Levi, and Damon Lesmeister
    arXiv preprint, 2023
    HTML PDF