Facial Recognition Program

The Montgomery County Police Department is utilizing a new computer "Facial Recognition" program to asist detectives in the identification of wanted suspects.  Recently, Major Crimes Division Robbery detectives were able to make seven arrests for a home invasion robbery that occurred on January 9, 2009, based on the identification of a photograph through this unique computer program.  This program was funded by the Department of Homeland Security's Urban Area Security Initiative which supports local and state homeland security and anti-terrorism efforts.  Because of the this federal initiative, facial recognition investigative capabilities were implemented in the National Capital Region in 2007.  They now assist investigators in Maryland, Northern Virginia, and the District of Columbia.

The Facial Recognition software is based on the ability to recognize a face and then measure the various features of the face.  Every face has numerous, ditstinguishable landmarks, the different peaks and valleys that make up facial features.  These landmarks are known as "nodal points".  Each human face has approximately 80 nodal points.  Some of those measured by the software include:  distance between the eyes, width of the nose, depth of the eye sockets, shape of the cheekbones, and length of the jaw line.

These nodal points are measured creating a numerical code, called a "faceprint" representing the face in the database.  To be effective and accurate, the image captured needs to be of a face that was looking almost directly at the camera, with little variance of light or facial expression from the image in the database.  Montgomery County Police Captain Mitch Cunningham, Director of the Information Support and Analysis Division, is strongly encouraging all local businesses that utilize surveillance video to lower their camera positions in order to capture images of suspects engaging in criminal activity.  Captain Cunningham said, "How these systems are deployed by users is of concern to police since they have so much investigative potential.  If they are poorly installed it can severly limit their value."  For best practices for users of video surveillance, please download the Video Surveillance Survey to see if your video cameras are compliant.  Community Services Officers in all Districts are available to do video surveillance reviews and to offer the best advice for surveillance deployement.

This program was recently used successfully in the investigation of a home invasion robbery and stabbing that occurred on January 9, 2009, at approximately 9:01 p.m., in the 26400 block of Johnson Drive in Damascus.  

Through the course of that investigation it was determined that a group of suspects drove to the Damascus address to assault two victims in retaliation after a previous confrontation.  The suspects knocked at the victim's door.  When the door opened, the suspects forced their way in and one of the suspects began stabbing a 24-yearo-old male victim who answered the door.  A second victim, a 21-year-old male, heard the assault and tried to help.  The second victim was stabbed and beaten by the group.  He managed to escape and run to a neighbor's house.  A third victim, a 20-year-old female, tried to interrupt the assault and was beaten by one suspect.

Detectives were able to obtain a photograph of one of the suspects (not from surveillance video).  Using the new Facial Recognition program, they obtained a 100 percent match of the photo to one in the database.  The identification of that suspect, then led to the identification of six more suspects in the case.