Deep learning algorithms can be used to develop completely new fields of application. We work closely with our customers to develop the right solution for every requirement.
Durch Deep-Learning Lernalgorithmen können ganz neue Anwendungsfelder realisiert werden. Wir bieten unseren Kunden die passende Lösung für jede Anforderung.
AI functionalities are already included in most products - at no extra charge. This applies to cameras, recorders and video management systems (VMS). AI enables a drastic reduction in false alarms and storage requirements, the rapid search for specific events and the identification of specific objects or events (e.g. vehicles, license plates, people).
When using AI and Deep Learning algorithms, there are many different things to consider. It is very important to understand how the technology works in order to be able to realistically assess the possibilities and limits of the technology. Many things are now possible, but many are not.
Metadata is feature information extracted from a target object. In the security industry, there are currently three main types of metadata: human faces, human bodies, and vehicle metadata.
ANPR (Automatic Number Plate Recognition) is a technology for optical character recognition on images for reading license plates. ANPR applications include toll collection, traffic monitoring and safety, speed and travel time measurement, parking and access control, and more.
The number of people entering/leaving/passing certain areas within certain time periods can be counted - unimportant objects can be filtered out (shopping carts, animals, etc.)
The face detection function can be used to determine whether faces are present in the specified face image or video. If faces are present, the position and size of each face and the position of the main features of each face are specified. Using this information, the identity features of each face are then extracted and the human face is modeled. Each face model is compared to the face models stored in the known faces database to identify individual faces.
Monitoring systems that trigger false alarms in extreme weather conditions, e.g. leaves moving in the wind or moving animals, are not only annoying but also counterproductive for a company's operations. Dahua's Smart Motion Detection (SMD) is an ideal solution to reduce false alarms.
Dahua's intelligent search technologies (Quick- and Acu-Pick) can very rapidly identify people/vehicle targets of interest. While Quick only performs a purely visual comparison, Acu-Pick also uses video metadata to identify objects.
Compared to manual verification of SMD events, the intelligent search saves considerable labor costs and increases the accuracy of the results.
