Description Field of the art

The present invention regards the field of domestic alarm and monitoring systems. In particular, the system in question is part of systems provided with artificial intelligence.

Prior art

The latest developments of digital technology have allowed machines to communicate with each other thanks to the internet and learn from the experience. This has led to the consolidation of concepts of the IoT (Internet of Things) and artificial intelligence, thus opening avenues to the so-called web 3.0.

The present invention intends to apply this technological progress to the field of the domestic alarms and partly in the home automation industry, as regards the activation of devices for blocking emergencies or scaring away thieves and ill-intentioned people without the intervention of human beings.

Some international patents regard particularly effective alarm devices, which however reveal some drawbacks.“Verisure”, which has been providing optimal anti-theft alarm and anti- intrusion systems for homes and workplaces over the years, is a leader in the industry. The company owns several patents (EP 269 87 74; EP 264 01 10; EP 314 52 11; EP 321 13 79 and others) all regarding the same field.

However, the Applicant identified some shortcomings in the alarm systems currently available in the market, including those of “Verisure”. Firstly, they are designated for defending walls of a protected place and they are designed to be switched on/off by a user. These characteristics make the system vulnerable given that they are disconnected from the recognition of the person who is controlling the system. Secondly, said systems are ineffective when the threat comes from the internal of the protected place, i.e. from a person who has been authorises to enter.

Thus, an object of the present invention is to implement the alarm system used for defending against intrusions by means of an integrated system capable of learning the behaviour of the people living in the environment and sending signals to at least one user to verify whether the abnormal behaviour is to be considered a threat or not.

Description of the invention

Provided according to the present invention is an alarm system based on the biometric recognition of persons, provided with artificial intelligence which effectively solves the aforementioned problems.

Besides being new and innovative, the characteristic of recognising the people who live or spend time in the place under surveillance is also an additional feature to the previously known advantages of alarm systems for private environments.

Thus, the system is capable of recognising foreign subjects and, among the known ones, to distinguish among various categories based on authorisations to access the area placed under surveillance.

Another fundamental and innovative characteristic lies in the so-called“scalability”, i.e. the possibility to implement the system with actuator sensors for the maximum customisation of the surveillance service in question, based on the needs of the user and the characteristics of the space placed under surveillance.

More in detail, in its basic version the system comprises at least one video camera which films the environment to be placed under surveillance, preferably a plurality of video cameras for filming it from various viewpoints and be able to identify the people to the maximum.

The video cameras are connected, preferably in wireless mode, to a control unit provided with double power supply, both to the fixed mains power supply, by means of a cable, and provided with a rechargeable battery so that it can operate even in case of blackout.

Advantageously, said control unit has a connection to a local network for exchanging data with a dedicated server and also an internet connection so as to be able to communicate with the operational centres of the law or security enforcers and with the portable electronic device of the user which manages the system.

The user who manages the system (they can even be more than one manager) can use the dedicated “App” for configuring and managing the system according to the method described hereinafter.

Said system also comprises at least one microphone and a speaker, preferably a plurality of these devices, all connected to said control unit to record the sounds and possibly the conversations too that occur in the area under surveillance and send them to said dedicated server. In case of emergency the speaker will serve as an acoustic signal for the emission of pre-set alarm signals.

Advantageously, microphones and speakers can also be configured to allow the speakerphone communication between the remotely positioned manager user and the persons present in the area under surveillance.

Said dedicated server is connected to a memory in which at least the images and the audio data recorded by the video cameras and by the installed microphones are stored.

Preferably, the saving of the images and the other data will occur only in case of detection of alarming or suspicious situations.

In a version of the present invention, in order to avoid the excessive production of recorded material, given that once configured the system runs 24 hrs a day, the memory will store the data only for the 24 hours subsequent to the recording allowing the display thereof by the manager user. Should the user need to store the data for a longer period of time, for example so as to submit it to law enforcers to testify a theft, the useful segments may be stored in an external memory.

The biometric recognition and behaviour analysis software which operates within the dedicated server, which is the same one shared among all users using the service, represents the core innovation offered by the present patent.

The artificial intelligence with which the software is provided advantageously allows progressive learning during the entire period of operation of the system, so as to make the surveillance service more and more precise and targeted.

After an initial step of installing all hardware components of the system, which, besides the video cameras, control unit, connection to the dedicated server which is remotely positioned, microphones and speakers, may also include other devices described hereinafter there will follow a first small configuration of the system. During said configuration, the user who manages the system downloads and installs the dedicated app on the portable electronic device (smartphone) and carries out the association with the control unit.

A time span in which the system is ON, but not operative, follows; as a matter of fact, it is in a learning period of time lasting between 2 days and 30 days, preferably 15 days. During this learning, the software analyses all data detected by the devices installed in the place to be placed under surveillance to identify who are the users authorised to be present, distinguishing them into categories. Such categories can be, purely by way of example, the house owners, with unlimited authorisations, relatives or friends who are authorised in the presence of at least one owner of the house and strangers or intruders whose presence will automatically set the alarm off.

Besides the biometric data for recognising the various persons, the typical behaviour of each one of them, such as the voice timbre or other optional characteristics (recurrent gestures, walking style...) are also subject of learning.

At the end of the pre-established learning period, in a first verification step, using the“App” on the smartphone thereof, the manager user confirms the correctness of the detected information, or the user can proceed to the step of manually configuring the various detected users and the relative authorisations.

The system is highly customisable in this step. For example, the presence of some users can be limited to some time brackets or at pre-established days of the week or of the month. Some particular types of surveillance may be required for some persons, for example monitoring the behaviour of a caregiver with respect to the elderly, or children left at home alone. Once all the above has been set, the aforementioned system automatically proceeds to carry out the required monitoring through the hardware devices installed in the private environment subject of monitoring

It should be observed that the system subject of the present invention is particularly designated for private homes, but it can also be easily configured, with the same advantages, for offices and other private closed places in which access is limited to a sustainable number of people for the biometric recognition of the software.

In a further verification step, the software analysis all the images and the other data coming from the area under surveillance and, through behaviour analysis it is capable of detecting potentially or certainly hazardous situations.

Upon detecting such situations, a step for signalling to the manager user through said dedicated“App” follows. The manager user may view the images, listen to the sounds or views what is detected by other possible sensors and thus decide whether it is a true or false alarm.

Should it have been a false alarm, the system will continue monitoring.

Otherwise, depending on the type of emergency observed, after recording the response of the manager user, the system will trigger an alarm.

More in detail, the alarms are differentiated based on the type of emergency to be addressed. Thanks to the internet connection means the control unit will determine, depending on the cases, the call to the rescue means (ambulances, fire fighters or the like), the call to the public or private law enforcers (police, carabinieri or a private surveillance firm) and/or the activation of the siren or possible smoke bombs installed previously.

A further version of the present invention has the object of including among the advantages the portability of the system, i.e. the simplicity of displacement and re-installation from one place to another. This is made possible by the fact that the technological devices of the basic version integrated in the technological object. Thus, the video camera will have the control unit, the microphone and the speaker integrated. It will suffice to connect this multipurpose device to the source of electrical power supply and then proceed to the configuration like in the method described above.

Other hardware devices that can be installed in the area to be placed under surveillance include a plurality of sensors for various applications. Even said sensors will be connected, in wireless mode, to the control unit and will send the detected data to the dedicated server so as to be processed by the software and possibly recognise the risky situations.

Some examples of said sensors can be:

- inertial sensors, suitable to recognise impacts or breakage of doors, windows and shutters, glasses or any other possible access broken into in the area under surveillance;

- volumetric sensors, suitable to recognise strangers in the area under surveillance in hours in which their presence is neither expected nor authorised;

- flood sensors, suitable to identify an unusual amount of water in the area under surveillance;

- smoke sensors, suitable to identify the presence of cigarette smoke or the like, fire starting and gas leakage in the area under surveillance.

In order to address possible criticalities detected by said sensors, even a plurality of corresponding actuators may advantageously be installed and connected, in wireless mode, to the control unit. Said actuators potentially including:

- electrical switches for reversibly switching the lights and/or radio and/or televisions ON and OFF, to simulate the presence of people in the environment under surveillance at pre-established times or upon a remote control by the manager user;

- electrical switches for reversibly switching at least one domestic appliance ON and OFF at pre-established times or upon a remote control by the manager user;

- main switch for disconnecting electrical power supply in case of detection of fire or flooding;

- water valves for shutting off water supply in case of detection of flooding;

- opening and closing doors to authorise the entry of some recognised users to check whether the doors are closed when the environment under surveillance is uninhabited. In order to help to recognise users, who could be many in some cases and thus difficult to identify for the system, other technological devices which, by sending data to the server, can contribute towards the recognition of the users and the relative authorisations, can be advantageously installed. Such devices can be, by way of non-limiting example:

- a digital fingerprint reader;

- a retina reader:

- a keypad where the user can enter a pre-established code;

- a magnetic key;

- an NFC reader suitable to recognise the smartphone of the authorised user.

All these systems can be useful for authorising the entry of users even on days and time off the pre-set authorisations, after signalling and approval by the manager user.

Another advantageous characteristic of the present system includes the possibility for the manager user to deactivate the system for a predetermined time interval. In order to avoid the user forgetting such interval, the system will advantageously send a notification on the smartphone when it is about to restart.

The advantages provided by the present invention will be clear in light of the description outlined up to now. Furthermore, they will be more apparent due to the attached figures and relative detailed description. Description of the figures

The invention will be described hereinafter in at least one preferred embodiment, provided by way of non-limiting example, with reference to the attached figures, wherein:

- FIGURE 1 shows a block diagram of the technological components that interact in a particularly complex embodiment of the present invention.

- FIGURE 2 illustrates the flow chart underlying the basic operation of the system subject of recognition and the verification of risky situations in a private environment.

Detailed description of the invention Now, the present invention will be illustrated purely by way of non-limiting or non-binding example, with reference to the figures illustrating some embodiments regarding the present inventive concept.

Let us assume installing the system in question in an apartment housing a couple with children, planning to combine common anti-intrusion systems with monitoring the behaviour of the children and the baby-sitter taking care of the children in the afternoon too.

Both parents will be the managers 400 of the system and will install on the respective smartphones 410 the“App” dedicated for the service.

The manager users 400 will thus install three video cameras 200-201-202, one for each room, a retina reader at the entrance door, inertial sensors 230 at the windows and smoke sensors 230 in proximity of the kitchen and in any room of the house. Actuators 240 which open and close the lock of the house and other actuators 240 which shut off the gas supply in case of leakage will be provided in this installation step 1000. A plurality of microphones 210 and speakers 220 serving as sirens in case of emergency will also be installed, as a basic version of the system.

Lastly, the control unit 100 will be installed, through the connection thereof to the power mains by means of an electrical cable 130 and the network connections will be configured 1100. In particular, the configuration 1100 of the connection means 110 to the local network will allow to establish data exchange with a dedicated server 300 and with all the other aforementioned devices (microphone 210, speaker 220, video cameras 200-201-202, sensors 230, actuators 240) and the means 140 for connection to the internet network 500 will allow to configure the connections with a pre-established private surveillance firm 600 and with the portable electronic devices 410 of the manager users 400.

Next follows the learning step 1200, in which in 15 days the biometric recognition and behaviour analysis software 310 which operates in the server 300 analyses the images filmed by the video cameras 200-201-202 and the data detected by the various microphones 210 to recognise the somatic traits of the people living in the house, associating the voice timbre to each one of them and storing each person in the appropriate category in the memory 320. At the end of the 15 -day learning step 1200 the system carries out a first verification 1300 in which the manager users 400 may confirm the correctness of the detected information. In case of errors in the allocation of each person to a category of users or failure to recognise one or more persons, the managers 400 will carry out a step 1350 for the manual configuration of the various users and the relative authorisations to access the environment subject of surveillance.

More in detail, the two manager users 400, i.e. the parents, have unlimited access to the house, with entry authorised at any time and at any day. Children will also be equally granted unlimited access to the house, but upon entering the house with the parents absent a notification will be sent to the smartphones of the parents 400.

The babysitter is instead authorised in specific time brackets, outside of which access will not be permitted. In case of absence in said time brackets instead, the fact will be notified to the parents.

Thus, the system will be activated in a monitoring step 1400 which will occur 24/7 to check any risks and continue improving the knowledge of the software 310 as concerns the detected people. During the absence of the two manager users 400 each unusual detection perceived by the sensors will be notified through the App to the respective electronic devices 410.

Thus, in said monitoring step 1400 all images filmed by the video cameras 200-201-202 and all other data detected by the microphones 210 and sensors 230 are advantageously stored in the memory 320 of the server 300 and made accessible for 24 hours after which they will be destroyed.

In a second verification step 1500 the software 310 processes the data, carrying out the analysis regarding the behaviour and risks substantially in real time.

In case of detection of a potential risky or emergency of any kind, the control unit 100, through the means 140 for connection to the internet 500 with which it is already provided, sends a signal to the manager users 400 by means of the respective portable electronic devices 410. Let us assume that the smoke sensor 230 detects an unusual presence of smoke in one of the rooms of the house. Through the smartphone 410 one of the managers 400 sees that the source of smoke is a cigarette lit by the babysitter in the balcony whose smoke was partly detected by a sensor 230 in the house.

Through a procedure to be carried out by the“App”, the manager 400 reports that this is a false alarm; the system records the response 1700 and continues monitoring 400 without further consequences.

In another case, the system records that the babysitter has moved away from the environment subject of surveillance and simultaneously a glass of the window has been broken. The presence of a stranger in the house is also detected.

The manager user will send the response 1700 which signals an emergency situation arising from an intruder in the house.

The alarm 1800 of the system will then go off and automatically activate the alarm signals 1830 or a siren emitted by said speakers 220 and through a call 1820 to the law enforcers 600 to arrest the intruder.

It should be observed that the operation of the system is guaranteed also in case of power failure due to the fact that the control unit 100 is provided with a common rechargeable battery 120.

Lastly, it is clear that the invention described up to now may be subjected to modifications, additions or variants obvious to a man skilled in the art, without departing from the scope of protection outlined by the attached claims.