[go: up one dir, main page]

WO2022190142A1 - Système de détection, de reconnaissance et de lutte contre les actes de violence physique ou d'atteinte physique. - Google Patents

Système de détection, de reconnaissance et de lutte contre les actes de violence physique ou d'atteinte physique. Download PDF

Info

Publication number
WO2022190142A1
WO2022190142A1 PCT/IT2021/000012 IT2021000012W WO2022190142A1 WO 2022190142 A1 WO2022190142 A1 WO 2022190142A1 IT 2021000012 W IT2021000012 W IT 2021000012W WO 2022190142 A1 WO2022190142 A1 WO 2022190142A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
kinetic
mobile device
garment
sensors
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/IT2021/000012
Other languages
English (en)
Inventor
Stefano BELTRAMO
Enrico Di Stefano
Nicole OSELLA
Silvia ACTIS PERINO
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Be St Srl
Original Assignee
Be St Srl
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Be St Srl filed Critical Be St Srl
Publication of WO2022190142A1 publication Critical patent/WO2022190142A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
    • G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
    • G08B21/043—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis detecting an emergency event, e.g. a fall
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
    • G08B21/0438—Sensor means for detecting
    • G08B21/0446—Sensor means for detecting worn on the body to detect changes of posture, e.g. a fall, inclination, acceleration, gait
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
    • G08B21/0438—Sensor means for detecting
    • G08B21/0453—Sensor means for detecting worn on the body to detect health condition by physiological monitoring, e.g. electrocardiogram, temperature, breathing
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B25/00—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems
    • G08B25/01—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium
    • G08B25/016—Personal emergency signalling and security systems

Definitions

  • a system for detecting, recognizing and combating acts of physical violence or physical harm A system for detecting, recognizing and combating acts of physical violence or physical harm.
  • the found object of the invention relates to personal security systems for the prevention and intervention against perpetrated violent crimes, both in the domestic and public area, to the detriment of weaker persons and unable to defend themselves.
  • the invention also relates to security systems for the prevention of accidental physical damage. Especially complies with wearable devices for the protection and intervention in the case of accidents, physical harm and physical violence against the wearer even if it is, temporarily or permanently inhibited or unable, to a physical and/or psychological level, to defend themselves or to ask for help.
  • the invention concerns an innovative and proactive system for the direct collection of kinetic signals and specific biometric data, able to promptly recognize, autonomously and without the intervention of the monitored person, possible situations of danger or violence, and intervene forwarding an alarm message through different channels and communication protocols; said alarm message being sent to a plurality of entities and prefixed users who have the possibility to intervene to stop violence and/or identify the person who perpetrated it.
  • the proposed invention can be conveniently used to combat accidents or violent actions against any kind of minority (ethnic, religious, etc%), as a means of intervention against bullying in schools and public places of aggregation, to combat violence against the elderly at home or hospital facilities, and as a means of detection and intervention against violence on women.
  • the proposed system is based on a specific wearable technology, integrated into specially designed garments using conductive fabrics to realize flexible sensors able to detect a different multitude of kinetic and biometric data, for example, heartbeat, sweating, temperature, Galvanic Skin Response (GSR) and, in particular, the level of pressure applied on the body of the wearer.
  • GSR Galvanic Skin Response
  • Said system is accompanied by a removable ECU, powered by a rechargeable battery, which collects and processes said signals, normalizes and forwards them to a paired mobile device via wireless communication.
  • the ECU is equipped with communication modules that use Bluetooth, Wi-Fi, and NFC communication protocols.
  • a mobile device could be, for example, a smartphone or a tablet on which it is possible to install and perform an application of data analysis and actions consistent with the analyzed context.
  • the system, and in particular the mobile application deploys techniques and methods of machine learning, based on supervised models, with a subjective probabilistic approach of classification, able to adapt itself and optimize its output according to the type of input data.
  • the neural structure used is strictly dependent on the type of event and on the data that are inserted in the input layer: the identification of a situation of violence/danger involves the monitoring of heterogeneous variables such as, for information only, non-exclusive, biometric signals of the user (heartbeat, breathing rate, temperature, sweating, GSR), kinetic signals (accelerations, rotations, position in space) and stochastic signals, impulse, pressors detected by kinetic sensors.
  • biometric signals of the user heartbeat, breathing rate, temperature, sweating, GSR
  • kinetic signals accelerations, rotations, position in space
  • stochastic signals impulse, pressors detected by kinetic sensors.
  • the use of common frequency statistical methods would be impossible for two reasons: the frequency methods need datasets, which means they require very high data volumes, that multiple contexts do not allow, and require the knowledge of the phenomenon, impossible to achieve in an area where each "event" has its intrinsic characteristics and differs from all others.
  • Every violent action, assault, or overt act implies its boundary conditions, such as the biometric parameters of the victim that change from person to person under the influence of emotions, state of mind, character, etc..., preventing the use of objective statistical techniques.
  • the invention aims to overcome the limits of the actual systems dedicated to personal safety which have two major constraints: they assume the capacity of voluntary activation of the alarm signal and they are based on the use of an additional object to have always at hand, with needs of easy and immediate availability in time of need and assured functioning, therefore the need for electric charging and protection from accidental damage.
  • the invention also completely removes the problem of having to voluntarily activate the alarm, thanks to the use of flexible integrated sensors in the fabrics of the garments and of a virtual application that uses the electronic components present in modern smartphones to process the data acquired in real-time.
  • the invention in its software component, can be integrated with the devices on the market, without the need for replacement or modification.
  • Portable push-button acoustic bollards wherein the pressure from the user activates the alarm, a high volume sound generated by the same device to disturb the assailant and attract the attention of people who can intervene.
  • These devices are often battery- powered and they can be carried in pockets, as well as in bags, or hooked to a set of keys.
  • GPS anti-aggression devices that interface directly with the GPS (system) of smartphones via Bluetooth connection and a dedicated app. Pressing the button activates the alarm signal through a text message, mail or phone call, including the user’s geographical location.
  • the technology evolution has enriched the mobile devices with accessory features that go beyond the simple possibility of making calls: the current smartphones, in fact, are comparable to small computers, for their calculation capacity and computing power, being able to process data, even complex ones, send and manage emails, documents and files of different kind; they are also equipped with sensors of different types, with global positioning systems, GPS, and with GSM/UMTS data transmission systems, Bluetooth protocols and data sharing by proximity.
  • the operative systems of the phones can perform a multitude of programs, so-called Applications or Apps, which provide additional features, using the resources of smartphones as, for example, the use of CPU and GPU for the processing of complex data in real-time, the use of GPS to record the position of the device or even short-range communication (Bluetooth) between different phones within a common area of action to create a mesh network in which to pass, on all connected devices, a precoded message.
  • Applications or Apps which provide additional features, using the resources of smartphones as, for example, the use of CPU and GPU for the processing of complex data in real-time, the use of GPS to record the position of the device or even short-range communication (Bluetooth) between different phones within a common area of action to create a mesh network in which to pass, on all connected devices, a precoded message.
  • Bluetooth short-range communication
  • BEAST SENSOR a device with 3 accelerometers, 3 gyroscopes and 3 magnetometers, easy to wear on the wrist and on the upper part of the back. Freely usable during training at the gym, connects via Bluetooth to Android and loS smartphones.
  • a system according to the present invention based on wireless communication to maximize flexibility and feasibility, also, must be able to make up any problems related to wireless communications, to ensure always, and under any circumstance, the transmission of the alarm.
  • a prevention system must be able, once confirmed physiological damage, to record audio/video data then to safely and securely transfer it to a storage environment and store it for as long as necessary, protecting it from tampering, theft, and/or attempts of cancellation;
  • a system according to the present invention bases the forwarding of the request for help on wireless communications between a paired device and the recipient; however, at the base of the invention, there is also the creation of a mesh network between devices on which the system management software is installed.
  • the creation of a network between devices minimizes the risk of blocking the alarm signal, as it is able to make up for any malfunctions, breakdowns, or failures of a single device.
  • the acquisition and sampling of each individual signal shall be net of electromagnetic interference on the same frequency of use, attributable to other electromagnetic devices and/or to the user’s own behavior.
  • the garment will be able to distinguish localized impacts and pressures on the user’s body, filtering the signal of muscle movement.
  • the data acquisition, processing and forwarding systems must be minimally invasive and shall not cause discomfort to the user; the garment must be, therefore, comfortably wearable in all circumstances and without limitations.
  • the electronic subsystems will be fully integrated, flexible and miniaturized not to interfere with normal and daily use of the garment.
  • the devices can be tablets and smartphones. None of the features of the application slows down, stops, or is not performed because it is too burdensome for the device itself;
  • the program is able to identify the user’s activities and states.
  • the program can recognize the state of motion or quiet of the user (standing, sitting, walking, running, dancing, etc.), the emotional state of the user (agitation, stress, tranquillity, fear, excitement, etc.), the impulsive activities of the user (jumps, falls, blows or trauma); distinguishing a safe situation from one of danger;
  • the program is able to forward an alarm signal to alert.
  • Conditions may be, by way of example non-binded nor limited to, beatings, assaults, falls, accidents, malaise, etc.
  • the program can alert people who can, in different ways, intervene to change the situation and/or provide assistance; it shall contain the data of one or more recipients to whom the alarm signal shall be transmitted.
  • Said alarm signal must be sent by different forwarding methods.
  • Some forwarding methods can be, non-binded nor limited to, pre-compiled text messages, pre-recorded voice calls, mail with preset text, etc.
  • the network is useful to overcome any problems of transmission of the device generating the alarm.
  • the program can search other devices in case of malfunctions and/or failure or, more generally, circumstances of impediment that block the direct transmission of the alarm from the device that generates it.
  • the program searches other devices within its range that have a copy of the program itself (that belong to the same "community") and exploits their hardware resources for the forwarding of said signal.
  • the search and connection with similar devices within the range of activity from which the alarm starts, simultaneously generate an alert in every device reached.
  • the research and connection ensure that other people can detect the aggression and promptly intervene further reducing rescue times.
  • a program according to the following invention shares the data, logs, and information needed to forward the alarm to those who own the same program.
  • the alarm can call those who are at the same place of aggression or within a limited perimeter around it.
  • the communication is encrypted and protected.
  • the program can forward the recorded material, with protected and encrypted communication, to a cloud server where such material will be stored.
  • the recorded material may be useful for any subsequent use as evidence of the crime.
  • the program if there are impediments to direct forwarding, is able to create a mesh network with other compatible devices to complete the upload of the acquired material. • It is able to execute the tasks, analyses and procedures of the previous points in total autonomy. The program does not require the intervention of the user who suffers the assault. Commonly, the victim of violence may, temporarily or permanently, psychologically and/or physically, not be able to request help and rescue directly, and therefore prevented from requesting help and rescue directly.
  • the program uses self-learning algorithms and sharing of data. Compatible devices are those on which a copy of the program is installed.
  • a cloud platform that acts as a repository of acquired recordings, also allowing it to obtain an updated history of logs and surveys of clothing.
  • a system according to the present invention implements the following steps:
  • Initial automatic calibration in the stage of dressing, the electronics inserted inside the clothes will acquire the parameters of the user by the biometric sensors and the status of pressure and motion sensors, setting these values as "initial system setting".
  • the sensor status refers to the pre-stretch and compression rate that the sensors themselves undergo when the garment is worn by a person. By doing so, the system is always self-referential and can detect abnormal parameters more effectively than an objective calibration performed during the assembly phase, with sensors completely inactive;
  • Pairing the program, once started, will perform pairing with the garment, verifying the association keys assigned during construction and assembly. When pairing is completed, the program will acquire from the garment the initial parameters and settings that will allow the system calibration and subsequent optimized use. In this phase, the machine learning model is engaged in the recognition and classification of the initial parameters of the garment, clustering them and using them as a dynamic dataset for subsequent analysis;
  • Data acquisition the sensors, distributed in the garment, record, in real-time and seamlessly, signals from the user’s body and the control unit standardizes and prepares them for forwarding to the mobile program.
  • the firmware on the garment analyzes and maps the signals to eliminate interference; the approach differs according to the type of signal: for biometric filters, the method is based on a high pass and low pass filters (by way of example non-binding, nor limiting, 40 and 50 Hz filters to exclude muscle movement) and on the extrapolation of the signal formants, subject to significant variations depending on the psychophysical state of the user.
  • the method is based on differential analysis of multiple channels in parallel and about the initial calibration threshold; this method allows, according to the sensor and the type of signal you want to sample, to exclude useless deformations (by way of example nonbinding, nor limiting, for pressure signals if the x and y-axis channel differential it is not null, its value will be discarded);
  • Reading of standardized data the mobile program, received the data packets, processes them to allow analysis algorithms to read them and perform matching with its own "experience", acquired during initial training, and updated regularly, looking for similarities useful for the classification of the event;
  • the new input data are compared with the internal dataset of the program to classify the context and define the current situation; for guidance only non-binding, nor limiting, the biometric data are compared with a set of standard situations (state of agitation, fear, stress, excitement, etc.), tabulated and loaded into the program, to find similarities.
  • Predefined pressure data on body areas are compared with pre-configured alert thresholds and related to other kinetic data (accelerations, falls, sudden changes of relative position, etc.). The combination of the results of these operations defines, probabilistically, the outline of the situation analyzed.
  • the output of this step is therefore classification of the contingent situation, for guidance only, “normal situation”, of “potential danger” or “violence”, with a detailed indication of what is happening; by way of example non-binding, nor limiting, the output can indicate a situation of normality and tranquility (physical activity or dance), a risk situation (fear, rush, agitation) or of assault (beatings, attempted rape, etc.);
  • Start audio/video recording in case of non-verification by the user or of output compatible with violence, a program, according to this invention, immediately activates the recording of audio and video files, exploiting the hardware of the device on which it is installed and is executed, to collect material to document, in the appropriate locations, the violence suffered by the user.
  • Fig. 1 shows a general scheme of the system with the representation of connections between garment, mobile device, cloud platform and types of alarm sent;
  • Fig. 2 and Fig. 3 represent the front and back of a garment, according to this invention, with the indication of areas used for the insertion of sensors and ECU;
  • Fig. 4 and Fig. 5 represent the different kinetic and biometric sensors used and the possible positioning on the garment;
  • Fig. 6 represents a preferred form of a kinetic pressure sensor
  • Fig. 7 and Fig. 8 represent the circuit of conductive wires, integrated into the garment
  • Fig. 9, Fig. 10, Fig. 11, Fig. 12, and Fig. 13 represent the ECU with a protective case and the relative connections for charging and anchoring to the garment;
  • Fig. 14, Fig. 15, Fig. 16, and Fig. 17 represent in detail two possible configurations for fixing the ECU on the garment;
  • Fig. 18 shows the block diagram of the ECU
  • Fig. 19 represents the block diagram of the method used by the ECU to acquire and transmit data
  • Fig. 20 represents the block diagram of the machine learning method for data analysis and for the forwarding of any warning signal
  • Fig. 21 represents the logic analysis of the data used by the mobile device
  • Fig. 22 shows a mesh network between mobile devices for forwarding alarm messages.
  • a system (100) is represented to detect, recognize and combat acts of physical violence or physical harm; according to said system, the data relevant to the classification and recognition of an act of violence, of a malaise or a dangerous situation are collected discreetly and confidentially using a garment (101) characterized by signal acquisition areas (102) housing kinetic and/or biometric sensors. The signals of these sensors are collected and pre-processed by an ECU (103) and, after a first standardization, the relevant data are sent via a wireless communication channel
  • a mobile device (105) equipped with an application that implements techniques of artificial intelligence and machine learning.
  • Said software being usable to analyze and classify signals collected in the signal acquisition areas (102) of this garment (101) to recognize the context, the situation and type of interaction in place. If the results of the classification are compatible with violence, malaise, or dangerous circumstances, the software starts recording visual audio material that certifies what is happening and sends an alarm signal (106), with various communication methodologies and protocols, to contacts and numbers pre-set by the user at first startup.
  • the system searches within the range of the communication modules of the mobile device (105) for additional mobile devices compatible with it, so devices bearing another installation of the same application, through which to alert entities and pre-recorded contacts. Also, once the alarm (106) has been forwarded, the recorded audio-video material is forwarded to a cloud server (107) for storage and subsequent recovery as evidence of the violent action or incident.
  • the forwarding of the alarm (106) by the mobile device (105), once identified a situation of violence, is transmitted, in a preferential manner not limiting, using the following communication methods and protocols, according to what the user set at the first start of the software and compatibly with the technical possibilities of the mobile device (105):
  • All forwarded messages will transmit data of the person undergoing violence, the request for intervention and the geolocation of the mobile device (105) at the time of sending the alarm signal (106).
  • the communication with the cloud storage platform (107) of the data takes place through the same channels and protocols, with a preference for the GSM/LTE protocol before the Internet.
  • the system (100) includes a garment (101) represented, as an example, by a short-sleeved body integrating appropriate and convenient signal acquisition areas (102); said areas being equipped with kinetic and biometric sensors whose signals are collected through an integrated circuit in the fabric conveying signals to a collection node (111) and subsequently processed by the control unit (103).
  • a garment (101) represented, as an example, by a short-sleeved body integrating appropriate and convenient signal acquisition areas (102); said areas being equipped with kinetic and biometric sensors whose signals are collected through an integrated circuit in the fabric conveying signals to a collection node (111) and subsequently processed by the control unit (103).
  • a short sleeve bodysuit is dictated by the desire to create a garment of easy fit and usability: in the form of realization and preferential use of this invention, indeed, a short-sleeved bodysuit can be conveniently worn daily below normal clothing, in every season and occasion, thus providing a secure, continuous and invisible monitoring system in the eyes of any assailant.
  • said garment (101) may consist of a body with long sleeves or sleeveless, in combination with evening or ceremony dresses.
  • said garment (101) may be a fitness set consisting, for example, of a sleeveless top and a pair of long leggings, to wear for indoor and outdoor physical activity.
  • the above-mentioned signal acquisition areas (102) are intended to house sensors used to acquire kinetic and/or biometric signals, said signals being subsequently collected, analyzed, processed and used for the monitoring and identification of potentially dangerous situations.
  • Biometric repository points the biometric signals collected by the system need precise and unambiguous collection points ensuring the acquisition of a useful, clean and usable signal.
  • Kinetic repository points the points for the collection of kinetic and pressure signals were chosen based on a study of the dynamics of violence, of the interactions between the aggressor and the victim, the most exposed parts of the body, and those instinctively used for protection.
  • the acquisition areas (102) equipped with sensors to detect said kinetic and biometric signals are as follows:
  • the upper armband (biceps and brachialis triceps);
  • the forearm area is added, used instinctively to protect the head and body in case of aggression.
  • kinetic signals are collected by kinetic pressure sensors (112) distributed on the garment (101) as follows: • N. 8 sensors (112) positioned along the lower edge of the pectoral fins in the lower breast area, 4 per side;
  • N. 2 sensors (112) are located along the spine, starting from the median scapular line to climb towards the neck.
  • IMU inertial platform
  • the inertial platform is inserted directly into the ECU board (103);
  • the biometric signals collected are:
  • ECG signal (echocardiogram): collected using 10 biometric ECG sensors (114) located in signal acquisition areas (102) corresponding to the abdominal loops, on the clavicular midline, on the sides of the third sternal, and along the lower margin of the left pectoral, from the sternum to the axillary cable;
  • Respiratory frequency signal as follows: the respiratory frequency and amplitude are detected through a biometric amplitude sensor and respiration frequency (115) consisting of a resistive strain gauge inserted into the bodysuit. It is a band of conductive fabric integrated at the height of the chest, and more particularly in the lower insertion band of the pectoral fins, just above the diaphragm; said band of conductive tissue undergoes deformations as a result of respiratory acts; such deformations, which occur both on the longitudinal (increase in length) and on the transverse axes (reduction of width) result in a quadratic increase in the strength of the band itself.
  • said conductive band is made of metallic technical fabric, obtained, for guidance only, not binding, with the use of a mix of silver-coated synthetic fibres (polyamide, polyester, elastomer, etc.) and insulated by a TPU coating.
  • silver-coated synthetic fibres polyamide, polyester, elastomer, etc.
  • TPU coating insulated by a TPU coating.
  • silver or other conductive metal yarn sufficiently ductile;
  • the biometric temperature sensor (116) is located in the central part of the chest strap, in the vicinity of the xiphoid process; in an alternative form, said sensor is located on one of the sleeves of the body, near the inner portion of the wrist; in at least one other alternative form, the sensor is located on a side, at cost level;
  • GSR Galvanic Skin Response
  • the biometric sensor GSR - Galvanic Skin Response (117), in the form of preferential implementation of the invention, is placed on the lower edge of the right chest, near the axillary cable.
  • a garment (101), according to this invention, is equipped with a small size ECU (103), including:
  • Said system is used to carry out the following tasks:
  • the kinetic (112), (113) and biometric (114), (115), (116), (117) sensors described above are fully integrated into the fabric of the garment and themselves made of conductive fabric, they need an external power supply, which they receive from the above-mentioned control unit (103).
  • the kinetic sensors (112) used are piezoresistive types while those used for biometric signals are capacitive. All kinetic and biometric sensors have an average consumption of about 2.5 mW each and they are powered at 5V by a lithium-ion rechargeable battery, integrated into the ECU (103).
  • the total power consumption in normal operating conditions is 1 Wh, with 2 Wh peaks at times of matching device search and data submission. Consequently, to ensure proper operation and protection to the user for at least 4 hours of continuous operation, the battery has a capacity of 1,200 mAh.
  • the battery may be smaller in size, to the advantage of the miniaturization of the system and the comfort of use, but with a more limited functioning and autonomy in time.
  • said kinetic pressure sensors (112) are made of conductive fabric, to ensure flexibility and maximum adherence to the body during all uses; this fabric is made of synthetic fiber coated with silver or other conductive and ductile material and insulated from the external environment through a micrometric coating of thermoplastic polyurethane (TPU).
  • TPU thermoplastic polyurethane
  • the catchment circuit is created by sewing, according to the geometry shown in Fig.6, an insulated conductive wire (118), compared to the fabric and the external environment, via a non-conductive polymer micrometric coating.
  • I electric current
  • this conductive wire (118) it is characterized, in the phase of inactivity and pre-load, by a certain resistance value; this value is recorded as the calibration threshold for each use of a garment according to the invention.
  • the kinetic pressure sensor (112) When the kinetic pressure sensor (112) is subjected to a strain load, particularly along the z-axis perpendicular to the construction plane of the circuit made of conductive wire (118), its resistance varies with quadratic proportionality compared to the deviation from the initial condition; the initial and subsequent measurements, compared with each other and with a table of reference values, with appropriate additions over time, allow to determine the nature of the event that generated pressure on said kinetic pressure sensors (112).
  • the pressure sensor (112) may not be integrated into the fabric as in Fig.6, but may be applied on it: in that variant, the kinetic pressure sensor consists of a flexible and highly deformable substrate such as, by way of example non-binding, nor limiting, polymer rubber or silicone rubber, inside which there is the sensor circuit made of different ductile and flexible conductive material.
  • the sensors both pressors (112) and biometrics (114), are made using silver nanowires in a micrometric polymer matrix.
  • the integrated circuit in the garment is illustrated in detail: in the form of the preferential implementation of this invention, the circuit and the connections of the kinetic (112), (113) and biometric sensors (114), (115), (116), (117) with the control unit (103) are made through the use of conductive wires (119) sewn and integrated into the fabric of the garment (101). These wires (119) convey all signals to a collection node (111) where the signals are centralized and are further transmitted by additional conductive wires (119) to the ECU (103).
  • These conductive wires (119) are made, byway of example non-binding, nor limiting, of synthetic fiber, elastomeric, integrated with metallic powders, ductile metals, or other similar materials, and are coated with an insulating coating made of polymeric material, TPU or similar.
  • the electrical signals taken on the surface of the body are subject to multiple interferences and noise due not only to the state of motion of the body itself but also from the surrounding environment as sources of electromagnetic radiation, wireless data transmission, etc.
  • the circuit therefore, needs to be isolated to ensure the high quality and cleanliness of the signal; at the same time, however, insulation should not be impactful and invasive to not affect the wearability and comfort of the garment.
  • the circuit insulation system uses the external insulating layer of conductive wires (119) used to make the connections and the conductive fabric itself: the manufacture of the garment (101) involves the use of materials typical of the textile industry (tulle, cotton + elastomer, etc.) with conductive fabric inserts.
  • the conductive fabric In the signal acquisition zones (102) the conductive fabric is in direct contact with the skin of the user, while in the rest of the garment (101), which includes circuit areas, it is placed on the outside, an insert sewn above the material normally used. So, the conductive fabric assumes the function of electromagnetic shielding for most of the emissions that are found in common environments frequented daily and combined with the insulating layer that envelops the cable ensures adequate shielding of the circuit.
  • the connection between the conductive wires (119) of the circuit and the kinetic and/or biometric sensors takes place, in the case of synthetic fiber wires, by micro-crimping of the circuit itself on the sensor terminal, typically a wire in a conductive material (119); in the case of insulated wire, by microsaldation of the sensor terminals on the cable.
  • both crimping and welding can be replaced by two-layer bonding where the first layer, innermost, is made of conductive material; by way of example non-binding, nor limiting, a conductive silicone adhesive, and the second, external, is made of insulating material, for example, a silicone adhesive.
  • Said electrical pressure contacts will therefore include a "female" part (123) placed on the lower portion of the case (121) while the corresponding "male” contacts (124) are anchored on the garment (101).
  • the case (120) containing the ECU (103) is equipped with a rechargeable battery via a connector (125), which allows connection to power banks or other electronic charging devices of common use for the mobile device (105).
  • a preferable form for attaching the case (120) to that garment (101) is presented.
  • the male contacts (124) are located on one side, at the rear of the iliac crest.
  • the positioning of the case (120) of the ECU ( 103) in this area gives greater comfort and daily usability to the invention, increasing its potential spread and ease of use among the population.
  • Concerning the attached drawings, and in particular, to Fig. 16 and 17 an alternative form of positioning and attachment of this case (120) to the garment (101) is represented.
  • the case (120) is placed at the height of the xiphoid process, that is, the lower extremity of the sternum, near the abdominal midline.
  • control unit (107) is not removable as in the cases depicted and, consequently, the electrical pressure contacts (123) on the lower portion (121) are replaced by connectors typical of electronic production; in this form, the terminals of the circuit embedded in the fabric are replaced by flex PCB and flat connectors that couple with those present on the ECU; in at least one other form of alternative construction, PCBs flex are replaced by flex bus, with their connectors.
  • the preamplification subsystem according to this invention is able to acquire and manage both kinetic/pressure signals and biometric signals, with voltage values on the sensor itself typically between 0.5 and 4.0 mV. In the preferred form of realization there is a preamplifier with the following characteristics:
  • a simplified block diagram (126) of the ECU (103) is represented.
  • signals relating to kinetic sensors (112), (113) and biometric sensors (114), (115), (116), (117) are acquired under "daily" user activity conditions that may include waking and activity, sleep, etc. Said conditions determine the possible overlap of multiple sources of interference on the acquisition of the signal itself.
  • an ECU integrates physical and digital filters that mitigate these effects by allowing a clean, stable and usable signal to be acquired.
  • an ECU according to this invention, has the following filters:
  • the A/D converter used in the example presented is a standard subsystem, commonly inserted in electronic devices, consisting of an integrated n. 12-channel input, 8 singles and 4 differentials, with 12-bit resolution, and n. 2-channel output, with 11 -bit resolution, with a maximum sampling rate of 10,000 samples/sec.
  • Concerning Fig. 19 of the attached drawings the system of acquisition and forwarding of signals related to kinetic sensors (112), (113) and biometric sensors (114), (115), (116), (117) is represented.
  • the system (100) provides that the user after wearing the garment (101) must activate a matching software that will run in the background on the mobile device (105) on which it was installed. Said software will acquire (if necessary) the data of the aforementioned kinetic and biometric sensors according to the following methodology:
  • the input signals from the sensors are compared with the tabulated values: o With input value ⁇ to the threshold values, the data are discarded, because they are considered useless for the analysis as attributable to a standard safety condition; o With input value > to the threshold values, the data is forwarded to the software running on the mobile device (105) combined for subsequent analysis.
  • the module that implements the wireless communication channel (104) is the consumption bottleneck of the ECU (103); its consumption is the highest among all types of components onboard the card, therefore its optimized use allows to size the battery properly and to minimize weights and dimensions to the advantage of fit and comfort of use. For this reason, minimizing the transmissions of the wireless communication channel (104), avoiding the forwarding of superfluous data and not attributable to situations of violence or danger, is a fundamental step of optimization.
  • the processor needs to be engaged as little as possible in data analysis and only when the data may be attributable to situations of danger or violence. Also in this case the aim is fully achieved with the strategy of pre-analysis and forwarding implemented in the ECU (103).
  • the thresholds indicate the limit value beyond which the signal is no longer physiological and referable to a condition of safety and tranquility. They provide only one level (Sb); o Time limits: indicate the duration of recorded events and signals (range threshold) and the number of their iterations (single value threshold) over time, t, taken into account for signal integration. In other words, said thresholds (St) indicate the duration of events and how many times they are repeated during a certain observation period.
  • the program starts the search for compatible clothing. Search and connect via Bluetooth Low Energy, BLE wireless communication protocol.
  • the program performs the pairing at each start going to look, in the first instance, for the first garment (101) that has been matched.
  • the program activates an alert to the user with a support request. If in the range of reception of the communication module there is another compatible device (101), never combined with the program, a request for connection to the new garment is activated (101).
  • the pairing is done freely with the first compatible garment within range of detection, without permanent storage of its coupling keys.
  • Comparison of biometric data o If the data are less than the threshold values, with an open-screen program, they are displayed and then stored as log files for use in a training dataset aimed at improving the recognition and classification efficiency of the system. In case the active program works in the background, the display does not take place, data are saved to the application memory on the mobile device (105) and used for the creation and expansion of the training dataset; o If the data are above the threshold, they move on to the next analysis task that covers the duration and number of iterations over time;
  • Comparison of kinetic data o Comparison with Sc1 : if they are smaller, with an on-screen active program, they are displayed and subsequently stored and used as training datasets; with the active program in the background, they are directly saved and included in the training database; if greater than Sd, they are compared with Sc2; o Comparison with Sc2: if smaller, they switch to the next time analysis task; if higher, they are compared with Sc3; o Comparison with Sc3: if the data are smaller, they move on to the next time analysis task; if the data are larger, they pass directly to the classification task performed by the machine-learning sub-program;
  • Comparison of data duration in this task, the time markers of the recorded data are compared with those of presets set by the programmer; o Biometric data: in the case of over-threshold detection, a timestamp is added to the raw data, a marker, indicating how long that signal has remained above the threshold. The marker is compared with tabular values used to cluster the data for subsequent classification. For example, an increase above the threshold of the value of the heartbeat and the respiration of some second can be clusterized in the sporadic events (a reading error, a fright, etc.); an equal increase in time, in the order of minutes, is clustered as possible physical activity or state of agitation or fear; o Kinetic data: kinetic data are accompanied by two-time markers:
  • Said markers are compared with a cross-analysis between the two markers and the tabulated and consequently clustered values; for example, an acceleration above the threshold of a few milliseconds that is not repeated, associated with a deviation of the gyroscope axis with similar temporal characteristics, is clustered as a single accidental event (accidental impact, random thrust, etc.); events of very low duration and high amplitude, repeated over time, can be clustered as possible traumatic events, physical activity, etc.
  • the output of this analysis needs to provide already standardized data to the machine learning model for the final classification of events.
  • Classification of the event employs a machine learning algorithm.
  • a scheme (129) of the logics of analysis of the used data is represented.
  • Said logics based mainly on a method of approximation of Bayesian Inferences subjective statistics, and more particularly the method of variational inferences, as they ensure better scalability with large models, especially for large-scale neural networks in deep learning applications.
  • the variational inference method used is capable of efficiently representing weights with a limited number of parameters.
  • the method presented is comparable to the category of FNNs, Feedforward Neural Networks, and approximates the posterior distributions of the network weights as Gaussian distributions associated with variational parameters.
  • model training takes place using CPU/GPU systems before the parameters for weight distributions and two types of parallel random number generators are used: o A RLF-GRNG linear response logic and binomial distribution generator; o A Gaussian Wallace random number generator, Bayesian Neural Networks- oriented, with a recursive method.
  • the outputs of the sensor comparisons or datasets (130) enter the statistical inferential motor (132) where they are classified by taking advantage of the "knowledge” provided by a reference model dataset (131), said dataset (131) being constantly updated with the log files generated by the program itself.
  • the result of the classification is a prediction (133) a probability value indicating whether given signals of a given event, can be indexed as a violent/dangerous event, a malaise, a normal situation, etc.
  • This step takes place by comparing the clustering carried out in the previous steps with the model dataset (131) inside the program: by way of example, non-binding, nor limiting, an event in which there are changes above the threshold of biometric signals, for a certain time without any kind of threshold exceedance of kinetic data, is indexed as a potential malaise and classified by comparison with samples of this category.
  • the actual classification takes place, whose outputs arise in system reactions to the event: o
  • Malaise an event indexed as such, has two possible outputs of classification such as confirmation of malaise or "false positive”.
  • the first threshold concerns random events, limited in time by duration, number of repetitions and amplitude, as an example, non-binding, nor limiting, accidental falls or thrusts, "friendly" physical contact (a pat on the shoulder, jokes between friends ...), etc.
  • the system output is the display of data on screen, if the program is active, resulting in the storage of event logs. In the case of application working in the background the data are not displayed but directly saved in the system memory for later use;
  • the second threshold defines events of moderate/high intensity that may be attributable to situations of danger or violence, as well as user activity situations (contact sports, high impact and intensity physical activity, dancing, etc.).
  • the program through haptic and sound signaling, requires the intervention of the user to clarify the type of event.
  • the data displayed on screen, are recorded in an event log and stored for later use; in case of a negative response from the user, of confirmation of violence, or non-response from the user, the system prepares for the next steps.
  • the third threshold the highest, classifies the event as directly dangerous to the user and starts the alarm activation procedures.
  • the system in its preferred form, starts recording audio/visual material, exploiting the hardware of the device on which it is installed, to document the current event.
  • the program accesses and activates the microphones of the device to record what is happening; similarly accesses and activates the system cameras to record a video in real time.
  • the audio/visual material is recorded, in its preferred form, for 4 minutes; after registration, the files are uploaded to a dedicated repository, on a cloud server, accessible only by the user.
  • audio/video material may be recorded for a longer time.
  • the audio/visual material may not be recorded.
  • the purpose of the recording is to provide evidence of what is happening, useful in the moments after the event to acquire useful information to intervene and, in later times, to document, in the appropriate locations, what happened.
  • the mobile program in a system according to the present invention, at the first start and for all subsequent starts until the action is completed, the mobile program requires that you provide contact details to which you can forward any alert; by way of example, non-binding, nor limiting, such contacts may be telephone numbers, e-mail addresses, social accounts, etc.
  • the program requests that the text of the message be forwarded to the recipients.
  • the program sends the message specified by the user, accompanied by her/his personal details and by the geolocation of the device at the time of sending, to the recipients, consistent with the type of contact set, using one or more of the following protocols: o BLE communication; o LTE communication; o Wi-fi communication, o NFC communication.
  • a simplified scheme for creating a mesh network (134) between compatible devices is represented; said network being used for the forwarding of alarm messages under particular conditions and in particular if the alarm signal cannot be forwarded due to system malfunctions or failure of one or more communication modules, the program implements alternative deployment strategies.
  • the program installed on the mobile device (105) will search, using proximity (NFC) and reduced-field communication modules (BTLE, Wi-fi), for devices compatible with the mobile device (105) within the range of coverage of the device on which it is installed and that generated the alarm.
  • Compatible device means, for guidance only, a mobile device, a smartphone or a tablet, on which another copy of the same program is installed.
  • a program according to this invention establishes a connection and pairing with them to start a secure and encrypted communication; this results in a mesh network in which each device, or network node, operates as an external data gateway, and as a bridge to reach other compatible devices and share data with them.
  • the program will forward in encrypted, anonymous and not viewable form by the owners of such devices, all data related to emergency contacts and alarm message (106) of the generating mobile device (105).
  • the iterations of the program in the associated arrangements will initiate attempts to send the alarm to newly imported contacts until one of them is successful.
  • a command will be generated that will block attempts on all other devices.
  • a notification is generated, haptic and acoustic, on coupled devices, to allow their owners to intervene quickly in support of the user.
  • the invention applies to the area of personal monitoring and security, and of vulnerable groups and/or workers, performing tasks with a high risk of assault, and in particular, the field of devices to prevent domestic and public violence and accidents with physical damage: in the field of prevention and intervention, a system according to the present invention is able to monitor, on an ongoing, silent and completely anonymous basis, the situation of the user, is able to recognize autonomously a situation of danger, violence and / or malaise and to intervene automatically bypassing the limits of voluntary activation, typical of the art known.

Landscapes

  • Health & Medical Sciences (AREA)
  • Emergency Management (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Health & Medical Sciences (AREA)
  • Gerontology & Geriatric Medicine (AREA)
  • Engineering & Computer Science (AREA)
  • Cardiology (AREA)
  • Psychology (AREA)
  • Psychiatry (AREA)
  • Computer Security & Cryptography (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Social Psychology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Physical Education & Sports Medicine (AREA)
  • Physiology (AREA)
  • Pulmonology (AREA)
  • Alarm Systems (AREA)
  • Collating Specific Patterns (AREA)

Abstract

L'invention concerne un système de détection, de reconnaissance et de lutte contre les actes de violence physique ou d'atteinte physique. Selon ce système, les données se rapportant à la classification et à la reconnaissance d'un acte de violence, d'un malaise ou d'une situation dangereuse sont recueillies discrètement et confidentiellement au moyen d'un vêtement; ce vêtement est caractérisé par des zones appropriées d'acquisition de signaux équipées de capteurs cinétiques et biométriques dont les données sont recueillies et prétraitées par une unité de commande et, après une première standardisation (uniformisation), envoyées par l'intermédiaire d'un canal de communication sans fil à un dispositif mobile; ce dispositif étant équipé d'une application logicielle qui met en œuvre des algorithmes classification et un apprentissage automatique/apprentissage par machine. Ce logiciel est utilisé pour analyser et classifier les données recueillies par des capteurs situés dans lesdites zones d'acquisition de signaux afin de reconnaître le contexte, la situation et le type d'interaction en place. Si les résultats de la classification sont compatibles avec une violence, un malaise ou une situation dangereuse, le logiciel démarre l'enregistrement d'un contenu audiovisuel qui authentifie ce qui se passe et émet un signal d'alarme, avec divers procédés et protocoles de communication, contacts et numéros préréglés par l'utilisateur au cours du premier démarrage. Si, en raison de pannes, de dysfonctionnements et/ou d'empêchements de nature technique, la transmission d'une alarme directe n'est pas possible, un système selon la présente invention recherche, à la portée des modules de communication du dispositif mobile, des dispositifs mobiles supplémentaires compatibles avec celui-ci, à savoir des dispositifs mobiles dotés d'une autre installation du même logiciel, par l'intermédiaire desquels alerter des entités et un contact préalablement inscrit. Après une alarme, le contenu audiovisuel enregistré est transmis à un serveur en nuage en vue d'un stockage et d'une récupération ultérieure en tant que preuve de l'action violente ou de l'incident.
PCT/IT2021/000012 2021-03-12 2021-03-22 Système de détection, de reconnaissance et de lutte contre les actes de violence physique ou d'atteinte physique. Ceased WO2022190142A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
IT102021000005999A IT202100005999A1 (it) 2021-03-12 2021-03-12 Apparato e metodo per il rilevamento, il riconoscimento e l’intervento per contrastare la violenza fisica a danno di persone.
IT102021000005999 2021-03-12

Publications (1)

Publication Number Publication Date
WO2022190142A1 true WO2022190142A1 (fr) 2022-09-15

Family

ID=75977788

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IT2021/000012 Ceased WO2022190142A1 (fr) 2021-03-12 2021-03-22 Système de détection, de reconnaissance et de lutte contre les actes de violence physique ou d'atteinte physique.

Country Status (2)

Country Link
IT (1) IT202100005999A1 (fr)
WO (1) WO2022190142A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20240242579A1 (en) * 2023-01-17 2024-07-18 Torrey Pines Logic, Inc. Passive personal laser detector warning safety device

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012176193A1 (fr) * 2011-06-20 2012-12-27 Cardio-Healthwatch Innovative Solutions Ltd. Système indépendant de surveillance sanitaire et d'alerte, non interférent, susceptible d'être porté
WO2019222846A1 (fr) * 2018-05-22 2019-11-28 Myant Inc. Procédé de détection et de communication de données biométriques et de communication bidirectionnelle avec une plateforme de capteur à base de textile
KR20200121624A (ko) * 2019-04-16 2020-10-26 한지운 직물 압력 센서가 구비된 스마트 의류 및 이를 이용한 낙상 대응 시스템

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012176193A1 (fr) * 2011-06-20 2012-12-27 Cardio-Healthwatch Innovative Solutions Ltd. Système indépendant de surveillance sanitaire et d'alerte, non interférent, susceptible d'être porté
WO2019222846A1 (fr) * 2018-05-22 2019-11-28 Myant Inc. Procédé de détection et de communication de données biométriques et de communication bidirectionnelle avec une plateforme de capteur à base de textile
KR20200121624A (ko) * 2019-04-16 2020-10-26 한지운 직물 압력 센서가 구비된 스마트 의류 및 이를 이용한 낙상 대응 시스템

Also Published As

Publication number Publication date
IT202100005999A1 (it) 2022-09-12

Similar Documents

Publication Publication Date Title
Dian et al. Wearables and the Internet of Things (IoT), applications, opportunities, and challenges: A Survey
US10631760B2 (en) Method for prediction, detection, monitoring, analysis and alerting of seizures and other potentially injurious or life-threatening states
Dai et al. PerFallD: A pervasive fall detection system using mobile phones
US20150130613A1 (en) Selectively available information storage and communications system
JP2018149355A (ja) 監視のための装置および方法
US20060282021A1 (en) Method and system for fall detection and motion analysis
KR102831595B1 (ko) 위험 상황을 판단하고, 사용자의 안전을 관리하는 시스템
KR20130051922A (ko) 심리장애(ｐｓｙｃｈｏｌｏｇｉｃａｌ ｄｉｓｏｒｄｅｒｓ) 치료를 위한 장치 및 방법
US11647954B2 (en) Ear device for heat stroke detection
CN205050303U (zh) 一种智能人体瘫倒监测装置
US20200037904A1 (en) Systems, Devices, and/or Methods for Managing Health
KR20180028358A (ko) 단말기 및 그 제어 방법
CN107846990B (zh) 智能内裤及其报警控制方法
Bharathiraja et al. Real-time fall detection using ESP32 and AMG8833 thermal sensor: A non-wearable approach for enhanced safety
CN204719923U (zh) 基于可穿戴设备的遇险手机分级自动报警装置
US20170086754A1 (en) System and method for measuring and detecting physiological appearances in humans
CA3039538A1 (fr) Systeme d'alerte
CN111161509B (zh) 脑电波触发的非接触式报警装置及其报警检测方法
WO2022190142A1 (fr) Système de détection, de reconnaissance et de lutte contre les actes de violence physique ou d'atteinte physique.
KR20090123041A (ko) 헬멧 형 유비쿼터스 안전관리 시스템 및 서비스
Ianculescu et al. Improving the Elderly’s Fall Management through Innovative Personalized Remote Monitoring Solution
KR101364638B1 (ko) 화생방 상황에 대응하기 위한 환경 검출 시스템, 장치 및 방법
US20240066310A1 (en) Garment change and wash detection in a wearable cardioverter defibrillator (wcd)
CN110974187A (zh) 用于被监护人的健康监测系统
KR101774702B1 (ko) 응급 상황 자동 신고 시스템 및 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21726468

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 17/01/2024)

122 Ep: pct application non-entry in european phase

Ref document number: 21726468

Country of ref document: EP

Kind code of ref document: A1