Project Overview
StickyBeak AI represents a breakthrough in conservation technology, seamlessly merging Internet of Things connectivity with artificial intelligence to revolutionise wildlife monitoring across Tasmania. Named after the curious currawong, this intelligent system transforms the conventional camera trap experience into an efficient, data-rich conservation tool that amplifies the impact of citizen scientists.
At its core, StickyBeak AI makes use of an interconnected network of camera traps deployed by over 450 private landholders throughout Tasmania's diverse landscapes. These motion-activated cameras function as remote sensing nodes, capturing wildlife activity and automatically transmitting data to a centralised platform where AI algorithms process and categorise images with remarkable precision.
The system addresses the critical bottleneck in wildlife monitoring: the overwhelming volume of imagery that requires human review. By automatically filtering out blank frames triggered by environmental factors and identifying native species, StickyBeak AI has already analysed over 774,000 images and saved more than 900 hours of volunteer time, equivalent to 23 working weeks.
What makes StickyBeak extraordinary is its potential to learn and improve continuously as more data flows through the system. We work closely with the University of Tasmania, where we are striving to create a seamless future path for citizen scientists to feed improvement directly back into the AI powering StickyBeak. Unlike static platforms, this IoT ecosystem becomes more intelligent and accurate with each image processed, creating a virtuous cycle of improvement that enhances conservation outcomes.
Project Commissioner
Project Creator
Team
Tim Askey - Creative Lead Developer
Kavie Mo - Backend Developer
Glen Bains - Subject matter expert Tasmanian Land Conservancy
Project Brief
Imagine a network of hundreds of digital eyes scattered across Tasmania's wilderness, farms and backyards, silently watching for signs of wildlife. Now imagine those eyes connected to a brain that can instantly distinguish between a curious pademelon, a passing wombat, or simply leaves dancing in the wind. This was the vision that inspired StickyBeak AI—a revolutionary IoT system that transforms ordinary camera traps into networked conservation intelligence.
The Tasmanian Land Conservancy approached Ionata Digital with an ambitious challenge: design a system that could harness the enthusiasm of citizen scientists while solving the data processing bottleneck that threatened to overwhelm their successful WildTracker program. With over 450 landholders collectively capturing hundreds of thousands of wildlife images, the manual review process had become unsustainable, with volunteers spending countless hours sorting through empty frames rather than contributing to meaningful conservation work.
Our brief was to create a seamless experience that would integrate sophisticated artificial intelligence into the existing ecological monitoring workflow without requiring technical expertise from users. The solution needed to automatically filter out empty images triggered by environmental factors, assist with species identification, and scale efficiently as the program grew—all while maintaining the engaging, community-driven ethos that made WildTracker successful.
The ultimate goal is to create an IoT ecosystem where every connected camera becomes not just a passive recorder but an active contributor to conservation science, empowering everyday Tasmanians to participate meaningfully in protecting the island's unique biodiversity.
Project Innovation/Need
Tasmania’s extraordinary wildlife faces increasing threats from habitat loss, climate change, and invasive species. Monitoring these pressures at scale is beyond what professional conservation teams can manage alone. The WildTracker program meets this challenge by empowering private landholders to become citizen scientists. However, its very success created a major bottleneck: processing the flood of camera trap images—over 700,000 to date—became unsustainable, with nearly one-third of images containing no wildlife at all.
StickyBeak was developed to address this critical need. By integrating AI into the image-sorting process, StickyBeak dramatically reduces the tedious time volunteers spend manually reviewing empty frames, removing the bottleneck, restoring enthusiasm and unlocking ecological insights hidden in vast data stores (>700,000 to date). Yet most existing identification models lack training on Tasmania’s fauna. StickyBeak is specifically trained with local data and algorithms to recognise species such as the eastern barred bandicoot and Tasmanian devil with high accuracy.
Looking ahead, our goal is to create a seamless, closed-loop system—allowing citizen scientists to not only collect data, but actively improve StickyBeak’s performance through their interactions. We will also develop direct integration pathways with statewide datasets, enabling real-time insights into species distribution and landscape-scale wildlife trends.
StickyBeak is more than an AI tool—it’s transforming the relationship between citizen scientists and conservation data. By continuing to improve efficiency, accuracy, and engagement, we are building a smarter, more responsive system that places both people and wildlife at the heart of WildTracker’s future.
Design Challenge
StickyBeak delivers a transformative user experience by seamlessly integrating sophisticated artificial intelligence into the familiar wildlife monitoring workflow. The design philosophy centres on making complex technology invisible, allowing users to focus entirely on their connection with wildlife rather than the machinery behind it.
For the citizen scientist in the field, the experience begins with their existing camera trap setup, no additional hardware required. Once images are captured, users upload them through the intuitive WildTracker interface, where StickyBeak immediately begins processing in the background. Within moments, the system automatically filters out empty frames and suggests species identifications for the remaining images, presenting results in a clean, visually-oriented gallery view.
The interface uses subtle visual cues to distinguish between AI-processed and human-verified identifications, maintaining transparency while guiding users through the verification process. Each identification includes a confidence rating, empowering users to make informed decisions about accepting or correcting AI suggestions. This collaborative approach creates a sense of partnership between human and machine intelligence, respecting the expertise of experienced naturalists while reducing their workload.
As users confirm or correct identifications, the system learns from these interactions, creating a personalised experience that adapts to the specific wildlife patterns on individual properties. Landholders receive periodic insights about wildlife activity on their land, delivered through engaging visualisations that transform raw data into meaningful stories about the ecosystems they're helping to monitor.
The streamlined workflow has significantly reduced tedious sorting tasks, removing empty images while enhancing the rewarding aspects of wildlife discovery.
Future Impact
StickyBeak AI represents an advancement in conservation technology design, demonstrating how artificial intelligence can seamlessly integrate into existing workflows to amplify human expertise rather than replace it. The tool's success in processing over 774,000 images while saving 900+ volunteer hours establishes a new standard for community-driven conservation platforms.
The system's future potential as a closed-loop learning mechanism can fundamentally transform how citizen scientists contribute to conservation science. As users verify and correct AI identifications, StickyBeak has the potential to become increasingly intelligent, creating a collaborative partnership between human expertise and machine learning that benefits conservation efforts globally. This approach positions individual landholders as active contributors to conservation science rather than passive data collectors.
Other potential future developments include direct integration with national biodiversity databases, enabling species distribution mapping and landscape-scale wildlife trend analysis.
StickyBeak's design philosophy of making sophisticated technology invisible creates a template for AI integration across environmental monitoring applications. The tool's ability to maintain user engagement while dramatically improving efficiency demonstrates how thoughtful design can solve complex conservation challenges through technology adoption rather than technological complexity.
As climate change accelerates threats to global biodiversity, StickyBeak's scalable approach to wildlife monitoring becomes increasingly critical. The tool's proven ability to engage communities while generating scientific-quality data positions it as a model for conservation technology worldwide, transforming how we understand and protect wildlife through intelligent design.
NEXT Futures - Environmental Sustainability
Projects that encourage a positive environmental impact through understanding and implementation of the Sustainable Development Goals.
More Details