CCTV Analytics

Client: European retail business​

Technology​

Facial Recognition, Queue Detector, People Counter, Activity Visualizer and Analytics

PostgreSQL, ASP.NET Web Services, Web API, IIS service for hosting the Web Application and Web API, Windows service for data synchronization Web Application, React JS

Services

Dashboard development

Challenges

  1. High inefficiency in defining customer clusters by gender and age breakdowns resulting in slow follow-ups across departments and locations due to variability in operations.
  2. Difficult implementation of immediate response programs.
  3. Lack of understanding of customer behavior.
  4. Multiple delays in feature approval leading to loss of potential business.

Solution​​

Using stream cameras and video analytics servers (which include Facial Recognition, Queue Detector, People Counter, Activity Visualizer and Analytics system package for retail), rinf.tech has developed this solution to map each client’s needs and solve even the most demanding requests.

The rinf.tech proposed architecture is the following:

  • PostgreSQL – fully managed database
  • ASP.NET Web Services and Web API
  • IIS service for hosting the Web Application and Web API


Furthermore, we employed:

  • Windows service for data synchronization Web Application – React JS frontend developed user interface application, in which the users perform actions like setting the store, viewing and comparing data on various criteria.
  • Third party software for face recognition analysis, queue and people count data.
  • Third party cameras with face recognition, people count and heatmap processing functionalities.


The main features include: people counting, gender and age breakdown, in depth visitor analysis, interval comparison & selection, direction and heatmap analysis, timeline comparison, queue time and waiting number of people.

  • cameras with people counting capabilities
  • face recognition systems
  • third party systems with push notifications events

Methodology

Problem definition

Rinf.tech assembled a team of technology experts to understand the problem areas to be addressed to derive proper analytics.

Testing the focus group

Based on the requirements & solutions discussed, mockups with single process flows were created. The mockups helped the client in visualizing the system even before it was built. This helped those who were unaware of the workings of IT systems to give feedback & suggestions for further improvement.

Designing the Solution
The solution architect created and introduced a configurable dashboard based on defined rules and conditions. This dashboard was scalable to meet the client’s ever-evolving reporting needs.
Application Launch
After the client’s sign-off on the dashboard, the analytics solutions were launched and tested for effectiveness and efficiency at client sites with a pilot run in a supervised group of participants.

Results

The retail analytics application helps our client:

  • 90% Increase in customer insights – Get better insight about the customers that enter your store with gender and age breakdown
  • 50% Increase of the immediate response rate Improve your service by analyzing queue data and act immediately
  • 60% Rise in accuracy of customer patterns by using direction and heatmap analysis
  • New feature: sends signals to the entrances of buildings, allowing each person In based on number of people Inside.
0 K
Optimized lines of code
0 K
Hours of work
(development, QA, design)
0 K
Targeted location

Other Similar Projects

Want to work together?

Let’s talk.