New to Magnet.me? Sign up now to get your own personalised job matches.

Sign up

New to Magnet.me? Sign up now to get your own personalised job matches.

Sign up

Classification and segmentation of vegetables in Deep Learning detections

Internship in Den Haag, the Netherlands
Posted

Degree level

Master

Job function

Engineering

Required language

English (Fluent)

Work experience

0 to 10 years

Salary

€300

Find your job faster with a profile

Directly see your job matches and find a job that's perfect for you.

Ever thought your knack for technology could change the way we feed the world? At VanBoven, we make agriculture more data-driven by developing predictive solutions using drone imagery and AI. We currently have an internship position open, so grab your chance and join us on our mission.

Job Description

At VanBoven we have developed proprietary Deep Learning technology to automatically analyze agricultural drone imagery. In fields with over 150k plants, our algorithms are able to identify every single one of them. We determine every plants’ size, and compare it with previous recordings to establish growth patterns. When harvest time comes around we take it to the next level and determine which broccolis, lettuces and cabbages are ready to be harvested.

Our models output bounding boxes describing the location and shape of vegetables, like shown in the images above. These boxes, however, provide limited shape information and some detections might even be false. Can you train a machine to inspect the identified regions and (a) confirm whether the detection is correct and (b) determine the shape of the identified vegetable?

A summary of your work

  • Develop a method to assess the likelihood of a detection being a false positive.
  • Develop a method to determine the shape of the identified vegetable (image segmentation).
  • Setup experiments to evaluate your work.
  • Optimize your solution to finetune performance.

What do we expect from you

  • Msc. level thinking with analytical focus and hands-on mentality.
  • Not afraid to dive into research literature.
  • Experience with classic computer vision and deep learning.
  • Experience with Python and relevant libraries (e.g. OpenCV, keras, tensorflow, PyTorch).
  • Available for at least 5 month.
  • Fluency in English - Dutch is a bonus.

What we offer

  • €300/month full-time internship reimbursement.
  • Hands-on learning about Deep Learning in industry.
  • Flexible work hours and location.
  • Travel allowance.
  • Being part of a quickly growing team

Interested?

Cool - Please send your CV and a short motivation to kaz@vanboven-drones.nl and we'll take it from there. Thanks!

Share:
Logo VanBoven
About VanBoven
to company page

Bij VanBoven werk je aan een duurzame toekomst voor de land- en tuinbouw waarin data, drones en machine learning een belangrijke rol spelen. VanBoven analyseert agrarische drone-opnamen en past machine learning algoritmen toe om de oogst per individuele plant te voorspellen. De inzichten die hieruit volgen worden overzichtelijk aan de boer én zijn afnemers getoond waarmee de gehele waardeketen, van 'farm to fork', wordt geoptimaliseerd.

Engineering
Den Haag
Active in 1 country
8 employees
70% men - 30% women
Average age is 25 years

What employees are saying

Juliette

Data Analysis Intern

Juliette

Tijdens mijn stage wilde ik graag leren werken met Python. Ik kreeg de ruimte om, samen met medeoprichter Eric, een opdracht vast te stellen waarin ik mezelf kon ontwikkelen op dat gebied. Zo heb ik geleerd wat ik wilde én een leuke tijd gehad.

Martijn

Thesis Student

Martijn

Bij VanBoven wordt er naar je geluisterd. Het team staat open open voor een andere invalshoek of nieuwe aanpak. Zo krijg ik de ruimte zelf te bepalen welke oplossing het best is en die ook uit te voeren.