Lanaccess video surveillance systems with Deep Learning for the development of Smart Cities

Video surveillance systems with Deep Learning (DL) technology will be the key in the future for the development of Smart Cities. But, today, there are already companies and institutions that are benefiting from Lanaccess solutions with its advanced video analytics to solve critical use cases in cities.

Our surveillance systems with Deep Learning artificial intelligence are distinguished by LAVA algorithms.

What do LAVA algorithms offer to Smart Cities?

Lanaccess’ exclusive video analytics, LAVA ANALYTICS , can be applied to perform the following functions:

• Analyze the behavior of people and vehicles to send alarms ( LAVA-PROTECT ) and count events (LAVA-COUNT).
• Control the occupation of squares and public areas ( LAVA-DENSITY ).
• Count bicycles and scooters in specific areas (LAVA-OBJECTS).

Behavior detection analytics are defined based on different rules: line crossing (directional or bidirectional), double line crossing (directional or bidirectional), group crossing (directional or bidirectional), ROI entry, ROI exit, permanence in ROI (loitering) or presence in ROI (occupation).

Rules that can be applied specifically to a particular person or to particular vehicles in their widest sense: cars, vans, trucks, buses and, very soon, motorcycles as well.

Key use cases for the development of Smart Cities

This analytic, in all its variants and with all its rules, can be applied in different situations that are essential for the growth of Smart Cities :

1.- Road safety

The LAVA-PROTECT analytic is responsible for generating an alarm when it detects a vehicle in a place where parking is prohibited, as well as a pedestrian crossing, a bus stop, among others.

 This analytics also sends alerts without human intervention when it detects that a vehicle is at a level crossing, making a prohibited turn or driving in the opposite direction.

2.- Citizen security

In this use case, the LAVA-DENSITY analytics work, which detects crowds and concentrations of unauthorized personnel in a specific space: squares, parks or gardens.

 It can also be in charge of protecting the perimeter of municipal buildings such as museums, schools, libraries, sports centers to prevent vandalism and theft.

 In addition, this set of algorithms is capable of generating alerts when several people are loitering in a specific area.

 3.- Mobility

The LAVA-COUNT analytics counts vehicles at a pre-established location to detect traffic jams and sends an alarm if the defined threshold is exceeded. It also analyzes the fluidity of an area in order to improve the control and planning of mobility in a city.

This same analytic allows to classify the vehicles that circulate in a city according to whether they are motorcycles, passenger cars, vans or trucks to obtain valuable information on the road traffic of a municipality and thus be able to make better decisions.

In addition, it is responsible for counting the bicycles and scooters that circulate in the city, whether or not there is a bike lane, to improve mobility planning by learning about the use of alternative and environmentally friendly transport.

Finally, this set of algorithms provides roundabouts with intelligence so that they provide relevant information about traffic, as well as a classification of the vehicles that access the roundabout and the exit routes.

Lanaccess provides video surveillance solutions with Deep Learning that offer advanced analytics capabilities so that public institutions can develop Smart Cities and improve the quality of their citizens.