Skip to contents

Classify Landsat QA bands

Source: R/classifyQA.R
classifyQA.Rd

extracts five classes from QA band: background, cloud, cirrus, snow and water.

Usage

classifyQA(
  img,
  type = c("background", "cloud", "cirrus", "snow", "water"),
  confLayers = FALSE,
  sensor = "OLI",
  legacy = "collection1",
  ...
)

Arguments

img

SpatRaster. Landsat 8 OLI QA band.

type

Character. Classes which should be returned. One or more of c("background", "cloud", "cirrus","snow", "water").

confLayers

Logical. Return one layer per class classified by confidence levels, i.e. cloud:low, cloud:med, cloud:high.

sensor

Sensor to encode. Options: c("OLI", "TIRS", "ETM+", "TM", "MSS").

legacy

Encoding systematic Options: c("collection1", "pre_collection"). Default is "collection1" for the Landsat Collection 1 8-bit quality designations. Use "pre_collection" for imagery downloaded before the Collection 1 quality designations were introduced

...

further arguments passed to writeRaster

Value

Returns a SpatRaster with maximal five classes:

classvalue
background1L
cloud2L
cirrus3L
snow4L
water5L

Values outside of these classes are returned as NA. If confLayers = TRUE then a RasterStack with one layer per condition (except 'background') is returned, whereby each layer contains the confidence level of the condition.

Confidencevalue
low1L
med2L
high3L

Details

By default each class is queried for *high* confidence. See encodeQA for details. To return the different confidence levels per condition use confLayers=TRUE. This approach corresponds to the way LandsatLook Quality Images are produced by the USGS.

See also

encodeQA decodeQA

Examples

library(terra)
#> terra 1.7.83
qa <- rast(ncol = 100, nrow=100, val = sample(1:2^14,  10000))

## QA classes
qacs <- classifyQA(img = qa)
## Confidence levels
qacs_conf <- classifyQA(img = qa, confLayers = TRUE)