A few hundred thousand people are reported missing each year in the U.S. alone. While majority of the cases are solved there remains a couple of thousand or so cases that remain open each year. As of January 2015, the number of total, open missing persons cases was about 85,000.
It occurs to me in this era of vast computing resources, prolifiration of social media, public cameras and machine learning that there might be a technological solution that may aid in the recovery of missing persons, here in the U.S. and around the world.
Facebook already has algorithms that can identify people by their faces. Microsoft just released Face APIs that can tell a lot from just a picture of a person, e.g. how old they might be, their sex, identify the same person in multiple pictures, etc. Better yet, Microsoft is making these application programming interface (API) available to the public for free (at least while they are in beta), which means anyone can use it to write facial recognition programs. Mat Velloso (granted, a Microsoft employee) used it to a Twins Or Not? in about 4 hours!
Consider a computer program that takes a picture of a missing person and compares it against the vast repository of publically available images and videos of people to find a match. This will likely be too large a task for an individual computer or even a large datacenter. So, employ something like BOINC, a grid computing system that parcels out a large task to several hundreds, if not millions of computers around the world, whose owners have volunteered their idle CPU cycles for this task.
I am no computer scientist so I am probably over-simplifying the solution. But there is nothing here that doesn't already exist in one form or another and I am sure if people smarter than me put their minds to it, they could come up with an elegant solution that could help many people and families reunite.
Such a program could be used to find victims of human trafficking, natural disasters, brain injuries or other disorders that may cause them to get lost and not be able to find their way home. I think it would be worth doing.