Fruit Detection and Yield Mass Estimation from a UAV Based RGB Dense Cloud for an Apple Orchard

Research output: Journal contributionsJournal articlesResearchpeer-review

Authors

  • Marius Hobart
  • Michael Pflanz
  • Nikos Tsoulias
  • Cornelia Weltzien
  • Mia Kopetzky
  • Michael Schirrmann

Precise photogrammetric mapping of preharvest conditions in an apple orchard can help determine the exact position and volume of single apple fruits. This can help estimate upcoming yields and prevent losses through spatially precise cultivation measures. These parameters also are the basis for effective storage management decisions, post-harvest. These spatial orchard characteristics can be determined by low-cost drone technology with a consumer grade red-green-blue (RGB) sensor. Flights were conducted in a specified setting to enhance the signal-to-noise ratio of the orchard imagery. Two different altitudes of 7.5 m and 10 m were tested to estimate the optimum performance. A multi-seasonal field campaign was conducted on an apple orchard in Brandenburg, Germany. The test site consisted of an area of 0.5 ha with 1334 trees, including the varieties ‘Gala’ and ‘Jonaprince’. Four rows of trees were tested each season, consisting of 14 blocks with eight trees each. Ripe apples were detected by their color and structure from a photogrammetrically created three-dimensional point cloud with an automatic algorithm. The detection included the position, number, volume and mass of apples for all blocks over the orchard. Results show that the identification of ripe apple fruit is possible in RGB point clouds. Model coefficients of determination ranged from 0.41 for data captured at an altitude of 7.5 m for 2018 to 0.40 and 0.53 for data from a 10 m altitude, for 2018 and 2020, respectively. Model performance was weaker for the last captured tree rows because data coverage was lower. The model underestimated the number of apples per block, which is reasonable, as leaves cover some of the fruits. However, a good relationship to the yield mass per block was found when the estimated apple volume per block was combined with a mean apple density per variety. Overall, coefficients of determination of 0.56 (for the 7.5 m altitude flight) and 0.76 (for the 10 m flights) were achieved. Therefore, we conclude that mapping at an altitude of 10 m performs better than 7.5 m, in the context of low-altitude UAV flights for the estimation of ripe apple parameters directly from 3D RGB dense point clouds.

Original languageEnglish
Article number60
JournalDrones
Volume9
Issue number1
Number of pages19
DOIs
Publication statusPublished - 16.01.2025

Bibliographical note

Publisher Copyright:
© 2025 by the authors.

    Research areas

  • apple trees, fruit detection, structure from motion (SfM), unmanned aerial vehicle (UAV), yield estimation
  • Environmental Governance

DOI

Recently viewed

Publications

  1. Assessment of cognitive load in multimedia learning with dual-task methodology
  2. FROM THE EDITORS ERRORS IN ORGANIZATIONS
  3. The end of certainties
  4. The role of learning strategies for performance in mathematics courses for engineers
  5. Clusteranalyse als Methode zur Strukturierung großer Datenmodelle
  6. Evaluating the (cost-)effectiveness of guided and unguided Internet-based self-help for problematic alcohol use in employees
  7. Geometric Properties on the Perfect Decoupling Disturbance Control in Manufacturing Systems
  8. Semi-Supervised Generative Models for Multi-Agent Trajectories
  9. HPLC and chemometrics-assisted UV-spectroscopy methods for the simultaneous determination of ambroxol and doxycycline in capsule.
  10. Changes in processing characteristics and microstructural evolution during friction extrusion of aluminum
  11. Tschick
  12. Implementing the Kyoto Protocol without Russia
  13. Homogenization for a non-local coupling model
  14. Developing robust field survey protocols in landscape ecology
  15. Can guided introspection help avoid rationalization of meat consumption?
  16. Participation in multi-level policy implementation: exploring the influence of governance culture
  17. Microstructure and mechanical properties of Mg-3Sn-1Ca reinforced with AlN nano-particles
  18. Dynamic Semantic Web Content for Museum Guides
  19. Decision-making models for Robotic Warehouse
  20. Culture as an Engine of Local Development Processes
  21. Using (Quantitative) Structure-Activity Relationships in Pharmaceutical Risk Assessment
  22. Key criteria for developing ecosystem service indicators to inform decision making
  23. Going beyond certificates
  24. System versus Intention
  25. One step forward, two steps back
  26. Techno-economic assessment of non-sterile batch and continuous production of lactic acid from food waste
  27. Analysis of Dynamic Response of a Two Degrees of Freedom (2-DOF) Ball Bearing Nonlinear Model
  28. Practical Formalist
  29. Co-EM Support Vector learning
  30. Estimation of minimal data sets sizes for machine learning predictions in digital mental health interventions
  31. Organizing Half-Things: Knowing, Theorizing and Studying Atmospheres
  32. The differential effects of self-view in virtual meetings when speaking vs. listening
  33. Implementation of EU labour law directives by way of national collective agreements
  34. Timing and fragmentation of daily working hours arrangements and income inequality
  35. A web- And mobile-based intervention for comorbid, recurrent depression in patients with chronic back pain on sick leave (get.back)
  36. Desynchronization of the Public and the Private
  37. Temporal and thermodynamic irreversibility in production theory
  38. You Are Where You Eat: A Theoretical Perspective on Why Identity Matters in Local Food Groups