Page 106 - Security Today, March 2018
P. 106
THE REDEMPTION
OF VIDEO ANALYTICS
VIA DEEP LEARNING TECHNOLOGY
Simply say the words “video analytics,” and you trigger a deep groan that echoes throughout the video surveillance industry. Analytics, it’s safe to say, has suffered the stigma of a poor reputation. They just haven’t performed in real life to the level that end users want, resellers hope for, and Hollywood purports them to do.
One of the biggest challenges with traditional analytics products is that their algorithms are unable to distinguish between objects and behaviors that a human being would have no problem classifying. In the process of identifying human activity, the algorithms would, at times, tag animals or moving shadows or trees waving in the wind. And while not every end user requires an analytics solution to be 100% accurate all of the time, the goal is to be as close to infallible as possible.
1 | ScanSource Networking and Security Focus SPRING 2018
