The service consists in processing images (or video) provided by the client in terms of automatic object detection and statistical summary of the detection results.
The classes of detected objects are determined depending on the scenario previously developed together with the client and the available technical possibilities. The performed classification allows for the subsequent collective statistical analysis of the detection in order to e.g. search for patterns, anomalies, etc.
The service uses modern technologies, based on convolutional neural networks (implemented by industry-leading machine learning frameworks, such as PyTorch or Tensorflow), additionally supported by proprietary image processing methods. In addition, the service has
access to large computing power provided by PSNC, which allows to process large volumes of data in a relatively short time.
The result of the service is a detailed report containing predefined statistics and indicators along with their interpretation. An additional generated product can be visualizations in the form of charts or videos with synchronized statistical parameters, adjusted to the client's requirements.