Plant part segmentation for robotic fruit harvesting using deep learning

Study level

Vacation research experience scheme

Topic status

We're looking for students to study this topic.


Dr Chris Lehnert
Lecturer in Robotics and Autonomous Systems
Division / Faculty
Science and Engineering Faculty


For a robot to successfully harvest fruit it needs to understand what parts of the plant should be avoided, such as the main stem or branches of a plant.

The aim of this project is to create a deep learning system for detecting and segmenting key parts of plants for use in robotic fruit harvesting. This work will look at developing a semantic segmentation deep learning system that can identify the key parts of the plant for use on a QUT-developed robotic fruit harvester.

Research activities

As part of the research project, you will be involved in:

  • training deep convolution neural networks for semantic segmentation
  • testing deep neural networks for performance on real world environments.


On completion of the project, we expect to have created:

  • a semantic segmentation method for classifying and segmenting different parts of plants
  • a performance analysis of the semantic segmentation method.

Skills and experience

To be considered for the research project, you should have skills and/or experience in:

  • Python programming
  • understanding of deep neural networks
  • machine learning techniques such as supervised learning
  • computer vision techniques, such as image-based object classification and image-based object segmentation.



Contact the supervisor for more information.