Is there a Model for Stenograhy Image Recovery

I do not have the skills to create such a model currently. Training a Model on Stenography Images that have images coded into them and then having the model create the encoded images would be the idea. I wonder if Stenography could be used in other ways as well. For instance, encoding an image into an image could be a way to categorize images or to create connections between images. Just some thoughts do with them as you, please. But follow up if you find anything interesting. If anyone is interested in a Stenography Data base though let me know though.