Slashdot Mirror
Police To Begin iPhone Iris Scans
cultiv8 writes "Dozens of police departments nationwide are gearing up to use a tech company's already controversial iris- and facial-scanning device that slides over an iPhone and helps identify a person or track criminal suspects. The smartphone-based scanner, named Mobile Offender Recognition and Information System, or MORIS, is made by BI2 Technologies in Plymouth, Massachusetts, and can be deployed by officers out on the beat or back at the station. An iris scan, which detects unique patterns in a person's eyes, can reduce to seconds the time it takes to identify a suspect in custody. This technique also is significantly more accurate than results from other fingerprinting technology long in use by police, BI2 says. When attached to an iPhone, MORIS can photograph a person's face and run the image through software that hunts for a match in a BI2-managed database of U.S. criminal records. Each unit costs about $3,000."
Not it will not. Portable DNA matching? You have been watching a little too much CSI my friend. It still takes a lab full of equipment, consumables and trained professionals to create a DNA profile and compare it to a sample or database. Even then, it's not nearly as accurate as you've been lied to believe. See, they don't actually sequence your DNA that would take too long, so they only do a profile. Which, while it might match a sample from the crime scene, it does not positively identify any one person. Only a class of people.
For instance, if the sample from the crime scene came from a white male of non-jewish decent, then that sample profile will match something like 15% of all white males of non-jewish decent. Even more, it will match all the males in a particular blood line. (depending on sample).
I think you are referring to Kyllo v United States which ruled exactly the opposite of what you have stated. The court concluded that using infrared cameras to scan homes for leaking heat is a search and thus requires a warrant under the fourth amendment. The basis for the court's opinion was very similar to the grandparent post.
The ruling did indeed involve Thomas who joined the majority opinion in a 5-4 decision. Quite frankly I would consider any court that reverses the ruling to be less sane.
Mod parent up.
DNA testing is far more useful to prove exclusion with great confidence, because the 'match' size is only one to one. But - that does NOT mean we can extend this methodology to prove a positive match against a database with thousands of random entries. If a sample does not match the suspect, then it can generally be shown that the sample did NOT originate from the suspect - the inverse however, is not necessarily true.
When samples do match to some degree, it can only be shown that there is some likelihood that a sample came from the suspect - a likelihood that is completely dependent on the search criteria and database size when matching the sample to a list of 'suspects'. The bigger the database, the less reliable the result.
For example, if a there are three unrelated people in a room, and I take a sample from each one of them, then a lab could determine with extreme confidence which sample goes with which person. If I do the same with 10000 random people in the room, the probability of correctly identifying a given sample falls dramatically - in fact, it is likely that the sample will 'match' hundreds of people.
As can be seen from the above examples, Gataca / [insert favourite TV crime show] style DNA matching is still far from realistic with current technology.