What Is Facial Recognition

 Facial recognition systems are bio-metric technologies capable of identifying or verifying a person from a digital image or a video frame from a video source. Facial recognition can identify a person by analyzing patterns based on the person’s facial textures and shape.

Facial recognition is said to be started in the 60s when Woodrow Wilson Bledsoe, one of the founders of artificial intelligence, manually classified faces by hand using a device (RAND tablet) to input horizontal and vertical coordinates.  The metrics then were placed in a database and when a new photograph was given to the system it could retrieve the image from the database matching the individual. Nowadays facial recognition solutions have already become a part of our everyday routine. This technology is used both for personal security (e.g,. in mobile devices or ATM machines) and national security (e.g., in the airports, music fests and many other purposes). 

How does it all work? Facial recognition software isolates a face from the rest of the background in the image or video. The software first recognizes the face, then measures different facial features. It recognizes these features as nodal points. A human face consists of 80 nodal points. Historically facial recognition software relied on a 2D image to identify or verify another 2D image from the database.  However, today algorithms are much more powerful and uses a 3D model, which is more reliable.

What is the future of facial biometrics? The US government  market  —  which  includes  federal,  state  and  local  law enforcement — is expected to soar from $136.9 million in 2018 to $375 million by 2025, according to an estimate by market research firm Grand View Research. If we take the global market, the figures are even more impressive – Facial Recognition Market Report, published by Allied Market Research, forecasts that the global market is expected to garner $9.6 billion by 2022, registering a CAGR of 21.3% during the period 2016-2022. 

REAL CASES

Facial recognition technology has already played a significant role in solving and preventing crime. Let’s look at some cases.

Generally, in the US facial recognition is already in heavy use. The Georgetown report stated facial recognition has helped the NYPD crack about 2,900 cases in more than five years of using the technology. The  New  York  Police  Department  has  become increasingly vocal regarding the appropriate use of this technology. Trained facial examiners review hundreds of candidate photos when attempting to identify an individual, corroborating video or photo evidence with other photos of the potential suspect sourced   from   previous arrests,  social   media   posts,  or  other   databases   available to   the investigators. 

In Europe facial biometrics use is more rigorously controlled. However in 2016, the “man in the hat” responsible for the Brussels terror attacks was identified thanks to FBI facial recognition software.

Moreover, the South Wales Police implemented it at the UEFA Champions League Final in 2017. Facial recognition will be a major topic for the 2020 Olympic Games in Tokyo, Japan.

Facial recognition technology and an information-sharing program have also proved instrumental in helping police in China to identify and track missing children. This powerful technology has already proven to be effective all over the globe.

FACIAL RECOGNITION TECHNOLOGY IN OXYGEN FORENSIC® SOFTWARE

In July 2019 Oxygen Forensic®’s JetEngine module, built into Oxygen Forensic Detective offered the innovative ability for investigators to categorize human faces using built in facial recognition technology. The facial recognition component is available in the Faces section to all users at no additional charge. The unique features include:

  • Industry leading accuracy (as measured by NIST)
  • Detailed face analytics (gender, race, emotion, and more)
  • Immediate categorization and matching (5 faces/second)
  • Support for massive volumes of data

Using the built-in facial recognition technology investigators will spend less time looking through thousands of photos or videos in a single mobile, cloud or drone extraction and now better equipped to analyze the aggregated visual data from hundreds of devices!

Here are some possible use cases:

  • Identification of known individuals from images captured from mobile devices and cloud services in investigative cases.  Clustering or grouping all faces in similarity “buckets” for easy identification of all images/videos with the subject(s).
  • Assistance in locating endangered children, human trafficking, or other crimes by searching across all the images/videos in cases within Oxygen Forensic® JetEngine showing grouping of known images uploaded and those with high confidence levels of being the same person(s).
  • Drones capture images and videos.  This data can be extracted and analyzed which can assist in theater to identify possible known terrorists.
Image of faces section
Faces section in Oxygen Forensic® JetEngine