As a PhD student in this project on Autonomous Bovine Behaviour Analysis, you will develop and study autonomous methods for the detection of cow-health issues based on colour and depth (RGB-D) cameras. The aim is to develop an early-warning system that signals health issues before the cow really gets in trouble. In this project, we will specifically utilize the latest deep-learning technologies to detect the individual cows, track their posture over time, and classify their behaviour. This behaviour is related to the cow's health and welfare. The focus will be on the development of a monitoring tool to prevent lameness in cattle to increase welfare and profitability of dairy cows. The solution should pinpoint those animals that are prone to develop diseases and are in need of immediate care by the farmer. The project is part of a research cooperation existing of Dutch experts in the area of deep learning and image processing, funded by the NWO-TTW program "Efficient Deep Learning (EDL)".

What you tell on birthday parties
I apply artificial neural networks to develop a "data-driven cow-whisperer" that senses the wellbeing of cows.
What we really need
What we really are looking for is a PhD-student who is eager to investigate state-of-the-art deep-learning methods for video processing of cows and relate this to their health and production data. You should have demonstrable experience with computer vision and machine learning. Besides a strong technical background, an affinity with livestock farming is desirable. You need to be able to work independently, but also in collaboration with other PhDs in the EDL project, as well as with stakeholders in the field of livestock farming.
A typical day at the office
Your days at the office range from frustrating farm data-downloads to energy-boosting acceptance letters of your ground-breaking articles, and anything in between! Today's challenge is to use Efficient Deep Learning to identify the posture of cows in camera images. First check of the day: is your on-farm data correctly collected? You start to think of an experiment to test your recently developed image-processing algorithm. You explore the preliminary results and over a mug of coffee, you discuss your ideas about deep neural-networks with Gert and Rik who have been dealing with these issues each in their specific domains. At lunch-time, you'll join Marjolein, Eldert and Sam to have lunch in the main building, and take a relaxing de-tour back to your office. The fresh air makes you consider developing a probabilistic model that predicts the cows' posture more robustly. But first you start a discussion with a fellow student as your sparring-partner, after which you sit back to implement the method. Good job! You have shown that the method is able to accurately predict 88.5% of the lameness cases. Now you start planning how you can concisely write your results into a scientific paper and how the application can be used by the members from the users committee of the project.

You (soon will) have a MSc degree in Computer Science, Artificial Intelligence, Computational Science or closely related area. You have demonstrable experience with computer vision and machine learning. Ideally, you have expertise in deep learning, but PhD-level courses are available if such expertise is missing. A keen interest in (precision) livestock farming is a preference. You should be able to work well in teams. Additional responsibilities of a PhD student include a limited amount of teaching and supervision of BSc and MSc students on topics related to the PhD work. Successful candidates are expected to write scientific papers for international peer-reviewed journals, present his/her work in international conferences, and finalize the work in writing a PhD thesis.

• We offer a full-time position (38 hours), initially for 1 year after which a go/no go decision will be taken on extension with another three years.
• You will be appointed under the CAO of the Dutch Universities.
• Gross salary per month € 2.222,-- in the first year for a fulltime appointment.
• In addition we offer a holiday allowance of 8% and an end-of-the year bonus of 8.3% of your annual salary.
• Academic training in the WIAS-program at the world's best-ranked University in Agriculture. Good academic working atmosphere and an inspiring university campus where there is still space to breath fresh-air.
• A stakeholder that is eager to implement your work in a practical situation

Wageningen University & Research

The mission of Wageningen University & Research is "To explore the potential of nature to improve the quality of life". Within Wageningen University & Research, nine specialised research institutes from the Wageningen Research Foundation and Wageningen University have joined forces to help answer the most important questions in the domain of healthy food and living environment.

With approximately 30 locations, 5,000 employees, and 10,000 students, Wageningen University & Research is one of the leading organisations in its domain worldwide. An integrated approach to problems and the cooperation between various disciplines are at the heart of the unique approach of Wageningen.

For further information about working at Wageningen University & Research, take a look at thespecial career site.

You will be working at the Wageningen University in the Farm Technology Group, one of the world's leading research groups on precision livestock farming and applications of computer vision in agriculture in general. You will be supervised by two assistant professors, Gert Kootstra having expertise in computer vision and artificial intelligence and Rik van der Tol with specific domain knowledge of the application. Eldert van Henten will be your promotor and will guide you in your scientific training. It is your job to merge their views and experiences into an excellent PhD-thesis in such a way that the total outcomes are beyond the sum of the parts. The project is part of the NWO-TTW program Efficient Deep Learning (EDL) with over 20 PhD students appointed at different Dutch universities, with whom you can collaborate and from whom you can learn.