We are using scans of all sorts to identify all sorts of things. For example, see
https://www.cbsnews.com/news/valeriana-maya-city-discovered-mexico-campeche-jungle/ .
But, those discoveries are based on huge amounts of data that are taken without necessarily intending to find the things that are later discovered by careful analysis, usually by humans who are looking for something in particular.
With the development of "artificial intelligence" that can scan more rapidly and sometimes more effectively than humans, the effort by humans shifts from looking at the data themselves to developing the AI algorithms to sift through the data to look of patterns of interest.
We are also using things like thermal imaging to find people lost in forests, when we know what to look for and roughly where to look.
So, yes, a lot can be done, and doing some of those things is already in-progress.
If you really want to get involved looking for something that interests you, you would make more progress by narrowing down the parameters you are interested in and looking for people who are already involved in doing that.
We are never going to be able to know
everything about what is on and under Earth's surface, even with satellite data for the entire surface. The question about dinosaur fossils is an example of something that we might be able to narrow down the best places to look, but we would not be able to find individual skeletons deep in rocks, because they are not really different materials from those rocks. Of course, if they were visible on the surface in a recognizable form, they could be seen with photography that had enough detail. But think about how hard it would be to design an algorithm to look at high resolution pictures of the entire surface of the continents and be able to recognize whatever the actual fossils happened to look like
without getting confused by anything else that might look similar, such as the bones of a road-killed deer lying in the woods. The problem is that humans would have to look through far too many "false positive" findings by the algorithm to maybe find an actual fossil actually exposed on the surface.