Electrolysis of water is simple and can be done anywhere, but it requires a lot of energy. However, since water is virtually free, the question then revolves around how much it will cost in electricity to produce a given quantity of hydrogen. I don't have the numbers for that.

In the past, I've written about the difficulty of developing a hydrogen infrastructure. However, it may be practical to have electrolysis-based processing at the refueling stations, with the necessary compression and storage facilities. Doing that would eliminate the need for massive production facilities, pipelines, etc. If electrolysis-based processing is too expensive, perhaps they can develop small facilities to split the hydrogen from natural gas or something similar.

As for our current petroleum based infrastructure, it has evolved over more than 100 years at huge expense. A single oil tanker costs many million dollars and can carry 100,000 tons or so of crude oil per trip. It takes a lot of energy to cross the ocean using a tanker. The oil has to be processed at refineries which also take quite a bit of energy, then transported to the gas stations around the country using pipelines, trains, and trucks.

All of that requires energy, but the economy of scale and improvements in technology have made it pretty inexpensive. If the hydrogen processing can be safely and economically accomplished at neighborhood stations, then we'd have a huge savings in infrastructure costs. While splitting hydrogen from water using electrolysis requires a lot of energy, I'd like to see a cost per mile analysis based on propelling a hydrogen powered vehicle (combustion or fuel cell). Suppose a given gasoline vehicle costs 5 cents per mile to carry a specific load. How much per mile would the hydrogen cost to move the same load?

As for your idea of using UV, I simply don't know. Hydrogen is bonded pretty strongly to oxygen in water, so I don't know if UV would help split it or not.