Linda Mills
Doctor of Philosophy, (Engineering)
Study Completed: 2016
College of Sciences
Citation
Thesis Title
Analysis of spectral response patterns of kiwifruit orchards using satellite imagery to predict orchard characteristics of commercial value before harvest
Read article at Massey Research Online: 
The dry matter content of kiwifruit has a large influence on the taste of kiwifruit at market and as a result is a driver of orchard management. Throughout the growing season the destructive ''90 Kiwifruit Protocol'' is used to predict kiwifruit dry matter at harvest. Satellite imagery has been used as an alternative to determine crop characteristics, although not for kiwifruit, but of the 20 statistics previously generated none have been successful. Ms Mills used data from multispectral satellite images of kiwifruit orchards to plot the dry matter of kiwifruit maturity areas in a three-dimensional colour space using a normalised red-green-blue-infrared vector. Visualisation of results showed that dry matter was ordered along an axis, which was identified by regression. Predictions resulting from this regression line were comparable to those of the ''90 Kiwifruit Protocol'' and had potential to predict kiwifruit dry matter content at harvest from very early in the growing season.
Supervisors
Dr Rory Flemmer
Dr Huub Bakker