Posts Tagged ‘masking’

Image Masking with the HTML5 Canvas

Image masking with the HTML5 canvas is easier than I thought it might be. This shows you the main idea and two examples.

If you’d like to cut to the chase, as they say, look at this example and its source code. The circular outline is an image. The Kandinsky painting is a rectangular image that is made to fill the circular outline. We see the result below:

The Kandinsky painting fills a blurry circle.

The Kandinsky painting fills a blurry circle.

The key is the setting for the canvas’s globalCompositeOperation property. Like me, if you had seen any documentation for this property at all, you might have thought that it only concerned color blending, like the color blending options in Photoshop for a layer (the options usually include things like ‘normal’, ‘dissolve’, ‘darken’, ‘multiply’, ‘color burn’, etc). But, actually, globalCompositeOperation is more powerful than that. It’s for compositing images. Image masking is simply an example of compositing. Studying the possibilities of globalCompositeOperation would be interesting. We’re just going to use a couple of settings in this article. The definition we read of “compositing” via Googling the term includes this:

“Compositing is the combining of visual elements from separate sources into single images, often to create the illusion that all those elements are parts of the same scene.”

We’re going to use the “source-atop” setting of  globalCompositeOperation. The default value, by the way, is “source-in”.

The basic idea is that if you want image F to fill image A, you draw image A on a fresh canvas. Then you set  globalCompositeOperation to “source-atop”. Then you draw image F on the canvas. When you do that, the pixels in the canvas retain whatever opacity/alpha value they have. So, for instance, any totally transparent pixels remain totally transparent. Any pixels that are partially transparent remain partially transparent. Image F is drawn into the canvas, but F does not affect the opacity/alpha values of the canvas.

Here is an example where a Kandinsky painting is made to fill some canvas text:

Click the image and then view the source code.

Click the image and then view the source code.

I’m working on some brushes for Aleph Null 2.0 that are a lot like the brushes in dbCinema: the brushes ‘paint’ samples of images.