ABSTRACT
This chapter provides a foundational study, using a time series clustering technique, followed by purity analysis, to demonstrate how many plants with the same genotype share the same temporal behavior in the formation of the clusters. It also provides a discussion on an eigenvalue-based analysis of the phenotypes, extracted from plant image sequences to demonstrate the temporal variation of phenotypes regulated by genotypes. The chapter then provides discussion on time series prediction in the context of image-based plant phenotyping analysis. It describes basic concepts of time series, and discusses a time series analysis algorithm, based on clustering and angular histograms for application in plant phenotyping. The chapter focuses on time series prediction, using neural networks, along with the most widely used prediction performance measures. It introduces time series analysis based on eigenvalues and presents the experimental analyses on a real phenotypic dataset.
