Facial Recognition
I’ve been talking about intelligent video applications including license plate recognition. Another specialized application with huge potential across many vertical markets with almost unlimited uses is facial recognition.
Other advanced security measures like biometrics that can recognize various features of the hand are also still on the cutting edge of technology but even they are more widely used more than facial recognition software.
An emerging high-profile application, facial recognition has the potential for greater accuracy than other advanced systems and is easier to integrate into a surveillance system.
Uses
The benefits of facial recognition can be realized in a virutally endless amount of applications across many markets.
Police
Police can receive alerts when people on their watch list are seen in public or specific areas of interest.
Access Control
Facial recognition can be used to grant access to restricted areas only to specific individuals.
Investigations
Search for people in recorded surveillance video. For example, casino’s could identify patrons who have been banned.
Border Patrol
Customs can use facial recognition to more accurately identify individuals of interest and help speed the process of manually checking citizens documentation.
What’s the process?
Similar to license plate recognition but with a major difference. Instead of being able to define what the system should look for beforehand such as specific combinations of letters and numbers, facial recogntion systems need to build a database of faces in order to recognize someone passing by.
Other than that, the process uses these steps:
- Find the person: When not standing still, the first step the system implements is finding the person in the image.
- Face finding: Next, the face is isolated from the rest of the body in the image.
- Identify facial features: Typical parts of the face (eyes, nose, mouth, chin), skin/hair color) are identified and a unique pattern of the person’s face is constructed.
- Matching: The last step is matching the facial pattern with those in the database to identify an actual person.
Challenges
The obstacles facing an efficient facial recognition system are substantial but the rewards can be great. Even perfect lighting and weather conditions can’t overcome the challenge of getting a good look at someone. That is, people are always moving around and will rarely look directly into a camera. Still, 3D facial recogntion systems are being introduced and the overall technology is continually being improved upon.
A reliable system may still be years away but the technology exists today. The Panasonic i-PRO line of SmartHD IP cameras and network video recorders offer a limited face matching technology. This Panasonic facial recognition system only works when the IP camera and NVR are paired together. Product examples that offer this are:
- Panasonic WJ-NV200 Network Video Recorder: Real Time Face Matching function by matching registered face images with a face displayed on live images.
- Panasonic WV-SW395 PTZ Dome IP Camera: Face Super Dynamic ensures clear images of faces, and a face-detection function detects the position of human faces and sends the information by XML or video stream.
- Panasonic WV-SC385 PTZ Dome IP Camera: Face Super Dynamic ensures clear images of faces, and a face-detection function detects the position of human faces and sends the information by XML or video stream.
1 Comment
I would like to state that face recognition technology is no more limited to only such high profile places. It can be easily seen in day to day life, as well. Today, any company having hundred or more employees use face recognition time attendance, access control, and visitor management in India.