Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
Application of Principal Component Analysis to Hyperspectral Data for Potassium Concentration Classification on Peach leaves
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: 2022 ASABE Annual International Meeting 2200490.(doi:10.13031/aim.202200490)
Authors: Megan Io Ariadne S. Abenina, Joe Mari Maja, Matthew Cutulle, Juan Carlos Melgar, Haibo Liu
Keywords: Hyperspectral Imaging, Precision Agriculture, Principal Component Analysis
Abstract. Hyperspectral imaging (HSI) is a technology utilized in agriculture. This system could be used to monitor the overall health of plants, vegetation, and plant or pest disease detection. Aside from HSI, NIR, chlorophyll fluorescence, and thermography are some techniques being used in precision agriculture to efficiently measure plant nutrients and detect early signs of diseases and prevent them from spreading. As sensing technology advancement expands, measuring nutrient levels and disease detection also progresses. The objective of this dissertation is to determine if there is a correlation between the potassium content and leaf image gathered from the hyperspectral camera. Nine leaves were selected randomly from multiple trees with varying nutrient levels of high, medium, and low potassium content. The Senop HSC-2 hyperspectral camera was used to scan the nine leaves. The samples were submitted to the Clemson Agricultural Service Laboratory for nutrient analysis. A Principal Component Analysis (PCA) was used to reduce the dataset due to the number of bands captured by the system (215 bands). Based on the results of the loading and scores of the scanned leaves, it showed the consistency of the scanned samples. The 500-520, 550, 630-640, and 690 wavelengths proved to have consistent peaks from all trees with varying potassium concentration. Based on the initial findings, it can be seen that there is a correlation between the potassium content of the sampled leaves and the wavelengths. Creating a model for predicting the potassium content using the mentioned wavelengths will be the next step to be conducted.
(Download PDF) (Export to EndNotes)