Remote Sensing: Change Detections
Many change detection methods are being utilized in this project
to identify areas of change over time and characterize landscape
dynamics. These methods include:
- Channel/Scene Integration: Near-infrared channels
(e.g., channel 4 of Landsat TM) from scenes of different periods
composited as a qualitative method of assessing regional change.
- Multidate Composite: Multiscene data stack representing
different time periods used as the input or feature set for an
unsupervised classification for defining change and no-change
spectral clusters.
- Principal Components Analysis: Change is assessed
through the derivation of eigenvectors that relate spectral channels
from scenes collected at different time periods to generate
components, and eigenvalues that indicate the percent variance
associated with each of the defined components.
- Image Algebra: Two channels of the same spectral region
and wavelengths for two different time periods are ratioed, and
image differencing is achieved by subtracting the spectral responses
of one date from that of the other.
- Binary Mask: Uses a multidate image composite recoded
into a binary mask consisting of areas, which have changed and not
changed between two dates.
- Post-Classification: Classification of two scenes from
different dates assessed on a pixel-by-pixel basis and reported
through a change matrix.
Change trajectories are also being implemented. Change trajectories
are analogous to panel data analysis in which survey responses are
mapped over time through multiple observations. However, here the unit
of observation is the pixel, and its "life history" is constituted
by the values derived from the component images of a satellite
time-series.
