Proof Obama could have forged birth document

It really is amazing how many people don't think Barack Obama could've forged his online Certification of Live Birth. Digital forgery is nothing new and has been around for years.

Before this, no one has either been bold and/or dumb enough to take it to such a magnitude in attempting to fool the masses.

Myself, I have done some digital forgery. As a programmer, I never had time to play with graphics much, but still wanted my stuff to look nice. So what did I do? I just did an image search on the Internet and found banners, logos and borders I liked. Most of them had text already on them, so I just replaced that text with my own. Good as new, you can't tell the difference, at least not from afar.

But, zoom in and whoa Nelly, that ugly dithered, granulated, washed-out coloring of pixels between the text on the forged graphic. Needless to say, it annoyed me. But then I thought to myself, "Well I couldn't tell until I zoomed in a lot, and most people don't zoom in on the graphics, so no one will notice."

Sure enough, I even received compliments on a few! I had the last laugh, I guess.

You see, on a non-forged image with text, the text is crisp. There is no "color bleeding" of the text onto the background. I'll explain why this is, but it may get a bit technical. Please try to visualize the concepts laid out here and apply what you learn.

I'm sure you have heard of image editing programs like Adobe Photoshop. These programs allow you to create and manipulate images. To allow you to edit any part of an image, the program reserves its own file extension to keep track of all the properties of the image. In this example, Adobe Photoshop has the file extension .PSD.

Image properties and attributes are things like how many pixels the image has, along with what each color of a pixel is. When you pull an image into Photoshop for editing, it detects the file format. If it is PSD, the editor knows everything about the image. You can add text to the background and it will make perfect calculations as to what pixels should be colored to create the text, giving you a clean, clear-cut, color transition from the text to the background.

Over the years, image editors have become more sophisticated. Now most allow you to directly scan images into the editing program as the native file format to minimize "image loss." In other words, the image editor will take an inventory of the scanned image's attributes such as shape, size, colors and more. It does an OK job, but there is and always will be "image loss" because the image being scanned isn't native to the editor.

It's great if you have an image where you do not need to alter or replace text. The image looks fine, but if you try to remove and/or replace the text on the original scanned image, lo and behold, you are warned about "image loss." Now you get to it see up close and personal (but only if you zoom in).

The image editor creates a separate layer to super-impose that text onto the original foreign image layer. Think of an image layer as a separate, yet transparent, new image that will lie over the top of an existing one. Only if the text and background are part of the same image layer, will you not have "pixilation" abnormalities. Only the device that originally created the image can know the true properties of that layer. This is why the image editor automatically creates a new layer, because it doesn't know enough about the existing one.

If you type the new text onto the scanned or imported image from a foreign source or of a different image format, it looks pretty good. But, as stated before, if you zoom in, you will see this "glow" and pixilated discoloration around the text.

The glow is created because the image editor makes a guess as to what the original existing image layer's pixel locations are. Remember, when you create/type text, the editor selects the pixels it will color to make that text. When you scanned or imported the non-native image into the editor, it made a calculation as to what the pixel size and locations were, relative to the properties and size dimensions it discovered. It isn't perfect though, so it makes a guess based on what it found selects the pixels on the new layer to color.

The distorted glow is the difference in "image loss" or mismatching between the two layers. Think of it as the remainder of a math equation. Everything doesn't compute perfectly, so it guesses and does a pretty good job, but you still get the glow. If the text were part of the same layer as the original image, there would be no glow, because the math is nearly perfect in calculating which pixels to color.

Now, apply what you have learned. Of the two images (A and B) below, what image is forged and what one is not?

Image A: