WO2019144709A1

WO2019144709A1 - Method for constructing dream reproducing model, dream reproducing method, and devices

Info

Publication number: WO2019144709A1
Application number: PCT/CN2018/119764
Authority: WO
Inventors: 王倩
Original assignee: 阿里巴巴集团控股有限公司
Priority date: 2018-01-29
Filing date: 2018-12-07
Publication date: 2019-08-01
Also published as: TWI714926B; CN108363487A; CN108363487B; TW201933049A

Abstract

A method for constructing a dream reproducing model, a dream reproducing method, and devices. The dream reproducing method comprises: obtaining brainwave data of a user in a sleep state (302); performing feature extraction on the obtained brainwave data, to obtain a feature value of the brainwave data (304); inputting the obtained feature value of the brainwave data into the dream reproducing model, to obtain a corresponding output value (306); and determining, from the correlation, a perceptible object having the highest similarity with the output value, to generate a dream reproducing result (308).

Description

Dream reconstruction model construction method, dream reproduction method and device

Technical field

The embodiments of the present disclosure relate to the field of computer application technologies, and in particular, to a method for constructing a dream reproduction model, a method and apparatus for reproducing a dream.

Background technique

Modern medicine believes that dreams are caused by various stimulating factors in the body, such as psychology, physiology, pathology and environmental factors, which are applied to specific cortex of the brain during sleep. The form can be image, sound, thought, feeling and so on. Studies have shown that a beautiful dream can bring a subjectively more pleasant feeling to a certain extent. People can even find inspiration in solving dreams in real dreams, but because the dream is not controlled by the subjective consciousness of the human body, the human body sleeps. When the state wakes up, it usually does not remember a complete and clear dream.

Summary of the invention

For the above technical problem, the embodiment of the present specification provides a method for constructing a dream reproduction model, a method and device for reproducing a dream, and the technical solution is as follows:

According to a first aspect of the embodiments of the present specification, a method for constructing a dream reproduction model is provided, the method comprising:

Obtaining at least one set of correspondences between the perceptible objects and the brain wave data when the user perceives the perceptible objects;

Performing feature extraction on each of the corresponding correspondences to obtain a training sample set, wherein each of the training samples uses the extracted feature value of the brain wave data as an input value to extract the perceptible object The eigenvalue is a tag value;

The training sample is trained by using a supervised learning algorithm to obtain a dream reproduction model. The dream reproduction model takes the feature value of the brain wave data as an input value, and uses the feature value of the perceptible object as an output value.

According to a second aspect of the embodiments of the present specification, a method for dream reproduction is provided, the method comprising:

Obtaining brain wave data of the user in a sleep state;

Performing feature extraction on the obtained brain wave data to obtain characteristic values of the brain wave data;

Inputting the obtained feature value of the brain wave data into the dream reproduction model to obtain a corresponding output value;

From the correspondence, a perceptible object having the highest similarity to the output value is determined to generate a dream reproduction result.

According to a third aspect of the embodiments of the present specification, a device for constructing a dream reproduction model is provided, and the device includes:

a data obtaining module, configured to obtain at least one set of correspondences between the perceptible object and the brain wave data when the user perceives the perceptible object;

a sample obtaining module, configured to perform feature extraction on each group of the corresponding relationships, respectively, to obtain a training sample set, wherein each of the training samples extracts the characteristic value of the brain wave data as an input value, to extract the The feature value of the perceptible object is a tag value;

a sample training module, configured to train the training sample by using a supervised learning algorithm, to obtain a dream reproduction model, wherein the dream reproduction model uses the feature value of the brain wave data as an input value, and uses the feature value of the perceptible object as output value.

According to a fourth aspect of the embodiments of the present specification, a dream recreating apparatus is provided, the apparatus comprising:

The brain wave acquisition module is configured to obtain brain wave data of the user in a sleep state;

a feature extraction module, configured to perform feature extraction on the obtained brain wave data to obtain a feature value of the brain wave data;

An output module, configured to input the obtained feature value of the brain wave data into the dream reproduction model to obtain a corresponding output value;

And a reproducing module, configured to determine, from the correspondence, a perceptible object having the highest similarity with the output value, to generate a dream reproduction result.

According to a fifth aspect of embodiments of the present specification, a computer apparatus includes a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the program when the program is executed A method of constructing any of the dream recurrence models provided by one or more embodiments of the specification.

According to a sixth aspect of embodiments of the present specification, a computer apparatus includes a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the program when the program is executed Any one of the embodiments provides a method of dream reproduction.

According to the technical solution provided by the embodiments of the present disclosure, by obtaining at least one set of correspondences between the perceptible objects and the brain wave data when the user perceives the perceptible objects, feature extraction is performed on each group of correspondences to obtain a training sample set. Wherein, each training sample extracts the characteristic value of the brain wave data as an input value, and the characteristic value of the perceptible object is a tag value, and the training sample is trained by using a supervised learning algorithm to obtain a dream reproduction model, The dream recurrence model takes the eigenvalue of the brain wave data as the input value, and the eigenvalue of the sensible object is the output value. Subsequently, the dream recurrence model can realize the user's brain wave data reproduction in the sleep state. Dreamland, satisfying the user experience.

The above general description and the following detailed description are merely exemplary and explanatory, and are not intended to limit the embodiments.

Moreover, any of the embodiments of the present specification does not need to achieve all of the above effects.

DRAWINGS

In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings to be used in the embodiments or the description of the prior art will be briefly described below. Obviously, the drawings in the following description are only It is a few embodiments described in the embodiments of the present specification, and other drawings can be obtained from those skilled in the art based on these drawings.

FIG. 1 is a schematic diagram of an application scenario for implementing dream replay according to an exemplary embodiment of the present disclosure; FIG.

2 is a flowchart of a method for constructing a dream reproduction model according to an exemplary embodiment of the present specification;

FIG. 3 is a flowchart of a dream reproduction method according to an exemplary embodiment of the present specification; FIG.

4 is a block diagram of an embodiment of a device for constructing a dream reproduction model according to an exemplary embodiment of the present specification;

FIG. 5 is a block diagram of an embodiment of a dream recreating apparatus according to an exemplary embodiment of the present specification; FIG.

FIG. 6 is a schematic diagram showing a hardware structure of a more specific computing device provided by an embodiment of the present specification.

Detailed ways

In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present specification, the technical solutions in the embodiments of the present specification will be described in detail below with reference to the accompanying drawings in the embodiments of the present specification. The examples are only a part of the embodiments of the specification, and not all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this specification should fall within the scope of protection.

Modern medicine believes that when the human body sleeps, various stimulating factors inside and outside the body, such as psychological, physiological, pathological, environmental factors, etc., are produced by the specific cortex of the brain. That is to say, when the person is dreaming, the cerebral cortex is excited. The state, which will produce brain waves, and the brain waves have a certain degree of correspondence with human consciousness activities, for example, when a person sees two different images, or hears two different melody music, the nerves of the cerebral cortex The activities are different, so that the brain waves generated are different. Similarly, when people dream of different scenes, the neural activity of the cerebral cortex is different, and the generated brain waves are also different. Based on this, the present invention proposes to realize the use of brain wave data. The dream reappears.

Please refer to FIG. 1 , which is a schematic diagram of an application scenario for implementing dream reproduction according to an exemplary embodiment of the present disclosure. As shown in FIG. 1, the user 110, the brain wave sensor 120, and the computer 130 are included, wherein the brain wave sensor 120 is worn on the head of the user 110 for collecting brain wave data of the user 110, and collecting the brain. The radio wave data is sent to the computer 130. Specifically, when the user 110 is in the awake state, the brain wave sensor 120 collects the brain wave data of the user 110 when the user 110 perceives the perceptible object, for example, when the user 110 sees an image. The brain wave data and the related information of the perceptible object corresponding to the brain wave data are sent to the computer 130, and the computer 130 performs training based on the received brain wave data and the related information of the perceptible object to obtain a dream reproduction model. The dream recurrence model can take the brain wave data related information as input and the object-related information as the output. It can be understood by those skilled in the art that in order to obtain a dream reproduction model, several samples are needed, that is, it is required to collect brain wave data generated by several users when they perceive different perceptible objects.

After the training to obtain the dream reproduction model, the brain wave sensor 120 can be used to collect the brain wave data generated by the user in the sleep state, and then the brain wave sensor 120 transmits the brain wave data to the computer 130, and the computer 130 can be based on the dream. The reproduction model outputs relevant information of the perceptible object corresponding to the electroencephalogram data, and then the dream reproduction result can be generated based on the related information of the perceptible object. After the user 110 wakes up, the above-mentioned dream reproduction result can be viewed by the computer 130 to realize the "re-warming" dream.

Based on the application scenario shown in FIG. 1, the present specification shows that the following embodiments are respectively described from the construction of the dream reproduction model and the realization of the dream reproduction based on the dream reproduction model.

First, describe this aspect of the construction of the Dream Reproduction Model:

Referring to FIG. 2, a flowchart of a method for constructing a dream reproduction model according to an exemplary embodiment of the present disclosure may include the following steps:

Step 202: Obtain a correspondence between at least one set of brainwave data including the perceptible object and the user when perceiving the perceptible object.

In the embodiment of the present specification, the perceptible object may be a single image or an image frame intercepted in the video. It can be understood by those skilled in the art that the essence of the single image and the image frame are images. For convenience of description, in the embodiment of the present specification, the perceptible object may be an image.

In an embodiment, the set of perceptible objects may be preset, for example, the set includes 1000 different images, and each perceptible object in the set of perceptible objects is sequentially provided to the user 110, for example, in a specific environment. The each perceptible object is played in a slideshow, and when the user 110 is provided with the perceptible object, the brainwave data of the user 110 when the perceptible object is perceived is synchronously acquired. Through such processing, each time the brain wave data is collected, a correspondence relationship between the perceptible object and the brain wave data when the user 110 perceives the perceptible object is obtained, for example, 1000 correspondences are obtained.

In an embodiment, the user 110 may be provided with a perceptible object according to a preset rule, for example every 5 seconds, and continuously collected during the entire process of providing the first perceptible object to the last perceptible object. The brain wave data of the user 110, and then, according to the extraction rule corresponding to the preset rule, for example, according to the acquisition time of the brain wave data, intercepting a piece of brain wave data every 5 seconds, and then establishing brain wave data and perceptible The correspondence between objects. Through such processing, finally, a plurality of correspondences including the perceptible object and the brain wave data when the user 110 perceives the perceptible object can be acquired.

In an embodiment, a video may be provided to the user 110, and the brain wave data of the user 110 is continuously collected during the whole process of the user 110 viewing the video, and then the image in the video is intercepted at the same time interval. Frames, as well as brainwave data, can then establish a correspondence between brainwave data and perceptible objects.

It should be noted that the two embodiments described above are only used as two alternative implementations. In an actual application, there may be other manners to obtain at least one set of the object that includes the perceptible object and the user when perceiving the perceptible object. Corresponding relationship of brain wave data, for example, when the user 110 performs activities according to the will of the user, the brain wave data of the user 110 is collected, and the retinal imaging of the user 110 is synchronously acquired, and the brain wave data and the retinal imaging can be established based on the acquisition time. Corresponding relationship, the retinal imaging can be equivalent to the perceptible object perceived by the user.

It will be understood by those skilled in the art that, according to the above description, the brain wave sensor 120 illustrated in FIG. 1 may also have the function of acquiring retinal imaging, or by another separate wearable smart chip (not shown in FIG. 1). It is responsible for collecting the retinal imaging of the user 110, which is not limited in this embodiment of the present specification.

Step 204: Perform feature extraction on each group correspondence, and obtain a training sample set, wherein each training sample uses the extracted feature value of the brain wave data as an input value, and the extracted feature value of the perceptible object is obtained. Tag value.

In the embodiment of the present specification, for each set of correspondences obtained in step 202, feature extraction is performed on the perceptible object in each set of correspondences and the brain wave data in which the user perceives the perceptible object, to obtain a training sample set. Each training sample in the training sample set includes the feature value of the extracted brain wave data and the extracted feature value of the perceptible object, and based on the related description of the application scenario shown in FIG. 1 above, the actual dream is heavy. In the present process, the perceptible object perceived by the user 110 is determined by the collected brain wave data of the user 110 in the sleep state. Therefore, each of the above training samples uses the characteristic value of the brain wave data as an input value to extract. The eigenvalue of the perceived object is the tag value.

Extract the characteristic values of brain wave data:

In an embodiment, by the mathematical concept of complex transformation, the real signal of any frequency can be expressed as a sum of a series of periodic functions, and the process of expressing the real signal as a series of periodic functions is to analyze the real signal. The process of each cycle is equivalent to the composition of the real signal. Based on this, the present specification proposes a complex transformation of brain wave data, for example, complex transformation of brain wave data by Fourier transform, and brain wave The data is represented as a sum of at least one complex variable function, and the at least one complex variable function can be used as an eigenvalue of brain wave data, for example, the characteristic value of the mentioned brain wave data is (a ₁ f ₁ (sinx), a ₂ f ₂ (sinx), a ₃ f ₃ (sinx)).

It should be noted that the manner of extracting the feature values of the brain wave data described above is only an optional implementation manner. In practical applications, the feature values of the brain wave data may also be extracted by other means, for example, by correlation. The eigenvalues of the brainwave data are extracted by analysis, AR parameter estimation, Butterworth low-pass filtering, genetic algorithm, etc. The specific type of the extracted feature values can be determined by an actual algorithm, for example, extracted by Butterworth low-pass filtering algorithm. The eigenvalue is the square of the signal amplitude, and the eigenvalue extracted by the AR parameter estimation algorithm is the power spectral density, which is not described in the embodiment of the present specification.

Extract the feature values of the perceptible object:

Taking the perceptible object as an example, in an embodiment, color statistics can be performed on the perceptible object, and the number of pixels corresponding to each color value in the perceptible object is obtained, and the obtained number of pixel points is expressed as 2 ^N -dimensional vector, where N is the number of color bits of the image, that is, the 2 ^N -dimensional vector can be used as the feature value of the perceptible object, for example, the extracted feature values are (y ₁ , y ₂ , y ₃ , ... y _2^N ).

Further, considering that the number of color bits of different images may be different, for example, an 8-bit image and a 16-bit image, the dimension of the extracted feature values is different, and the training is unified for the subsequent training of the training samples. The color statistics of images with different color bits can be mapped to a unified vector space. Here, "unified" means that the dimensions of the vectors obtained based on the color statistics result are the same.

In addition, it should be noted that the larger the dimension of the vector, the higher the complexity of training the training sample and the larger the calculation amount. Therefore, in the embodiment of the present specification, the object characteristics are guaranteed. When the fineness of the value satisfies the user's expectation, a vector space with a small number of dimensions can be set as much as possible.

It should be noted that, in practical applications, for images of different color digits, the images may be first set to the same color number, and then the feature extraction is performed according to the above description. The step of mapping each color statistical result to the unified vector space is performed.

Step 206: Training the training sample by using a supervised learning algorithm to obtain a dream recurrence model. The dream recurrence model takes the feature value of the brain wave data as an input value, and uses the feature value of the perceptible object as an output value.

In the embodiment of the present specification, the training sample obtained in step 204 can be trained by using a supervised learning algorithm to obtain a dream reproduction model, and the dream reproduction model uses the characteristic value of the brain wave data as an input value to perceive the object. The eigenvalue is used as the output value. It can be understood that the training-to-dream reproduction model can be understood as a functional relationship between the input value and the output value, wherein the output value is affected by all or part of the input value, and therefore, the output value and the input value The functional relationship between them can be exemplified as follows:

y=f(x ₁ ,x ₂ ,...x _M )

Wherein, x ₁ , x ₂ , ... x _M represent M input values, that is, characteristic values of M brain wave data, and y represents an output value, and may also perceive the feature value of the object, and may specifically be in the perceptible object. The proportional relationship between the number of pixels corresponding to each color value.

It should be noted that the form of the dream reproduction model can be selected according to actual training needs, such as a linear regression model, a logistic regression model, and the like. The embodiment of the present specification does not limit the selection of the model and the specific training algorithm.

In addition, it should be noted that different users may have different perceptions of the same perceptible object. Therefore, in this embodiment, different dream re-creation models are separately constructed for different users; further, the same user is in his or her own psychology and physiology. In different states, the sensing ability of the same perceptible object may be different. Therefore, in the embodiment of the present specification, different dream recurrence models are separately constructed for different time segments of the same user. In addition, in the actual application, there may be other implementation manners, for example, the same dream reproduction model is constructed for different users, and the embodiment of the present specification does not specifically limit this.

It can be seen from the foregoing embodiments that the technical solution provided by the embodiments of the present disclosure performs feature extraction for each group of correspondences by obtaining at least one set of corresponding relationships between the perceptible objects and the brain wave data when the user perceives the perceptible objects. Obtaining a training sample set, wherein each training sample extracts the characteristic value of the brain wave data as an input value, and the eigenvalue of the object is a tag value, and the training sample is trained by using a supervised learning algorithm to obtain a dream Recurring the model, the dream recurrence model takes the eigenvalue of the brain wave data as an input value, and the eigenvalue of the sensible object is an output value, and subsequently, the brain can be realized by using the user in a sleep state through the dream reproduction model The radio wave data reproduces the user's dream and satisfies the user experience.

So far, the relevant description of the construction of the dream reproduction model is completed.

Secondly, it describes the realization of dream reappearance based on the dream recurrence model:

Referring to FIG. 3, a flowchart of a method for reproducing a dream according to an exemplary embodiment of the present disclosure may include the following steps:

Step 302: Obtain brainwave data of the user in a sleep state.

In the embodiment of the present specification, the brain wave data of the user in the sleep state can be obtained by the brain wave sensor 120 illustrated in FIG. 1 according to a preset rule, for example, every one minute, or every two minutes.

Step 304: Perform feature extraction on the obtained brain wave data to obtain characteristic values of brain wave data.

For a detailed description of this step, refer to the related description in step 204 in the foregoing embodiment shown in FIG. 2, and details are not described herein again.

Step 306: Input the obtained feature value of the brain wave data into the dream reproduction model to obtain a corresponding output value.

It can be seen from the dream re-creation model described in the embodiment shown in FIG. 2 that, in this step, the feature value of the brain wave data extracted in step 304 can be input into the dream reproduction model to obtain a corresponding output value, and the output is obtained. The value can be a feature value of the perceptible object.

Step 308: From the correspondence, determine a perceptible object having the highest similarity with the output value to generate a dream reproduction result.

In the embodiment of the present specification, the feature value of each perceptible object in the training sample set may be similarly calculated with the output value in step 306, and the feature value of the perceptible object with the highest similarity to the output value may be determined, and then, Then, in the corresponding relationship described in the embodiment shown in FIG. 2 above, the perceptible object having the highest similarity with the output value may be determined, and the dream reproduction result may be generated based on the determined perceptible object.

For a specific process of obtaining the feature values of each perceptible object in the training sample set, refer to the related description in the embodiment shown in FIG. 2 above, which will not be described in detail herein.

Further, the dream reproduction result can be displayed to the user, for example, the determined plurality of images are played in a slide show in the order of the acquisition time of the brain wave data.

It can be understood by those skilled in the art that the above-mentioned perceptible object having the highest similarity with the output value may be one or more, which is not limited in the embodiment of the present specification.

In the above description, the specific manner of calculating the similarity between the output value and the feature value of the perceptible object may be a Euclidean distance algorithm, a cosine similarity calculation algorithm, and the like, which are not limited in the embodiment of the present specification.

It can be seen from the above embodiments that the technical solution provided by the embodiment of the present invention obtains the feature value of the brain wave data by obtaining the brain wave data of the user in a sleep state, and obtains the feature value of the brain wave data, and inputs the feature value into the dream state. Reproducing the model, obtaining the corresponding output value, and then determining the perceptible object having the highest similarity to the output value from the correspondence between the pre-acquired object containing the perceptible object and the brainwave data when the user perceives the perceptible object In order to generate dreams to reproduce the results, thereby realizing the user to "relive" the dream based on the dream reappearing results.

So far, a description has been completed on the realization of dream reproduction based on the dream reproduction model.

Corresponding to the embodiment of the method for constructing the dream reproduction model, the embodiment of the present specification further provides a device for constructing a dream reproduction model, which is shown in FIG. 4, which is a dream reproduction model according to an exemplary embodiment of the present specification. A block diagram of an embodiment of a device, which may include a data acquisition module 41, a sample acquisition module 42, and a sample training module 43.

The data obtaining module 41 may be configured to obtain at least one group of correspondences between the perceptible object and the brain wave data when the user perceives the perceptible object;

The sample obtaining module 42 may be configured to perform feature extraction on each group of the corresponding relationships to obtain a training sample set, wherein each of the training samples extracts the characteristic value of the brain wave data as an input value to extract The feature value of the perceived object is the tag value;

The sample training module 43 can be configured to train the training sample by using a supervised learning algorithm to obtain a dream reproduction model, wherein the dream reproduction model uses the feature value of the brain wave data as an input value to perceive the characteristics of the object. The value is used as the output value.

In an embodiment, the data acquisition module 41 may include (not shown in FIG. 4):

Providing a sub-module for sequentially providing each perceptible object in the preset perceptible object set to the user;

And a collecting submodule, configured to synchronously collect brain wave data when the user perceives the perceptible object when the user is provided with the perceptible object.

In an embodiment, the sample acquisition module 42 can include (not shown in FIG. 4):

a first decomposition sub-module, configured to perform complex transformation decomposition on brain wave data in each of the corresponding correspondences, and represent the brain wave data as a sum of at least one complex function;

a first determining submodule configured to use the at least one complex variable function as a feature value of the brain wave data.

In an embodiment, the perceptible object is an image, and the sample acquisition module 42 may include (not shown in FIG. 4):

a statistic sub-module, configured to perform color statistics on each image in the corresponding relationship, to obtain a number of pixels corresponding to each color value in the image;

The second determining sub-module is configured to represent the obtained number of pixel points as a 2N-dimensional vector, where N is a color number of bits of the image.

In an embodiment, the apparatus may also include (not shown in Figure 4):

A mapping module for mapping color statistical results of images having different color bits to a unified vector space.

In an embodiment, different dream recurrence models are constructed separately for different users.

It can be understood that the data acquisition module 41, the sample acquisition module 42, and the sample training module 43 are three functionally independent modules, which may be simultaneously configured in the device as shown in FIG. 4, or may be separately configured in the device. Therefore, the structure shown in FIG. 4 should not be construed as limiting the embodiment of the present specification.

In addition, the implementation process of the functions and functions of the modules in the foregoing apparatus is specifically described in the implementation process of the corresponding steps in the method for constructing the dream reproduction model, and details are not described herein.

Corresponding to the above embodiment of the dream reproduction method, the embodiment of the present specification further provides a dream recreating device. Referring to FIG. 5, it is a block diagram of an embodiment of a dream recreating device according to an exemplary embodiment of the present specification. The brain wave acquisition module 51, the feature extraction module 52, the output module 53, and the reproduction module 54 may be included.

The brain wave acquisition module 51 can be used to obtain brain wave data of the user in a sleep state;

The feature extraction module 52 may be configured to perform feature extraction on the obtained brain wave data to obtain a feature value of the brain wave data;

The output module 53 can be configured to input the obtained feature value of the brain wave data into the dream reproduction model to obtain a corresponding output value;

The reproducing module 54 may be configured to determine, from the correspondence, a perceptible object having the highest similarity with the output value to generate a dream reproduction result.

In an embodiment, the feature extraction module 52 can include (not shown in FIG. 5):

a second decomposition sub-module, configured to perform complex transformation decomposition on the obtained brain wave data, and represent the brain wave data as a sum of at least one complex variable function;

And a third determining submodule configured to use the at least one complex variable function as a feature value of the brain wave data.

In an embodiment, the reproduction module 54 can include (not shown in FIG. 5):

a fourth determining submodule, configured to determine a reference feature value of each perceptible object in the correspondence relationship;

a calculation submodule, configured to separately calculate a similarity between the output value and a reference feature value of each perceptible object;

A fifth determining sub-module for determining a perceptible object having the highest similarity to generate a dream reproduction result.

It can be understood that the brain wave acquisition module 51, the feature extraction module 52, the output module 53, and the reproduction module 54 are four functionally independent modules, which can be simultaneously configured in the device as shown in FIG. 5, or separately. The configuration shown in FIG. 5 is not to be construed as limiting the scope of the embodiments of the present specification.

In addition, the implementation process of the functions and functions of the modules in the foregoing apparatus is specifically described in the implementation process of the corresponding steps in the above-mentioned dream reproduction method, and details are not described herein again.

Corresponding to the embodiment of the method for constructing the dream reproduction model, the embodiment of the present specification further provides a computer device including at least a memory, a processor, and a computer program stored on the memory and operable on the processor, wherein the processing The method for constructing the aforementioned dream recurrence model is implemented when the program is executed, and the method includes: obtaining at least one set of correspondences between the at least one set of the perceptible object and the brain wave data when the user perceives the perceptible object; Performing feature extraction on each set of the corresponding relationships, obtaining a training sample set, wherein each of the training samples extracts the feature value of the brain wave data as an input value, to extract the feature value of the perceptible object a tag value; training the training sample with a supervised learning algorithm to obtain a dream recurrence model, wherein the dream recurrence model takes the eigenvalue of the brain wave data as an input value, and uses the eigenvalue of the perceptible object as an output value .

Corresponding to the above embodiment of the dream reproduction method, the embodiment of the present specification further provides a computer device including at least a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the The foregoing method for reproducing a dream is implemented, the method includes at least: obtaining brain wave data of a user in a sleep state; performing feature extraction on the obtained brain wave data to obtain a feature value of the brain wave data; Entering the feature value of the brain wave data into the dream reproduction model to obtain a corresponding output value; and determining, from the correspondence, the perceptible object having the highest similarity with the output value to generate a dream reproduction result .

FIG. 6 is a schematic diagram showing a hardware structure of a more specific computing device provided by an embodiment of the present specification. The device may include a processor 610, a memory 620, an input/output interface 630, a communication interface 640, and a bus 650. The processor 610, the memory 620, the input/output interface 630, and the communication interface 640 implement a communication connection between the devices via the bus 650.

The processor 610 can be implemented by using a general-purpose CPU (Central Processing Unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits for performing correlation. The program is implemented to implement the technical solutions provided by the embodiments of the present specification.

The memory 620 can be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, or the like. The memory 620 can store the operating system and other applications. When the technical solutions provided by the embodiments of the present specification are implemented by software or firmware, the related program codes are saved in the memory 620 and executed by the processor 610.

The input/output interface 630 is used to connect an input/output module to implement information input and output. The input/output/module can be configured as a component in the device (not shown in Figure 6) or externally to the device to provide the corresponding functionality. The input device may include a keyboard, a mouse, a touch screen, a microphone, various types of sensors, and the like, and the output device may include a display, a speaker, a vibrator, an indicator light, and the like.

The communication interface 640 is used to connect a communication module (not shown in FIG. 6) to implement communication interaction between the device and other devices. The communication module can communicate by wired means (such as USB, network cable, etc.), or can communicate by wireless means (such as mobile network, WIFI, Bluetooth, etc.).

Bus 650 includes a path for transferring information between various components of the device, such as processor 610, memory 620, input/output interface 630, and communication interface 640.

It should be noted that although the above device only shows the processor 610, the memory 620, the input/output interface 630, the communication interface 640, and the bus 650, in a specific implementation, the device may also include necessary for normal operation. Other components. In addition, it will be understood by those skilled in the art that the above-mentioned devices may also include only the components necessary for implementing the embodiments of the present specification, and do not necessarily include all the components shown in the drawings.

Corresponding to the embodiment of the method for constructing the dream reproduction model, the embodiment of the present specification further provides a computer readable storage medium, where the computer program is stored, and when the program is executed by the processor, the foregoing dream reproduction model is constructed. method. The method at least includes: obtaining at least one set of correspondences between the perceptible objects and the brainwave data of the user when the user perceives the perceptible objects; performing feature extraction on each set of the corresponding relationships, respectively, to obtain a training sample set, wherein And each of the training samples uses the extracted feature value of the brain wave data as an input value, and the extracted feature value of the perceptible object is a tag value; and the training sample is trained by using a supervised learning algorithm, A dream reproduction model is obtained. The dream reproduction model takes the feature value of the brain wave data as an input value, and uses the feature value of the perceptible object as an output value.

Corresponding to the above embodiment of the dream reproduction method, the embodiment of the present specification further provides a computer readable storage medium, where the computer program is stored, and when the program is executed by the processor, the foregoing dream reproduction method is implemented. The method at least includes: obtaining brain wave data of a user in a sleep state; performing feature extraction on the obtained brain wave data to obtain a feature value of the brain wave data; and inputting the feature value of the obtained brain wave data into the The dream reproduces the model to obtain a corresponding output value; from the correspondence, the perceptible object having the highest similarity with the output value is determined to generate a dream reproduction result.

Computer readable media includes both permanent and non-persistent, removable and non-removable media. Information storage can be implemented by any method or technology. The information can be computer readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory. (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disk read only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape storage or other magnetic storage devices or any other non-transportable media can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary storage of computer readable media, such as modulated data signals and carrier waves.

It can be clearly understood by those skilled in the art that the embodiments of the present specification can be implemented by means of software plus a necessary general hardware platform. Based on such understanding, the technical solution of the embodiments of the present specification may be embodied in the form of a software product in essence or in the form of a software product, which may be stored in a storage medium such as a ROM/RAM. Disks, optical disks, and the like, including instructions for causing a computer device (which may be a personal computer, server, or network device, etc.) to perform the methods described in various embodiments of the embodiments of the present specification or embodiments.

The system, device, module or unit illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product having a certain function. A typical implementation device is a computer, and the specific form of the computer may be a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email transceiver, and a game control. A combination of a tablet, a tablet, a wearable device, or any of these devices.

The various embodiments in the specification are described in a progressive manner, and the same or similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment. The device embodiments described above are merely illustrative, and the modules described as separate components may or may not be physically separated, and the functions of the modules may be the same in the implementation of the embodiments of the present specification. Or implemented in multiple software and/or hardware. It is also possible to select some or all of the modules according to actual needs to achieve the purpose of the solution of the embodiment. Those of ordinary skill in the art can understand and implement without any creative effort.

The above is only a specific embodiment of the embodiments of the present specification, and it should be noted that those skilled in the art can make some improvements and refinements without departing from the principles of the embodiments of the present specification. Improvements and retouching should also be considered as protection of embodiments of the present specification.

Claims

A method for constructing a dream reproduction model, the method comprising:

Obtaining at least one set of correspondences between the perceptible objects and the brain wave data when the user perceives the perceptible objects;

Performing feature extraction on each of the corresponding correspondences to obtain a training sample set, wherein each of the training samples uses the extracted feature value of the brain wave data as an input value to extract the perceptible object The eigenvalue is a tag value;

The training sample is trained by using a supervised learning algorithm to obtain a dream reproduction model. The dream reproduction model takes the feature value of the brain wave data as an input value, and uses the feature value of the perceptible object as an output value.
The method according to claim 1, wherein the obtaining at least one set of correspondences between the perceptible object and the brainwave data of the user when the user perceives the perceptible object comprises:

Providing each perceptible object in the preset set of perceptible objects to the user in turn;

When the user is provided with the perceptible object, brain wave data of the user when the perceptible object is perceived is synchronously acquired.
The method according to claim 1, wherein the feature extraction is performed for each group of the corresponding relationships, including:

Performing complex transformation decomposition on the brain wave data in each of the corresponding correspondences, and expressing the brain wave data as a sum of at least one complex variable function;

The at least one complex variable function is used as a feature value of the brain wave data.
The method according to claim 1, wherein the perceptible object is an image, and the extracting the feature for each set of the corresponding relationship comprises:

Performing color statistics on the images in each of the corresponding correspondences to obtain the number of pixels corresponding to each color value in the image;

The number of obtained pixel points is expressed as a 2N-dimensional vector, where N is the number of color bits of the image.
The method of claim 4 further comprising:

Map color statistics of images with different color bits to a uniform vector space.
According to the method of claim 1, different dream reproduction models are separately constructed for different users.
A dream reproduction method based on the dream reproduction model according to any one of claims 1 to 6, the method comprising:

Obtaining brain wave data of the user in a sleep state;

Performing feature extraction on the obtained brain wave data to obtain characteristic values of the brain wave data;

Inputting the obtained feature value of the brain wave data into the dream reproduction model to obtain a corresponding output value;

From the correspondence, a perceptible object having the highest similarity to the output value is determined to generate a dream reproduction result.
The method according to claim 7, wherein the feature extraction of the obtained brain wave data is performed to obtain characteristic values of the brain wave data, including:

Performing complex transformation decomposition on the obtained brain wave data, and expressing the brain wave data as a sum of at least one complex variable function;

The at least one complex variable function is used as a feature value of the brain wave data.
The method according to claim 7, wherein the determining, from the correspondence, the perceptible object having the highest similarity with the output value to generate a dream reproduction result comprises:

Determining a reference feature value of each perceptible object in the correspondence relationship;

Calculating a similarity between the output value and a reference feature value of each perceptible object;

The perceptible object with the highest similarity is determined to generate a dream reproduce result.
A device for constructing a dream reproduction model, the device comprising:

a data obtaining module, configured to obtain at least one set of correspondences between the perceptible object and the brain wave data when the user perceives the perceptible object;

a sample obtaining module, configured to perform feature extraction on each group of the corresponding relationships, respectively, to obtain a training sample set, wherein each of the training samples extracts the characteristic value of the brain wave data as an input value, to extract the The feature value of the perceptible object is a tag value;

a sample training module, configured to train the training sample by using a supervised learning algorithm, to obtain a dream reproduction model, wherein the dream reproduction model uses the feature value of the brain wave data as an input value, and uses the feature value of the perceptible object as output value.
The apparatus according to claim 10, wherein the data acquisition module comprises:

Providing a sub-module for sequentially providing each perceptible object in the preset perceptible object set to the user;

And a collecting submodule, configured to synchronously collect brain wave data when the user perceives the perceptible object when the user is provided with the perceptible object.
The apparatus according to claim 10, wherein the sample acquisition module comprises:

a first decomposition sub-module, configured to perform complex transformation decomposition on brain wave data in each set of the correspondence relationship, and represent the brain wave data as a sum of at least one complex variable function;

a first determining submodule configured to use the at least one complex variable function as a feature value of the brain wave data.
The apparatus according to claim 10, wherein the perceptible object is an image, and the sample obtaining module comprises:

a statistic sub-module, configured to perform color statistics on each image in the corresponding relationship, to obtain a number of pixels corresponding to each color value in the image;

The second determining sub-module is configured to represent the obtained number of pixel points as a 2N-dimensional vector, where N is a color number of bits of the image.
The apparatus of claim 13 further comprising:

A mapping module for mapping color statistical results of images having different color bits to a unified vector space.
According to the apparatus of claim 10, different dream reproduction models are separately constructed for different users.
A dream recreating apparatus based on the dream reproduction model according to any one of claims 10 to 15, the apparatus comprising:

The brain wave acquisition module is configured to obtain brain wave data of the user in a sleep state;

a feature extraction module, configured to perform feature extraction on the obtained brain wave data to obtain a feature value of the brain wave data;

An output module, configured to input the obtained feature value of the brain wave data into the dream reproduction model to obtain a corresponding output value;

And a reproducing module, configured to determine, from the correspondence, a perceptible object having the highest similarity with the output value, to generate a dream reproduction result.
The apparatus of claim 16, the feature extraction module comprising:

a second decomposition sub-module, configured to perform complex transformation decomposition on the obtained brain wave data, and represent the brain wave data as a sum of at least one complex variable function;

And a third determining submodule configured to use the at least one complex variable function as a feature value of the brain wave data.
The apparatus according to claim 16, wherein the reproducing module comprises:

a fourth determining submodule, configured to determine a reference feature value of each perceptible object in the correspondence relationship;

a calculation submodule, configured to separately calculate a similarity between the output value and a reference feature value of each perceptible object;

A fifth determining sub-module for determining a perceptible object having the highest similarity to generate a dream reproduction result.
A computer device comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, wherein the processor executes the program to implement the method of any one of claims 1 to method.
A computer device comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, wherein the processor executes the program to implement the method of any one of claims 7 to 9 method.