This is Algorithm Central

A new and growing community of algorithm developers, and their creations.

Browse Algorithms

pdal/readers_las 1.8.0
Readers.LAS

The LAS Reader supports reading from LAS format files, the standard interchange format for LIDAR data. The reader does NOT support point formats containing waveform data (4, 5, 9 and 10). This algorithm is ported from the excellent PDAL - Point...

pdal/filters_range 1.8.0
Filters.Range

The range filter applies rudimentary filtering to the input point cloud based on a set of criteria on the given dimensions. This algorithm is ported from the excellent PDAL - Point Data Abstraction Library. For more information please refer to...

pdal/writers_gdal 1.8.0
Writers.Gdal

The GDAL writer creates a raster from a point cloud using an interpolation algorithm. Output is produced using GDAL and can use any driver that supports creation of rasters. A data_type can be specified for the raster (double, float, int32, etc.). If...

pdal/writers_las 1.8.0
Writers.Las

The LAS Writer supports writing to LAS format files, the standard interchange file format for LIDAR data.

pdal/filters_dem 1.8.0
Filters.DEM

filters.dem uses a raster to keep data within a range of the raster cell. For example, atmospheric or MTA noise in a scene can be quickly removed by keeping all data within 100m above and 20m below a pre-existing elevation model. This algorithm is...

pdal/filters_greedyprojection 1.8.0
Filters.Greedyprojection

The Greedy Projection filter creates a mesh (triangulation) in an attempt to reconstruct the surface of an area from a collection of points. GreedyProjectionTriangulation is an implementation of a greedy triangulation algorithm for 3D points based...

pdal/readers_gdal 1.8.0
Readers.GDAL

The GDAL reader reads GDAL readable raster data sources as point clouds. Each pixel is given an X and Y coordinate (and corresponding PDAL dimensions) that are center pixel, and each band is represented by “band-1”, “band-2”, or “band-n”. Using the...

pdal/readers_matlab 1.8.0
Readers.Matlab

The Matlab Reader supports readers Matlab .mat files. Data must be in a Matlab struct, with field names that correspond to Dimensions names. No ability to provide a name map is yet provided. Additionally, each array in the struct should ideally...

pdal/run_pipeline 1.8.0
Run Pipeline

Execute a pdal pipeline as defined by an input JSON string. This algorithm is ported from the excellent PDAL - Point Data Abstraction Library. For more information please refer to https://pdal.io.

pdal/writers_matlab 1.8.0
Writers.Matlab

The Matlab Writer supports writing Matlab .mat files. The produced files has a single variable, PDAL, an array struct. This algorithm is ported from the excellent PDAL - Point Data Abstraction Library. For more information please refer to https://pdal.io.

pdal/writers_ply 1.8.0
Writers.Ply

The ply writer writes the polygon file format, a common file format for storing three dimensional models. The writer emits points as PLY vertices. The writer can also emit a mesh as a set of faces. filters.greedyprojection and filters.poisson create a...

pdal/filters_outlier 1.8.0
Filters.Outlier

The Outlier filter provides two outlier filtering methods: radius and statistical. These two approaches are discussed in further detail below. It is worth noting that both filtering methods simply apply a classification value of 7 to the noise...

pdal/filters_info 1.8.0
Filters.Info

The info filter provides simple information on a point set as metadata. It is usually invoked by the info command, rather than by user code. The data provided includes bounds, a count of points, dimension names, spatial reference, and points meeting a...

pdal/filters_smrf 1.8.0
Filters.SMRF

Filter ground returns using the Simple Morphological Filter (SMRF) approach outlined in: https://pdal.io/references.html#pingel2013 This algorithm is ported from the excellent PDAL - Point Data Abstraction Library. For more information please refer...

stratus2019/ndvi 0.0.1
NDVI

Create NDVI image from input image.

stratus2019/open_geotiff 0.0.1
Open Geotiff

Open a Geotiff image and generate a Geotiff image object using the resippy codebase.

stratus2019/apply_colormap 0.0.1
ColorMap GreyScale

Apply a colormap using seaborne to greyscale geotiff image (from resippy).

stratus2019/chips_from_highest_detection_results 0.0.1
chips from highest detection results

Generates image chips from an image given a set of scores and upper left hand pixel coordinates (y, x values), and writes them out to a directory.

stratus2019/filter_image_chips 0.0.1
filter image chips

This algorithm filters training and val chip data by deleting the files specified in a text file. The directory where files flagged for deleting is specified by a line in the text file using the following convention relative_dir=<relative...

stratus2019/load_rgb_image_data 0.0.1
load rgb image data

Loads RGB data from disk given a base data directory and relative path to the file. Only the first 3 channels are preserved, so if the image contains an alpha channel it will be removed.

stratus2019/set_base_data_dirs 0.0.1
set base directories

This is a convenience algorithm that sets base directories that are root directories for data that is to be processed, and results to be saved. Once set, these directories can be referenced by other algorithms in a processing chain. It is useful...

stratus2019/train_cnn 0.0.1
train neural network

Trains a convolutional neural network. It is assumed that data is broken into directories with the names "train" and "val" each of those directories contain subdirectories which hold images of a particular class. The subdirectory should have a name...

stratus2019/create_training_chips 0.0.1
create training chips

training chips from overhead imagery. This algorithm scans training and validation directories for image data and corresponding images that are point masks. Positive (target) mask names are the base image filename, followed by "Annotated_Cars". ...

stratus2019/load_cnn_for_training 0.0.1
load cnn for training

loads a convolutional neural network for training. This removes the top layer of the network, so it is ready to be trained for any number of image classes.

stratus2019/load_cnn_for_scoring 0.0.1
load cnn for scoring

loads a convolutional neural network for scoring. This will load an entire, saved model, and include the top layer, so the neural network is ready for inference. It uses Keras and has been tested with a tensorflow backend.

pdal/filters_stats 1.8.0
Filters.Stats

The stats filter calculates the minimum, maximum and average (mean) values of dimensions. On request it will also provide an enumeration of values of a dimension. The output of the stats filter is metadata that can be stored by writers or used...

stratus2019/make_tile_cache 0.0.1
Make Tile Cache

Build a tile cache suitable for viewing imagery on the web.

opencv/imread 4.0.0
imread

Open an image with opencv

opencv/imwrite 4.0.0
imwrite

Saves an image to a specified file. The function imwrite saves the image to the specified file. The image format is chosen based on the filename extension (see cv::imread for the list of extensions). In general, only 8-bit single-channel or...

opencv/gaussianBlur 4.0.0
GaussianBlur

Blurs an image using a Gaussian filter. The function convolves the source image with the specified Gaussian kernel.

opencv/Canny 4.0.0
Canny

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts. Finds edges in an image using the Canny algorithm with custom image gradient.

opencv/resize 4.0.0
Resize

Resizes an image. The function resize resizes the image src down to or up to the specified size. Note that the initial dst type or size are not taken into account. Instead, the size and type are derived from the src,dsize,fx, and fy.

stratus2019/score_image_with_cnn 0.0.1
score image with cnn

Scores an image given a convolutional neural network using a sliding windowed approach. The user can specify the window size (in pixels), the target chip size (image size used by the trained neural network), and percent overlap to use for the sliding...