The goal of plimanshiny is to provide a Shiny interface for the pliman package
The development version of plimanshiny can be installed from
GitHub using the
pak package:
install.packages("pak")
pak::pkg_install("NEPEM-UFSC/pliman")
pak::pkg_install("NEPEM-UFSC/plimanshiny")Note: If you are a Windows user, you should also first download and install the latest version of Rtools.
To start using plimanshiny, you need to load the package and call the
app with run_app().
library(plimanshiny)
run_app()The app will initiate and display the following home screen. It comprises four primary modules for the analysis of orthomosaics. Initially, you need to import the mosaic (1), then proceed to either build or import the shapefile (2). Additionally, you have the option to compute vegetation indexes (3), and ultimately, analyze the mosaics (4).
Before importing the mosaic, several settings (1-3) can be configured. Firstly, configure the mosaic bands (1). By default, the ‘BGR’ system (bands 1, 2, and 3) is used, with Red-Edge and NIR bands assigned to layers 4 and 5. You can also define the upper and lower quantiles for color stretching configuration (3). Lastly, set the maximum number of pixels to render in the map or plot. Choosing a larger value will result in high-resolution leaflet plots, but the rendering time may considerably increase. Utilizing the browse button (4), you can search for and upload the mosaic file, with a maximum upload size of approximately 2GB. There are three options for visualizing the mosaic: ‘rgb’ for an RGB plot, ‘mapview’ for an interactive map, or ‘bands’ to plot individual layers (5). If the mosaic needs to be cropped to the region of interest before analysis, you can switch it “on” (6). This option allows users to crop the mosaic using the ‘Draw Rectangle’ tool if necessary. Finally, you can save the mosaic (7).
After the mosaic has been uploaded, relevant informations are shown in a pop-up.
The “shapefile” module can be utilized for both building and importing a shapefile (1). When constructing a shapefile, users have the option to configure the shape style, generating either a grid plot or a “free” shape. To delineate the plot area, utilize the “Draw Polygon” tool. The number of rows and columns can be specified, and a buffering factor (5) is available. It’s important to note that negative values will remove areas from the plot edge. Once the polygon is completed, the built plots will be displayed. Click on “shapefile finished” (3) to conclude the plot creation process.
After the plots have been created, you can edit them by toggling the “Edit the drawn plots?” switch. In this step, you have the flexibility to reshape, move, or delete individual plots. To finalize the editing process, check the “Edition finished” box.
If a shapefile is already available, you can import it (1). It is also possible to define a palette to color the plots if any variables are available in the shapefile.
Calculating vegetation indices is straightforward. Users need to choose either built-in indexes (1) or personalize their own (2). A list of personalized indexes can be provided by separating each index with a comma (“,”). It’s important to note that the index must be computed considering the bands (R, G, B, RE, NIR) or the names of the mosaic’s layers. Finally, users can select an index for synchronization with the basemap (3).
The “Analyze” module works its magic. Users can first set the option to define a summarization function for the extracted indexes (1), with the mean being the default. Following that, options for plot or individual segmentation are provided. If neither plots nor individuals are segmented, the indexes (computed in the “Index” module) are extracted, and the mean is computed for each shape (defined in the “Shapefile” module).
When “Segment individuals” is selected (2), various options for defining
how the segmentation is performed become available. Firstly, users must
define an index to be used for segmentation (3), set a threshold value
(5), decide whether to invert the segmentation (7), and specify the
tolerance and extension arguments for watershed segmentation (9). For
additional details on this, refer to
analyze_objects()
function. Objects with less than 15% of the overall mean area are
removed by default, a factor controlled by the “Lower noise” option
(11). To reduce noise, it’s also possible to apply median filtering to
the mask (12). When all is right, click “Analyze the mosaic!”
After the mosaic has been analyzed, you can view a summary of the results at the plot and individual levels. Attributes for plots and individuals can be selected to be plotted on the map (2).
Finally, the results can be exported to shapefiles (5) or sent to a variable in the R environment (6).
- Results exported directly to the R environment
