ChatGPT解决这个技术问题 Extra ChatGPT

使用 javascript 画布调整图像大小(平滑)

我正在尝试使用画布调整一些图像的大小,但我对如何平滑它们一无所知。在 Photoshop、浏览器等上。他们使用了一些算法(例如双三次、双线性),但我不知道这些算法是否内置在画布中。

这是我的小提琴:http://jsfiddle.net/EWupT/

var canvas = document.createElement('canvas');
var ctx = canvas.getContext('2d');
canvas.width=300
canvas.height=234
ctx.drawImage(img, 0, 0, 300, 234);
document.body.appendChild(canvas);

第一个是正常调整大小的图像标签,第二个是画布。请注意画布是如何不那么光滑的。我怎样才能实现“平滑”?

与<img />相比,画布中结果模糊的问题是什么?如果是这样,很可能您有视网膜屏幕(DPR >= 2),您可以放大画布以实现与 <img /> 中看到的类似结果,例如像这样 - jsfiddle.net/c8uo3pda/4

佚名

您可以使用向下步进来获得更好的结果。大多数浏览器在调整图像大小时似乎use linear interpolation rather than bi-cubic

更新在规范中添加了质量属性,imageSmoothingQuality 目前仅在 Chrome 中可用。)

除非选择不平滑或最近邻,否则浏览器将始终在缩小图像后对图像进行插值,因为此功能用作低通滤波器以避免混叠。

双线性使用 2x2 像素进行插值,而双三次使用 4x4,因此通过逐步进行,您可以在使用双线性插值的同时获得接近双三次结果,如结果图像中所示。

var canvas = document.getElementById("canvas"); var ctx = canvas.getContext("2d"); var img = new Image(); img.onload = function () { // 设置与图像成比例的大小 canvas.height = canvas.width * (img.height / img.width); // 第 1 步 - 调整大小为 50% var oc = document.createElement('canvas'), octx = oc.getContext('2d'); oc.width = img.width * 0.5; oc.height = img.height * 0.5; octx.drawImage(img, 0, 0, oc.width, oc.height); // 步骤 2 octx.drawImage(oc, 0, 0, oc.width * 0.5, oc.height * 0.5); // 第三步,调整到最终尺寸 ctx.drawImage(oc, 0, 0, oc.width * 0.5, oc.height * 0.5, 0, 0, canvas.width, canvas.height); } img.src = "//i.imgur.com/SHo6Fub.jpg";

如果差异较小,您可以根据调整大小的剧烈程度跳过第 2 步。

在演示中,您可以看到新结果现在与图像元素非常相似。


@steve heh,有时会发生这些事情:) 对于图像,您通常可以通过设置 css BTW 来覆盖它。
Ken,第一个结果效果很好,但是当我更改图像时,您会发现它太模糊了 jsfiddle.net/kcHLG 在这种情况下和其他情况下可以做什么?
@steve,您可以将步骤数减少到 1 或无(对于某些图像,这很好用)。另请参阅与此类似的 this answer,但在这里我添加了一个锐化卷积,以便您可以在缩小后使生成的图像更清晰。
@steve 这里是对 Bill 的修改,只使用了一个额外的步骤:jsfiddle.net/AbdiasSoftware/kcHLG/1
@neaumusic 代码是 OPs 代码的延续。如果你打开小提琴,你会看到 ctx 被定义。为了避免误解,我在这里内联了它。
M
Miro

由于 Trung Le Nguyen Nhat's fiddle 根本不正确(它只是在最后一步中使用原始图像)
我编写了自己的通用小提琴来进行性能比较:

FIDDLE

基本上是:

img.onload = function() {
   var canvas = document.createElement('canvas'),
       ctx = canvas.getContext("2d"),
       oc = document.createElement('canvas'),
       octx = oc.getContext('2d');

   canvas.width = width; // destination canvas size
   canvas.height = canvas.width * img.height / img.width;

   var cur = {
     width: Math.floor(img.width * 0.5),
     height: Math.floor(img.height * 0.5)
   }

   oc.width = cur.width;
   oc.height = cur.height;

   octx.drawImage(img, 0, 0, cur.width, cur.height);

   while (cur.width * 0.5 > width) {
     cur = {
       width: Math.floor(cur.width * 0.5),
       height: Math.floor(cur.height * 0.5)
     };
     octx.drawImage(oc, 0, 0, cur.width * 2, cur.height * 2, 0, 0, cur.width, cur.height);
   }

   ctx.drawImage(oc, 0, 0, cur.width, cur.height, 0, 0, canvas.width, canvas.height);
}

见过的最被低估的答案。
作为魅力!非常感谢!
这应该是公认的答案。只需确保将“宽度”替换为所需的宽度值即可。
另请注意,此方法将创建带有透明图像的伪像,因为稍后对 octx.drawImage 的调用不会完全覆盖较小矩形中的像素
f
fisch2

我创建了一个可重用的 Angular 服务来为感兴趣的人处理高质量的图像/画布大小调整:https://gist.github.com/transitive-bullshit/37bac5e741eaec60e983

该服务包括两个解决方案,因为它们都有自己的优点/缺点。 lanczos 卷积方法以速度较慢为代价获得更高质量,而逐步缩小方法可产生合理的抗锯齿结果,并且速度明显更快。

示例用法:

angular.module('demo').controller('ExampleCtrl', function (imageService) {
  // EXAMPLE USAGE
  // NOTE: it's bad practice to access the DOM inside a controller, 
  // but this is just to show the example usage.

  // resize by lanczos-sinc filter
  imageService.resize($('#myimg')[0], 256, 256)
    .then(function (resizedImage) {
      // do something with resized image
    })

  // resize by stepping down image size in increments of 2x
  imageService.resizeStep($('#myimg')[0], 256, 256)
    .then(function (resizedImage) {
      // do something with resized image
    })
})

很抱歉 - 我已经更改了我的 github 用户名。刚刚更新了链接 gist.github.com/transitive-bullshit/37bac5e741eaec60e983
我看到了角度这个词,我有那种有趣的感觉
m
marc_s

虽然其中一些代码片段很短且有效,但遵循和理解它们并非易事。

由于我不喜欢 stack-overflow 中的“复制粘贴”,因此我希望开发人员能够理解他们推送到软件中的代码,希望您会发现以下内容对您有用。

DEMO:使用 JS 和 HTML Canvas Demo fiddler 调整图像大小。

您可能会找到 3 种不同的方法来调整大小,这将帮助您了解代码的工作原理以及原因。

https://jsfiddle.net/1b68eLdr/93089/

可以在 GitHub 项目中找到完整的演示代码和您可能希望在代码中使用的 TypeScript 方法。

https://github.com/eyalc4/ts-image-resizer

这是最终代码:

export class ImageTools {
base64ResizedImage: string = null;

constructor() {
}

ResizeImage(base64image: string, width: number = 1080, height: number = 1080) {
    let img = new Image();
    img.src = base64image;

    img.onload = () => {

        // Check if the image require resize at all
        if(img.height <= height && img.width <= width) {
            this.base64ResizedImage = base64image;

            // TODO: Call method to do something with the resize image
        }
        else {
            // Make sure the width and height preserve the original aspect ratio and adjust if needed
            if(img.height > img.width) {
                width = Math.floor(height * (img.width / img.height));
            }
            else {
                height = Math.floor(width * (img.height / img.width));
            }

            let resizingCanvas: HTMLCanvasElement = document.createElement('canvas');
            let resizingCanvasContext = resizingCanvas.getContext("2d");

            // Start with original image size
            resizingCanvas.width = img.width;
            resizingCanvas.height = img.height;


            // Draw the original image on the (temp) resizing canvas
            resizingCanvasContext.drawImage(img, 0, 0, resizingCanvas.width, resizingCanvas.height);

            let curImageDimensions = {
                width: Math.floor(img.width),
                height: Math.floor(img.height)
            };

            let halfImageDimensions = {
                width: null,
                height: null
            };

            // Quickly reduce the dize by 50% each time in few iterations until the size is less then
            // 2x time the target size - the motivation for it, is to reduce the aliasing that would have been
            // created with direct reduction of very big image to small image
            while (curImageDimensions.width * 0.5 > width) {
                // Reduce the resizing canvas by half and refresh the image
                halfImageDimensions.width = Math.floor(curImageDimensions.width * 0.5);
                halfImageDimensions.height = Math.floor(curImageDimensions.height * 0.5);

                resizingCanvasContext.drawImage(resizingCanvas, 0, 0, curImageDimensions.width, curImageDimensions.height,
                    0, 0, halfImageDimensions.width, halfImageDimensions.height);

                curImageDimensions.width = halfImageDimensions.width;
                curImageDimensions.height = halfImageDimensions.height;
            }

            // Now do final resize for the resizingCanvas to meet the dimension requirments
            // directly to the output canvas, that will output the final image
            let outputCanvas: HTMLCanvasElement = document.createElement('canvas');
            let outputCanvasContext = outputCanvas.getContext("2d");

            outputCanvas.width = width;
            outputCanvas.height = height;

            outputCanvasContext.drawImage(resizingCanvas, 0, 0, curImageDimensions.width, curImageDimensions.height,
                0, 0, width, height);

            // output the canvas pixels as an image. params: format, quality
            this.base64ResizedImage = outputCanvas.toDataURL('image/jpeg', 0.85);

            // TODO: Call method to do something with the resize image
        }
    };
}}

P
Paul Roub

我不明白为什么没有人建议createImageBitmap

createImageBitmap(
    document.getElementById('image'), 
    { resizeWidth: 300, resizeHeight: 234, resizeQuality: 'high' }
)
.then(imageBitmap => 
    document.getElementById('canvas').getContext('2d').drawImage(imageBitmap, 0, 0)
);

效果很好(假设您为图像和画布设置了 ID)。


因为它没有得到广泛支持caniuse.com/#search=createImageBitmap
73% 的用户支持 createImageBitmap。根据您的用例,它可能已经足够好了。只是 Safari 拒绝添加对它的支持。我认为值得一提的是一个可能的解决方案。
不错的解决方案,但不幸的是它不适用于 Firefox
d
dansch

我创建了一个库,允许您在保留所有颜色数据的同时降低任何百分比。

https://github.com/danschumann/limby-resize/blob/master/lib/canvas_resize.js

您可以在浏览器中包含该文件。结果看起来像 Photoshop 或图像魔术,保留所有颜色数据,平均像素,而不是取附近的并丢弃其他的。它不使用公式来猜测平均值,而是取精确的平均值。


我现在可能会使用 webgl 来调整大小
T
Trung Le Nguyen Nhat

基于 K3N 的答案,我通常为任何想要的人重写代码

var oc = document.createElement('canvas'), octx = oc.getContext('2d');
    oc.width = img.width;
    oc.height = img.height;
    octx.drawImage(img, 0, 0);
    while (oc.width * 0.5 > width) {
       oc.width *= 0.5;
       oc.height *= 0.5;
       octx.drawImage(oc, 0, 0, oc.width, oc.height);
    }
    oc.width = width;
    oc.height = oc.width * img.height / img.width;
    octx.drawImage(img, 0, 0, oc.width, oc.height);

更新 JSFIDDLE 演示

这是我的ONLINE DEMO


这不起作用:每次调整画布大小时,它都会清除其上下文。你需要2张画布。在这里,它与直接使用最终尺寸调用 drawImage 相同。
V
Viacheslav Panev

我通过使用画布比例解决了这个问题,我的情况下的图像质量变得非常好。

所以首先我缩放画布内的内容:

ctx.scale(2, 2)

然后用 css 横向扩展 canvas 标签:

#myCanvas { transform: scale(0.5); }

D
Diyaz Yakubov

我写了一个小的 js-utility 来在前端裁剪和调整图像大小。这是 GitHub 项目上的 link。您也可以从最终图像中获取 blob 以发送它。

import imageSqResizer from './image-square-resizer.js'

let resizer = new imageSqResizer(
    'image-input',
    300,
    (dataUrl) => 
        document.getElementById('image-output').src = dataUrl;
);
//Get blob
let formData = new FormData();
formData.append('files[0]', resizer.blob);

//get dataUrl
document.getElementById('image-output').src = resizer.dataUrl;

w
webargus

这是我的代码,我希望它对 SO 社区中的某个人有用:

您可以将目标图像尺寸作为参数包含在脚本调用中。这将是您的图像宽度或高度的结果值,以较大者为准。较小的尺寸会调整大小,保持图像纵横比不变。您还可以在脚本中硬编码您的默认目标大小。

您可以轻松更改脚本以满足您的特定需求,例如您想要输出的图像类型(默认为“image/png”),并决定您想要调整图像大小以获得更精细的结果的百分比(参见代码中的 const percentStep)。

   const ResizeImage = ( _ => {

const MAX_LENGTH = 260;     // default target size of largest dimension, either witdth or height
const percentStep = .3;     // resizing steps until reaching target size in percents (30% default)
const canvas = document.createElement("canvas");
const canvasContext = canvas.getContext("2d");
const image = new Image();

const doResize = (callback, maxLength) => {

    // abort with error if image has a dimension equal to zero
    if(image.width == 0 || image.height == 0) {
        return {blob: null, error: "either image width or height was zero "};
    }

    // use caller dimension or default length if none provided
    const length = maxLength == null  ? MAX_LENGTH : maxLength;

    canvas.width = image.width;
    canvas.height = image.height;
    canvasContext.drawImage(image, 0, 0, image.width, image.height);
    // if image size already within target size, just copy and return blob
    if(image.width <= length && image.height <= length) {
        canvas.toBlob( blob => {
            callback({ blob: blob, error: null });
        }, "image/png", 1);
        return;
    }

    var startDim = Math.max(image.width, image.height);
    var startSmallerDim = Math.min(image.width, image.height);

    // gap to decrease in size until we reach the target size,
    // be it by decreasing the image width or height,
    // whichever is largest
    const gap = startDim - length;
    // step length of each resizing iteration
    const step = parseInt(percentStep*gap);
    //  no. of iterations
    var nSteps = 0;
    if(step == 0) {
        step = 1;
    } else {
        nSteps = parseInt(gap/step);
    }
    // length of last additional resizing step, if needed
    const lastStep = gap % step;
    // aspect ratio = value by which we'll  multiply the smaller dimension
    // in order to keep the aspect ratio unchanged in each iteration
    const ratio = startSmallerDim/startDim;

    var newDim;          // calculated new length for the bigger dimension of the image, be it image width or height
    var smallerDim;     // length along the smaller dimension of the image, width or height
    for(var i = 0; i < nSteps; i++) {
        // decrease longest dimension one step in pixels
        newDim = startDim - step;
        // decrease shortest dimension proportionally, so as to keep aspect ratio
        smallerDim = parseInt(ratio*newDim);
        // assign calculated vars to their corresponding canvas dimension, width or height
        if(image.width > image.height) {
            [canvas.width, canvas.height]  = [newDim, smallerDim];
        } else {
            [canvas.width, canvas.height] = [smallerDim, newDim];
        }
        // draw image one step smaller
        canvasContext.drawImage(canvas, 0, 0, canvas.width, canvas.height);
        // cycle var startDim for new loop
        startDim = newDim;
    }

    // do last missing resizing step to finally reach target image size
    if(lastStep > 0) {
        if(image.width > image.height) {
            [canvas.width, canvas.height]  = [startDim - lastStep, parseInt(ratio*(startDim - lastStep))];
        } else {
            [canvas.width, canvas.height] = [parseInt(ratio*(startDim -lastStep)), startDim - lastStep];
        }
        canvasContext.drawImage(image, 0, 0, canvas.width, canvas.height);
    }

    // send blob to caller
    canvas.toBlob( blob => {
        callback({blob: blob, error: null});
    }, "image/png", 1);

};

const resize = async (imgSrc, callback, maxLength) => {
    image.src = imgSrc;
    image.onload = _ => {
       doResize(callback, maxLength);
    };
};

return { resize: resize }

})();

用法:

ResizeImage.resize("./path/to/image/or/blob/bytes/to/resize", imageObject => {
    if(imageObject.error != null) {
      // handle errors here
      console.log(imageObject.error);
      return;
    }
    // do whatever you want with the blob, like assinging it to
    // an img element, or uploading it to a database
    // ...
    document.querySelector("#my-image").src = imageObject.blob;
    // ...
}, 300);

K
Kamesh S
export const resizeImage = (imageFile, size = 80) => {
    
    let resolver = ()=>{};

    let reader = new FileReader();

    reader.onload = function (e) {
        let img = document.createElement("img");
        img.onload = function (event) {
            // Dynamically create a canvas element
            let canvas = document.createElement("canvas");

            canvas.width=size;
            canvas.height=size;

            // let canvas = document.getElementById("canvas");
            let ctx = canvas.getContext("2d");

            // Actual resizing
            ctx.drawImage(img, 0, 0, size, size);

            // Show resized image in preview element
            let dataurl = canvas.toDataURL(imageFile.type);

            resolver(dataurl);
        }
        img.src = e.target.result;
    }

    reader.readAsDataURL(imageFile);

    
    return new Promise((resolve, reject) => {
        resolver = resolve;
    })
};