Our work

Description

Computer Vision

One of our customers, a CCTV system integration company, requested us to develop a system to count people and measure the flow of people in an underground metro. The purpose of the system is to improve platform load and passenger queues. The video surveillance system had been already deployed there and RTSoft should propose a hardware and software add-on at a very competitive price. 

RTSoft helps customers build AI + IoT applications through integration of computer vision, data analysis and augmented reality.  Our experience in embedded platforms (Intel Movidius, Nvidia Jetson, Google Coral) and dedicated ML frameworks (Openvino, TensorRT, TFLite) made it possible to fulfill this project.

System features:

  • Visitor headcount

  • Crowd headcount

  • Queue detector

  • GDPRA compliance 

Key facts

  • On premises video server
  • AI/ML software: RTSoft OMGE
  • Software recognition accuracy
  • OTA software updates: RTSoft IoT Rollouts

Solution

In the early stages of the project RTSoft built people recognition algorithms and tested them onsite on several CCTV cameras in collaboration with the customer’s technical team to identify a possible set of use cases and draft requirements for the solution. Since the end-solution was supposed to be implemented based on the RTSoft OMGE proprietary platform it was quite a simple task to rapidly design data flows, develop and bring up the system, and then scale it to multiple CCTV cameras. Then RTSoft developed a deployment ready docker with all key components inside:

  • Video frame capture

  • People recognition

  • Graphical user interface 

  • Target reports and statistics

  • OTA software updates.

Finally, the system was successfully installed and passed through performance, load balancing, and acceptance tests and was implemented as a software product by our customer.

Benefits

Technologies

More projects

Skip to content