===============================IMPORTANT==============================
If you want to use it within ES5 JS project under the window object, you should access it by window.ImageCompressor.ImageCompressor. Now ImageCompressor is module, so that strange access sequence appeared.
However within ES6 JS project after first import of the module, ImageCompressor will be available at window.ImageCompressor
=====================================================================
Image compressor uses canvas in the background for resizing operations. Result image has no dimensional distortions, original image is being compressed with the original aspect ratio.
Original image is centered in the result image. Free space can appear aside of result image due to possible different aspect ratios of original and compressed images.
Such free space is filled with the #FFFFFF color for image/jpeg mime type. For image/png free space aside result image filled transparently.
Demo
Image compressor uses step-down algorithm for scaling down images. That means an image gets compressed for some discrete steps. Within each step image gets compressed by a half(original size : width, height, size after first step is width / 2, height / 2 ). Such stepping allows limited interpolation on compressed image and gives more similar to original image result
npm install image-compressor
import { ImageCompressor } from 'image-compressor';
const imageCompressor = new ImageCompressor;
const compressorSettings = {
toWidth : 100,
toHeight : 100,
mimeType : 'image/png',
mode : 'strict',
quality : 0.6,
grayScale : true,
sepia : true,
threshold : 127,
vReverse : true,
hReverse : true,
speed : 'low'
};
...
imageCompressor.run(imageSrc, compressorSettings, proceedCompressedImage);
function proceedCompressedImage (compressedSrc) {
// ...
}Image compressor has one useful method: run(imageSrc, compressorSettings, proceedCompressedImage)
imageSrc : Source of the image to compress/resize. Enabled dataURL or image src url. Be careful with CORS restrictions!
compressorSettings : Object with the compression settings. Enabled fields :
compressorSettings.mode : Values strict, stretch. Strict mode disables dimensional distortions, stretch mode stretches or squeezes original image to fit result sizes
compressorSettings.toWidth : Width in pixels of the result (compressed/stretched) image, default : 100
compressorSettings.toHeight : Height in pixels of the result (compressed/stretched) image, default : 100
compressorSettings.mimeType : Mime type of the result (compressed/stretched) image, allowed values image/png, image/jpeg, default : "image/png"
compressorSettings.quality : Quality of the result (compressed/stretched) image, allowed values 0-1 with step 0.1. So 0.5 or 0.8 - correct values , 0.35 or 2 - incorrect values, default : 1
compressorSettings.grayScale : If you need to apply grayscale filter to pixels of the compressed image, just set this parameter to true. Default : false
compressorSettings.sepia : If you need to apply sepia filter to pixels of the compressed image, just set this parameter to true. Default : false
compressorSettings.threshold : Pixel intensity edge to set black and white. Pixels that has intensity larger then number specified in this setting will be converted to white color, if smaller - pixel become black Default : false
compressorSettings.vReverse : Applies vertical reverse to the result image Default : false
compressorSettings.hReverse : Applies horizontal reverse to the result image Default : false
compressorSettings.speed : Compression speed. Allowed values "low", "high". In the case of the "low" value quality lossless algorithm is being applied(slower, many steps of compression), "high" value compresses an image just in one step(faster but with the large delta between original and compressed sizes result image has poor quality). Default : "low"
If toWidth or toHeight did not specified, compressor will use original aspect ratio of an image and will combine it with available toWidth or toHeight
So for example if an original image size will be 2000x1000 and specified toWidth : 500 result toHeight will be calculated as 500 * 1000 / 2000
At least one of two above settings must be set to make compressor working correctly
You can apply only one filter from available(grayscale, sepia, threshold) to the compressed image. If selected both of them at a time, compressor will use only grayscale filter
Vertical and horizontal reverses can be applied together. If selected both of them, result image will be reversed diagonally.