常用的Node.js图片处理模块，有node-image、sharp和jimp。node-image很轻量，但是处理gif有点问题，更新还有点慢了。sharp跨操作系统部署的时，常常还需要编译一下。所以，jimp也是一个很好的选择。

介绍：

An image processing library for Node written entirely in JavaScript, with zero native dependencies.

看这个官方介绍也说了，完全使用Node写的的库，无原生依赖。 支持图片类型bmp、gif、jpeg、png和tiff。

安装：

npm install --save jimp

示例代码

Promise will never resolve if callback is passed

回调必须执行通过，否则规约将永远不会编程完成状态。

var Jimp = require('jimp');
// open a file called "lenna.png"
Jimp.read('lenna.png', (err, lenna) => {
 if (err) throw err;
 lenna
  .resize(256, 256) // resize
  .quality(60) // set JPEG quality
  .greyscale() // set greyscale
  .write('lena-small-bw.jpg'); // save
});

使用 Promises:

Jimp.read('lenna.png')
 .then(lenna => {
  return lenna
   .resize(256, 256) // resize
   .quality(60) // set JPEG quality
   .greyscale() // set greyscale
   .write('lena-small-bw.jpg'); // save
 })
 .catch(err => {
  console.error(err);
 });

Jimp.read 的基础用法

The static Jimp.read method takes the path to a file, URL, dimensions, a Jimp instance or a buffer and returns a Promise:

静态函数 Jimp.read可以接受多种类型的参数，比如：文件、链接、Jimp实例或者一个缓存，返回一个Promise对象。

Jimp.read('./path/to/image.jpg')
 .then(image => {
  // Do stuff with the image.
 })
 .catch(err => {
  // Handle an exception.
 });

Jimp.read('http://www.example.com/path/to/lenna.jpg')
 .then(image => {
  // Do stuff with the image.
 })
 .catch(err => {
  // Handle an exception.
 });

Jimp.read(jimpInstance)
 .then(image => {
  // Do stuff with the image.
 })
 .catch(err => {
  // Handle an exception.
 });

Jimp.read(buffer)
 .then(image => {
  // Do stuff with the image.
 })
 .catch(err => {
  // Handle an exception.
 });

Methods

常用方法

Once the promise is fulfilled, the following methods can be called on the image:

/* Resize */
image.contain( w, h[, alignBits || mode, mode] );  // scale the image to the given width and height, some parts of the image may be letter boxed
image.cover( w, h[, alignBits || mode, mode] );   // scale the image to the given width and height, some parts of the image may be clipped
image.resize( w, h[, mode] );   // resize the image. Jimp.AUTO can be passed as one of the values.
image.scale( f[, mode] );     // scale the image by the factor f
image.scaleToFit( w, h[, mode] ); // scale the image to the largest size that fits inside the given width and height

// An optional resize mode can be passed with all resize methods.

/* Crop */
image.autocrop([tolerance, frames]); // automatically crop same-color borders from image (if any), frames must be a Boolean
image.autocrop(options);     // automatically crop same-color borders from image (if any), options may contain tolerance, cropOnlyFrames, cropSymmetric, leaveBorder
image.crop( x, y, w, h );     // crop to the given region

/* Composing */
image.blit( src, x, y, [srcx, srcy, srcw, srch] );
                 // blit the image with another Jimp image at x, y, optionally cropped.
image.composite( src, x, y, [{ mode, opacitySource, opacityDest }] );   // composites another Jimp image over this image at x, y
image.mask( src, x, y );     // masks the image with another Jimp image at x, y using average pixel value
image.convolute( kernel );    // applies a convolution kernel matrix to the image or a region

/* Flip and rotate */
image.flip( horz, vert );     // flip the image horizontally or vertically
image.mirror( horz, vert );    // an alias for flip
image.rotate( deg[, mode] );   // rotate the image clockwise by a number of degrees. Optionally, a resize mode can be passed. If `false` is passed as the second parameter, the image width and height will not be resized.

/* Colour */
image.brightness( val );     // adjust the brighness by a value -1 to +1
image.contrast( val );      // adjust the contrast by a value -1 to +1
image.dither565();        // ordered dithering of the image and reduce color space to 16-bits (RGB565)
image.greyscale();        // remove colour from the image
image.invert();          // invert the image colours
image.normalize();        // normalize the channels in an image

/* Alpha channel */
image.hasAlpha();           // determines if an image contains opaque pixels
image.fade( f );         // an alternative to opacity, fades the image by a factor 0 - 1. 0 will haven no effect. 1 will turn the image
image.opacity( f );        // multiply the alpha channel by each pixel by the factor f, 0 - 1
image.opaque();          // set the alpha channel on every pixel to fully opaque
image.background( hex );     // set the default new pixel colour (e.g. 0xFFFFFFFF or 0x00000000) for by some operations (e.g. image.contain and

/* Blurs */
image.gaussian( r );       // Gaussian blur the image by r pixels (VERY slow)
image.blur( r );         // fast blur the image by r pixels

/* Effects */
image.posterize( n );       // apply a posterization effect with n level
image.sepia();          // apply a sepia wash to the image
image.pixelate( size[, x, y, w, h ]); // apply a pixelation effect to the image or a region

/* 3D */
image.displace( map, offset );  // displaces the image pixels based on the provided displacement map. Useful for making stereoscopic 3D images.

Resize modes

图片缩放的模式

The default resizing algorithm uses a bilinear method as follows:

image.resize(250, 250); // resize the image to 250 x 250
image.resize(Jimp.AUTO, 250); // resize the height to 250 and scale the width accordingly
image.resize(250, Jimp.AUTO); // resize the width to 250 and scale the height accordingly

项目地址：

https://github.com/oliver-moran/jimp/tree/master/packages/jimp