Using Machine Learning/Artificial Intelligence to assist crop protection decisions on kiwifruit

Data mining/Machine Learning algorithms are used to develop models to forecast the outcome of leafroller pest monitoring decisions on ‘Hayward’ kiwifruit crops in New Zealand. Using industry spray diary and pest monitoring data gathered at an orchard block level for compliance purposes, 80 attributes (independent variables) were created in three categories from the spray diary data: (1) individual insecticide applications applied during 2-week time windows, (2) groups of insecticide applications within time periods prior to or after fruit set and (3) orchard management attributes. Five machine learning algorithms (Decision Tree, Naïve Bayes, Random Forest, AdaBoost, Support Vector Machine) and one statistical method (Logistic regression) (classifiers) were used to develop models to forecast insecticide application decisions for leafroller control, by predicting whether pest monitoring results were above or below a spray threshold. Models to forecast 2011 spraying decisions were trained on 2008 and 2009 data and tested on 2010 data. Forecasts were made for spray and no-spray decisions based upon pre-determined acceptable rates of precision (proportion of correct decisions in test results). Orchard blocks in which a forecast could not be made to a prescribed degree of precision were recommended to be monitored, which is the normal practice. Spray decisions could not be forecast to an acceptable degree of precision, but decisions not to spray were successfully forecast for 49% of the blocks to a precision of 98% (AdaBoost) and 70% of the blocks to a precision of 95% (Naïve Bayes). Models with as few as four attributes gave useful forecasts, and orchard management attributes were the most important determinants of model forecasting accuracy. The potential for this methodology to assist with pest spray forecasting using customised data sets is discussed.


  • Geoff Holmes
  • Dale Fletcher
  • Peter Reutemann
  • Hisham Abdel Qader
  • Corey Sterling


Big Data with ADAMS.

P. Reutemann and G. Holmes (2015). Big Data with ADAMS. Proceedings of the 4th International Workshop on Big Data, Streams and Heterogenous Source Mining: Algorithms, Systems, Programming Models and Applications. Editors: Wei Fan, Albert Bifet, Quiang Yang, Philip S. Yu. Vol. 41:5-8. pdf slides

The use of data mining to assist crop protection decisions on kiwifruit in New Zealand

M.G. Hill, P.G. Connolly, P. Reutemann and D. Fletcher (2014). The use of data mining to assist crop protection decisions on kiwifruit in New Zealand. Computers and Electronics in Agriculture, Vol 108, pp 250-257. doi

Scientific Workflow Management with ADAMS

Reutemann, Peter and Vanschoren, Joaquin (2012). Scientific Workflow Management with ADAMS. Proceedings of the Machine Learning and Knowledge Discovery in Databases (ECML-PKDD), Part II, LNCS 7524, pp. 833–837. doi

An application of data mining to fruit and vegetable sample identification using Gas Chromatography-Mass Spectrometry

Geoffrey Holmes, Dale Fletcher, and Peter Reutemann (2012). An application of data mining to fruit and vegetable sample identification using Gas Chromatography-Mass Spectrometry. Proceedings of the International Congress on Environmental Modelling and Software (IEMSS), Leizpig, Germany, 2012. pdf

Precision Horticulture & Automation

Application of machine intelligence in the service of primary industry automation. The group specialises in the application of machine vision techniques, time-of-flight imaging as well as spectral signature analysis in the horticulture and related sectors.


We develop algorithms that is robust in realistic operating environment:

  • The decision making process of the algorithm should be transparent and amenable to human interpretation.
  • The algorithm should be sufficiently lightweight for mobile deployment


  • Ye Chow Kuang
  • Melanie Ooi
  • Michael Cree
  • Lee Streeter
  • Shen Hin Lim


Algorithm Development

  • Fusion of convolution feature extraction with ADTree


  • Pesticide-free autonomous lawn care
  • Non-invasive plant growth monitoring
  • Robotic vine management
  • Autonomous asparagus harvester
  • Autonomous apple harvester
  • Autonomous kiwifruit harvester


  • W.K.Tey, Y.C.Kuang, M.P.-L.Ooi, J.J.Khoo, Automated quantification of renal interstitial fibrosis for computer-aided diagnosis: A comprehensive tissue structure segmentation method, Computer Methods and Programs in Biomedicine, Computer Methods and Programs in Biomedicine, 155, pp.109-120, 2018
  • J.J.Khoo, W.K.Tey, V.Tan, S.W.Peter, Y.C.Kuang, M.P.-L.Ooi, Visual Quantification of Renal interstitial fibrosis: Inter and Intra-observer variations, Pathology, 49, pp. S130, 2017
  • H.K.Sok, M.P.-L.Ooi, Y.C.Kuang, Multivariate Alternating Decision Trees, Pattern Recognition, 50, pp.195-209, 2016
  • H.K.Sok, M.P.-L.Ooi, Y.C.Kuang, Sparse Alternating Decision Tree, Pattern Recognition Letters, 60-61, pp.57-64, 2015