README.md

InstaR

Travis-CI Build Status

Coverage Status

A Collaborative Software Development Project

Date: February 9, 2018

Overview

According to a study by Yahoo Labs, “Filtered photos are 21 percent more likely to be viewed and 45 percent more likely to be commented on. Have you ever wondered how you could transform your images using filters similar to Instagram in R?

We present this package that performs digital image processing on .png images. It encompasses functions ranging from transformations like a simple flip, playing with color hues (rgb2gray) to 2D convolutions using a simple kernel matrix to do some interesting things! We have started with quite basic but diverse functions and hope to advance and add more with time.

Functions

Blur

This function performs convolution to de-emphasizes differences in adjacent pixel values. It has an averaging effect removing detail and noise resulting in blurring of the image.

Flip

This is a transformation function which flips the image either horizontally or vertically.

Greyscale

This function converts an RGB image to grayscale.

Non transparent background .png images required

Note: See Usage section below

R ecosystem

"A picture paints a thousand words", however, a well-constructed image speaks even more than that without having to rely on a written description. We want to explore the elements of filters and their implementation in R. A similar package called "magick" exists in R which has standard filters like blur, sobel among others. We have started with a few basic functions but want to work towards building more advanced filters similar to the ones provided by Instagram.

Installation

To install InstaR, follow these instructions:

  1. Please check if devtools has been installed. If not, open the console and input the following: install.packages("devtools") to install devtools from CRAN. Also, check the package dependencies down below.
  2. Input the following into the console:
devtools::install_github("UBC-MDS/InstaR", build_vignettes = TRUE)
  1. You are all set to use go!

Usage

Non transparent background .png images required

library(InstaR)

1.flip(input_path, direction, output_path)

Arguments:

2.blur(input_path, output_path)

Arguments:

3.greyscale(input_path, output_path)

Arguments:

Package dependencies

png

testit

tableMatrix

Collaborators:

Bhatnagar, Tarini

Guo, Xin (Alex)

Nikel, Indiana



UBC-MDS/InstaR documentation built on May 25, 2019, 1:35 p.m.