sharp

What is sharp?

The sharp npm package is a high-performance Node.js module for resizing, converting, and manipulating images. It is built around the libvips image processing library, which allows it to handle large images and perform operations quickly and with a low memory footprint.

What are sharp's main functionalities?

Image Resizing

Resizes an image to the specified width and height.

sharp('input.jpg').resize(300, 200).toFile('output.jpg', (err, info) => {});

Format Conversion

Converts an image from one format to another, such as JPEG to PNG.

sharp('input.jpg').toFormat('png').toBuffer().then(data => {});

Image Rotation

Rotates an image by a specified degree.

sharp('input.jpg').rotate(90).toBuffer().then(data => {});

Extracting Image Regions

Extracts a region of the image starting at the left and top offsets and with the specified width and height.

sharp('input.jpg').extract({ left: 100, top: 100, width: 300, height: 200 }).toFile('output.jpg', (err, info) => {});

Image Overlay

Overlays an image on top of another using composition.

sharp('input.jpg').composite([{ input: 'overlay.png', gravity: 'southeast' }]).toFile('output.jpg', (err, info) => {});

Adjusting Image Quality

Adjusts the quality of an image, useful for optimizing the file size.

sharp('input.jpg').jpeg({ quality: 80 }).toBuffer().then(data => {});

Other packages similar to sharp

imagemagick

ImageMagick is a software suite to create, edit, compose, or convert bitmap images. It can read and write images in a variety of formats. It is more comprehensive than sharp but can be slower and more memory-intensive.

jimp

An image processing library for Node written entirely in JavaScript, with zero native dependencies. Jimp is more accessible due to its pure JavaScript nature but generally performs slower than sharp.

gm

GraphicsMagick for node.js, which is an image processing library that is a fork of ImageMagick. It is similar to ImageMagick in functionality and also provides a comprehensive set of image manipulation features but may not be as fast as sharp.

README

sharp

adj

1. clearly defined; distinct: a sharp photographic image.
2. quick, brisk, or spirited.
3. shrewd or astute: a sharp bargainer.
4. (Informal.) very stylish: a sharp dresser; a sharp jacket.

The typical use case for this high speed Node.js module is to convert a large JPEG image to smaller JPEG images of varying dimensions.

It is somewhat opinionated in that it only deals with JPEG images, always obeys the requested dimensions by either cropping or embedding and insists on a mild sharpen of the resulting image.

Under the hood you'll find the blazingly fast libvips image processing library, originally created in 1989 at Birkbeck College and currently maintained by the University of Southampton.

Performance is 4x-8x faster than ImageMagick and 2x-4x faster than GraphicsMagick, based mainly on the number of CPU cores available.

Prerequisites

• Node.js v0.8+
• node-gyp
• libvips-dev 7.28+

sudo npm install -g node-gyp
sudo apt-get install libvips-dev

When installed as a package, please symlink vips-7.28.pc (or later, installed with libvips-dev) as /usr/lib/pkgconfig/vips.pc. To do this in Ubuntu 13.04 (64-bit), use:

sudo ln -s /usr/lib/x86_64-linux-gnu/pkgconfig/vips-7.28.pc /usr/lib/pkgconfig/vips.pc

Install

npm install sharp

Usage

var sharp = require("sharp");

crop(inputPath, outputPath, width, height, callback)

Scale and crop JPEG inputPath to width x height and write JPEG to outputPath calling callback when complete.

Example:

sharp.crop("input.jpg", "output.jpg", 300, 200, function(err) {
  if (err) {
    throw err;
  }
  // output.jpg is a 300 pixels wide and 200 pixels high image
  // containing a scaled and cropped version of input.jpg
});

embedWhite(inputPath, outputPath, width, height, callback)

Scale and embed JPEG inputPath to width x height using a white canvas and write JPEG to outputPath calling callback when complete.

sharp.embedWhite("input.jpg", "output.jpg", 200, 300, function(err) {
  if (err) {
    throw err;
  }
  // output.jpg is a 200 pixels wide and 300 pixels high image
  // containing a scaled version of input.jpg embedded on a white canvas
});

embedBlack(inputPath, outputPath, width, height, callback)

Scale and embed JPEG inputPath to width x height using a black canvas and write JPEG to outputPath calling callback when complete.

sharp.embedBlack("input.jpg", "output.jpg", 200, 300, function(err) {
  if (err) {
    throw err;
  }
  // output.jpg is a 200 pixels wide and 300 pixels high image
  // containing a scaled version of input.jpg embedded on a black canvas
});

Testing

npm install --dev sharp
npm test

Performance

AMD Athlon 4x core 3.3GHz 512KB L2

• imagemagick x 5.55 ops/sec ±0.45% (31 runs sampled)
• gm x 10.31 ops/sec ±3.57% (53 runs sampled)
• epeg x 27.79 ops/sec ±0.12% (69 runs sampled)
• sharp x 31.52 ops/sec ±8.74% (80 runs sampled)

AWS t1.micro

• imagemagick x 1.36 ops/sec ±0.96% (11 runs sampled)
• sharp x 12.42 ops/sec ±5.84% (64 runs sampled)

AWS m1.medium

• imagemagick x 1.38 ops/sec ±0.45% (11 runs sampled)
• sharp x 12.66 ops/sec ±5.54% (65 runs sampled)

AWS c1.medium

• imagemagick x 2.10 ops/sec ±0.67% (15 runs sampled)
• sharp x 18.97 ops/sec ±10.54% (52 runs sampled)

AWS m3.xlarge

• imagemagick x 4.46 ops/sec ±0.33% (26 runs sampled)
• sharp x 28.89 ops/sec ±7.75% (74 runs sampled)