Camfi

Camfi enables long-term, non-invasive population monitoring and high-throughput behavioural observation of low-flying insects using images and videos from wildlife cameras with automated detection and annotation via Mask R-CNN.

Key Features:

• Automated Insect Detection: Detects and annotates flying insect targets in still images or video clips using the Mask R-CNN framework.
• Transfer Learning: Can be trained to detect and annotate insects in a few hours through transfer learning.
• Large-Scale Monitoring Support: Facilitates very large monitoring programs with high throughput.
• Remote Deployment: Operates in remote areas without requiring vehicle access to field sites.

Scientific Applications:

• Insect Ecology and Conservation: Supports ongoing efforts to understand the behaviour and ecology of declining insect populations.
• Agronomy: Can be applied to agricultural monitoring of flying insect activity and abundance.
• Behavioural Observation: Enables high-throughput behavioural observation of low-flying insects.

Methodology:

Camfi employs wildlife cameras to obtain images and videos of low-flying insects, then applies the Mask R-CNN framework for automatic detection and annotation of flying insect targets, with training accelerated through transfer learning.