Fig. 2From: A method for real-time classification of insect vectors of mosaic and brown streak disease in cassava plants for future implementation within a low-cost, handheld, in-field multispectral imaging sensorFlowchart of Phase-2 processing. The flowchart depicts the steps undertaken to process the MSI data and classify the viral vectors. The green boxes represent the input data streams: trimmed spectra—low noise spectral data, i.e. 900–1030 nm wavelengths in this case; raw spectra—data as taken from the instruments; body-only spectra—as extracted from the spatial data. The yellow and blue boxes depict the feature reduction and classification stage respectively. Abbreviations used: incremental Principle Component Analysis (iPCA), Support Vector Clustering (SVC)Back to article page