friessn / rsdb-data

RSDB distribution with example datasets

RSDB distribution with example datasets

This distribution contains RSDB, example datasets and example R-scripts.
If issues with RSDB arise please refer to our documentation and to our troubleshooting page.

The example datasets are already inserted into RSDB and additionally the original example dataset files are in folder example-data.
Example R-scripts are located in folder example-data/r_script. (--> documentation of processing by R-scripting)

Download this distribution as zip-file

Or download this distribution with git:

git clone --depth 1 -b master https://github.com/environmentalinformatics-marburg/rsdb-data.git

This RSDB distribution is runnable on Windows and on Ubuntu.

Run on Windows 10 64bit

Extract the zip-file on a short path without spaces in it, e.g. to C:/rsdb

Doubleclick win_server.cmd to run RSDB server. Stop it by closing the console window.

No dependencies are needed as java and gdal for Windows 10 64bit are included.

Files of pattern win_*.cmd are executable for Windows.

Run on Ubuntu

Extract the zip-file.

Dependencies Java (default-jdk) and GDAL libgdal-java are needed which are not included this distribution for Ubuntu. Refer to the documentation.

Install dependencies and mark sh-files as executable.

sudo apt-get install default-jdk
sudo apt install libgdal-java
chmod +x *.sh

Run RSDB server.

./server.sh

Press ctrl-c to stop RSDB server.

RSDB usage

Web-interface: Per default your local RSDB server is running at http://127.0.0.1:8081

If your local RSDB is running currently you can directly click on this link to open RSDB web-interface. (For freshly started RSDB server, wait a few seconds until "Server running.." is printed on the terminal.)

This URL can also be used for access by the RSDB R-package.

---

Example datasets

All example datasets are inserted in RSDB as layers.

You can use the original example data files located in subfolders of example-data to try out to insert data into RSDB.

---

Dataset lidar_forest_edge

This is a dataset of LiDAR data covering an edge of the forest.

Original dataset is located at example-data/lidar_forest_edge

Task for dataset import

web-interface Tools - Task - Category PointCloud - Task import

Insert task parameters. At the bottom the summary of the resulting task is shown which should be the following:

{ "task_pointcloud": "import", "pointcloud": "lidar_forest_edge", "source": "example-data/lidar_forest_edge", "storage_type": "TileStorage"}

Click Submit and wait a few seconds to finish the task.

Refresh you browser window (by pressing "F5") to update list of contained layers.

Task for rasterized layer generation

web-interface Tools - Task - Category PointCloud - Task rasterize

Insert task parameters. At the bottom the summary of the resulting task is shown which should be the following:

{ "task_pointcloud": "rasterize", "pointcloud": "lidar_forest_edge", "storage_type": "TileStorage" }

Click Submit and wait a few seconds to finish the task.

Refresh you browser window (by pressing "F5") to update list of contained layers.

---

Dataset plots_forest_edge

This dataset contains plot definitions as polygons in a GeoPackage file. The plot areas are covered by the lidar dataset lidar_forest_edge.

Original dataset is located at example-data/plots_forest_edge/plots_forest_edge.gpkg

---

Example processing

Processing R script is located at example-data/r_script (--> documentation of processing by R-scripting)

The script uploads a plot mask into RSDB as raster layer plots_forest_edge_indices at band 1 with title plot mask.

Aim of the following:
Based on the plot mask, pixels of resolution of 1 meter should be filled at plot areas with point cloud index calculation values by task index_raster.

The point cloud task index_raster can be specified and submitted on the web interface or on the command line. (see documentation)

Parameters:

task_pointcloud: "index_raster" // task name
pointcloud: "lidar_forest_edge" // source point cloud layer
rasterdb: "plots_forest_edge_indices" // target raster layer with mask at band 1
indices: "BE_H_MEAN" // indices to calculate for each pixel
mask_band: 1 // the previously created band 1 should be used

Following task specification will be generated if the parameter arguments have been typed into the web interface:

{ "task_pointcloud": "index_raster", "pointcloud": "lidar_forest_edge", "rasterdb": "plots_forest_edge_indices", "indices": [ "BE_H_MEAN" ], "mask_band": 1 }

This task specification can be submitted by the Submit button or at command line it can be typed directly and submitted.

(--> documentation of processing by R-scripting)