maipofields: Remote sensing of fruit trees

In alexanderbrenning/wiml: Interpreting machine-learning models in transformed space

Remote sensing of fruit trees

Description

This data set for multi-class classification contains inventoried fruit-tree crop types and multitemporal Landsat satellite data from the Maipo river basin in central Chile. And Landsat data is available for nine image dates from one growing season. This is a subset of the data used by Pena et al. (2015), who provide more detailed information on the data set (e.g., image dates). Note that this is different from the sperrorest::maipo data set shipped with the sperrorest package, which contains pixel-level data; this data set contains field-level data, i.e. one (averaged) observation per field (n=400).

Usage

maipofields

Format

A data frame with 400 rows and 67 variables:

class

Croptype "crop1" to "crop4" as a factor variable.

utmx,utmy

UTM x and y coordinates - not to be used as predictors.

b12,b13,b14,b15,b16,b17,b22,b23,b24,b25,b26,b27,b32,b33,b34,b35,b36,b37,b42,b43,b44,b45,b46,b47,b52,b53,b54,b55,b56,b57,b62,b63,b64,b65,b66,b67,b72,b73,b74,b75,b76,b77,b82,b83,b84,b85,b86,b87

bij is the value of Landsat band j in image i; images are numbered from early season = 1 to late season = 9; see Peña et al. (2017) for details, e.g. image dates.

ndvi01,ndvi02,ndvi03,ndvi04,ndvi05,ndvi06,ndvi07,ndvi08

ndvi0i is the NDVI value (representing vegetation vigour) for each image date i.

ndwi01,ndwi02,ndwi03,ndwi04,ndwi05,ndwi06,ndwi07,ndwi08

ndwi0i is the NDWI value for image date i, i.e. a moisture index (representing moisture content of plants).

Details

Objective: To predict crop type as the response variable based on the available optical remote sensing data.

Source

Pena, M.A., and Brenning, A. 2015. Assessing fruit-tree crop classification from Landsat-8 time series for the Maipo Valley, Chile. Remote Sensing of Environment, 171: 234-44. https://doi.org/10.1016/j.rse.2015.10.029

alexanderbrenning/wiml documentation built on Sept. 29, 2023, 4:45 a.m.