Facial recognition: Finding faces in a crowd         

By Dinesh Kandanchatha

Faces and facial expressions are one of the key ways that we identify people as being a threat or not a threat. From infancy we learn the expressions for anger, aggression, fear, nervousness, etc. We also learn the faces of people who are not a threat to us, like our parents and close friends. Understanding the intent and relationship of the people around us is critical to our security and the security of our families.

Facial recognition software is based on the premise that a computer can compare image (infrared, 3d or photograph) with an individual passing in front of a sensor and make the determination of threat or non-threat, based upon its database and its classification algorithm.

Facial recognition systems are divided into the following types:

Geometric/statistical recognition– This type of system uses algorithms to analyze the relative position, size, and/or shape of the eyes, nose, cheekbones, and jaw relative to each other. These facial features and expressions are then used to search for other images with matching features or feature combinations.  Other algorithms normalize a gallery of face images and then compress the face data, only saving the data in the image that is useful for face recognition. Most of these types of facial recognition systems try as much as possible to use template matching either through geometric or statistical matching.

3D-Matching– These systems use 3D sensors to capture information about the shape of a face. This information then is used to try to identify the distinctive features on the surface of a face, such as the contour of the eye sockets, nose, and chin. These 3D sensors work by projecting structured light onto the face across the light spectrum. The result is a much more resilient algorithm that is able to handle changes in lighting conditions, movement of the subject.

Thermal– These type of facial recognition systems try to detect the shape of the head through its heat signature, ignoring features such as  hats, glasses, makeup , etc. that  do not emit a heat signature. The challenge with thermal facial recognition systems is that they are affected by bright sunlight, and environmental conditions.

Facial recognition has been experimented for security applications since the 1960’s and continues to be an active area of experimentation and product development. As with many artificial intelligence tasks, the simplest are the hardest. We are just getting the technology to the competency of a baby… so there is still a lot of work to do before we can consistently find a face in a crowd.

Author: Dinesh Kandanchatha, Regular Contributor and President/CTO, Patriot One Technologies

Many of the articles within the media pages of the patriot1tech.com website are 3rd party in origin and have been included for informative purposes only. Decisions to include articles are solely based on the timely nature of the storyline as it applies to the security industry in general and to the proliferation of threats to public safety in particular. The inclusion of these articles does not imply that PATRIOT ONE its management, agents or employees endorses any statements expressed. The public is advised to fully investigate any contentious claims or assertions prior to arriving at any conclusions. Any hyperlinks included in these articles does not imply that PATRIOT ONE monitors or endorses these websites. Accordingly, PATRIOT ONE accepts no responsibility for such websites. Additional information regarding exclusions and liability limitations are outlined here.