README for Relative Digital Elevation Model Map Generation from GeoTIFF

Overview

This script generates depth maps from large (gigabytes) GeoTIFF orthophotos using monocular vision, a technique that estimates depth from a single image by leveraging spatial patterns and visual cues. It employs sliding window processing to handle large GeoTIFF files, producing a GeoTIFF Digital Elevation Model (DEM) as output.

Example Output

Below is an example of a Digital Elevation Model (DEM) generated by the script:

Large GeoTIFF Processing

This script is capable of processing large GeoTIFF files, as shown below:

Development Setup

1. Clone the Repository

git clone <repository-url>
cd <repository-directory>

2. Use the provided environment.yml file to set up a Python environment:

conda env create -f environment.yml

3. Activate the environment:

conda activate relative-height-digital-elevation-model

4. Download sample data

Download the required orthophoto data from here and place it in the data/orthophoto/ directory (or another directory of your choice).

Generating Relative Digital Elevation Model

python scripts/generate_depth_map.py \
  --input_file data/orthophoto/small_dense_area.tif \
  --output_file data/dem/dem_small_dense_area.tif

Optional Arguments

You can include additional arguments to customize the behavior of the script:

Argument	Default	Description
--model_type	DPT_Large	Choose MiDaS model type. Options include DPT_Large, DPT_Hybrid.
--window_size	512	Define the size of the sliding window (in pixels).
--overlap	0.7	Set overlap ratio between patches (0 to 1).
--show_output	False	Add this flag to visualize the orthophoto and depth map side-by-side.

How to cite?

This is part of research project "Dense Urban Area Mapping" by Nishon Tandukar. If you use this code please cite.

@misc{naxa_rh_dem,
  author = {Nishon Tandukar},
  title = {RH Digital Elevation Model Generator},
  howpublished = {\url{https://github.com/naxa-developers/rh-digital-elevation-model-generator}},
  note = {Accessed: 2024-11-18}
}