Thesis subject

Artificial intelligence for Crop Disease/Weed Detection in Extended Realities (MSc)

AI can play a crucial role in enabling rapid and accurate disease or weed identification for both indoor and outdoor crops, reducing the reliance on manual inspection and increasing the efficiency of response measures. Yet, the AI identification process needs communicating to the end user. For this, the advanced communication potential of extended realities could offer an ideal solution.

Short description

AI can play a crucial role in enabling rapid and accurate disease or weed identification for both indoor and outdoor crops, reducing the reliance on manual inspection and increasing the efficiency of response measures. Yet, the AI identification process needs communicating to the end user. For this, the advanced communication potential of extended realities could offer an ideal solution.

Extended realities (XR) is a catch-all term to refer to augmented reality (AR), virtual reality (VR), and mixed reality (MR). The technology is intended to combine or mirror the physical world with “digital content”, of which the user is able to interact with. XR can enhance the user experience and immersion by overlaying digital information onto the real world or creating entirely virtual environments.

This topic involves tasks to train Machine/Deep learning models using diverse datasets of crop disease images and devising a pipeline for its integration into XR applications, allowing farmers to use their smartphones to scan crops and receive instant disease diagnosis and treatment recommendations.

    Objectives

    • Review previous work and datasets on the application of machine/deep learning approaches for plant disease detection in XR
    • Integrate trained deep learning models using Python (e.g. pyTorch, Tensorflow, sci-kit learn) and Unity Software (support with the coding can be provided by the supervisor)
    • Simulate different crop conditions and test performance of trained models

      Tasks

      The work in this thesis entails:

      • Conduct background study and propose an integration plan for XR
      • Look into integrating the trained AI model in XR in unity
      • Design and develop a roadmap to integrate AI and XR
      • Using XR hardware (provided by the supervisors), simulate different crop conditions and test the performance and robustness of the trained deep learning models under various scenarios.

          Literature

          • Anastasiou, Evangelos, Athanasios T. Balafoutis, and Spyros Fountas. "Applications of extended reality (XR) in agriculture, livestock farming, and aquaculture: A review." Smart Agricultural Technology (2022): 100105.
          • Naftali Slob, William Hurst, Rick van de Zedde, Bedir Tekinerdogan, Virtual reality-based digital twins for greenhouses: A focus on human interaction, Computers and Electronics in Agriculture, Volume 208, 2023, 107815, ISSN 0168-1699


          Requirements

          • Courses: Programming in Python (INF-22306), (Optional), Data Science Concepts (INF-34306), Artificial Intelligence (INF-36306) or Machine Learning (FTE-35306), software engineering
          • Required skills/knowledge: Basic data analytics/machine learning and willingness to learn new software tools, interest about XR, unity software.

            Key words: Artificial Intelligence, Data analytics, Deep learning, Immersive Technology, Information Systems, Information technology, Internet of Things, IoT Sensors

            Contact person(s)