Page 127 - Security Today, September 2018
P. 127
intervention, it’s able to extract as many features from the target as possible, including abstract nose, the cleft in his chin, and his oddly shaped ears.) The more features there are, the more accurate the can bring include achieving comparable or even better-than-human pattern recognition accuracy, strong anti-interference capabilities, and the ability
of the time.
and developing algorithms that can perform successful deep learning facial recognition.
capabilities for data storage behind this solution,
and recorders, so no PC server is needed,” reports
support, all human faces are supported.”
This type of high-level facial recognition is able to improve security and investigation scenarios, of
ScanSource Networking and Security Focus FALL 2018 | 14
