I have this code that for some reason it should draw canvas many times (using setTimeout):
function drawImage() {
var img = new Image();
var canvas = document.getElementById("event");
var context = canvas.getContext("2d");
img.src = "../img/event.png";
img.onload = function () {
context.drawImage(img, 0, 0);
}
setTimeout(drawImage, 3000);
}
but it cause dark edge as draw more times as this:
Is it possible to prevent edges become dark in repetitively drawing?
As long as the image has an alpha-channel you cannot avoid this when re-painting it over and over without some mechanism to reset the alpha values drawn.
The reason for this is that the alpha-value for that pixel will accumulate so when you draw a non-opaque edge (or anti-aliased edge) on top of each other the value for the alpha channel will be added to the value already drawn resulting in that the edge will be more and more visible.
There are fortunately a couple of ways to avoid this:
A) If you want to keep the alpha-channel in the image use clearRect before drawing the image.
context.clearRect(0, 0, img.width, img.height);
context.drawImage(img, 0, 0);
This will clear the canvas and the alpha channel.
ONLINE DEMO
B) If alpha channel is not important save out the image without any alpha-channel (use PNG with transparency off or use JPEG).
Also a note to your loop in the example - this is not a good way to redraw an image as you are initiating a load/cache check as well as image decoding each time.
You can modify your code as this (adopt as you please):
var canvas = document.getElementById("event");
var context = canvas.getContext("2d");
var img = document.createElement('img'); // due to chrome bug
img.onload = drawImage; // set onload first
img.src = "../img/event.png"; // src last..
function drawImage() {
context.clearRect(0, 0, img.width, img.height);
context.drawImage(img, 0, 0);
setTimeout(drawImage, 3000);
}
Related
I have this image drawn to a HTML5 canvas:
What I want to do is apply color to just a part of it.
The part where I want to apply color is defined by the following overlay image:
So, basically, I would like to guide my coloring by the overlay. So where the overlay pixels meets the main image pixels I should apply a color on the main image. At least that's how I see it working.
Notice that the overlay matches the whole image except for the lacing.
The catch is that I would like to retain the main image texture while applying the color. You can see that it has a leather texture and a "real" feel which I want to keep.
Can you please show me some methods of achieving this or share some thoughts?
Thank you!
globalCompositeOperation is your friend here.
Basically, you draw your overlay, then you set the gCO to 'source-atop' composite mode, which will make all your future drawings to only stay where there were already opaque pixels drawn, so it is important that your overlay has transparent parts.
So then you just fill a rectangle of your desired command, and finally you draw your original image, either behind, or blended to the new shape we just created.
var ctx = canvas.getContext('2d');
var loaded = 0;
function onload(){
if(++loaded === 2){
canvas.width = this.width;
canvas.height = this.height;
ctx.font = "40px sans-serif";
draw();
}
}
var original = new Image();
var overlay = new Image();
original.onload = overlay.onload = onload;
original.src = 'https://i.stack.imgur.com/vIKpI.png';
overlay.src = 'https://i.stack.imgur.com/10Tre.png';
// list of blending modes.
// Note that destination-over is a composite mode,
// which place the new drawings behind the already-there ones
var currentMode = 0;
var modes = ['destination-over', 'lighter', 'multiply', 'screen', 'overlay', 'darken',
'lighten', 'color-dodge', 'color-burn', 'hard-light', 'soft-light',
'exclusion', 'hue', 'saturation', 'color', 'luminosity' ];
function draw(){
// switch between different Blending modes
var mode = modes[currentMode];
currentMode = (currentMode+1)%(modes.length);
// clear previous
ctx.clearRect(0,0,canvas.width, canvas.height);
// draw our overlay
ctx.drawImage(overlay, 0,0);
// this will keep new drawings only where we already have existing pixels
ctx.globalCompositeOperation = 'source-atop';
ctx.fillStyle = 'red';
ctx.fillRect(0,0,canvas.width, canvas.height);
// now choose between the list of blending modes
ctx.globalCompositeOperation = mode;
// draw our original image
ctx.drawImage(original, 0,0);
// go back to default
ctx.globalCompositeOperation = 'source-over';
// just so we can know which one is shown
ctx.fillStyle = 'black';
ctx.fillText(mode, 40,40)
// do it again
setTimeout(draw, 1000)
}
canvas{
width: 100%;
}
<canvas id="canvas"></canvas>
I'm very new to Html5 and I was wondering if someone could shed some light on this:
<script type="text/javascript">
window.onload = function () {
var canvas = document.getElementById('canvas'); //682 x 111 pixel canvas
var context = canvas.getContext('2d');
var image = new Image();
image.src = "/Content/ImageTestOne/logo-for-dissolve.png"; //682 x 111 pixel image
image.onload = function () { context.drawImage(image, 0, 0); drawFrame(); };
function drawFrame() {
window.requestAnimationFrame(drawFrame, canvas);
imageData = context.getImageData(0, 0, canvas.width, canvas.height);
//Do something to some pixels here that persists over time
context.clearRect(0, 0, canvas.width, canvas.height);
context.putImageData(imageData, 0, 0);
};
};
</script>
According to my limited knowledge of Html5 this code should do nothing except continually display the "image". But instead the image quite rapidly burns out to almost white which suggests that the imageData is changed slightly each time it is either read from or written to the canvas...
Basically I wanted to fade the image where the mouse was located so that a background image shows through as the mouse is moved around. Is there a way around this or am I going to have to become a little more creative with the process? Is there anyway I can manipulate the "image" ImageData rather than getting it from the canvas each time?
Thanks in advance, I've tried using jpg and png and loading into DOM rather than via image.src but they all have the same issue.
Using the latest Chrome btw.
Here is the setup for the requestionAnimationFrame to handle a range of browsers with a fail over to SetTimeout:
(!window.requestAnimationFrame)
{
window.requestAnimationFrame = (window.webkitRequestAnimationFrame ||
window.mozRequestAnimationFrame ||
window.oRequestAnimationFrame ||
window.msRequestAnimationFrame ||
function (callback) {
return window.setTimeout(callback, 1000/60);
});
}
Here is the code for the canvas
<canvas id="canvas" width="682" height="111"></canvas>
That's all the code for this.
putImageData() and getImageData() can be lossy. There's a note in the spec about this:
Due to the lossy nature of converting to and from premultiplied alpha
color values, pixels that have just been set using putImageData()
might be returned to an equivalent getImageData() as different values.
See also this related question:
Why does HTML Canvas getImageData() not return the exact same values that were just set?
I have tried also to apply this to my game where in im going to manipulate the selected pixels to have effect but It doesn't give me the expected result
here is some sample code that i used to manipulate the pixel to change
get image information and store
var img = context.getImageData(0,0, width, height)
var imgdata = img.data
var len = imgdata.length
loop to all data and manipulate pixel information
var i = 0;
for(i; i<leng; i++) {
var red = imgdata[i]
var green = imgadata[i+1]
var blue = imgdata[i+2]
var alpha = imgdata[i+3]
imgdata[i] = new value
imgdata[i+1] = new value
imgdata[i+2] = new value
imgdata[i+3] = new value
}
context.putImageData(img, 0,0)
then create animation frame to see effect
requestAnimationFrame is an experimental feature (june 2012) that uses time based frame access. The reason for this is avoid latency in animations.
I suggest you take a look at this Moz article.
I'm trying to add a clear button to erase the canvas of union draw
I thought clearRect should be able to do that. I tried with:
function clearCanvas() {
clearRect(0,0,canvas.width,canvas.height);
}
or
function clearCanvas(x,y,w,h) {
ctx.clearRect(x,y,w,h);
}
...but it doesn't work, why?
The problem with using ctx.clearRect(0,0,canvas.width,canvas.height) is that if you have modified the transformation matrix you likely will not be clearing the canvas properly.
The solution? Reset the transformation matrix prior to clearing the canvas:
// Store the current transformation matrix
ctx.save();
// Use the identity matrix while clearing the canvas
ctx.setTransform(1, 0, 0, 1, 0, 0);
ctx.clearRect(0, 0, canvas.width, canvas.height);
// Restore the transform
ctx.restore();
Edit:
Chrome responds well to: context.clearRect ( x , y , w , h ); but IE9 seems to completely ignore this instruction.
IE9 seems to respond to: canvas.width = canvas.width; but it doesn't clear lines, just shapes, pictures and other objects.
So if you have a canvas and context created like this:
var canvas = document.getElementById('my-canvas');
var context = canvas.getContext('2d');
You can use a method like this:
function clearCanvas(context, canvas) {
context.clearRect(0, 0, canvas.width, canvas.height);
var w = canvas.width;
canvas.width = 1;
canvas.width = w;
}
Edit2:
use this code to check if the browser supports it:
function checkSupported() {
canvas = document.getElementById('canvas');
if (canvas.getContext){
ctx = canvas.getContext('2d');
// Canvas is supported
} else {
// Canvas is not supported
alert("We're sorry, but your browser does not support the canvas tag. Please use any web browser other than Internet Explorer.");
}
}
And to make this code execute when the web page loads, adjust the body tag so it reads like this:
<body onload="checkSupported();">
I have a flash animation I am trying to convert to HTML5. Now I have taken out all the images. For example in the hand animation, I have taken images of all hand images. I have made the canvas with the base drawing but I don't know how to replace those images frame by frame.
function draw(){
var canvas = document.getElementById('canvas');
if(canvas.getContext){
// canvas animation code here:
var ctx = canvas.getContext('2d');
var lhs = new Image();
lhs.src = "images/left_hnd_1.png";
lhs.onload = function(){
ctx.drawImage(lhs, 293, 137);
}
} else {
// canvas unsupported code here:
document.getElementById('girl').style.display = "block";
}
}
Now I have three more frame for this image. left_hnd_2.png, left_hnd_3.png & left_hnd_4.png. I would've used one image but the difference in frames is way too much for it to be done with one image. How can I animate this with the time differences I want.
Any ideas would be greatly appreciated. Thanks!
Try this:
var imgNumber = 1;
var lastImgNumber = 4;
var ctx = canvas.getContext('2d');
var img = new Image;
img.onload = function(){
ctx.clearRect( 0, 0, ctx.canvas.width, ctx.canvas.height );
ctx.drawImage( img, 0, 0 );
};
var timer = setInterval( function(){
if (imgNumber>lastImgNumber){
clearInterval( timer );
}else{
img.src = "images/left_hnd_"+( imgNumber++ )+".png";
}
}, 1000/15 ); //Draw at 15 frames per second
An alternative, if you only have 4 images, would be to create a single huge image with all four in a 'texture atlas', and then use setTimeout or setInterval to call drawImage() with different parameters to draw different subsets of the image to the canvas.
This worked for me as well! For some reason, it didn't work when I had used the OP's opening code: function draw(){
However when I used: window.onload = function draw() { the animation plays on the canvas. I'm also using about 150 PNG images with an Alpha channel so this is a great way to bring 'video' or create composites to the iPad/iPhone. I confirm that it does work on iPad iOS 4.3.
Using the HTML5 <canvas> element, I would like to load an image file (PNG, JPEG, etc.), draw it to the canvas completely transparently, and then fade it in. I have figured out how to load the image and draw it to the canvas, but I don't know how to change its opacity.
Here's the code I have so far:
var canvas = document.getElementById('myCanvas');
if (canvas.getContext)
{
var c = canvas.getContext('2d');
c.globalAlpha = 0;
var img = new Image();
img.onload = function() {
c.drawImage(img, 0, 0);
}
img.src = 'image.jpg';
}
Will somebody please point me in the right direction like a property to set or a function to call that will change the opacity?
I am also looking for an answer to this question, (to clarify, I want to be able to draw an image with user defined opacity such as how you can draw shapes with opacity) if you draw with primitive shapes you can set fill and stroke color with alpha to define the transparency. As far as I have concluded right now, this does not seem to affect image drawing.
//works with shapes but not with images
ctx.fillStyle = "rgba(255, 255, 255, 0.5)";
I have concluded that setting the globalCompositeOperation works with images.
//works with images
ctx.globalCompositeOperation = "lighter";
I wonder if there is some kind third way of setting color so that we can tint images and make them transparent easily.
EDIT:
After further digging I have concluded that you can set the transparency of an image by setting the globalAlpha parameter BEFORE you draw the image:
//works with images
ctx.globalAlpha = 0.5
If you want to achieve a fading effect over time you need some kind of loop that changes the alpha value, this is fairly easy, one way to achieve it is the setTimeout function, look that up to create a loop from which you alter the alpha over time.
Some simpler example code for using globalAlpha:
ctx.save();
ctx.globalAlpha = 0.4;
ctx.drawImage(img, x, y);
ctx.restore();
If you need img to be loaded:
var img = new Image();
img.onload = function() {
ctx.save();
ctx.globalAlpha = 0.4;
ctx.drawImage(img, x, y);
ctx.restore()
};
img.src = "http://...";
Notes:
Set the 'src' last, to guarantee that your onload handler is called on all platforms, even if the image is already in the cache.
Wrap changes to stuff like globalAlpha between a save and restore (in fact use them lots), to make sure you don't clobber settings from elsewhere, particularly when bits of drawing code are going to be called from events.
Edit: The answer marked as "correct" is not correct.
It's easy to do. Try this code, swapping out "ie.jpg" with whatever picture you have handy:
<!DOCTYPE HTML>
<html>
<head>
<script>
var canvas;
var context;
var ga = 0.0;
var timerId = 0;
function init()
{
canvas = document.getElementById("myCanvas");
context = canvas.getContext("2d");
timerId = setInterval("fadeIn()", 100);
}
function fadeIn()
{
context.clearRect(0,0, canvas.width,canvas.height);
context.globalAlpha = ga;
var ie = new Image();
ie.onload = function()
{
context.drawImage(ie, 0, 0, 100, 100);
};
ie.src = "ie.jpg";
ga = ga + 0.1;
if (ga > 1.0)
{
goingUp = false;
clearInterval(timerId);
}
}
</script>
</head>
<body onload="init()">
<canvas height="200" width="300" id="myCanvas"></canvas>
</body>
</html>
The key is the globalAlpha property.
Tested with IE 9, FF 5, Safari 5, and Chrome 12 on Win7.
This suggestion is based on pixel manipulation in canvas 2d context.
From MDN:
You can directly manipulate pixel data in canvases at the byte level
To manipulate pixels we'll use two functions here - getImageData and putImageData.
getImageData usage:
var myImageData = context.getImageData(left, top, width, height);
The putImageData syntax:
context.putImageData(myImageData, x, y);
Where context is your canvas 2d context, and x and y are the position on the canvas.
So to get red green blue and alpha values, we'll do the following:
var r = imageData.data[((x*(imageData.width*4)) + (y*4))];
var g = imageData.data[((x*(imageData.width*4)) + (y*4)) + 1];
var b = imageData.data[((x*(imageData.width*4)) + (y*4)) + 2];
var a = imageData.data[((x*(imageData.width*4)) + (y*4)) + 3];
Where x is the horizontal offset, y is the vertical offset.
The code making image half-transparent:
var canvas = document.getElementById('myCanvas');
var c = canvas.getContext('2d');
var img = new Image();
img.onload = function() {
c.drawImage(img, 0, 0);
var ImageData = c.getImageData(0,0,img.width,img.height);
for(var i=0;i<img.height;i++)
for(var j=0;j<img.width;j++)
ImageData.data[((i*(img.width*4)) + (j*4) + 3)] = 127;//opacity = 0.5 [0-255]
c.putImageData(ImageData,0,0);//put image data back
}
img.src = 'image.jpg';
You can make you own "shaders" - see full MDN article here
You can. Transparent canvas can be quickly faded by using destination-out global composite operation. It's not 100% perfect, sometimes it leaves some traces but it could be tweaked, depending what's needed (i.e. use 'source-over' and fill it with white color with alpha at 0.13, then fade to prepare the canvas).
// Fill canvas using 'destination-out' and alpha at 0.05
ctx.globalCompositeOperation = 'destination-out';
ctx.fillStyle = "rgba(255, 255, 255, 0.05)";
ctx.beginPath();
ctx.fillRect(0, 0, width, height);
ctx.fill();
// Set the default mode.
ctx.globalCompositeOperation = 'source-over';
I think this answers the question best, it actually changes the alpha value of something that has been drawn already. Maybe this wasn't part of the api when this question was asked.
Given 2d context c.
function reduceAlpha(x, y, w, h, dA) {
let screenData = c.getImageData(x, y, w, h);
for(let i = 3; i < screenData.data.length; i+=4){
screenData.data[i] -= dA; //delta-Alpha
}
c.putImageData(screenData, x, y );
}
Set global Alpha draw the object that has opacity then set back to normal.
//////////////////////// circle ///////////////////////
ctx.globalAlpha = 0.75;
ctx.beginPath();
ctx.arc(x1, y1, r1, 0, Math.PI*2);
ctx.fillStyle = colour;
ctx.fill();
ctx.closePath();
ctx.globalAlpha = 1;
How i made it..on canvas i first draw rect in a selfrun function 0,0,canvas.width,canvas.height as a background of canvas and i set globalAlpha to 1 .then i draw other shapes in ather own functions and set their globalAlpha to 0.whatever number they dont affect each other even images.
Like Ian said, use c.globalAlpha = 0.5 to set the opacity, type up the rest of the settings for the square, then follow up with c.save();. This will save the settings for the square then you can c.rect and c.fillStyle the square how you want it. I chose not to wrap it with c.restore afterwards and it worked well
If you use jCanvas library you can use opacity property when drawing. If you need fade effect on top of that, simply redraw with different values.
You can't. It's immediate mode graphics. But you can sort of simulate it by drawing a rectangle over it in the background color with an opacity.
If the image is over something other than a constant color, then it gets quite a bit trickier. You should be able to use the pixel manipulation methods in this case. Just save the area before drawing the image, and then blend that back on top with an opacity afterwards.