This course is not scheduled yet.
In the age of information, we have access to an unprecedented level of details of the animal life on Earth. The possibilities of exploration and investigation of our planet seem, at times, limitless. At the same time the increasing biodiversity crisis calls for more data and more detailed investigation on the relationships between animal behaviour and movement within the environment. Understanding how animals move across landscapes, how the landscape (i.e. environment) changes over time, and the intersection between those topics will allow us to solve long-standing puzzles in ecology.
To address these fundamental questions on animal movement and environmental change, researchers have been using a plethora of technologies to gather relevant data. In this course, attendees will learn the fundaments of various animal tracking and remote sensing techniques and how to process and to analyse the resulting large datasets.
Participants will learn about the usage of animal tracking technologies such as GPS, automated radiotracking, geolocators, camera traps, radiofrequency identification readers, and accelerometers. In the course, we will focus on a unifying aspect of all these methods, namely the processing of these extremely large, raw datasets to produce structured, analysable, and visualizable datasets. This is common ground for researchers using all of these techniques. As such, participants will also learn about the nature of animal tracking data on different scales, i.e., filtering and summarizing data, error estimations, calculating locations and space use including social network analysis i.e., environment changes, land use, species distribution using machine learning, in particular deep learning for visual data. Participants will be introduced to foundational deep learning concepts for the detection of animals in the wild using deep convolutional neural networks (CNNs) applied to camera images.
List course objectives
- Provide students with the theoretical framework on - and practical understanding of state-of-the-art tracking technologies
- Have students understanding the challenges on the usage of such tools
- Provide students with the practical knowledge on analyzing spatial data and Deep Convolutional Neural Networks.
- Describe and assess different tracking technologies
- Determine most suitable technologies for their projects
- Identify data accuracy in tracking datasets
- Define and apply Spatial data analysis of wild animals
- Define and apply Deep Neural Networks & Deep Convolutional Neural Networks
PhDs and postdocs and others using animal tracking and monitoring techniques
in-depth, post graduate
A minimum of 10 and a maximum of 25 participants
2.5 days given every two years
|Home work/ self- study||Reading articles that are provided by trainers|
|Credit points||0.8 ECTS|
|Course content coordinator||Filipe Cunha (firstname.lastname@example.org)|
|Name lecturers||Filipe Cunha, Chris Tyson, Aneesh Chauan & Marc Naguib|
|In-person||Wageningen University campus, Wageningen, the Netherlands.
Note: Room and building will be communicated to registered and confirmed participants about one week before the start of the course.
- History of Tracking
- Why do we track animals?
- Examples of usage of animal tracking technologies
- Challenges of animal tracking technologies
- Where to gather data online?
- How accurate is the available data?
- What is accuracy?
- Movebank (and similar data repositories)
- Getting data from the wild into databases
- Determining its accuracy
- The different tracking systems and their limitations
- Real-time GPS
Analysis of spatial data from automatized radio tracking (workshop)
- Cellular Tracking Technology
- Why deep learning: Motivating cases
- Deep Neural Networks: Foundations of deep learning and key concepts
- Deep Convolutional Neural Networks: Breakthroughs of the last decade
- Object detection and instance segmentation
- Introduction the problem
- Introduction to the tools to be used for the workshop
- Data preparation
- Transfer learning
- Training your own network
- Tuning hyperparameters
- Performance metrics
- Diving deep into what the networks learn
- Closing statements
Fee (includes study and training material, coffee/tea and lunches)
1) Reduced fee: PhD candidates of Wageningen University doing a Training and Supervision Plan (TSP)
2) University fee: All other PhD candidates / postdocs and staff of Wageningen University
|3) Other researchers||€600|
You may cancel free of charge up to 1 month before the start of the course. After this date you will be charged with the University fee, unless:
- You can find someone to replace you in the course and supply the course coordinator with the name and contact information of your replacement. In this case you will be charged a €50,- cancellation fee.
- You are a PhD of Wageningen University with a valid reason to cancel (illness or death in the family 1st or 2nd degree). In this case you will be charged the reduced fee and your supervisor/PI must send a mail indication the reason for cancellation.
Note: WUR PhD candidates have priority. If the course is fully booked, PhDs from other universities will be invited to join the course based on registration date.
Information: For more information please contact: email@example.com, tel: +31317486836 or firstname.lastname@example.org