Pocket Lab: Turn Your Smartphone into a 3D Plant Scanner
- Beschreibung
Affordable, non-invasive monitoring of plant structures is essential for precision farming and modern agricultural management. Conventional 3D phenotyping systems, such as LiDAR or structured light scanners, are expensive and complex.
This project explores a hands-on, low-cost method to create accurate three-dimensional(3D) models of small potted plants using only a smartphone camera and freely available open-source tools. Participants will capture overlapping photographs of plants and process these images with an open-source photogrammetry workflow to produce a high-density 3D point cloud.
To enrich the analysis, students will integrate a lightweight artificial intelligence (AI) component: a pre-trained or small custom model will automatically extract traits, such as any anomalies or diseases.
- Beschreibung des interdisziplinären Teils des Projekts
- This project brings together knowledge and methods from several disciplines to create a hands-on, innovative learning experience. From agricultural sciences, participants will learn about plant growth, traits, and phenotyping needs. From computer science and engineering, they will use photogrammetry to build 3D models and apply artificial intelligence (AI) for automatic trait detection. At the same time, data science and image processing skills will be developed through working with open-source tools and analyzing plant structures. By combining agriculture, technology, and AI, the project demonstrates how interdisciplinary collaboration can make precision farming more affordable, accessible, and sustainable.
- Projektzeitraum
- Wintersemester 2025/2026
- Bewerbungszeitraum
- 13. bis 27.10.2025
- Durchführung
- semesterbegleitend
- Details zu Projektzeitraum und Durchführung
The project will be conducted during the winter semester. Smartphone image acquisition and simple reference measurements will take place at the Institute of Agricultural Engineering(440e). Data processing, AI-based trait extraction and analysis will be carried out on the institute's computers.
- Studienfach
- offen für alle Studienfächer
- Betreuende
- Khandoker Ahammad, Prof. Dr. Joachim Müller
- Institut
- Institut für Agrartechnik (440) (Plant Phenotyping,Precision Agriculture,Computer Vision, 3D Modeling,Artificial Intelligence (AI))
- Sprache
- deutsch/englisch
- Teilnehmendenanzahl
- min. 1, max. 1
- Arbeitsaufwand
-
ca. 150 Stunden pro Teilnehmende:r
| 5
ECTS-Punkte
Arbeitsaufwand (Stunden und ggf. ECTS) sind ungefähre Angaben. Die tatsächlich vergebenen ECTS-Punkte ergeben sich aus der tatsächlich geleisteten Arbeit.
- Für dieses Projekt ist kein Motivationsschreiben des Studierenden erforderlich
- Projektart
- experimentell
- Lernziele
-
Die Teilnehmende lernen in diesem Projekt:
In this project, participants will learn to:
- Formulate and process a scientific research question on plant phenotyping.
- Understand the principles of multi-view photogrammetry and 3D model creation.
- Capture and process smartphone images for 3D reconstruction.
- Use open-source software for data processing and trait analysis.
- Apply a lightweight AI model to detect traits.
- Anmerkungen für Studierende
- Motivated and interested in the topic.
- It is possible to integrate the project into a bachelor thesis
- Schlagworte
- 3D phenotyping, open-source photogrammetry, AI-assisted trait extraction, low-cost plant monitoring, non-invasive plant measurement