Kamm, Matthew; Reed, J Michael (2018): Results from confusion matrices of supervised classification of habitat at 12 American kestrel nest sites in Massachusetts [dataset]. Tufts University, PANGAEA, https://doi.org/10.1594/PANGAEA.884609,

---

In: Kamm, M; Reed, JM (2018): Assessment of classification accuracy of habitat types in UAV aerial photos around American kestrel nest box sites in Massachusetts, USA [dataset publication series]. PANGAEA, https://doi.org/10.1594/PANGAEA.884669

These data are the overall accuracies and kappa coefficients of several different supervised classification operations using program ENVI to identify areas of Grass, Herbaceous growth, Woody growth, Bare Ground, and Human-modified habitat in UAV aerial photos taken around 12 American Kestrel nest box sites located around Massachusetts, USA. Sites are numbered 1-12. "Self.Acc" or "Self.Kap" indicate overall accuracy or kappa coefficients when the classifier was trained with image data from the site itself. "4.Kap" or "9.Acc" indicate results when the classifier was trained with image data from Site 4 or Site 9, rather than the site itself. The "Kernels" tabs contain results using Classification Aggregation with a minimum kernel size, in pixels, indicated by the column titles, ex: "768.Kap" is the kappa coefficient when the minimum kernel size in ENVI Classification Aggregation was set to 768 pixels.