I agree with PJ. Cars (building them, racing them, studying their history...you name it) have been in my blood since I was about 2 weeks old and my father put a model of a then-current Indy car beside me and snapped a photo. That was in 1947. It's been all downhill since then. 'Course I've done the same thing with my grandkids. I gave 'em race car models at about the same age.
I have shelves full of books that go back at least 40 years, a database of well over 150 thousand images, cataloged and indexed for quick lookup (if I only had something like Google Search By Image for my own collection...) and thousands of photos that I've shot myself since 1999 when I started photographing car shows and hot rod cruises, then did the freelance photographer thing for a while both with sports cars and on the Bonneville Salt Flats. I still do the salt flat thing, but otherwise I'm retired; my heart and cardiovascular system are pretty worn out. (My primary website has many of these photos and I've done my best to identify the cars, at least in the car show "cruisin" section by year, make and model...I also have some of the photos I shot with my first SLR camera, a Pentax Spotmatic, going back to 1968-1971.)
To make a a website/collection like you've described, I suggest doing a thorough search of the internet and make a list of all the sites that already attempt to do this. Some are pretty good, others are riddled with errors or have gaping holes in the information provided...like country of origin.
From there, I'd suggest designing a database that will allow a logical method of searching and presenting the content without endless hours of maintenance. (I was a software developer for the last 15 years of my career, and I also did some database design.) As you're doing this, consider how you want the information presented. I start by breaking things into several major categories: production cars, race cars, hot rods and kustom cars, trucks and so on. Then within each major section I break things down in what I think is the most logical way. For production cars, it's by country of origin. For race cars, it's a combination of marque...but I have a sub-section for Formula cars...broken down by type: F1, F2, F5000 and so on. I also have several sections for US racing series...Indy/sprint and midget; US sports car racing; stock car racing. Then I have another section for Land Speed Racing that includes all countries.
I suppose I could restructure this, since I'm not using a database manager, just some image viewing (ACDSee) and cataloging software (I use a program called, "WhereIsIt".)
One thing I had to do was build a "high-performance" computer (8 processor cores, 16 gigabyte memory, 11 terabytes of disk, high-end video card with 2 large flat screen monitors), since computers that I'd had previously just weren't up to the task. I'd suggest looking at your anticipated need for storage and performance and then double it. This is another case where "too much is just about right."
Before you even consider any of this, ask yourself if you plan to make any money from it and how that might be accomplished. If you DO intend to have a commercial site, make sure that you have the copyright to all the images you plan to show. And even then you may find yourself pestered with people who demand that you remove the photos of their cars...I can quote you the US law about this in depth, but basically it says that if you present a photo of "property" (which has no rights to privacy of its own) that can't be easily associated with an individual and the photo was shot on a location that's accessible to the public, the owner can't demand payment, nor removal of the photo, commercial website or not. However, if you have photos of some racing series, that can become a problem, since some series have people who attempt to sell photos of their racing events...and then the threats of lawsuits and other legal issues begins. I've been fortunate not to be in that position, but I know people who have.
Now. If you still want to go ahead with this and the magnitude of the project isn't overwhelming and you've seen what other people have done, then make sure you have virtually all your time (you may have to give up sleeping and working to do it) available to devote to it. While you're at it, consider what language(s) you plan to present it in and decide how to address this issue. One option here might be to specialize in one country, marque or genre. There are many of those on the web now, but I'm sure there's a niche that someone hasn't covered.
So, like PJ, I won't suggest that you give up. But I've tried to outline some of the issues that should be addressed before continuing. I'd also suggest sticking around here, solving puzzles (it's amazing how much one can learn doing that) and/or writing articles. The people here have a huge storehouse of information and are generally pretty good about sharing it.
Again, check in my profile for my primary website (Ray the Rat's Chevy Asylum) and see how I've done things. It may give you some insight of what to do and what NOT to do.
Best wishes for success in your project,