David Selinger is the founder of Deep Sentinel, a technology company that develops smart home surveillance systems that combine next-gen cameras, artificial intelligence (AI), and human intervention. The surveillance technology founder organically arrived in the home security space when he was motivated to dig deeper after a close neighbor was the victim of a home invasion.

With an extensive background in AI, Selinger came to the quick and startling realization that home security systems were woefully behind the technology curve and he set out to change that. Selinger discusses the state of home security before his company entered the space. He details how literally 25 years of innovation in traditional home security systems has amounted to not much more than wireless sensors, WIFI connectivity, and improved motion detection. Deep Sentinel seeks to fully integrate AI by using algorithms to analyze multiple security camera footage, which will then identify patterns in visual data, flag them, and predict crime before it even happens.

The surveillance technology founder discusses rapid advancements that have been made in security that were bolstered by deep learning technology promulgated by the gaming industry. Selinger discusses how the complex intersection of data, computing power, AI, and mathematics is ushering in a new era of advanced home security. He details how convolutional neural networks are an invaluable tool for machine learning that can greatly improve image recognition and classification.

A convolutional neural network is a class of intensive, feed-forward synthetic neural networks adept with analyzing visual imagery.

Selinger lays out some of his company’s security technology and explains how advanced motion detection that decreases security ‘noise,’ essentially defined as false alarms (dogs, wind, etc.

), allows for better real-time intervention for prevention of crime.

Selinger’s system can push the data and information to their trained security personnel within 10 seconds. And the combination of real-time data, integrated lights, and high decibel speakers which allow for immediate communication with the would-be criminal, often prevents or stops the crime as it is in progress. While perimeter security is the immediate goal, facial recognition technology will be coming in the near future.

Podcast: Play in new window | Download | Embed