AI Case Study

Deep Sentinel identifies neighbourhood security threats using machine vision that can detect potential intruders and other events

Using computer vision and machine learning, Deep Sentinel wants to make security systems intelligent. Next gen cameras will be constantly monitoring the neighbourhood to learn about normal happenings and familiar faces in the neighbourhood. Activities are classified into threat or normal and law enforcement is alerted as fit.

Industry

Public And Social Sector

Security

Project Overview

"Deep Sentinel hopes to not only alter you to threats, but also to issue some kind of deterrent to scare away bad actors. That could be anything from turning on automatic sprinklers to sending in a remote piloted drone.

Deep Sentinel engineers have equipped the security Hub with AI which can interpret incoming alerts, and filter out ones which are inconsequential. For example, when a squirrel decides to run along the front porch, or the leaves start falling from the trees — your Deep Sentinel system will stay quiet.

If, however, the system detects behavior associated with criminal activity, an alert will be sent immediately to the Sentinels and to the customer. This means your alerts truly mean something, and you have little risk of tuning them out. This also means that law enforcement takes your alert seriously, and sends real time response.

The company will be using the technology the auto industry will employ for self driving cars and social media companies use to identify people and objects inside of photos."

Reported Results

Research; Results not yet available. However, the company claims 99% accuracy in detecting threats.

Technology

Function

R And D

Product Development

Background

Deep Sentinal wants to improve traditional security monitoring systems by adding artificial intelligence.

Benefits

Data