tools

Danesfield Tools

This directory contains several command line tools for executing the Danesfield algorithms. The run_danesfield.py tool runs each component of the Danesfield system end-to-end. Each of the following subsections covers a single tool.

Note that several of the tools require model files not included in this repository. These model files may not be publicly available at the time of writing. Please contact Kitware directory to request access to the models.

Run Danesfield

This script runs each of the Danesfield system components in an integrated end-to-end pipeline. The inputs / outputs of the script are controlled by a single configuration file. A template configuration file (input.ini) can be found in this repositories root directory.

Tools

• run_danesfield.py

Prerequisites

A completed configuration file. The configuration file requires paths to source materials, paths to model files, and parameters for some algorithms.

Usage

python run_danesfield.py \
       <input_configuration_file>

Segmentation by Height

Tools

• segment_by_height.py

Prerequisites

Usage

python segment_by_height.py \
       <dsm_image_path> \
       <dtm_image_path> \
       <output_mask_path> \
       --msi <msi_image_path>

Can optionally pass in OpenStreetMap road data as a geojson file to include roads in the output mask.

python segment_by_height.py \
       <dsm_image_path> \
       <dtm_image_path> \
       <output_mask_path> \
       --msi <msi_image_path> \
       --road-vector <road_vector_geojson_path> \
       --road-rasterized <output_road_raster> \
       --road-rasterized-bridge <output_road_bridge_raster>

Columbia Building Segmentation

Tools

• building_segmentation.py

Prerequisites

Download the files in this folder. When running the script, specify the path to this folder using the --model_dir argument and the common prefix of the model files using the --model_prefix argument, i.e. "Dayton_best".

Usage

python building_segmentation.py \
    --rgb_image <rgb_image_path> \
    --msi_image <msi_image_path> \
    --dsm <dsm_image_path> \
    --dtm <dtm_image_path> \
    --model_dir <model_directory> \
    --model_prefix <model_prefix> \
    --save_dir <output_directory> \
    --output_tif

UNet Semantic Segmentation

Tools

• kwsemantic_segment.py

Prerequisites

Download the pretrained model here. A default configuration file is included in this repository at danesfield/segmentation/semantic/test_denseunet_1x1080_retrain.json.

Usage

python kwsemantic_segment.py \
    <config_file_path> \
    <model_path> \
    <rgb_image_path> \
    <dsm_image_path> \
    <dtm_image_path> \
    <msi_image_path> \
    <output_directory> \
    <output_filename_prefix>

Material Classification

Material classification from Rutgers University.

Authors

• Matthew Purri (matthew.purri@rutgers.edu)

Input

• Orthorectified image (GeoTIFF)
• Image metadata (.IMD or .tar)

Output

• Orthorectified material segmentation map (GeoTIFF)
• Colorized material segmentation map (PNG)

Tools

• material_classifier.py

Prerequisites

Download RN18_All.pth.tar. When running the script, specify the path to this file using the --model-path argument.

Usage

python material_classifier.py --image_paths <image_paths> --info_paths <info_paths> --output_dir <output_dir> --model_path <model_path> --cuda

PointNet Geon Extraction

PointNet Geon Extraction provided by Columbia University.

Authors

• Xu Zhang (xu.zhang@columbia.edu)

Input

• Building point cloud (las text)

Output

• Building point cloud with root type labels (las text)

Tools

• roof_segmentation.py

Prerequisites

Download the files in this folder. When running the script, specify the path to this folder using the --model_dir argument and the common prefix of the model files using the --model_prefix argument, i.e. "dayton_geon".

Usage

python roof_segmentation.py \
    --model_dir=<path_to_model_dir> \
    --model_prefix=<model_prefix> \
    --input_pc=<path_to_input_pointcloud> \
    --output_txt=<path_to_output_pointcloud> \
    --output_png=<path_to_output_graphic> \

Curve Fitting

Curve fitting provided by Columbia University.

Authors

• Xu Zhang (xu.zhang@columbia.edu)

Input

• Point cloud with roof type labels (las text)

Output

• Point cloud of remaining (planar) points (las text)
• Geon file of fitted curves (npy)

Tools

• fitting_curved_plane.py

Prerequisites

Usage

python fitting_curved_plane.py \
    --input_pc=<path_to_input_pointcloud> \
    --output_png=<path_to_output_png> \
    --output_txt=<path_to_output_remainingpoints_pointcloud> \
    --output_geon=<path_to_geon_output>

Geon to mesh

Script to convert geon file to mesh provided by Columbia University.

Authors

• Xu Zhang (xu.zhang@columbia.edu)

Input

• Geon file (npy)
• DTM file (tif)

Output

• Mesh file (ply)

Tools

• geon_to_mesh.py

Prerequisites

Usage

python geon_to_mesh.py \
    --input_geon=<path_to_input_geon_npy> \
    --input_dtm=<path_to_input_dtm> \
    --output_mesh=<path_to_output_mesh_ply>

Roof geon extraction

Wrapper script for running Purdue's point cloud segmentation and reconstruction code, and Columbia's roof segmentation and geon fitting code in the right sequence.

Authors

• Bo Xu (xu1128@purdue.edu)
• Xu Zhang (xu.zhang@columbia.edu)

Input

• P3D Point cloud (las)
• Threshold CLS file (tif)
• DTM file (tif)
• Roof segmentation model dir / prefix

Output

• Mesh files (ply, obj)
• Geon JSON (json)

Tools

• roof_geon_extraction.py

Prerequisites

Download the files in this folder. When running the script, specify the path to this folder using the --model_dir argument and the common prefix of the model files using the --model_prefix argument, i.e. "dayton_geon".

Usage

python roof_geon_extraction.py \
    --las=<path_to_p3d_pointcloud> \
    --dtm=<path_to_input_dtm> \
    --cls=<path_to_input_threshold_cls> \
    --model_prefix=<prefix_for_model_files> \
    --model_dir=<directory_containing_model_files> \
    --output_dir=<path_to_output_directory>

Get road vector

Fetches road vector data from OpenStreetMap for an AOI, and converts to GeoJSON.

Input

• Latitude / Longitude bounds
• Output directory

Output

• Road vector data (GeoJSON)
• Original OSM data (osm)

Tools

• get_road_vector.py

Usage

python get_road_vector.py \
       --left <left_bound> \
       --bottom <bottom_bound> \
       --right <right_bound> \
       --top <top_bound> \
       --output-dir <path_to_output_directory>

Buildings to DSM

Renders a DSM or CLS from a DTM and polygons representing buildings.

Input

• DTM file (tif)
• Building polygons (vtp or list of obj paths)

Output

• DSM or CLS file (tif)

Tools

• buildings_to_dsm.py

Usage

python buildings_to_dsm.py \
       <path_to_dtm> \
       <path_to_output_file> \
       --input_obj_paths <list_of_obj_paths>

Run Metrics

Wrapper script around JHU/APL's Core3D scoring software found here. Given a directory of ground truth files with a common prefix, score our output files.

Input

• DSM file (tif)
• DTM file (tif)
• CLS file (tif)
• MTL file (tif)

Output

• Output scores (json)

Tools

• run_metrics.py

Usage

python run_metrics.py \
       --output-dir <output_directory_path> \
       --ref-dir <path_to_reference_files> \
       --ref-prefix <reference_files_prefix> \
       --dsm <dsm_file_to_score> \
       --cls <cls_file_to_score> \
       --mtl <mtl_file_to_score> \
       --dtm <dtm_file_to_score>

Orthorectify

Orthorectify a source image using a DSM, DTM, and RPC.

Input

• Source image (tif)
• DSM file (tif)
• DTM file (tif)
• RPC file (txt)

Output

• Orthorectified image (tif)

Tools

• orthorectify.py

Usage

python orthorectify.py \
       <source_image_path> \
       <DSM_path> \
       <output_image_path> \
       --dtm <DTM_path> \
       --raytheon-rpc <RPC_path>

Texture Mapping

Textures building models using pre-processed source imagery.

Input

• Cropped and pansharped source imagery (tif)
• Model files (obj)
• DSM file (tif)
• DTM file (tif)

Output

• Textures (png)
• Textured models (obj)
• Textured ground (obj)

Usage

python texture_mapping.py \
       <DSM_path> \
       <DTM_path> \
       <output_directory_path> \
       <occlusion_mesh_path> \
       --crops <list_of_cropped_and_pansharpened_image_paths> \
       --buildings <list_of_model_paths_to_texture>

Third-party tools

Core3D JSON data representation and parser

A data representation and meshing utility for CORE3D deliverables. See https://github.com/CORE3D/data_rep_c3d.

Includes the following tools:

• json2obj.py
• meshIO.py
• paramCounter.py
• primitiveMeshGen.py