Video cameras and other types of recorders are not always very effective. In some cases, it’s not enough to have the entire crime recorded in order to accuse the suspect in court. It is not even enough to identify the criminal in cause. There are many problems that can occur during the identification process, among which the clarity of the video recording, the distance from the subject or the possibility that he was disguised.
Important law enforcement agencies have a strong database of tattoos and distinctive marks after which the tracing and finding the suspect is being made. Precisely because of the dimensions of the database there can be further problems consisting in matching a search key word to a tattoo. While this operation seems to be quite simple at a first sight, the entire process of finding a matching tattoo to the description is, in fact, a provoking and daunting investigation in itself. The text-based searching engines complicate the operation even more, whenever the symbol of the tattoo is uncommon. What could go perfectly for a lion, a rose or a gun is a possibly lost job for an encrypted specific symbol.
There is, however, a newer program whose concept belongs to has found a solution to match the image on the recording with the possible correspondences on the database. The program is very efficient because it works by identifying essential common points of the two compared images, called key-points. The name of the program is TattooID and the technique of the key-points is usually infallible, the same used in the fingerprints examination software programs. The main advantage of the program is the elimination of the human factor and the lack of subjectivity of the software in analyzing and matching, which has proven to be four times more effective than the classical word based searching engine.
Such a much needed breakthrough is, in the first place, a great complementary platform for the situations in which face recognition is impossible, as well as effective. Another characteristic of the program is that it could even help the ongoing researches and investigations by managing to match and put together the distinctive marks that a band or gang have tattooed as signature, belonging and trust sign. Tattoos are one of the most common signs to designate membership inside a mafia organization.
The engine could also help the police by identifying false identities based on the color and pixel recognition through the biometric software. , a biometric tech company has licensed the software, which also provides face-matching and fingerprints recognition along with this new biometric characteristics.
Starting with 2014 the Federal Bureau of Investigation is going to use the Next Generation Identification (NGI) automating the process of identifying or eliminating suspects by biometric software. The Biometric Center of Excellence (BCOE) will be able to operate the program of image matching with or without any descriptive data, while if the descriptive information is irrelevant the software is going to ignore it in the matching process. Applying the research with the image-based matching technology, the database is also going to determine the symbols and other cryptic information as it is going to suggest possible meanings of their association.
Victim identification is as important as suspect identification through image matching technologies, especially given the fact that many of the bodies decompose quickly, in adverse conditions, while identifying through tattoos remains the only viable option because of the depth of the ink layer in the skin.
