IT202100013115A1 - Image capture method - Google Patents

Description

activating a camera shutter

PRIVATE AND CONFIDENTIAL

Image capture method

The invention relates to an image acquisition method and a corresponding device and system for acquiring an image, the system comprising said apparatus. The invention also relates to a program which implements the image acquisition method and to a storage medium comprising said program.

The photography ? loved by many people, especially nowadays thanks to the possibility? to take a picture at any time using everyday devices other than traditional cameras, such as mobile phones, tablets or laptops. Often, people want to photograph a subject in a special situation, but are not always able to capture the so-called "magic moment" to make the image memorable. In fact, in many cases, in addition to the technical aspects, such as the correct exposure to the light, the focus of the camera or the right background, the best images are the result of the emotional sensations that the photographer feels in that particular situation when he observes the subject . Emotional sensations are often unconscious sensations that an experienced photographer can usually handle for the purpose of taking a picture. On the other hand, for most people with no special skills? photographic, ? difficult to recognize these sensations and correlate them with the activation of the photographic shot.

In the prior art, ? I notice that different methods and systems recognize emotions. In particular, the recognition of emotions ? a technology to predict people's emotional states based on user responses such as verbal or facial expressions. This technology can be applied in various fields, such as health care, game and education. To help these applications, the technology recognizes emotions in real time and naturally as the user experiences them. For this purpose, ? It is possible to employ wearable devices that rely on bioelectrical signals that indicate physiological responses of the user.

However, all of these known methods and systems are not efficient enough since they are limited in detecting and analyzing only the passive or involuntary reactions of a user (i.e. emotional sensations as a consequence of the visual stimulus) leaving out other important aspects such as the voluntary reactions to said stimulus in terms of cognitive factors. Furthermore, prior art methods and systems are not configured to bypass the conscious and biomechanical steps that lead the user to press the button to take the picture and change the photographic parameters (e.g. zoom, focus, etc.) to capture the image, or part of the image (for example due to the zoom of a part of the image), of the subject you are observing.

Examples of the present disclosure seek to address or at least alleviate the above problems.

According to a first aspect of the invention, there is provided an image acquisition method that uses an image acquisition apparatus comprising an imaging device having an image sensor for generating image data, the method comprising: displaying, using a imaging device, images captured by the imaging device based on the image data; the generation of attribute data from a plurality? of biological signals measured by a plurality? of biological sensors, the attribute data being indicative of attributes of a user at a time when the user is viewing images displayed by the image reproducing device; the generation of data of significance? based on the data attributes and the association of the data attributes from which the data of significance are generated? to each captured image; controlling the photographic parameters of the image acquisition apparatus based on the attribute data;

the analysis of the data of significance? to determine whether displayed images should be captured and archived; and storing the image data of the displayed images together with the attribute data in a storage device on the basis of the significance data? analysed.

In a second aspect of the invention, there is provided a program comprising processor-readable instructions which, when implemented by a processor, cause that the processor implements the method as defined above.

In a third aspect of the invention, a storage medium comprising the program as defined above is provided.

In a fourth aspect of the invention there is provided an image acquisition apparatus, comprising: an imaging device comprising an image sensor operable to generate image data; an image playback device operable to display images acquired by the imaging device based on the image data; a processor operable to: generate data attributes from a plurality? of biological signals measured by a plurality? of biological sensors, the attribute data being indicative of attributes of a user at a time when the user is viewing images displayed by the image reproducing device; generate data of significance? based on the data attributes and associate the data attributes from which the data of significance are generated? to each captured image; controlling the photographic parameters of the image acquisition apparatus based on the attribute data; analyze the data of significance? to determine whether displayed images should be captured and archived; and storing the image data of the displayed images together with the attribute data in a storage device (7) according to the significance data? analysed.

In a fifth aspect of the invention, there is provided an image acquisition system comprising: a plurality of? of biological sensors; And

an image acquisition apparatus as defined above.

Other aspects and features are defined in the appended claims.

Examples of the description can make it possible to use attribute data, i.e. not only physiological attributes of the user, but also cognitive and emotional attributes, to adjust photographic parameters (e.g. focus, zoom in/out, exposure in the light...) of the image capturing apparatus, such as a camera, and to capture an image, to be stored with the corresponding image data.

Examples of the disclosure will now be described, by way of example only and with reference to the accompanying drawings in which like reference numerals refer to like parts, wherein:

Figure 1 shows an image acquisition method flowchart according to an example.

Fig. 2 shows a schematic representation of the image acquisition apparatus and the image acquisition system according to an example.

FIG. 3 shows a schematic representation of the data flow for processor operation according to an example.

Fig. 4 shows a schematic process diagram for triggering a take of an imaging device using the image acquisition method according to an example.

Figure 5 shows a system for tuning the model parameters.

Figure 6 shows a schematic diagram of the process for coping with asynchronous signals.

Figure 7 shows a schematic diagram of the process of triggering a take of an imaging device with a single user and a plurality of users. of additional users.

Figure 8 shows a flowchart of an image acquisition system with a mobile phone as the imaging device.

Figure 9 shows a flowchart of an image acquisition system with a photographic device as the imaging device.

Figure 10 shows a schematic representation of a computer system.

An image acquisition method, an image acquisition apparatus and an image acquisition system are described. In the following description, some specific details are presented in order to provide a comprehensive understanding of the examples of the description. will result? however, it will be apparent to one skilled in the art that these specific details are not to be employed to implement the examples of the disclosure. Conversely, specific details known to the person skilled in the art are omitted for the sake of clarity in presenting the examples.

Figures 1 , 2 and 3 are used to describe the steps of the method 100 for acquiring an image using the corresponding apparatus 1. The image acquiring apparatus 1 comprises an imaging device 2 having an image sensor 3 for acquiring a image and for generating image data, and an image reproduction device 4 for receiving the image data and for displaying the image. Furthermore, the apparatus 1 comprises a processor 5 connected to both the imaging device 2 and the image reproduction device 4 for processing information data derived from these two devices. Processor 5 ? also connected to other devices, as a plurality? of biological sensors 6 and a storage device 7. It should be noted that an image acquisition system 9 comprises the aforementioned apparatus 1 and the plurality of biological sensors 6.

In step S101 of method 100, the imaging device 4 is used to display the images acquired by the imaging device 2 based on the image data derived from the imaging device 2. At this time, a user is observing the displayed images by the imaging device 4. The biological sensors 6 are used to measure the biological signals of the user, and the attribute data is generated in step S102 using said signals. In particular, the attribute data is indicative of attributes of a user at a time when the user is viewing the images, and/or a subset of the image, displayed by the image reproduction device 4. In step S103, the data of significance are generated on the basis of the attribute data and the attribute data are associated from which the significance data is generated? to each captured image. Are the photographic parameters of the image acquisition apparatus 1 checked based on the attribute data in step S104 and the significance data? are analyzed in step S105 to determine whether the displayed images are to be acquired and stored. In step S106, the image data of the displayed images is stored together with the attribute data in a storage device 7 based on the significance data? analysed. Attribute data can be stored independently of the displayed images. Storage device 7 can? be part of the apparatus 1 or pu? be part of system 9. Alternatively, storage device 7 can? also be located remotely from the apparatus 1 and the system 9, for example in a remote server in a cloud platform.

The operation of the processor 5 of Figure 2 is better clarified in Figure 3 . According to this figure, the imaging device 2 ? operable to acquire image data (ID) which is displayed, for example through a screen. One or more? acquired images (CI) are shown on the screen by a display module 51. The imaging device 2, which ? a photographic type device (e.g. cell phone, camera, etc.) and related hardware/software for selecting the multiplicity? of shooting parameters (such as zoom in / out, focus, exposure, crop, tilt, filtering, etc.) pu? represent the physical implementation of the apparatus 1.

With the term ?imaging device? it is intended to cover any type of device that may cause an image to be reproduced, such as an active/passive screen, front/rear projection apparatus, direct-to-the-eye or direct-to-the-lens projection.

A set of 6 biological sensors allows the acquisition of the activity? biological (that is to say biomechanics, neurophysiological and cognitive) of a user (or more? users) and communicates with the apparatus 1 with a modality? in real time, which can? be considered a so-called mode? synchronous. For example, Figure 3 shows the use of four sensors, namely a sensor for measuring heart rate, a sensor for measuring pupil diameter, a gaze detector and a microphone. Each sensor? operable to generate a corresponding biological signal (BS). The user interacts with the imaging device 2 through a real-time processing system of signals extracted from the biological sensors 6. asynchronous, for example due to biological or hardware/software sensing latency, a given data processing allows for a virtual synchronization, in which there is? first a post-processing of raw data and then predictions based on previously acquired data. In particular, to remedy synchronism problems, the method can? further comprising temporal synchronization of the biological signals with each other and/or with the image data. Here, a mode? asynchronous can? be considered a situation where the timing of the biological signals does not match the timing of the image display. It should be noted that for real and future (artistic, emotional…) post-processing and post-analysis purposes, all data can be stored as associated metadata. In this way, in addition to a real-time forecast of the signals, ? It is possible to subsequently synchronize the signals in order to process and analyze the signals at a later time.

In order to receive biological signals from one or more? biological sensors, in an example the apparatus can? include a communication interface.

The attribute data (AD), indicative of the user's attributes at a time when the user is observing the displayed images, are generated from the biological signals by means of an attribute data generation module 52 and the significance data (SD) are generated by associating the attribute data (AD) with each acquired image (CI) by means of an association module 53. The attribute data (AD) is also used to control the photographic parameters of the imaging device 2 by means of a control module 54. It should be noted that photographic parameters can be controlled using dedicated actuators, such as an assembly of optical, digital and/or mechanical components. At this point, the data of significance? (SD) are analyzed by an analysis module 55 and the image data (ID) is stored together with the attribute data (AD) based on the result of the analysis of the significance data? (SD) in a storage module 56. As mentioned above, the image data is stored together with the user-derived attribute data. According to one example, the attribute data comprises the signal data of the measured biological signals. For example, the attributes of the user when he is observing the images displayed may refer to the heart rate, which part of the image he is looking at, whether he makes vocalizations in reaction to viewing the image, etc. Also, will you appreciate it? that the user's attributes could refer to his cognitive processes and/or emotions when viewing the images.

Rather than the interaction between the user and the imaging device 2 which occurs through interactions of a biomechanical nature (e.g. pressing the button to take the picture, rotating the lens to zoom in/out of the focus, adjustment commands/touch screen, etc.), the examples of the present description allow a user to interact with the apparatus, for example through an artificial intelligence engine (AI) which transforms signal data into cognitive information emotional and decision-making. For example, if a biological sensor 6 such as an eye tracking sensor detects a prolonged fixation of a certain portion of the framed scenery, the apparatus 1 can? move its focus to that portion by applying a zoom in of magnitude proportional to the duration of the fixation. It should be noted that this is not understood as an explicit command from the human operator ? that does not ? trained, or needs to be trained, to learn how to change the mode? to obtain this result. On the other hand, the apparatus 1 pu? process the acquired data to determine the level of significance? associated with such physiological data and consequently determine the correct/correlated set of trigger parameters and activate an output related to the decision to trigger or not trigger. In fact, the set of biological signals is used as a multivariate parameter for the classification of the triggering of the photo: trigger event or non-trigger event. There? it means that a particular cognitive biological state corresponds to the execution of a photographic shot. It should be noted that the user ? still allowed to use imaging device 2 in mode? standard, for example deciding when to press the shutter button directly. In this case, the user's emotional/cognitive metadata is still associated with the photo shot for future/post-processing purposes unless an explicit command not to proceed is set. Although this feature? can be implemented using artificial intelligence, will be appreciated? that any appropriate form of machine learning could be used, for example, based on trained datasets. Will we make it? also account that this functionality? it could be implemented with decision logic, such as fuzzy logic, or any other appropriate logic.

The term ?biological sensors? intends to indicate a set of multiple physiological, behavioral and environmental sensors that allow the acquisition of signals or data directly and/or indirectly related to the physiological, cognitive and emotional attributes of the local user. Consequently, attribute data represents metadata containing information regarding different attributes (e.g., physiological, cognitive, emotional) of the user.

In one example, to improve the effectiveness of the method, the significance data? can include a plurality? of significance levels, each associated with a respective biological signal from which the attribute data was generated. Consequently, the analysis of the data of significance? can? understand the generation of a level of significance? combined on the basis of levels of significance? for each of the biological signals. Also, storing image data together with attribute data can understand the storage of the displayed images in the storage device together with the attribute data based on a comparison of the level of significance? combined and a triggering threshold level. There? does it mean that the level of significance? combined could be greater than, less than, equal to, or some function of the threshold level in order to store the displayed images together with the attribute data. In an example, if the level of significance? combined exceeds the threshold level of activation, the method pu? provide an automatic setting of photographic parameters and trigger the shooting of an image without the user having to press a dedicated shutter actuator. However, it is emphasized that all data and information acquired by the image acquisition apparatus can be archived regardless of the level of significance achieved. In this way, examples of the description allow both to conduct experimental analyzes on the acquired data to generate new further and additional models of photographic triggering (i.e. ex-post analyses), and to reuse the archived data to generate new photographic snapshots based on shutter activation patterns other than the one selected during the real-time session. The level of significance? can? therefore be decided ex-post, where ex-post means in near real time and/or at any subsequent moment. In principle, it can be re-examined years later in light of new findings. In other words, the level of significance? can? also be zero at the present moment, since? Not ? excluded that the level of significance? can be determined differently, i.e. above a certain threshold in the future (near real time years away) when new knowledge is acquired/discovered which gives the same archived data a different level of significance.

In another example, the generation of significance data? and the analysis of the significance data? they can be based on a predetermined shooting pattern. For example, the shooting model can? be a model of artificial intelligence/machine learning. In particular, the predetermined shooting pattern can? be generated on the basis of biological signals acquired by a user or by a plurality? of users when viewing a set of model training images for predetermined shooting model training. In this way, the model can act as an activation system for a photographic shot, capable of capturing a precise scenario observed by the user at a precise moment in time corresponding to a precise identifiable and quantifiable biological state through the identification of a pattern in the signal values and in those extracted by themselves.

In one example, the activation threshold level ? determined according to the predetermined shooting pattern. In this way, the threshold level can? be determined automatically and possibly continuously updated by said model. Also, in order to fine-tune the predetermined trigger pattern, the method can? understand the transmission of image data of displayed images along with their associated local biological signals to a database of shot patterns. For example, the predetermined shooting pattern can? be generated using machine learning based on one or more? data set, e.g. generated from previously generated data attributes that are associated and/or related to physical operating parameters of the imaging apparatus such as the time the image was taken? shot, exposure settings, zoom control, focus, etc. Examples are described in more detail pi? after you.

In this context, can be introduced the concept of pattern of significance? multiparameter of the recorded signals. In other words, the values of the signals used to activate the trigger are discretized according to some thresholds (i.e. signal thresholds or activation threshold levels), which identify the boundaries (states) of significance? of each signal. For example, the range of values of each signal can? be divided into three ranges: significance? low, significance? mean and significance high. It should be noted that a high value of significance? indicates a favorable signal condition for triggering. Assuming a sensor dashboard that allows the acquisition and processing of six signals, ? possible to construct a matrix of dimensions 6x3. This matrix, updated at each timestamp, represents the configuration of the signal values discretized through the values 1 and 0, as shown in table I.

Table I: levels of significance for each biological signal

Each signal is assigned a weight. Then, a value is assigned to each level of significance for each signal. These values represent the importance of the signal in providing information on the emotional-cognitive state. By attributing a weight to each level of significance for each signal, the above matrix is transformed into a matrix of significance? as follows (Table II).

Table II: significance matrix

Once the values of the individual items of the significance matrix have been obtained, they are added together providing a numerical value, called significance? overall (OS). It should be noted that the significance overall varies over time, following the synchronous sampling rate of the signals. Each time the array is updated, it is compared to a reference value, i.e. the activation threshold level. If the significance exceeds the threshold level, the system (e.g. processor 5) provides an automatic setting of shooting parameters (e.g. focus, exposure, etc.) and activates photo shooting without the user having to press the shutter button.

Although Tables I and II provide examples of levels of significance? and matrices of significance, you will appreciate? That ? Can you use any number of levels of significance? and any number of signals, for example depending on the model and desired control of the apparatus. Will you appreciate it? also that the levels of significance? they could be different to control different aspects of the apparatus. For example, can weights of greater significance be given? signals associated with eye gaze and where the user is looking when controlling things like zoom and focus. As another example, when trigger activation (such as shutter activation) is determined, it can? be given more weight to signals associated more with emotion, such as heart rate.

The process 200 for activating a click is summarized and illustrated in Figure 4 . surroundings using an imaging device 2, i.e. a photographic system. The information on the activity biological signals of the user are measured by biological sensors 6. The raw biological signals are extracted and synchronized with each other and with frame sequences acquired by the imaging device 2 in an acquisition and filtering module 201. A matrix of significance? ? built in a module 202 of the matrix of significance? and significance data is generated, based on the measured biological signals and model parameters, i.e. signal thresholds, significance level weights, and significance data. and the triggering threshold level, analyzed in the model parameters module 203. Is the significance calculated? overall and a comparison is made with the trigger threshold level in control module 204. A trigger by imaging device 2 is triggered by trigger module 205 based on a particular relationship between the significance value? significance and the triggering threshold level (for example, the overall significance value is greater than the triggering threshold level). Shooting parameters, i.e. zoom, focus, exposure, are automatically fine-tuned according to the synchronization and filtering process before shooting in module 206 of shooting parameters.

All data is recorded as photographic metadata in a given photographic format which contains not only the raw data of photographic signals, but also the raw data extracted from humans. This format allows for a variety of individual and collective post-processing tools, as well as training data to further improve the accuracy, sensitivity? and predictability? at the individual and collective user level. The signal thresholds, the weights of each level of significance? and the triggering threshold level (i.e. model parameters) can be adjusted using one of the following two types of decision logic.

In particular, the image acquisition system has a modular configuration comprising a user interface for displaying audio-visual stimuli, an operator interface for monitoring and managing each stimulus, and an arrangement of sensors for recording biological signals. In addition to managing the interactions between these three elements, an AI module automatically processes the acquired and synchronized biological signals recorded by a user during a visualization of an audiovisual stimulus (input) which defines biological dynamics and indices. The purpose of this processing? to return a revised visual representation of the stimulus (output) modified based on the subjective cognitive perception of the original stimulus, as well as? some parameters related to cognition. The output is generated by matching the information extracted from the user's biological cues during stimulus delivery and the stimulus itself, resulting in an augmented visual representation of the input. The signal processing procedure varies from signal to signal, but generally ? constituted by an initial phase in which a single signal is filtered to improve a particular behavior and by a second stage in which various signal analysis algorithms are applied in order to extract specific indices which briefly describe the trend of the signal. Both the filtered signals and the extracted indexes are considered as information couplers and can be considered metadata associated with that specific process. The system allows you to simultaneously acquire one or more? biological signals and the AI model processes each signal in order to obtain the desired output.

Differently from prior art approaches, the present apparatus and system aim not only to process and return indices reflecting some biodynamic behaviors of the user, but several indices extracted from the signals and the signals themselves are used as moderators of the user's subjective perception of the stimulus proposed. Indeed, through the particular stage of elaboration, ? possible to produce a cognitive reworked version of the input.

Figure 5 illustrates a process 300 employing the model parameter tuning system. According to a first logic, the model parameters are set manually 301. For example, the configuration of these parameters can? be identified by the producers of the apparatus 1 following experimental laboratory tests on the nature and information conveyed by the signals 302. Alternatively, the user has the possibility? to adjust the parameters by yourself 303.

According to a second logic, on the other hand, the different configurations of the model parameters can be identified by machine learning systems 304 trained using signals acquired by the user (or users) during their shooting sessions 305. In other words, by default each user who uses the device 1 pu? share their metadata - signals and personal information - in a database accessible to authorized parties. In addition or alternatively, the activity? similar to the clinical trial pu? be performed on a cohort of volunteer users recruited to determine these parameters with the appropriate level of statistical power 306. Depending on the cluster or grouping to which the users or sessions belong, ? It is possible to produce a different configuration 307. As a result, 308 the parameters of the model are determined.

According to another example, the user ? a local user who ? local to the image acquisition apparatus, and the attribute data comprises local attribute data which is associated with a time when the local user is observing the images displayed by the imaging device. In other words, for example, the user can? physically hold the image acquisition apparatus, which can? be a camera, smartphone, camcorder, etc.

In another example, the user ? a remote user who ? remotely from the imaging apparatus, the imaging device ? local to the remote user, and the attribute data comprises remote attribute data which is associated with a time when the remote user is viewing images displayed by the imaging device.

In yet another example of the apparatus, the user can? be a local user that ? local to the image acquisition apparatus, and the attribute data may comprise local attribute data which is associated with a time when the local user is observing the images displayed by the imaging device. In particular, the apparatus pu? include a communication interface operable to receive remote attribute data from a plurality? of remote user devices associated with several remote users, each remote user device having a respective remote imaging device for displaying images captured by the imaging device, and each remote user device being operable to generate remote attribute data from a plurality of remote user devices. of remote biological signals measured by a plurality? of remote biological sensors, the remote attribute data being indicative of attributes of the remote users at a time when the remote users are viewing images displayed by their respective remote user imaging devices. In this case, the significance data? can be generated based on local attribute data and remote attribute data.

In this context, the terms "local" and "remote" are to be understood as being within or outside a certain distance from a reference point. For example, a local user ? understood as a user in close proximity, or within a threshold distance, of a device, i.e. the imaging device. In this case, the user can? interact directly with the device, by touching it directly, through connecting means, such as a cable, or through a wireless communication, preferably a short-range wireless communication, such as Bluetooth, near-field magnetic induction? On the other hand, a remote user ? understood as a user away from, or outside a threshold distance of, a device, i.e. the imaging device. For example, the remote user can employ a real-time communication system (such as a local or widespread 5G telecommunication infrastructure) to acquire information from the image acquisition device and finally transfer data to a cloud and edge computing architecture connected to the image acquisition device Images. It should be noted that the imaging device ? preferably located in close proximity, i.e. "local", of the remote user to enable a playback of image data from the image acquisition device.

For example, the image capture device can be positioned to capture images of a wildlife scene, such as in an environmentally challenging location or where the presence of a user present would be disturbing to wildlife. Captured images can be transmitted from the image capture device to the image playback device located in a more convenient place. sure it won't disturb the wildlife.

In a configuration where there is a local user and a remote user, the method can Also include viewing the images captured by the imaging device on a plurality of? of remote user imaging devices that are associated with several remote users. Furthermore, the method can understand the generation of remote attribute data from a plurality? of remote biological signals measured by a plurality? of remote biological sensors, the remote attribute data being indicative of attributes of the remote users at a time when the remote users are viewing images displayed by their respective remote user imaging devices. In particular, the method pu? understand the fine-tuning of the predetermined model based on user input by the different remote users. There? can? help generate a model more? complete, effective and accurate why? data from different users and different images can be obtained more? easily.

This method 100 for acquiring images can? hence be employed in different scenarios.

In a first scenario (1:1 scenario), the user ? the only source of biological information to trigger the shot. In this individual dimension, the user is connected to biological sensors 6 and the apparatus 1 receives and processes data deriving from said sensors 6, which extract information from the user datum, to decide through the overall meaning value when and whether to activate a trip. The set of metadata recorded during each shooting session is archived in a personal database and the user himself can accept or modify the parameters of the model based on experimental observations, creativity? personal or based on the training of machine learning algorithms.

In a second scenario (1:N scenario), does triggering not only take place on the basis of the significance value? total of the single user using the device. In fact, in this scenario, different users (ie the protagonists viewers) view in real time what? that a primary user is observing (e.g. the concept of screen sharing). At the same time, each spectator protagonist ? equipped with several sensors that acquire biological signals and through a real-time communication system (for example, a local or widespread 5G telco infrastructure), transfer information to the cloud, and an edge-computing architecture is connected to the imaging device 2 of the main user. The information transferred by the protagonist spectators consists of the overall meaning value which is updated at each timestamp. Again, the model parameters can be adjusted according to one of the two logics described above. The shot, therefore, will not come? activated by the comparison of the significance? total of the main user with only the activation threshold level. Conversely, at each timestamp primary user imaging device 2 comprises a significance vector? collective, together with the individual values of the overall meaning of individual users.

Table III shows the significance vector? collective - five viewers (V) and one main user (MU) - at a specific time.

Table III: vector of significance collective

Using one of the two logics for setting up the model parameters, the MU can? decide not only the parameters of the model of the other spectators, but also how much importance to attribute to the single components of the vector of significance? collective, as shown in Table IV below.

Table IV: weighted significance vector collective

The sum of the individual components of the vectors pu? then be compared with an activation threshold level to activate a photographic shot based on emotional and cognitive reactions and processes acquired by a community? of people.

A third scenario (N:N scenario) allows a network of users to share the signals acquired while viewing the same reality among themselves (devices). observed (from multiple visual and biological perspectives). Each user can decide which type of fine-tuning of the adjustment parameters to opt for, depending on the photographic shot you want to activate.

A fourth scenario (user profiling of scenario neurocognitive research based on ML) can be applied to any of the three scenarios described above. Does this scenario allow for the extraction of metadata associated with each user for research and sensitivity improvement purposes? and accuracy (in general, performance) of the system, as well as? for the extraction of new socio-cognitive emotional patterns/models and the creation of new performances and characteristics of the individual and collective system. Once stored in a database, this data can be used to find common cognitive patterns among individuals and therefore new methods of activation based on social metadata.

It should be emphasized that in the event that the available technology does not allow the processing and sharing of biological data extracted from users in near real time, the structure of the flow of analysis of the significance? for the photo-activation pu? be implemented in several ways. In particular, the signals can be stored and analyzed in time slots synchronized with the single image frames taken by the imaging device 2. In this way, the signals can be processed retrospectively, isolating the frame corresponding to the exceeding of the trigger threshold . Alternatively, implementing a real-time signal prediction system can prevent the use of the decision logic described above, as shown in Figure 6. According to this process 400, can? 401 a new time window may be used to synchronize signals with image frames 405. In other words, this can? be considered an example of temporal synchronization of biological signals with each other and/or with image data. The new time window is used if the comparison between the significance? overall and the activation threshold value does not meet the requirements determined 402. On the other hand, if the comparison between the significance? and the trigger threshold value satisfies certain requirements 403, a shot 404 is performed in combination with image frame information.

As mentioned above, the generation and analysis of significance data? they are based on a predetermined shooting pattern. This model can be an AI/ML model performed a priori using a set of recordings which, processed for the purpose of information extraction (i.e. signal processing and feature extraction), serve as training (and test) sets to feed and adjust parameters of the model itself. Subsequently, in a second phase, the performance of the model and its structure can be improved/modified through updates resulting from an increase in the size of the training (and testing) set, rather than using new technologies to extract new signals or by the use of new and innovative mathematical methods to extract information from the signals themselves.

The term "recording" refers to the set of different biological signals acquired through a set of 6 non-invasive biological sensors, for example wearable sensors, by a user while taking a picture, starting from the moment the user zoom in to view a scenario through a lens until the user decides to capture what? which is framed through the photographic shot itself.

The purpose of the model building phase ? to generate an AI system capable of autonomously identifying one or more? patterns within the values of the biological signals, which suggest a relationship between the photographic shot and the user's biological-cognitive-emotional state. In this way, the model can act as a photographic trigger activation system, capable of capturing a precise scenario observed by the user at a precise moment in time corresponding to a precise identifiable and quantifiable biological state through the identification of a pattern in the signal values and in those extracted from the themselves. Furthermore, each recording ? associated with a defined number of quantitative and qualitative information relating to the user such as photographic experience, age, number of photos taken, etc. This information acts as a moderator for the statistical analysis, grouping the various recordings into specific clusters from which ? You can generate and moderate different outputs/shots (i.e. different model configurations depending on the type of output you want). For example, can be used a shooting model configuration trained on recordings of ultra-experienced photographers, or rather of photographers highly interested in natural landscapes. It should be noted that each pattern configuration has additional conformations in numbers equal to the number of combinations (without repetition and without concern for order) of extracted signals. For example, ? It is possible to exploit a setup that triggers a photograph based on the cluster of ultra-skilled photographers by analyzing only selected biological signals (for example, the photoplethysmographic signal (PPG) and the galvanic skin response signal (GSR)), rather than a superset of these signals.

In the training and model building phase, the algorithm is fed with data extracted from the signals of the individual recordings. In particular, it optimizes the size of the time window before the trigger used to isolate a part of the signals. Then, a combination of features extracted from those signals that best classify the trigger event based on the value patterns of the features is considered. Before this stage? therefore necessary to establish a primary database of different registrations.

The machine learning model ? then a supervised binary classification system: to shoot or not to shoot. This model can be represented by a random forest of classification through which, with the appropriate and refined development of the regulation parameters of the model itself, ? possible to obtain a predictive system capable of activating a photographic shot. For each cluster identified by grouping recordings based on moderators, a specific model configuration is generated, thus able to identify a specific pattern in the signal values, as well as? a particular condition of the person's emotional and cognitive state.

The result of the whole process? a set of templates that new users can use. Once trained and made available, these models are integrated into a software system embedded in a photographic device, for example in the device's firmware. Regardless of how it is used (ie choosing the specific cluster to refer to for the production of the photograph and then choosing the type of model), a user ? capable of using the apparatus 1 by simply equipping itself with biological sensors 6 capable of extracting the signals used for training the model or a subset thereof. In order to increase the performance of different model configurations, user recordings enriched with their personal information (i.e. moderators) can be archived to increase the size of the primary database. In this way, this database can be used periodically as a new database for new model building phases. The newly generated model configurations can then be integrated into photographic devices and regarded as system updates.

As mentioned above, a 9 image acquisition system can? be represented by the apparatus 1 for acquiring images combined with one or more? 6 biological sensors applied to a user. While the user observes real objects and/or landscapes around him (i.e. the same objects and landscapes captured by a photographic lens of the imaging device 2 in the form of images), the information about his activity? biologics are extracted from your body non-invasively. This is done using wearable sensors and/or sensors not directly connected to body parts such as sensor systems belonging to the IoT world.

Biological signals can be acquired using different non-invasive sensors 6 that exploit different technologies. Are these sensors 6 connected directly or indirectly with biomechanical, cognitive and emotional processes - conscious or unconscious? activated in humans or animals (for example, pets such as dogs or cats) by visual and other sensory stimuli. because every animal ? a sentient being, ? It is possible to trigger a photographic shot using the animal's "emotional" feedback alone or mixed with human feedback. In this way ? It's simple to scale the process using not only human cognitive-emotional feedback but more? of one, resulting from monitoring the emotion and cognition of N people.

In particular, the sensors 6 are incorporated into the triggering system ("shooting") (DSLRcam/smartphone) or can be distributed in the environment through the use of wired/wireless sensors (e.g. wearable sensors and/or IoT sensors that extract signals/data from other detection systems positioned in places where the user is potentially interested in taking pictures).

For example, a photographic camera can receive data from:

1. several non-invasive wearable sensors worn by a primary user handling the camera;

2. several non-invasive wearable sensors integrated into several technological dashboards mounted on other secondary users and connected wired/wireless with the primary user's camera; and/or

3. different IoT environmental sensors capable of translating the body language and the emotional-cognitive dimensions of different individuals (for example in a field/square/street), where there are different objectives. When the environmental and/or wearable sensor system detects a particular response in individual or crowd behavior, a photo is triggered by one or more people. goals. In other words, the system ? able to interpret and rationalize individual or collective cognitive-emotional behaviour.

The camera can be handheld or worn by the direct user, incorporated into other handheld systems or incorporated into wearable accessories (for example, hats, goggles, helmets, jackets, gloves, rings, bracelets), as well as? remotely operated by a human, animal or robotic operator (for example mobile robots, aerial or aquatic drones, vehicles in general including boats) being controlled in real time by an individual or by a multiplicity of connected users. For example, a drone or a dog? equipped with a camera and has the ability? to move (autonomously or guided). Based on the cognitive-emotional process elaborated by a system of technological sensors, the camera is activated to shoot the acquisition signal from an individual or a crowd.

In the following, an indicative list of sensors that can be used as biological sensors according to the present method, apparatus and/or system is mentioned. The list is not intended to be exhaustive in terms of relevant signals (other discoveries will provide evidence of new psycho-physiological parameters in the future affecting the scope of this patent). Can you? consider, for example, how the use of non-linear analysis tools (i.e. quantifying the complexity, the variability, of the signal) on the simple heartbeat signal allows the extraction of fundamental information on the neurological state of the person, even reaching to the detection of emotions through only the same signal. Another example is that new technologies are able to non-invasively measure and quantify biological variables that were previously unthinkable to measure (for example pupil diameter, SpO2). Furthermore, some discoveries will likely be generated by the platform described in the present description. In this way, the list of sensors is not intended to be exhaustive in terms of usability? (because in the future the miniaturization of electronics in sensing, processing, wireless connection, energy harvesting will favor the integration of a multiplicity of sensing systems and eventually improve the usability and experience activation system user ("click")). Just think, for example, of the innovation and progress of telephony (from the phone booth to the telecommunications system integrated into a smart watch).

The biological sensors 6 can comprise at least one of the following:

? GSR signal (Galvanic Skin Response) ? PPG signal (photoplethysmography, Photoplethismography)

? Position of gaze

? EEG signal (Electroencephalography, Electroencephalography) ? Facial expression

? ECG signal (electrocardiography, Electrocardiography)

? EMG signal (Electromyography)

? Pupil diameter

? Body movement language (inertial, haptic, gestural, etc.) ? Natural Spoken Language (Microphone)

? Natural Emotional Language (microphone)

? fNIRS signal (functional Near Infra-Red Spectroscopy)

? Blood oxygen level

? Biochemical parameters (eg cortisol, dopamine, norepinephrine) ? Geolocation signal (GPS, WiFi, Bluetooth, 5G)

? IoT environmental sensors (e.g. body language detectors)

As an example, an eye tracking sensor could acquire the gaze position signal and the pupil diameter signal, including the speed of the eye. of its contraction. In this case, it could be conceivable to detect the trend of oxygen concentration in the pupil, the level of dopamine from the speed? contraction of the pupil, the heartbeat from the behavior of eye pressure or the concentration of biochemicals in the retina from an ocular fluorescent excitation response. So it might be possible to establish a correlation between these parameters / trends extracted and the cognitive-emotional state since? a bio-imaging processing software pu? acquire and decode signals from molecules in the ocular blood. Like this particular example, the eye-tracking sensor could be embedded in the shutter system (for example, in the lens or screen of a DSLR or bridge camera in a smartphone, in the frame or lens of eyeglasses) and could track movement eyepiece by allowing the software system to record which details of the image have captured the user's attention.

fNIRS technology can give another example. Being a set of wearable sensors, it can't? be physically integrated into the system, but pu? communicate with it through wired/wireless communication protocols, easily allowing different users to interact with their business? acquired fNIRS with the camera.

So far, sensors have been described that acquire signals from a single user. However, ? conceivable a social platform by means of which the acquisition pu? be activated by a multiplicity? of digitally connected users present on the site or active remotely by means of their emotions or their cognitive processes, so that the input data can also include biological signals and information from a community? social (human and/or animal), which can? then interact with the system managed by the main user or by a human, animal, robotic being.

Figure 7 schematically illustrates an example of process 500, where a primary user (user #1) employs a photographic system with sensor elements 501. In a signal refinement system 502, the sensor dashboard extracts 502 different types of signals of quality raw (i.e. as they are collected) that are synchronized both with each other and with the sequence of frames captured by the photographic lens, in order to have a unique correspondence between each timestamp and the information recorded in the process. According to this example, is a plurality used? of biological sensors, such as GSR, PPG, pupil diameter, gaze position, EEG, fNIRS, facial expression…

The signals obtained from the synchronization process pass through a two-stage processing 503. Each signal is filtered 504 with appropriate and customized techniques aimed at reducing the noise typical of the signal or due to human artifacts. These filtered signals are subjected to numerical analysis techniques 505 to extract indices and signals representing trends or figures of merit explanatory of the cognitive and emotional state of the user. In particular, while the system continuously acquires frames and signals from the various sensors, the integrated software system filters the signals and analyzes them. The analysis consists in automatically searching for a specific pattern in the signal values, so that, once an optimal criterion is met, the system freezes the image and a photo is generated. Furthermore, all recorded physiological metadata is saved in the digital file (as currently happens with the "raw" format and other processed photographic formats, for example Jpeg, which contains data from light sensor arrays and other basic technical data derived from the configuration camera, associated GPS coordinates, etc.). Once extracted, the signal outputs are ready to be used to search for a trigger pattern for the photographic shot through a machine learning (ML) system based on the 506 classification model. This system receives the input outputs of the filtering phase , processing and analysis and, on the basis of previously adjusted setup parameters, identifies a time instant in correspondence with which the specific emotional-cognitive model is used with the consequent taking of a photograph (i.e. the action of immortalizing the frame for that instant of time) 507.

Similarly, ? Is it possible to consider an ecosystem of a plurality? of sensors to detect and collect emotional and cognitive information deriving from multiple? individuals (user n. 2, user n.3, ?user n. N) who interact with a main user (user n.1) in the way they see the same portion of reality? identified by the photographic lens. In fact, through the same signal refinement system, cognitive and emotional tendencies and figures of merit are extracted from different users who feed a hub, which contains emotional socio-feedbacks completely synchronized with those of the main user 508. Yes points out that not only a single user, but also a plurality? of users can? feed the ML 509 system as input.

Figures 8 and 9 both illustrate a flowchart describing a process employed in an image acquisition system, in which the imaging device ? a mobile phone (Figure 8) or camera device (Figure 9). The process of these figures basically corresponds to the process shown in Figure 7 for a single user.

In Figure 8, the user looks at a cell phone or smartphone 601. The smartphone ? equipped with 6 biological sensors to detect and collect raw biological data (e.g. PPG, GSR, facial expression, image frames, and eye tracking). The raw signals are synchronized with each other along with the frame sequence 602. The synchronized signals are then filtered. In particular, a first filtering process (ie data binning) 603 is performed. It should be noted that based on the measured biological signal, ? It is possible to use a second dedicated filtering process. For example, in the case of a PPG signal, a smooth sampling check is performed and then a filtering stage using a 604 type II (order IV) Chebyshev filter. The filtered PPG is then subjected to a peak detector based on envelope to evaluate the variability? heart rate (HRV). The other biological signals undergo the same or similar processes. Filtering stages are performed for the biological signals for each time sample 605. Synchronized image frames are used to control and adjust photographic parameters such as zoom and focus 606. These parameters are also controlled based on the results of the eye tracking signals in terms of fixation time and fixation position 607.

The filtered signals are fed into a trained ML algorithm 608 to identify a point in time for the shot 609. It should be noted that the photographic parameters are adjusted prior to the shot based on the analysis of the image frames.

In Figure 9 , the user employs an apparatus with a camera device instead of a smartphone. In this case, some biological signals, such as facial expression, are not taken into account. However, other parameters such as better 610 pupil analysis may be considered. Apart from this, the process of Figure 9 ? the same as in Figure 8. As a result, the same reference numerals are identified.

The machine learning training process is developed after previous exploratory sessions performed on subjects recruited for their ability to of manual photographic shutter. Will the system learn? from their cognitive processes carried out before and during the snapshot. In this way, it will not be the user to decide when to shoot, but his cognitive and biological process will guide? unconsciously the software towards the shot, emulating the biological and cognitive process of the photographer more? expert. This approach represents the first level of development, thus allowing a user to take a photo as a skilled photographer would using patterns from other photographers (ie grouped according to the user's desirability).

There? ? systematically in action with pi? of a single individual, i.e. a given self-defined set of individuals according to criteria of affinity? explicit or implicit, or the totality? of users are generating a set of big data that can be used to determine significant patterns, thus strengthening? the effectiveness of machine learning algorithms.

One level higher deep functioning of the system consists in the possibility? for a user to take a photo as he wishes without knowing/being aware of his cognitive-emotional processes, but allowing the shooting system to measure, quantify and define the cognitive emotional state of the user himself over time (i.e. "assisted photography ").

In order to improve the accuracy and efficiency of the system, a starting point can be be a limited number of cognitive paths resulting from an expert photographer. However, ? It is also interesting to further understand and explore how a common user (for example not necessarily having photographic experience) feels and reacts when faced with the acquisition of images by a camera, arriving at defining what kind of cognitive-emotional pattern would trigger a shot.

The concept ? that the system allows you to generate and use different "activation filters" which are represented by the way in which a particular user (for example the user himself, his friend, the expert photographer, etc.) takes a photo.

Additionally, users can be grouped into clusters to see how they behave and what types of information they aggregate. Once you get a large amount of metadata (for example from several thousand or millions of images), through ML clustering methods based on affinity criteria? (supervised or unsupervised) the user could cause the generation of an activation filter (predetermined shooting pattern) based on the cognitive and emotional reactions he wishes to emphasize.

Furthermore, starting from the data and/or metadata of individual users and their extracted information, ? Consider generating and powering a revolutionary social system (e.g. "Instagram of Neuroscience"), which is capable of doing the following:

1) share pre-processed contents with a social network of people promoting an informed and aware communication of the developed cognitive processes/paths; 2) produce new artistic, cognitive and emotional contents starting from the analysis of the cognitive-biological information extracted and associated with the subject who is watching a particular content of the social network. (for example a user observes a particular shared image and reacts biologically to it. The Instagram of Neuroscience is able to produce a tangible cognitive-emotional feedback of the visualization).

Figure 10 schematically shows a computer system for implementing methods of examples of the description. In particular, Figure 10 shows an example of a computing device 2000 that can be, for example, willing to implement one or more? examples of the methods described here. In the examples, the computing device 2000 includes the unit main 2002. The unit? main 2002 can? include a 2004 processor and 2006 system memory. In the examples, the 2004 processor can? include a 2008 processor core, a 2010 cache and one or more? registers 2012. In the examples, the 2008 processor core can? include one or more processing core and pu? understand a plurality of cores that can run a plurality? of threads. The 2004 processor can? be of any suitable type such as a microcontroller, microprocessor, digital signal processor, or some combination thereof, although it would be appreciated that other processor types can be used.

In the examples, the 2008 processor core can? understand one or more unit? of processing. In the examples, the 2008 processor core includes one or more? between a unit? floating point, a unit? arithmetic, a unit? of digital signal processing or a combination of these and / or a plurality? of other units? of elaboration, although it will be appreciated? that other units can be used? of processing. In the examples, cache 2010 can? understand a plurality such as a level one cache and a level two cache, although other appropriate cache arrangements may be used.

In the examples, the 2004 processor includes a 2014 memory controller that can be used to allow communication between the 2004 processor and the 2006 system memory via a 2016 memory bus. be implemented as an integral part of the 2004 processor, or can? be implemented as a separate component.

In the examples, the 2006 system memory can? be of any suitable type such as non-volatile memory (such as flash memory or read-only memory), volatile memory (such as random access memory (RAM)), and/or a combination of volatile and non-volatile memory volatile. In the examples, the 2006 system memory can? be arranged to store code for execution by the 2004 processor and/or execution-related data. For example, system memory can store 2018 OS code, 2020 application code, and 2022 program data. In the examples, the 2020 application code can understand a code to implement one or more? of the example methods described here, for example to implement the steps described above with reference to Figures 1, 8 and 9. Application Code 2020 can? be arranged to cooperate with 2022 program data or other media, e.g. to enable photographic parameters to be controlled or image data to be stored together with attribute data.

In the examples, computing device 2000 can? have features, functionality? or additional interfaces. For example, the unit main 2002 can? cooperate with one or more peripheral devices, for example to implement the methods described here. In the examples, the computing device 2000 comprises, as peripheral devices, an output interface 2024, a peripheral interface 2026, a storage device 208 and a communication module 2030. In the examples, the processing device comprises an interface bus 2032 arranged to facilitate communication between the unit? main 2002 and peripheral devices.

In the examples, the 2024 output device can? include output devices such as a drive? graphics processing unit (GPU) 2034 and a unit? of 2036 audio outputs arranged to be able to communicate with external devices such as a display and/or speaker, through one or more? suitable ports such as an audio/video (A/V) port. In the examples, the 2026 peripheral interface can? include a 2038 serial interface, a 2040 parallel interface, and one or more? 2042 input/output ports which can be operable to cooperate with the unit? main 2002 to allow communication with one or more? external input and/or output devices through I/O port 2042. For example, I/O port 2042 can? communicate with one or more input devices such as a keyboard, mouse, touch pad, voice input device, scanner, imaging capture device, video camera, and the like, and/or with one or more? output devices such as a 2D printer (e.g. a paper printer) or a 3D printer or other suitable output device. For example, signals can be received through I/O port 2042 and/or communication module 2030.

In the examples, the storage device can include 2044 removable storage media and/or 2046 non-removable storage media. For example, removable storage media may be random access memory (RAM), electrically erasable programmable read-only memory (EEPROM), flash memory read-only (ROM) or other memory technology, optical storage media such as compact discs (CDs), digital versatile disks (DVDs) or other optical storage media, magnetic storage media such as floppy discs, magnetic tape, or other magnetic storage. However, it will appreciate what can? any suitable type of removable storage media can be used. Non-removable storage media 2046 may include magnetic storage media such as a hard disk, or solid state hard disk, or other suitable media, although it will be appreciated what can? any suitable non-removable storage medium can be used. The 2028 storage device can allow access through the unit? main 2002 for example to implement the methods described here.

In the examples, the communication module can? include a 2048 wireless communication module and a 2050 wired communication module. be willing to communicate in mode? wirelessly via a suitable wireless communication standard, such as related to Wi-Fi, Bluetooth, near-field communication, optical communication (such as infrared), acoustic communication, or via a suitable mobile telecommunications standard. The wired communication module can? allow communication over a wired or optical link, such as via Ethernet or optical cable. However, it will appreciate what can? any suitable communication module can be used.

Referring to Figures 2 through 9, the display, filter, analysis, or snap module can for example be implemented by the processor. In the examples, one or more of the display, filtering, analysis or snap modules can be implemented using the unit? principal 2002, although it will be appreciated? that other suitable implementations can be used. In the examples, the filter, analysis or snap module can? be implemented by the unit? main 2002 in cooperation with the 2024 output device, although it will be appreciated? that other suitable implementations can be used.

Will you appreciate it? that, in the examples of the description, the elements of the described methods can be implemented in a computing device in any suitable way. For example, a conventional computing device can be suitable to perform one or more? of the methods described here by programming/adapting one or more? computer device processors. Consequently, in the examples, programming/adaptation can? be implemented in the form of a computer program comprising computer-implementable instructions stored on a data medium and/or carried by a signal-carrying medium, such as a floppy disk, hard disk, optical disc, drive? solid-state memory, flash memory, programmable read-only memory (PROM), random access memory (RAM), or any combination of these or other storage media or signal carrier media, or transmitted over a network such as a network wireless, Ethernet, the Internet, or any combination of these or other networks.

In other words, in the examples, a computer program can? understand computer-readable instructions that, when implemented on a computing device, do that the computing device performs a method according to the examples of the description. In the examples, a storage medium can? include, for example, the computer program as mentioned above. Will we make it? I also realize that other suitable IT architectures could be used, such as those based on one or more? parallel processors. Furthermore, at least some processing can be implemented on one or more? unit? graphics processing unit (GPU). Although the Computing Device 2000 is described as a general purpose computing device, it will be appreciated that this could be implemented in any appropriate device, such as mobile phone, smartphone, camera, camcorder, tablet device, server device, etc.? with modifications and/or adaptation, if appropriate, to the characteristics described above, for example according to the functionality? and the desired hardware features.

The present image acquisition method and related apparatus and system can be adopted in all structures that consider the interaction between a user and an audiovisual stimulus by developing an appropriate stimulus processing procedure, in which each user could share the stimulus processed and its associated metadata (e.g. signals and extracted information) with any other person in order to compare experiences and understand the dynamics and biological reactions of the community.

For example, in the field of photography, integrating the current apparatus/system into a camera would produce an intelligent camera. In this way, ? It is possible to understand the path of visual exploration of a user before and during the photo shoot, identifying the biological response through the capture of signals, their synchronization and analysis. All the output data and the respective extracted information could be archived and shared to understand and compare the cognitive process to which a photographic shot is subjected. Additionally, a smart camera could be equipped with an upgraded AI system that can figure out when to shoot automatically, using a computer learning algorithm powered by selected prior experiences (such as shared or personal). In this case, the AI shooting system relies on real-time metadata processed and recorded during viewfinder viewing.

At a level of interpretation more? high, without a specific application structure, ? It is possible to consider the arrangement of the sensors as a system that extracts additive stimulation information (metadata) with respect to the visualization of the stimulus, generating a space of multidimensional characteristics associated with the fruition of the stimulus. Depending on the nature of the particular field of application, the collection of a single augmented experience could be elaborated and then summarized in a communicative report, which would consequently change its form, content and purpose. This report could act as a personal social communicator feeding into a social network (e.g. an Instagram of the Neuroscience or as a science hub), where the associated metadata could be regarded as social information or knowledge discovery, respectively, depending on the artistic nature of the stimuli. scientific.

Therefore, this extensive sharing produces recursive enrichment of AI processing procedure, learning from different augmented experiences. The upgraded AI algorithm would provide a new and autonomous platform with dedicated automated processing procedure. This ? obtainable by considering a specific cluster of experiences as a set of learning, generating new metadata starting from those generated by users (metadata from metadata). Consequentially, ? possible to profile each user using not only the information collected from the questionnaire, but exploiting their biological cognitive response.

The system transforms biological parameters into a new photographic language, which could communicate information extracted from non-communicable signals to third parties. Consequently, another field of application can? be the visual art seen as a concept much more? ample. For example, a user can enjoy an image projected on a screen while several wearable sensors record its activity biology and its response. The visual stimulus and associated metadata can be processed into shareable results. On a further level of development, ? It is possible to consider a museum equipped with autonomous intelligent sensors and non-invasive wearables that acquire the biological response of users in response to a particular use of the work of art.

An advanced development would be to hyper-connect people with reality? creating an open-air museum in any location where the user identifies an artistic visual stimulus (i.e. augmented reality). Again, could the user be clad in suitable wearable sensors, sensitive clothing, smart glasses, and using a smartphone capable of acquiring further biological signals? You can record various metadata associated with that particular view. All the information acquired could be processed and shared in a community network? virtual.

Another scenario can be the elaboration of an audio-visual stimulus proposed to a patient for neurocognitive evaluation and rehabilitation. Could the output be compared between different subjects in order to evaluate the possibility? of cognitive impairment. Furthermore, it would be possible to generate an automated evaluation by keeping track of the user's previous stimulations and using a dedicated AI algorithm.

Fashion represents a further area of application of this method and of the apparatus/system. For example, a user could see an image that represents him dressed or made up according to his wishes. Could biological reactions be recorded in the form of signals and processed in order to produce a report that could be shared with the community? in order to receive feedback from other users. The updated processing of the AI would consist in the automatic purpose of a makeup and clothing configuration which, according to social and personal opinions, turns out to be beautiful. The application in the fashion area suggests the creation of a new web-based image search process, based on the user's feeling. To do this, it would be necessary to update the AI algorithm in such a way that it finds a correlation between the biological emotional response and the images that express that type of emotion.

Although a variety of of techniques and examples of such techniques, these are provided by way of example only and many variations and modifications of such examples will be apparent to the person skilled in the art and will fall within the spirit and scope of the present invention, which is defined by the appended claims and their equivalents.

TRANSLATION OF THE FIGURES

Claims (18)

1. Method of acquiring images (100) by using an image acquiring apparatus (1) which includes an imaging device (2) having an image sensor (3) for generating image data, the method (100) including: displaying (S101), by using an image reproducing device (4), images acquired by the imaging device (2) based on the image data; generation (S102) of the attribute data from a plurality? of biological signals measured by a plurality? of biological sensors (6), the attribute data being indicative of attributes of a user at a time when the user is observing the images displayed by the image reproduction device (4); generation (S103) of data of significance? on the basis of the attribute data, and association of the attribute data from which the data of significance? are generated with each acquired image; controlling (S104) the photographic parameters of the image acquisition apparatus (1) based on the attribute data; analysis (S105) of the data of significance? to determine whether displayed images should be captured and archived; And storage (S106) of the image data of the displayed images together with the attribute data in a storage device (7) according to the significance data? analysed. The image acquisition method (100) according to claim 1, wherein: the significance data? include a plurality of levels of significance, each associated with a respective biological signal from which the data attributes were generated; the analysis (S105) of the data of significance? includes the generation of a level of significance? combined on the basis of levels of significance? for each of the biological signals; And the storage (S106) of the image data, together with the attribute data, includes the storage of the displayed images in the storage device (7) together with the attribute data based on a comparison between the level of significance? combined and a triggering threshold level. The image acquisition method (100) according to claim 1 or claim 2, wherein the generation (S103) of the significance data? and the analysis (S105) of the data of significance? they are based on a predetermined shooting pattern. The imaging method (100) according to claim 3, wherein the predetermined shot pattern is generated based on biological signals acquired from a user or a plurality of images. of users when a set of model training images for predetermined shooting model training is displayed. The imaging method (100) according to claim 3 or claim 4, when including dependence on claim 2, wherein the trigger threshold level is determined according to the predetermined trigger pattern. The image acquisition method (100) according to any one of claims 3 to 5, comprising transmitting the image data of the displayed images together with their associated local biological signals to a database of shot patterns to fine-tune the pattern of predetermined shot. The image acquisition method (100) according to any preceding claim, wherein the user is a local user who ? local to the image acquisition apparatus (1), and the attribute data comprises local attribute data which is associated with a time when the local user is observing the images displayed by the imaging device (4). The image acquisition method (100) according to any preceding claim, wherein the user is a remote user who ? remotely from the image acquisition apparatus (1), the image reproduction device (4) ? local to the remote user, and the attribute data comprises remote attribute data which is associated with a time when the remote user is observing the images displayed by the imaging device (4). The image acquisition method (100) according to any one of claims 7 or 8, comprising: viewing the images captured by the imaging device on a plurality? of remote user imaging devices that are associated with multiple remote users; the generation of remote attribute data from a plurality? of remote biological signals (6) measured by a plurality? of remote biological sensors, the remote attribute data being indicative of attributes of the remote users at a time when the remote users are viewing images displayed by their respective remote user imaging devices (4). The imaging method (100) according to claim 9 when comprising dependence on any one of claims 3 to 6, comprising fine-tuning the predetermined pattern based on user input from the different remote users . The image acquisition method (100) according to any preceding claim, comprising the time synchronization of the biological signals with each other and/or with the image data. The imaging method (100) according to any preceding claim, wherein the attribute data comprises signal data of the measured biological signals. 13. A program comprising processor-readable instructions that, when implemented by a processor, do that the processor implements a method (100) according to any preceding claim. The storage medium comprising the program according to claim 13. 15. Image acquisition apparatus (1) comprising: an imaging device (2) comprising an image sensor (3) operable to generate image data; an image reproducing device (4) operable to display images acquired by the imaging device (2) based on the image data; a processor (5) operable for: generate data attributes from a plurality? of biological signals measured by a plurality? of biological sensors (6), the attribute data being indicative of attributes of a user at a time when the user is observing the images displayed by the image reproduction device (4); generate data of significance? based on the data attributes, and associate the data attributes from which the data of significance? are generated with each acquired image; controlling the photographic parameters of the image acquisition apparatus (1) based on the attribute data; analyze the data of significance? to determine whether displayed images should be captured and archived; archive the image data of the displayed images together with the attribute data in a storage device (7) according to the significance data? analysed. The image acquisition apparatus (1) according to claim 15, comprising a communication interface (8) operable to receive the biological signals from the one or more? biological sensors (7). The image acquisition apparatus (1) according to claim 15, wherein the user is a local user who ? local to the imaging apparatus (1), and the attribute data comprises local attribute data which is associated with a time when the local user is observing the images displayed by the imaging device (4), and the apparatus (1) includes: a communication interface operable to receive remote attribute data from a plurality? of remote user devices associated with several remote users, each remote user device having a respective remote imaging device (4) for displaying the images captured by the imaging device (2), and each remote user device being operable to generate the data remote attributes from a plurality? of remote biological signals measured by a plurality? of remote (6) biological sensors, the remote attribute data being indicative of attributes of the remote users at a time when the remote users are viewing images displayed by their respective remote user imaging devices (4), in which the data of significance? are generated based on local attribute data and remote attribute data. 18. Image acquisition system (9) comprising: a plurality? of biological sensors (6); And an image acquisition apparatus (1) according to any one of claims 15 to 17.