Creating alpha for UIImage

Question

I'm trying to create a feature in my app that allows the user to extract a specified area of an existing image, and save it as a png with alpha enabled.

I've put a UIView ontop of a UIImageView - the imageView displays the image, while you are drawing your mask on the transparent view. For drawing, I'm using UIBezierPath. The user are able to draw around the object, and the inside will temporarily fill in with black.

The user picks the image from photo roll, and it's presented in the underlying UIImageView as shown on the left image, and when the user has drawn a shape (which automatically closes), on the overlying UIView, it looks like the right image:

This is working as expected, but when the user then clicks "Crop", then the magic should start. So far, I've only been able to create a "mask" and save it as an image on the roll, as displayed here (never mind the aspect ratios, I'll fix that later):

This is just a normal image, created from the path/shape, with colors(black on white, not black on transparent). What I need is some way to use this "image" as the alpha channel for the original image.

I know that these are two completely separate things, that an alpha-channel isn't the same as an image, but I have the shape, so I would like to know if there's any possible way to "crop" or "alpha out" with my data. What I want to end up with, is a png of this cat's face, with the surroundings 100% transparent (or not even there), so that I can change the background, like this:

It's important to note that I'm not talking about showing a UIImage in a UIImageView with applied mask, I'm talking about creating a new image, based on an existing image, combined with another image that I want to somehow convert to the first image's alpha channel, thus saving one image like above, with transparency.

I'm not too familiar with handling the data of images, so I was wondering if anyone know how to create a new image based on two images, one acting as alpha for the other, when neither of the images necessarily have an alpha channel to begin with.

Answer 1

The method below will take your original image and the mask image (the black shape) and return a UIImage that includes only the content of the image covered by the mask:

+ (UIImage*) maskImage:(UIImage *)image withMask:(UIImage *) mask
{
    CGImageRef imageReference = image.CGImage;
    CGImageRef maskReference = mask.CGImage;

    CGImageRef imageMask = CGImageMaskCreate(CGImageGetWidth(maskReference),
                                             CGImageGetHeight(maskReference),
                                             CGImageGetBitsPerComponent(maskReference),
                                             CGImageGetBitsPerPixel(maskReference),
                                             CGImageGetBytesPerRow(maskReference),
                                             CGImageGetDataProvider(maskReference),
                                             NULL, // Decode is null
                                             YES // Should interpolate
                                             );

    CGImageRef maskedReference = CGImageCreateWithMask(imageReference, imageMask);
    CGImageRelease(imageMask);

    UIImage *maskedImage = [UIImage imageWithCGImage:maskedReference];
    CGImageRelease(maskedReference);

    return maskedImage;
}

The areas outside the mask will be transparent. You can then combine the resulting UIImage with a background color.