liamstaras/esri_heightmap

Convert an ESRI Grid ASCII (*.asc) file to a RAW heightmap

ESRI_Heightmap

Convert an ESRI Grid ASCII (*.asc) to a binary headerless TIFF (*.raw) for use as a heightmap.

Installation

Download the repository and Python 3. Then, run python3 setup.py in the directory with the code to install.

Usage

python -m esri_heightmap [-h] [--output_mode {RAW,TIFF,BOTH}]
                              [--directory]
                              path_to_input path_to_output

Process *.asc files into an image heightmap

positional arguments:
  path_to_input         path to *.asc file or directory containing *.asc files
                        on which to run processing
  path_to_output        path to output file without extension

optional arguments:
  -h, --help            show this help message and exit
  --output_mode {RAW,TIFF,BOTH}
                        output file format (default: BOTH)
  --directory           run the process on a directory of *.asc files

Use cases

Our use case is for getting LiDAR data into Unity as a terrain for VR interaction. Add yours here!

esri_heightmap's Issues

Move reusable code into functions

To allow this code to be reused by others, move as much as possible into function definitions

Deal with variation in NODATA_value

NODATA_value should be able to vary over many files and the code should deal with this. Currently, I don't need this feature, but it may be useful to anyone else in the future.

Create setup.py

Add setup.py and requirements.txt to follow standard Python conventions.

README.md formatting and verbosity

README.md is not correctly formatted. This needs correcting, as well as making generally more verbose.

Slow NaN elimination

How I deal with less authoritative data is inefficient: importing and processing the entire set and then performing a cellwise comparison for NaNs. Surely there must be a better way to do this?

Potential for multithreading?

The following image says it all - this could probably be optimized with proper multithreading. Running main program takes a long time and runs on a single core. This screenshot is of the program during the NaN elimination phase.

Related to #3