https://github.com/atsyplenkov/filtrs

Rust-boosted linear and spatial filtering in R

https://github.com/atsyplenkov/filtrs

Repository

Rust-boosted linear and spatial filtering in R

README.Rmd

---
output: github_document
---



```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%",
  dpi = 1000
)

requireNamespace("smoothr", quietly = TRUE)
requireNamespace("bench", quietly = TRUE)
```

# filtrs


[![R-CMD-check](https://github.com/atsyplenkov/filtrs/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/atsyplenkov/filtrs/actions/workflows/R-CMD-check.yaml)
[![Lifecycle: experimental](https://img.shields.io/badge/lifecycle-experimental-orange.svg)](https://lifecycle.r-lib.org/articles/stages.html#experimental)
[![CRAN status](https://www.r-pkg.org/badges/version/filtrs)](https://CRAN.R-project.org/package=filtrs)
![GitHub R package version](https://img.shields.io/github/r-package/v/atsyplenkov/filtrs?label=github)
![GitHub last commit](https://img.shields.io/github/last-commit/atsyplenkov/filtrs)


Rust-Boosted Linear and Spatial Filtering in R.

Currently, the package supports only the Whittaker-Eilers smoother as it is implemented in the `whittaker-eilers` [crate](https://crates.io/crates/whittaker-eilers). Based on that filter, a smoothing approach for spatial geometries (only single-part `LINESTRING` for now) is proposed.

## Installation

You can install the development version of `filtrs` like so:

``` r
# install.packages(remotes)
remotes::install_github("atsyplenkov/filtrs")
```

## Time-series filtering

For equally-spaced data, one can use the `fil_wt` function as an API to `whittaker_eilers::WhittakerSmoother`. It has two controls, `lambda` and `order`, and interpolates missing values by default. An extremely nice description of the smoother is written by [Andrew Bowell](https://www.anbowell.com/blog/the-perfect-way-to-smooth-your-noisy-data).

```{r timeseries}
library(filtrs)
## basic example code

data("airquality")

airquality$Ozone_smooth <- 
 fil_wt(as.double(airquality$Ozone), 10, 2)

plot(airquality$Ozone, type = "l", xlab = "Days", ylab = "Ozone")
lines(airquality$Ozone_smooth, col = "red", lwd = 2)

```

## Spatial filtering
Spatial data may not be equally spaced; it often has more nodes on the bends of lines and fewer on straight segments. To address this challenge, the distance between nodes—calculated using either Cartesian or Haversine methods, depending on the Coordinate Reference System (CRS)—is used as a positions vector in the background. The function `fil_wt_sf` is designed to smooth a single `LINESTRING`, proving particularly beneficial for processing high-frequency data, such as that obtained from satellite imagery.

```{r spatial}
library(sf)
library(smoothr)

file_path <- system.file("exdata/examples.gpkg", package = "filtrs")

lines <- 
  sf::st_read(file_path, layer = "gswe", quiet = TRUE) |>
  # Increase nodes count for smoother result
  smoothr::smooth("densify", max_distance = 10)

lines_wt <- 
  fil_wt_sf(lines, lamda = 10^-7, order = 3)

lines_smoothr <-
  smoothr::smooth(lines, method = "ksmooth",
                  smoothness = 21)

```


Plot’s code
```{r spatial_plot, eval=T, fig.show='hide'}
par(
  mar = c(0.5, 0.5, 0.2, 0.2),
  mfrow = c(1, 2),
  oma = c(0, 0, 0.2, 0.2)
)

plot(
  sf::st_geometry(lines),
  col = "grey30",
  lwd = 3.5
)
plot(
  sf::st_geometry(lines_wt),
  col = 'firebrick3',
  lwd = 2,
  add = TRUE
)

# Add the legend
legend(
  "bottomleft",
  legend = c("Original", "{filtrs}", "{smoothr}"),
  col = c("grey30", "firebrick3", "dodgerblue3"),
  lwd = c(3, 3)
)

plot(
  sf::st_geometry(lines),
  col = "grey30",
  lwd = 3.5
)
plot(
  sf::st_geometry(lines_smoothr),
  col = 'dodgerblue3',
  lwd = 2,
  add = TRUE
)
```




Similar filtering can be achieved via the [`smoothr`](https://github.com/mstrimas/smoothr) R package; however, the Rust-based approach outperforms in speed when high-order smoothing is needed. For example, `r nrow(sf::st_coordinates(lines))` nodes were filtered 30 times faster using the `filtrs` package rather than `smoothr`.

```{r benchmark}
bench::mark(
  filtrs = fil_wt_sf(lines, lamda = 10^-7, order = 3),
  smoothr = smoothr::smooth(lines, method = "ksmooth", smoothness = 21),
  check = F,
  relative = T
)
```

## Similar packages
As linear and spatial filters are pretty common, there is no shortage of analogs.

* **[smoothr](https://github.com/mstrimas/smoothr)** - Spatial feature smoothing written in base R. However, there is no Whittaker-Eilers smoother.

* **[phenofit](https://github.com/eco-hydro/phenofit)** - As part of vegetation phenology, Whittaker-Eilers and Savitzky-Golay time-series filtering is implemented using Rcpp. No spatial filtering.

* **[signal](https://cran.r-project.org/web/packages/signal/index.html)** - A set of signal processing functions originally written for 'Matlab' and 'Octave'. No spatial filtering.

Owner

GitHub Events

Issues and Pull Requests

Last synced: over 1 year ago