| Project/Area Number |
16KK0175
|
|---|
| Research Category |
Fund for the Promotion of Joint International Research (Fostering Joint International Research)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Insect science
|
|---|
| Research Institution | Okinawa Institute of Science and Technology Graduate University |
|---|
Principal Investigator |
Mikheyev Alexander 沖縄科学技術大学院大学, 生態・進化学ユニット, 准教授 (90601162)
|
|---|
| Project Period (FY) |
2016 – 2018
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥13,130,000 (Direct Cost: ¥10,100,000、Indirect Cost: ¥3,030,000)
|
|---|
| Keywords | animal behavior / social insects / machine learning / neural networks / social organization / genetic traits |
|---|
| Outline of Final Research Achievements |
This project aimed to use machine vision to track individual honey bees in the social insect colony.While humans identify individual bees and their behavior easily, they don’t have time to look at thousands of individuals over months of observation. Computers can do that, but identifying individuals in a densely packed environment is a challenging computational task. Here we developed new computational approaches for the tracking of unmarked individuals in a densely packed hive and used this methodology to track hive behavior over the course of weeks, providing the longest continuous observation of an undisrupted social insect colony to date. Training data for the main project was acquired through interaction with human agents. While data were being passively acquired for the main project, we used the same technological toolkit to solve another scientific problem-the web-based visualization of phylogenetic. The results of this work are now publicly available at https://phylogeny.io
|
| Academic Significance and Societal Importance of the Research Achievements |
Understanding collective interactions between individuals requires advances in efficient tracking methods. Methods developed in the course of this grant allowed for tracking unmarked individuals and will improve the implementation of monitoring frameworks.
|
