Stabilize the detection range of sensor to 1.5m (from the wall) and 20m (along the wall) ensuring its right working under any environmental condition.
Increase accuracy and efficiency of video-tracking system (being able to identify people in complex situation (among group of people) and bad illumination conditions (night, rain, snow, etc.).
Optimize of monitoring software to be used with a large number of sensors and cameras.
To ensure the desired range detection and a huge resilience in the system, the current hardware has to be redesigned in order to improve the performance of the sensor and the communication system.
Both the detection algorithm and tracking intruder algorithm must work under any environment situation in any part of the perimeter. For this reason, these algorithms have to be trained to consider different approaching to the fence under distinct environmental conditions. This training will require models that allow the algorithms to recognize boundary conditions and operate accordingly.
Due to the critical nature of this solution, it is necessary to ensure the proper operation of the system and its comfortable management. For this reason, in this task a cloud communications infrastructure will be defined, as well as, a friendly human machine interface (HMI) for a large number of cameras will be developed.
In order to ensure the successful in WP2, before to deploy the system in Ontech perimeter, this will be tested in a scaled indoor perimeter (60 meters) where the different influential parameters are known and where the approaches to the perimeter are controlled. The objective is to validate functionally the redesign developed.