Malika Ratnayake

Getting our @raspberrypi.com–based insect monitoring units ready for field deployment. 🐝🌸🌾

Welcome sign

Exciting first day at the National Geographic Explorers Festival 2025! Amazing to meet fellow explorers in person for the first time and hear about their inspiring work. #NatGeo

2025 starts with a methodology publication by Ștefan et al. on Utilising affordable smartphones and open-source time-lapse photography for pollinator image collection and annotation - Happy new year!
doi.org/10.26786/192...

Blue-banded bee

Finally spotted a blue-banded bee in my garden.
#nativebee #Australia

Malika Nisal Ratnayake, Lex Gallon, Adel N. Toosi, Alan Dorin
Motion-based video compression for resource-constrained camera traps
https://arxiv.org/abs/2405.14419

Could you please add me? Thanks.

A bee entering the bee hotel

My bee hotel just welcomed its very first guest! 🐝🌸 #BeeHotel #PollinatorFriendly

I would love to be added! Thank you :)

We’ve published the complete dataset, including honeybee videos, annotated images for YOLOv2, and extracted insect tracks, on @monashuniversity.bsky.social and @figshare.bsky.social

doi.org/10.26180/5f4...
(6/6)

Poster presented at CVPR2021

I presented this research at the #CV4Animals workshop held in conjunction with @cvprconference.bsky.social 2021. #CVPR

malikanr.com/images/CV4An...
(5/6)

Tracking individual honeybees among wildflower clusters with computer vision-facilitated pollinator monitoring Monitoring animals in their natural habitat is essential for advancement of animal behavioural studies, especially in pollination studies. Non-invasive techniques are preferred for these purposes as t...

We used the HyDaT algorithm to extract data on honeybee behaviour, including (a) movement, (b) heatmaps, (c) visible times, (d) speeds, and (e) turn-angles in two ground covers.

doi.org/10.1371/jour...
(4/6)

GitHub - malikaratnayake/HyDaT_Tracker: Tracking insect pollinators in complex dynamic environments Tracking insect pollinators in complex dynamic environments - malikaratnayake/HyDaT_Tracker

HyDaT also has an occlusion identification algorithm that uses background subtraction to identify insect occlusions and field-of-view exits.

HyDaT is available at @github.com repo github.com/malikaratnay...

(3/6)

Our algorithm, HyDaT (Hybrid Detection and Tracking), combines background subtraction with deep learning-based object detection #YOLO to accurately detect and track honeybees, even amidst environmental changes. It achieved an accuracy of 86.6% on a custom dataset.

(2/6)

Tracking individual honeybees among wildflower clusters with computer vision-facilitated pollinator monitoring Monitoring animals in their natural habitat is essential for advancement of animal behavioural studies, especially in pollination studies. Non-invasive techniques are preferred for these purposes as t...

Starting my journey on @bsky.app with a recap of my #PhD.
#1 instalment:

My 1st paper published in @plos.bsky.social
presented a #ComputerVision and #DeepLearning-based algorithm to track #honeybees in outdoor environments through long occlusions. 🐝📹

journals.plos.org/plosone/arti...

(1/6)

### NEW FUNDING OPPORTUNITY ###

Global Joint Synthesis Call

"𝗜𝗻𝘁𝗲𝗿𝗮𝗰𝘁𝗶𝗼𝗻𝘀 𝗯𝗲𝘁𝘄𝗲𝗲𝗻 𝗕𝗶𝗼𝗱𝗶𝘃𝗲𝗿𝘀𝗶𝘁𝘆 𝗮𝗻𝗱 𝗖𝗹𝗶𝗺𝗮𝘁𝗲 𝗖𝗵𝗮𝗻𝗴𝗲"

Funding of:

➡️ Working groups
➡️ Postdocs
@several synthesis centres/initiatives🔥❤️😍

PLEASE SPREAD WIDELY! THX!

www.idiv.de/research/sdi...

Thrilled to join Bluesky! 🦋

Here’s an artistic take on my work, courtesy of @openaidalle.bsky.social.

📷🐝🪰