Apparatus and method for assisting visually impaired persons

Abstract

An apparatus (1 ) for assisting visually impaired persons, comprising: - at least one image or video acquisition camera (2) positionable on a device wearable by a person; said camera (2) being configured to generate a video signal (3) in real time; - a processing unit (4) connected to the camera (2) and configured to: receive said video signal (3); process the images contained in said video signal (3); characterised in that said processing unit (4) is further configured to transduce said video signal into an audio and / or vibration signal, preferably a 3D sound, which can be played back on an audio or vibration playback device (6) associable with the user so that the latter can perceive such sound or vibration corresponding to the configuration of the environment and the position of the obstacles.

Description

[0001] "APPARATUS AND METHOD FOR ASSISTING VISUALLY IMPAIRED PERSONS"

[0002] * * * * *

[0003] Technical field

[0004] The present invention relates to an apparatus and a method for assisting visually impaired persons or for increasing the perception of the surrounding space and related obstacles of a person (not necessarily visually impaired). In detail, the apparatus relates to the field of perception of the surrounding environment and related obstacles by means of the generation of corresponding sounds and / or vibrations which are perceived by the person using the apparatus.

[0005] Prior art

[0006] According to the World Health Organization, about 285 million people are visually impaired and 39 million are blind. Moreover, this count increases by 7 million every year, and this ratio is expected to double by 2030.

[0007] The reduced ability to analyse the environment, and consequently the reduced capacity for interaction and mobility are among the most relevant consequences of vision loss.

[0008] Although guide dogs still represent a valuable and effective solution for reducing such consequences of the blind and visually impaired, especially outdoors, technological advances have led to an evolution of the techniques used to assist these users.

[0009] For example, satellite navigation solutions run on smartphones, or the use of ultrasonic sensors for avoiding obstacles and other solutions which are not described in detail here are known.

[0010] However, such known solutions have some disadvantages related to the fact that it is not yet possible to have an assistance system which is effective in everyday use. For example, with reference to satellite navigation systems, they allow guidance outside, but not inside, buildings, since the connection with the satellite is not always available. Furthermore, such systems have the disadvantage that they do not identify the obstacles present on the path and often the indicated path is not the optimal one for the visually impaired.

[0011] * * * * *

[0012] Objects of the invention

[0013] Therefore, the technical task of the present invention is to propose an apparatus and a method for assisting visually impaired persons which makes it possible to obviate the drawbacks of the prior art described above. In particular, an object of the present invention is to provide an apparatus and a method for assisting visually impaired persons which is more effective in terms of assistance in all situations (indoor, outdoor environments, presence of obstacles, etc.).

[0014] A further object of the present invention is to provide an apparatus and a method for assisting visually impaired persons which is easily worn by a user and which allows to simultaneously recognise the shape of the surrounding environment, thus facilitating movement in both indoor and outdoor scenarios.

[0015] * * * * *

[0016] Brief description of the drawings

[0017] The present invention is hereinafter further described in some of the preferred embodiments thereof, given for purely exemplary and non-limiting purposes with reference to the accompanying drawings, in which:

[0018] - figure 1 shows an operating diagram of the apparatus according to the present invention;

[0019] - figures 2a, 2b, 2c show, according to a schematic perspective view, the reconstruction steps of the surrounding environment; - figure 3 shows, according to a schematic perspective view, an obstacle identification step:

[0020] - figure 4 shows, according to a schematic perspective view, an identification step of the type of obstacles.

[0021] With reference to the drawings, they serve solely to illustrate embodiments of the invention for the purpose of better clarifying, in combination with the description, the inventive principles at the basis of the invention.

[0022] * * * * *

[0023] Detailed description of the invention

[0024] With reference to the accompanying drawings, an apparatus for assisting visually impaired persons has been indicated overall with the number 1 .

[0025] As mentioned above, such an apparatus could also be used to increase the perception of reality for people who are not necessarily visually impaired.

[0026] The apparatus 1 comprises at least one image or video acquisition camera 2 positionable on a device wearable by a person (e.g., glasses, shoes, etc.), in which the camera 2 is configured to generate a video signal 3 in real time. The apparatus 1 could comprise more than one camera 2 (in this case a software processing system is provided to synchronise the images with each other and estimate the distance of the camera from the scene by means of stereovision algorithms). Preferably, the camera(s) used is / are RGBD cameras.

[0027] Furthermore, there is a processing unit 4 connected to the camera 2 and configured to receive said video signal 3 and to process the images contained in said video signal 3. As will be seen below, such a processing unit 4 could be physically made in a single electronic device or in more than one spaced apart from each other.

[0028] In accordance with the present invention, such a processing unit 4 is configured to:

[0029] - extrapolate information from the images contained in said video signal 3 related to the shape and / or dimensions of the environment surrounding the user so as to define the perimeter and the openings of the environment (e.g., room);

[0030] - make a three-dimensional digital reconstruction of the surrounding environment on the basis of said extrapolated information (figures 2a, 2b, 2c). Preferably, the processing unit 4 is configured to make a three-dimensional digital reconstruction of the surrounding environment using a SLAM (Simultaneous Localization and Mapping) algorithm;

[0031] - extrapolate information from the images contained in said video signal 3 related to the obstacles (objects) present in the surrounding environment and associate the position of said obstacles within said three-dimensional digital reconstruction of the environment (figure 3). It should be noted that the position of the obstacles is detected with respect to the environment surrounding the user and defined by the previous three-dimensional reconstruction of the image;

[0032] - identify the position and orientation of the user within said three- dimensional digital reconstruction of the environment;

[0033] - generate a processed signal 5 containing said three-dimensional digital reconstruction of the environment including the position and orientation of the user therein and the position of said obstacles therein, so that the user can perceive said predetermined information of the three-dimensional digital reconstruction, by listening to the audio or perceiving the generated vibration.

[0034] The apparatus 1 further comprises an audio or vibration playback device 6 associable with the user so that the latter can perceive such sound or vibration. Preferably, such a playback device 6 comprises audio headphones or earphones for the playback of an audible sound (wearable on the ears), or headphones which transmit vibrations once worn.

[0035] The processing unit 4 is further configured to transduce the processed signal 5 into an audio or vibration signal 8, preferably into a sound with a three-dimensional effect. In detail, the processing unit 4 is configured to associate a set of respective audio or vibrational features with the information contained in said processed signal 5 according to a predefined logical map in which predetermined information of the three-dimensional digital reconstruction of the environment, position of the obstacles and position and orientation of the user correspond to respective predefined audio or vibrational features of frequency, power and time lag. The term "audio or vibrational features" means precisely the set of features such as frequency, power, signal amplitude, time lags, etc.

[0036] For example, a sound having a first frequency can correspond to the presence of a first wall to the left of the user. A second sound having a second frequency different from the first can correspond to the presence of a second wall to the right of the user.

[0037] The processing unit 4 is configured to

[0038] - generate an output signal 7 containing said set of respective audio or vibrational features which have been associated with the features of the environment according to said predefined logical maps;

[0039] - play back said output signal 7 by means of said audio or vibration playback device 6.

[0040] In other words, the processing unit 4 is configured to associate a set of respective audio or vibrational features with the information contained in said processed signal 5 according to a predefined logical map, so that said output signal 7 has predefined audio or vibrational features as a function of the position, distance and nature of the obstacles and the surrounding environment so that the user can perceive the position, distance and nature of the obstacles and the surrounding environment as a function of the audio or vibrational playback.

[0041] Preferably, the output signal 7 is an audio signal 8 of the three-dimensional or 3D effect type.

[0042] That is, in relation to the 3D sound generated, it can be said that a characteristic acoustic footprint is created which allows the receiver of the signal to distinguish the position, distance and nature of the surrounding obstacles and, in the event of indoor use, the perimeter of the room.

[0043] Preferably, the audio or vibration playback device 6 comprises two sound or vibration emission channels, typically right and left channel which are associated with the respective ears of the user. The processing unit 4 is configured to generate said output signal 7 containing audio or vibrational features which are different or with a time lag with respect to said two sound or vibration emission channels so as to generate said 3D sound. For example, the effect of spatial depth can be generated by a time lag between the right and left channel.

[0044] Furthermore, as already mentioned, the processing unit 4 comprises at least two distinct processing subunits 4 connected to each other, in which:

[0045] - a first processing subunit 4a is configured to process the images contained in said video signal 3, generating said processed signal 5;

[0046] - a second processing subunit 4b is configured to generate said output signal 7 containing said set of respective audio or vibrational features.

[0047] In practice, there will be a container containing the processing unit 4 which will comprise a motherboard equipped with CPU and optionally a volatile memory, non-volatile memory, sound card, network communication card(s) such as WiFi and / or Bluetooth and other optional sensors such as compass, accelerometers, GPS modules, etc.

[0048] However, the solution of using only one container must not be considered binding for the present invention. In fact, alternatively, the apparatus 1 can comprise two containers of which one comprises the first processing subunit 4a and the second comprises the second processing subunit 4a, connected to each other by radio or by cable and each, preferably, with hardware optimised for the purpose it is intended to achieve. Each subunit is provided with a battery which supplies the energy necessary for the operation thereof. In detail, the surrounding environment is preferably perceived by a camera 2 (e.g., Asus Xtion PRO LIVE) which provides RGBD at 30 fps with a resolution of 640x480 pixels and an estimate of the depth from 80 cm to 4 metres. The frames are then processed by a device of the processing unit

[0049] 4 with hardware acceleration for the execution of deep learning models in real time.

[0050] The system operates with a standard PC, but also with a portable embedded device.

[0051] At each time step, the RGBD frame is processed with the SLAM RTAB-Map (Real-Time Appearance-Based Mapping) algorithm to align the point clouds reconstructed from successive frames and gradually construct a complete representation of the environment and localise the environment and simultaneously localise the user therein.

[0052] The system approximates the layout of the room with a box, that is, the structure of the environment is mathematically modelled with six flat surfaces perpendicular to each other. At the beginning this model is adapted to the point cloud data using the least mean squares.

[0053] If the camera 2 is mounted at the head, the lower surface is the floor and the parallel plane thereof is the ceiling.

[0054] Initially, a maximum of three other planes can be considered, but often two are enough, as in the case of corridors where the front plane is too far away to be detected by the depth sensor.

[0055] In addition to walls, the end user must be aware of obstacles and objects in order to plan a path free of collisions.

[0056] A YoloV3 convolutional neural network is preferably used for detecting objects. The model processes the RGB channel and returns the bounding box coordinates and class label for each object recognised in the image. The bounding box is then mapped onto the depth channel to localise the placement of the object in 3D.

[0057] The 3D position of the centroid of the object is then assigned by calculating the median of the depth values of all the points of the bounding box and applying the coordinate of the selection rectangle and applying the transformation of the coordinates obtained with SLAM. This allows to estimate the position of the objects with respect to the user, and to store this information in the global frame of reference in order to be able to use it even when the object is no longer in the field of view. In parallel, a probabilistic map of the occupation of the environment is generated to identify anomalies in the layout of the room, such as stairs, unidentified objects, irregular walls, etc.

[0058] The 3D space is then divided into a uniform voxel grid and the probability of occupation is calculated as proportional to the number of points of the point cloud which are inside the voxel.

[0059] As for the audio playback device 6, it can also be incorporated in the main case or it can be housed in a separate case and connected by radio or by cable to the processing unit 4. In this case, the power supply of the audio playback device 6 will be ensured by a dedicated battery or by a cable connection to a container provided with a battery.

[0060] As already mentioned, the apparatus 1 can also comprise a distance sensor 9, arranged near the cameras, and configured to:

[0061] - detect the distance of the apparatus 1 from surrounding objects;

[0062] - generate a corresponding distance signal 10.

[0063] In such a case, the processing unit 4 is connected to said distance sensor 9 and is configured to:

[0064] - receive said distance signal 10;

[0065] - cross the contents of said distance signal with the contents of the video signal 3;

[0066] - generate said processed signal 5 as a function of the contents of said video signal 3 and said distance signal 10.

[0067] Furthermore, the processing unit 4 is further configured to:

[0068] - recognise the type of surrounding environment on the basis of said information extrapolated from the images;

[0069] - recognise the type of obstacles present in the surrounding environment as a function of the recognised environment and a predefined database of objects present in said environment. For example, if the processing unit 4 recognises that the environment is a kitchen (following a comparison with a pre-established database) the obstacles could be furnishing elements typical of a kitchen, for example: a table, a chair, etc. Therefore, the processing unit 4 is capable of estimating what type of furniture it is by detecting the dimensions thereof and comparing it with a predefined database.

[0070] Preferably, the processing unit 4 is configured to recognise the type of obstacles by means of a YOLO algorithm (deep learning algorithm).

[0071] Furthermore, the processing unit 4 is further configured to generate specific acoustic alarms as a function of the detected distance of the user from the perimeters of the surrounding environment or from the obstacles.

[0072] In addition, the apparatus 1 comprises a speech synthesis system and said processing unit 4 is further configured to generate a speech audio signal 8 containing some specific verbal information, which is played back through said speech synthesis system, as a function of the contents of said processed signal 5 or of said output signal 7.

[0073] Summarising the above, the apparatus 1 comprises the following components:

[0074] • One or more video acquisition cameras positioned on a wearable element which allows the cameras to move integrally with the user's head;

[0075] • Optional distance acquisition sensors placed in an integral position with the camera(s) 2 / and

[0076] • A hardware system which acts as a driver for the camera(s) 2 / and for any sensors;

[0077] • A case containing the actual computer (processing unit 4), which will contain a motherboard equipped with CPU, volatile memory, non-volatile memory, sound card, network communication card(s) such as WiFi and / or Bluetooth and other optional sensors such as compass, accelerometers, GPS modules, etc. The solution of using a single case to host the device must not be considered binding; in fact, in some circumstances it is considered more convenient to split the calculation processes into two groups (the group of processes aimed at processing the images and the group of processes which deal with the generation of sound) into two distinct cases connected to each other by radio or by cable and each with hardware optimised for the purpose it must achieve. The device(s) is / are provided with a battery which supplies the energy necessary for the operation thereof;

[0078] • The sound playback device 6 to which headphones or earphones of different types, including those with bone conduction, can be connected. This device can also be incorporated in the main case or it can be housed in a separate case and connected by radio or by cable. In this second case, the power supply will be ensured by a dedicated battery or by a cable connection to the case provided with a battery;

[0079] Therefore, the general operating principle of the system is based on technology which acquires RGB (or RGB-D) frame sequences and processes them, constructing a mapping of the surrounding environment and providing the localisation and orientation of the subject acquiring the images with respect to the starting point (SLAM algorithm). Such an algorithm is used in particular within delimited spaces and, through prolonged acquisition over time, allows to make a three-dimensional digital reconstruction of the space which is saved in the device's memory.

[0080] Parallel to such a system, a second algorithm operates which, exploiting the same RGB source, is responsible for identifying obstructing volumes and / or steps to climb (obstacles), as well as "absences" of information which translate into steps to descend or holes.

[0081] Lastly, a third algorithm, also used in parallel with the first two, deals with the recognition of objects in real time (YOLO). Such a feature is particularly useful in outdoor environments. The three aforesaid acquisition channels operate in parallel but can be individually deactivated in circumstances where the specific feature is not required. The main processing system creates, for each of these channels, a sequence of data packets characterised by a high level of abstraction and containing the metric and conceptual information which each channel envisages.

[0082] The packets thus created are sent on three distinct channels to the sound generation system which, receiving the information on three inputs (which depending on the version can be three different socket communication ports or three distinct wired connections), establishes the processing priority and / or contextuality, generating, on the basis of the information received, sounds whose architecture is established by predefined but parametrizable rules and whose independent transmission to the right and left channels of the playback device 6 allows to simulate a sound received from the external environment and for which it is easy to intuit direction of origin and probable distance.

[0083] In more detail, in relation to the 3D sound generated, it can be said that a characteristic acoustic footprint is created which allows the receiver of the signal to distinguish the position, distance and nature of the surrounding obstacles and, in the event of indoor use, the perimeter of the room.

[0084] The apparatus 1 can further comprise a possible power generation system which is driven by the ambulatory motion of the user. For example, a piezoelectric generator is used which is embedded in the heel of a shoe and which recharges the battery / batteries of the system. Two charging methods are envisaged: o Direct connection, i.e., directly charging the batteries used by the system o Indirect connection, i.e., charging batteries which are exchanged with those used by the system when the latter are discharged. Therefore, an object of the present invention is also a user-wearable device comprising the apparatus 1 described above, which is fully referred to below. In this case, at least one camera 2 and / or distance sensor 9 is directly mounted on a main body of said wearable device and faces a predefined user advancement direction.

[0085] Said main body can be defined by a part or portion of glasses, shoes, watch or anything else not expressly indicated herein.

[0086] The present invention further relates to a method for assisting visually impaired persons. Such a method is directly derived from what is disclosed above in relation to the apparatus 1 to which reference is made hereinbelow. The method, comprising the following operating steps:

[0087] - arranging at least one image or video acquisition camera 2 positionable on a device wearable by a person; said camera 2 being configured to generate a video signal 3 in real time;

[0088] - arranging a processing unit 4 connected to the camera 2 and configured to: receive said video signal 3; process the images contained in said video signal 3;

[0089] The method further comprises the following further operating steps:

[0090] - extrapolating information from the images contained in said video signal 3 related to the shape and / or dimensions of the environment surrounding the user;

[0091] - making a three-dimensional digital reconstruction of the surrounding environment on the basis of said extrapolated information;

[0092] - extrapolating information from the images contained in said video signal 3 related to the obstacles present in the surrounding environment and associating the position of said obstacles within said three-dimensional digital reconstruction of the environment;

[0093] - identifying the position and orientation of the user within said three- dimensional digital reconstruction of the environment; - generating a processed signal 5 containing said three-dimensional digital reconstruction of the environment including the position and orientation of the user therein and the position of said obstacles therein.

[0094] The method further envisages arranging an audio or vibration playback device 6 associable with the user so that the latter can perceive such sound or vibration. The method comprises the following further operating steps:

[0095] - associating a set of respective audio or vibrational features with the information contained in said processed signal 5 according to a predefined logical map in which predetermined information of the three-dimensional digital reconstruction of the environment, position of the obstacles and position and orientation of the user correspond to respective predefined audio or vibrational features of frequency, power and time lag;

[0096] - generating an output signal 7 containing said set of respective audio or vibrational features;

[0097] - playing back said output signal 7 by means of said audio or vibration playback device 6.

[0098] The present invention achieves the set objects.

[0099] In particular, the present invention allows a person who has lost or who has never had all or part of the sense of sight but who retains the sense of hearing to become autonomous in walking (in some circumstances also without a stick) and in recognising the surrounding environment and the elements which populate it. In practice, the apparatus 1 provides a three- dimensional acoustic signal capable of providing the visually impaired with an audio "image" representative of the surrounding volumes and, optionally, also with verbal information which provides information on the nature of the surrounding objects.

[0100] The use of the apparatus 1 compensates, in part, the disadvantage suffered by the blind person with respect to those who have such a sense, allowing them to perceive the environment and consequently behave in a manner not dissimilar to those who are capable of seeing.

[0101] THE AGENT

[0102] Ing. Davide Anselmi

[0103] (Reg. no. 1068BM)

Claims

C L A I M S1. An apparatus (1 ) preferably for assisting visually impaired persons, comprising:- at least one image or video acquisition camera (2) positionable on a device wearable by a person; said camera (2) being configured to generate a video signal (3) in real time; a processing unit (4) connected to the camera (2) and configured to: receive said video signal (3); process the images contained in said video signal (3); characterised in that said processing unit (4) is further configured to:- extrapolate information from the images contained in said video signal (3) related to the shape and / or dimensions of the environment surrounding the user;- make a three-dimensional digital reconstruction of the surrounding environment on the basis of said extrapolated information;- extrapolate information from the images contained in said video signal (3) related to the obstacles present in the surrounding environment and associate the position of said obstacles within said three-dimensional digital reconstruction of the environment;- identify the position and orientation of the user within said three- dimensional digital reconstruction of the environment;- generate a processed signal (5) containing said three-dimensional digital reconstruction of the environment including the position and orientation of the user therein and the position of said obstacles therein; said apparatus (1 ) further comprising an audio or vibration playback device (6) associable with the user so that the latter can perceive such a sound or vibration; said processing unit (4) is further configured to:- associate a set of respective audio or vibrational features with the information contained in said processed signal (5) according to a predefined logical map in which predetermined information of thethree-dimensional digital reconstruction of the environment, position of the obstacles and position and orientation of the user correspond to respective predefined audio or vibrational features of frequency, power and time lag;- generate an output signal (7) containing said set of respective audio or vibrational features;- play back said output signal (7) by means of said audio or vibration playback device (6) so that the user can perceive said predetermined information of the three-dimensional digital reconstruction, by listening to the audio or perceiving the generated vibration.

2. The apparatus (1 ) according to claim (1 ), wherein the processing unit (4) is configured to associate a set of respective audio or vibrational features with the information contained in said processed signal (5) according to a predefined logical map, so that said output signal (7) has predefined audio or vibrational features as a function of the position, distance and nature of the obstacles and the surrounding environment so that the user can perceive the position, distance and nature of the obstacles and the surrounding environment as a function of the audio or vibration playback.

3. The apparatus (1 ) according to claim (2), wherein said output signal (7) is an audio signal (8) of the three-dimensional effect type.

4. The apparatus (1 ) according to any one of the preceding claims, wherein said processing unit (4) comprises at least two distinct processing subunits (4a, 4b) connected to each other, wherein:- a first processing subunit (4a) is configured to process the images contained in said video signal (3), generating said processed signal (5);- a second processing subunit (4b) is configured to generate said output signal (7) containing said set of respective audio or vibrational features.

5. The apparatus (1 ) according to any one of the preceding claims, wherein said at least one acquisition camera (2) is an RGBD-type camera (2).

6. The apparatus (1 ) according to any one of the preceding claims, comprising a plurality of image or video acquisition cameras positionable on a device wearable by a person; said processing unit (4) being connected to the cameras and configured to: receive the respective video signals; cross the information related to both video signals and process the images contained in said video signals.

7. The apparatus (1 ) according to any one of the preceding claims, comprising a distance sensor (9), arranged near the cameras, and configured to:- detect the distance of the apparatus (1 ) from surrounding objects;- generate a corresponding distance signal (10); said processing unit (4) being connected to said distance sensor (9) and being configured to:- receive said distance signal (10);- cross the contents of said distance signal with the contents of the video signal (3);- generate said processed signal (5) as a function of the contents of said video signal (3) and said distance signal (10).

8. The apparatus (1 ) according to any one of the preceding claims, wherein said processing unit (4) is configured to make a three-dimensional digital reconstruction of the surrounding environment using a SLAM-type algorithm.

9. The apparatus (1 ) according to any one of the preceding claims, wherein said processing unit (4) is further configured to: recognise the type of surrounding environment on the basis of said information extrapolated from the images;recognise the type of obstacles present in the surrounding environment as a function of the recognised environment and a predefined database of objects present in said environment.

10. The apparatus (1 ) according to claim 9, wherein said processing unit (4) is configured to recognise the type of obstacles by means of a YOLO- type algorithm.

11. The apparatus (1 ) according to any one of the preceding claims, wherein said processing unit (4) is further configured to generate specific acoustic alarms as a function of the detected distance of the user from the perimeters of the surrounding environment or from the obstacles.

12. The apparatus (1 ) according to any one of the preceding claims, comprising a speech synthesis system, wherein said processing unit (4) is further configured to generate a speech audio signal (8) containing some specific verbal information as a function of the contents of said processed signal (5) or of said output signal (7).

13. The apparatus (1 ) according to any one of the preceding claims, wherein said audio or vibration playback device (6) comprises two sound or vibration emission channels; said processing unit (4) being configured to generate said output signal (7) containing audio or vibrational features which are different or with a time lag with respect to said two sound or vibration emission channels.

14. A device wearable by a user comprising the apparatus (1 ) according to any one of the preceding claims, wherein said at least one camera (2) and / or distance sensor (9) is directly mounted on a main body of said wearable device and faces a predefined user advancement direction.

15. A method preferably for assisting visually impaired persons, comprising the following operating steps:- arranging at least one image or video acquisition camera (2) positionable on a device wearable by a person; said camera (2) being configured to generate a video signal (3) in real time;- arranging a processing unit (4) connected to the camera (2) and configured to: receive said video signal (3); process the images contained in said video signal (3); characterised in that said method comprises the following further operating steps:- extrapolate information from the images contained in said video signal (3) related to the shape and / or dimensions of the environment surrounding the user;- make a three-dimensional digital reconstruction of the surrounding environment on the basis of said extrapolated information;- extrapolate information from the images contained in said video signal (3) related to the obstacles present in the surrounding environment and associate the position of said obstacles within said three-dimensional digital reconstruction of the environment;- identify the position and orientation of the user within said three- dimensional digital reconstruction of the environment;- generate a processed signal (5) containing said three-dimensional digital reconstruction of the environment including the position and orientation of the user therein and the position of said obstacles therein; the method further envisages arranging an audio or vibration playback device (6) associable with the user so that the latter can perceive such sound or vibration; the method comprises the following further operating steps:- associating a set of respective audio or vibrational features with the information contained in said processed signal (5) according to a predefined logical map in which predetermined information of the three-dimensional digital reconstruction of the environment, position of the obstacles and position and orientation of the user correspondto respective predefined audio or vibrational features of frequency, power and time lag;- generating an output signal (7) containing said set of respective audio or vibrational features; - playing back said output signal (7) by means of said audio or vibration playback device (6).

Images