I'm not sure you really understand digital photography/forensics.
It is true that digital photography can introduce "artifacts" - some of these are due to software/hardware malfunctions and compression others are just part of the digital photography process. These artifacts are permanent to the raw image and cannot be "recovered" via any program. The jagged edges you see are pixels and will occur any time you zoom far enough into a digital image - this is known as aliasing and/or mosaicing. The number of pixels in an image is determined by resolution - an image from a higher resolution camera will contain more pixels and thus you have to zoom further into the image to see those jagged edges.
All modern camera sensors perform some sort of anti-aliasing in which the pixels between 2 distinct colors will be a mix of both of those colors. This makes the pixelization somewhat less noticeable in a low-res photo but there are still jagged edges when zooming in. If the sensor has a color filter array this anti-aliasing effect is even stronger. Think about it this way, let's say I take a picture of a piece of paper that is part pure black and part pure white - without any aliasing performed all my pixels will be either pure black or pure white (in a perfect world with perfect lighting and perfect equipment). Anti-aliasing firmware will attempt to smooth the jagged edges out by extrapolating a color between the black and white pixels so you will see many shades of gray in between the pure black and pure white pixels if zoomed in far enough. These gray pixels won't be noticeable when viewing the image at full resolution though.
Too many episode of
CSI has led us to believe that we can zoom and enhance a photo to the point that something that did not exist in the original image will somehow be magically created. This is not the case. Yes, there is software that can alter pixel colors, reduce noise, and further blur pixel edges so that they are not so noticeable. I think you are looking for software that will blur your pixel edges so you do not see that jaggedness. But all that software is doing is simply extrapolating the current pixels to create new pixels and each kind of software will use a slightly different proprietary algorithm to do this. So one software program might make a highly pixelated object look like a gun while another might make it look like a knife. It's not an exact science and certainly not admissible in a court of law. It's just some estimated values inserted by a computer program.
That being said, there are some highly advanced techniques that can be accomplished using image data points that are admissible in court because they use sound science that is unambiguous and repeatable. I believe there were reports of the FBI and/or NASA helping LE out with some of these images and this is probably where they came into play. Here is a simple example - lets say that I have 30 frames of video of a license plate but cannot read any of the characters in the individual frames. Now I isolate those 30 individual plate images and make them identical in size then lay them on top of each other to create a new image based on the average or extreme of those pixels for each data point. This new image can possibly be read.
I didn't mean to correct you but I just wanted you to fully understand the limitations of digital photography/forensics and not all digital imagery artifacts are unwanted - they exist to make a more pleasing picture.