AI Smart is the wide and diverse constellation of apps available to monitor, supervise, and characterise events and the behaviour of people or vehicles.
The apps use the most modern and sophisticated scientific methodologies of deep learning and have as their primary task that of generating an event associated with a situation of interest (e.g. a vehicle driving on the wrong side of the road) with all the data of interest for perfectly characterising the event itself (in the previous example the number plate, the speed, the colour, the make and model of the car, the class it belongs to, the lane used, etc.).
The wide variety of available plugins is grouped into the following sections:

AI-Smart Retail

AI-Smart Retail includes all the plugins needed to monitor your point of sale, in the most universal sense of the term; it includes counting people at a gate, assessing the crowd occupying an area, the time density map as well as biometric profiling of people in the area.

AI-Smart Surveillance

AI-Smart Surveillance includes all plugins required for indoor and outdoor surveillance and security; it includes selective intrusion detection, it may be activated and/or excluded for single or multiple categories (person, animal, vehicle), detection of anomalous behaviour (abandoned luggage, loitering, falling to the ground, onset of panic) and alarm situations for fire and/or smoke. The platform also includes a plugin for monitoring ATMs in order to prevent attacks preceded by anomalous behaviour of the attackers.

AI-Smart Transportation

AI-Smart Transportation includes all the plug-ins needed in a modern smart city; it features intelligent traffic monitoring through 3D detection of pedestrians and vehicles with their complete identification by reading number plates, classifying them into categories, identifying the model and make and characterising their motion by estimating their speed and the lane they occupy. Detection of anomalous situations, such as accidents, congestions, traffic violations, pedestrian gatherings further enhance the solution.

AI-Smart Parking

AI-Smart Parking includes all the plugins needed to manage parking areas, both bordered and not. It includes real-time identification of available and occupied parking lots, and for the latter it collects information on the parked vehicle, incorporating number plate recognition, vehicle classification and counting of people who have left and/or entered each vehicle. It allows the detection of anomalous behaviour of pedestrians wandering in the car park with unclear intentions.

AI-Smart Total

If you don’t know what to choose and need all the apps, we have the solution for you. It is a special licence, AI-SMART TOTAL, which does not activate a single app, but gives you the possibility to activate all A.I. Tech video analysis apps for a single stream captured by a camera. For example, you can decide to use this special licence to activate during the day, on the cameras inside your shop AI-RETAIL-DEEP to count people and estimate crowds, and at night on the cameras outside your shop AI-INTRUSION, to detect intrusions. There are two versions of this licence, depending on which app you need to activate. In particular, with the AI-SMART TOTAL licence you can activate AI-People, AI-Crowd, AI-Heat, AI-Occupancy, AI-Heat+, AI-Intrusion, AI-Lost, AI-FaceDetect, AI-ATM, AI-Spill, AI-Masking, while with the AI-SMART TOTAL-PRO licence you can activate AI-Retail, AI-Retail-Deep, AI-People, AI-Crowd, AI-Crowd-Deep, AI-Heat, AI-Occupancy, AI-Bio, AI-Heat+, AI-Security, AI-Intrusion, AI-Lost, AI-Fire, AI-Smoke, AI-Fire+, AI-FaceDetect, AI-ATM, AI-Spill, AI-Parking, AI-Road3D, AI-Incident, AI-Road3D-Deep, AI-Incident-Deep, AI-LPR, AI-Violation.

Funzionalità comuni alle app

Regardless of the specific purpose of apps and the specific application areas in which they can be used, let’s discover some of the common functionalities.

Tampering

If your camera does not have this functionality, don’t worry. All apps can generate events when tampering events occur.

Masking

In all the apps masking of moving objects can be activated, in order to be compliant with all kinds of privacy requirements.

PTZ Handover

Do you need to control a PTZ camera when an event occurs? With the ability to trigger a sequence of HTTP calls, you can bring the camera to the desired pre-set, start the PTZ tracking (if available on the camera) and bring the PTZ camera back to the home position after a configurable amount of time.

AI-Dash-Embedded

This is the app’s built-in dashboard, which allows you to have the data for that specific camera at your fingertips at all times (even on camera, if you have an SD card).

Attivazione su base temporale

Apps can be activated on a time basis, only in specific time intervals, e.g. from 9 p.m. to 8 a.m. . In addition, for server or embedded apps, it is also possible to specify the day (e.g. Monday from 10 a.m. to 12 p.m., Tuesday from 6 p.m. to 10 p.m.).

Attivazione dinamica via software

In addition to time-scheduling, you can activate your apps dynamically, through software calls, via HTTP GET requests that allow you to operate in two modes: impulsive and level-based. In the first case, you have a single START request and the app will remain active for a certain predefined amount of time. In the second case, you have one START request and one STOP request to control the app.

Vantaggi

Quali sono i vantaggi nello scegliere i prodotti A.I. Tech? Scopriamone alcuni.

Accuratezza ed affidabilità

L’esperienza decennale sul mercato dell’analisi video dell’azienda, combinata con la trentennale esperienza del team di A.I. Tech nel settore dell’intelligenza artificiale e delle più recenti metodologie del deep learning, consentono di progettare e realizzare avanzati algoritmi, che risultano estremamente accurati e affidabili.

Notifiche su canali multipli

Integrare gli eventi generati dai prodotti di analisi video con sistemi di terze parti è fondamentale; per questo motivo, le app di A.I. Tech supportano nativamente oltre 20 meccanismi di notifiche verso sistemi di terze parti, che includono VMS di mercato, o protocolli standard (quali HTTP, MODBUS o MQTT).

Open ed integrabile

Grazie alla disponibilità di API pubbliche e gratuitamente disponibili, gli eventi generati dalle app di A.I. Tech possono essere integrate in qualsiasi sistema esterno.

Efficienza

L’ingegnerizzazione e l’ottimizzazione degli algoritmi consente di elaborare in parallelo un elevato numero di flussi video, oltre che di attivare anche più applicazioni in contemporanea direttamente a bordo camera. Infatti, nonostante le app siano basate su reti neurali profonde, queste sono realizzate dal team di A.I. Tech, ottimizzate per mantenere le prestazioni molto vicine a quelle delle migliori reti profonde (grandi e estremamente onerose da un punto di vista computazionale) ma abbattendo di centinaia di volte le risorse richieste.

Multipiattaforma

Oltre 20 sono le piattaforme supportate dalle app di A.I. Tech. Le app possono infatti essere installate direttamente a bordo camera (edge side), su server o in versione embedded, ossia integrato nella soluzione AI-APPLIANCE. Per scoprire i dettagli ti suggeriamo di consultare il seguente link

GDPR compliance

Tutte le app di A.I. Tech consentono di seguire un approccio privacy by design e by default (art. 25 GDPR), oltre che garantire l’adozione di tecniche di minimizzazione, anonimizzazione e pseudoanonimizzazione dei dati. La short declaration di A.I. Tech è disponibile al seguente link.

Semplice da configurare

La configurazione delle app di A.I. Tech può essere effettuata attraverso un cruscotto web, che rende la configurazione una operazione semplice e intuitiva. Sono inoltre disponibili on line, accessibili gratuitamente e in ogni momento, i manuali utenti e i corsi di formazione, che consentono di diventare in poco tempo esperti nell’utilizzo delle app di analisi video di A.I. Tech.

Supporto alla progettazione

A.I. Tech non abbandona i propri clienti e i propri partner in nessuna fase della proposizione. Grazie all’impiego del tool AI-DESIGNER, è possibile progettare un impianto di analisi video, determinando in modo automatico la distanza massima di rilevamento per gli oggetti di interesse di una specifica applicazione, in pochi semplici passi.

Menu