4 steps to achieving privacy and security goals for CCTV based crowd analysis

This short one minute read explains how DCM uses 4 steps to achieve privacy and security goals for CCTV based crowd analysis.

1. Architect from the ground-up to meet privacy/security needs and regulatory requirements
2. Avoid identification of individuals
3. Utilise in-line video and image processing to avoid storage
4. Leverage world class infrastructure
   
   

1. Architect from the ground-up to meet privacy/security needs and regulatory requirements
   Creating and maintaining a strict end-end logical separation of all CCTV video streams and derived crowd metrics allows the costs benefits of shared computing and networking infrastructure to be leveraged, whilst enforcing tight security and privacy requirements. This is achieved by enforcing installation-based access controls at every point in the data pipeline and utilising a secure architectural model.
   
   
2. Avoid identification of individuals
   The ability to compute insightful people metrics without identifying individuals is critical. To achieve this whilst maintaining high levels of accuracy, individuals can be uniquely catalogued for only the duration of computation by using an alphanumeric tag. Furthermore, these tags are unique to each view and cannot be used to track individuals across a site or different locations.
   
   
3. Utilise in-line video and image processing to avoid storage
   Utilising in-line processing of live video streams will negate the need to store imagery at any point before, during or after processing. However, to maintain the same level of privacy and security for historical analysis, video streams and sample images can be secured in long term storage maintained using the same logical separation employed during processing.
   Furthermore, storage should be encrypted with a customer unique key. Logical separation and encryption will ensure the archival material is only visible to and usable by the customer who owns that material.
   
   
4. Leverage world class infrastructure
   Using a provider with security as their number one priority is critical. AWS has developed powerful and flexible logical security controls that create strong isolation boundaries between customers.
   Customers leveraging AWS benefit from architecture built to satisfy the requirements of the most security-sensitive organisations in the world. The AWS approach to privacy and security supports the conclusion that the strict logical isolation between customer environments can be more effective and reliable than security seen in dedicated physical infrastructure.
   
   Dynamic Crowd Measurement leverages best practice for cloud privacy and security equal to, or superior to on-premises solutions, whilst providing significant advantages in availability, scalability, and lower cost.