CN115359914A

CN115359914A - Method, apparatus, and medium for discriminating psychological states based on quantization description vector

Info

Publication number: CN115359914A
Application number: CN202211033515.2A
Authority: CN
Inventors: 宋业臻; 肖维斌; 王荣全; 韩伟; 黄岩; 曲继新
Original assignee: Shandong Xinfa Technology Co ltd
Current assignee: Shandong Xinfa Technology Co ltd
Priority date: 2022-08-26
Filing date: 2022-08-26
Publication date: 2022-11-18
Anticipated expiration: 2042-08-26
Also published as: CN115359914B

Abstract

The application discloses a mental state distinguishing method, equipment and a medium based on a quantitative description vector, wherein the method comprises the following steps: displaying the pre-collected negative image data to a user; collecting multi-dimensional physiological indexes of a user when the user watches negative image data; determining to generate a multidimensional quantization description vector according to the negative image data in advance, and determining the classification and the classification grade of the negative image data obtained according to the multidimensional quantization description vector; and generating a judgment result according to the multi-dimensional physiological indexes acquired when the user observes negative image data of different classifications and grades and a preset judgment rule so as to distinguish the categories of the psychological state of the user. Before the negative image data are displayed to the user, the negative image data are quantized into multi-dimensional quantitative description vectors in advance, so that the effective dimensions in the image data can be accurately extracted and clustered, more reasonable rules can be formulated for the data, and the psychological states of the user can be accurately classified.

Description

Method, apparatus, and medium for discriminating psychological states based on quantization description vector

Technical Field

The present application relates to the field of data identification data representation, and in particular, to a method, an apparatus, and a medium for distinguishing a psychological state based on a quantization description vector.

Background

As the quality of life of people increases, in addition to physical conditions, people are also more concerned about psychological conditions, which may even lead to depressive disorders (also called depression). Psychological states are also often classified into various types, such as healthy state, sub-healthy state, and depressive disorder, and more particularly, depressive disorder is also classified into depressive disorder not caused by anxiety disorder, anxiety disorder-co-morbid depressive disorder, and the like.

In a traditional distinguishing mode, a psychiatric clinician usually distinguishes the distinguishing mode, the distinguishing process basically depends on subjective judgment of the clinician, and the final distinguishing result is easy to be abnormal.

Disclosure of Invention

In order to solve the above problem, the present application provides a method for distinguishing a psychological state based on a quantization description vector, including:

the method comprises the steps that pre-collected negative image data are displayed to a user, wherein the negative image data comprise predefined negative factors and can have negative influence on the emotion of at least part of people, and the negative influence is determined through the changes of nuclear magnetic resonance imaging data and brain wave data of the at least part of people when the negative image data are observed;

acquiring a multi-dimensional physiological index of the user when watching the negative image data;

determining to generate a multi-dimensional quantization description vector according to the negative image data in advance, and determining the classification of the negative image data and the grade under the classification according to the multi-dimensional quantization description vector;

generating a judgment result according to the multi-dimensional physiological indexes acquired when the user observes negative image data of different classifications and grades and a preset judgment rule so as to distinguish the categories of the psychological state of the user, wherein the categories of the psychological state at least comprise: anxiety disorders co-morbid depressive disorders, non-anxiety induced depressive disorders.

In one example, generating a multidimensional quantization description vector from the negative image data specifically includes:

for each negative image data, acquiring answer data of multiple persons to the negative image data according to multiple preset questions by a multi-person evaluation mode, and forming a group of multidimensional quantization description vectors { LT, VS, CT } according to the answer data, wherein the LT vector is a binary variable and represents binary judgment on loss and threat; the VS vector is a continuous variable, represents a valence index, and is provided with a highest score and a lowest score; the CT vector is a sentence content vector composed of natural language vocabularies and describes the content of the negative image data.

In an example, obtaining the classification of the negative image data and the grade under the classification according to the multi-dimensional quantization description vector specifically includes:

analyzing the CT vector through natural language processing to obtain a description subject word corresponding to the negative image data, wherein the description subject word is used for representing a category corresponding to the negative image data;

for the VS vector, averaging the scores of the negative image data by a plurality of users in advance;

and performing cluster analysis on the negative image data belonging to the same description subject term on the basis of the average term, and in the process of cluster analysis, enabling the intra-cluster distance to be smaller than the inter-cluster distance, and obtaining the optimized cluster number under the condition that the intra-cluster distance is minimized and the inter-cluster distance is maximized, wherein the optimized cluster number is used for representing the grade of the negative image data.

In one example, the presetting the plurality of questions comprises: whether the negative image data has been lost or will be dangerous, the level of discomfort to you for the scene in the negative image data, a summary of the main content of the negative image data using a dialog.

In one example, the multi-dimensional physiological metric includes: at least a plurality of dimensions of eye tracking data, facial expression information, heart rate data, heart rate variability data;

the classification includes: the method includes at least one of a loss, a stain-bamboo, a dangerous scene, and a social separation, wherein the loss, the stain-bamboo, and the social separation each include a plurality of levels, and the dangerous scene does not distinguish the levels.

In one example, the generating a determination result according to the multi-dimensional physiological index collected when the user observes negative image data of different classifications and grades and a preset determination rule specifically includes:

if the user observes the negative image data classified as suffering loss and having the highest grade and the multi-dimensional physiological indexes of the user accord with preset standards when observing the negative image data classified as social separation, judging that the psychological state of the user belongs to the non-anxiety induced depressive disorder; and/or

If the user observes that the user suffers from the slight soil and the multidimensional physiological indexes of the user all meet the preset standard when the user observes that the user is classified as the slight soil, judging that the psychological state of the user belongs to the non-anxiety induced depressive disorder; and/or

If the user observes the negative image data classified as the suffered loss and the multi-dimensional physiological indexes of the user accord with the preset standard when observing the negative image data classified as the suffered loss, judging that the psychological state of the user belongs to the anxiety disorder comorbid depressive disorder; and/or

And if the multi-dimensional physiological indexes of the user accord with the preset standard when the user observes the negative image data classified as the dangerous scene, judging that the psychological state of the user belongs to the anxiety disorder comorbid depressive disorder.

In one example, the eye tracking data includes a glance mode, an attention transfer duration;

the preset criteria include: the saccade mode is a fine machining saccade, the attention transfer duration is longer than a preset duration, the facial expression information is negative, the heart rate is increased, and the heart rate variability is reduced.

In one example, the negative image data includes image data, video data; the negative factors comprise natural factors and human social factors, wherein the natural factors comprise: at least one of overcast and rainy weather, rainstorm weather and flood weather; the social factors considered include: the lost object is at least one of a search, divorce, car accident, death and attack.

In another aspect, the present application further provides a mental state distinguishing apparatus based on a quantization description vector, including:

at least one processor; and (c) a second step of,

a memory communicatively coupled to the at least one processor; wherein, the first and the second end of the pipe are connected with each other,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform such as:

In another aspect, the present application further proposes a non-transitory computer storage medium storing computer-executable instructions configured to:

the method comprises the steps that pre-collected negative image data are displayed for a user, wherein the negative image data comprise predefined negative factors and can have negative influence on the emotion of at least part of people, and the negative influence is determined by the changes of nuclear magnetic resonance imaging data and brain wave data of the at least part of people when the negative image data are observed;

acquiring a multi-dimensional physiological index of the user when the user watches the negative image data;

The psychological state distinguishing method based on the quantization description vector can bring the following beneficial effects:

before the negative image data are displayed to a user, the negative image data are quantized into multi-dimensional quantitative description vectors in advance, so that the effective dimensions in the image data can be accurately extracted and clustered, more reasonable rules can be formulated for the data, and the psychological states of the user can be accurately classified.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

FIG. 1 is a flowchart illustrating a method for distinguishing psychological states based on quantized description vectors according to an embodiment of the present disclosure;

fig. 2 is a schematic flowchart of a psychological state distinguishing method based on a quantization description vector in a scene in an embodiment of the present application;

fig. 3 is a schematic diagram of a mental state distinguishing apparatus based on a quantization description vector in an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

As shown in fig. 1 and fig. 2, an embodiment of the present application provides a mental state distinguishing method based on a quantization description vector, including:

s101: and displaying pre-acquired negative image data to a user, wherein the negative image data comprises predefined negative factors and can generate negative influence on the emotion of at least part of people, and the negative influence is determined by the changes of the nuclear magnetic resonance imaging data and the brain wave data of the at least part of people when the negative image data is observed.

Negative image data may include image data, video data (typically 10-30 seconds in length), and the like. The negative factors negatively affect the mood of at least a part of the population, wherein the population may comprise users of different psychological states, such as health state, users with depressive disorders, etc., and the negative effects on the mood may be generated even for the population with health state after seeing the negative factors. The negative effects may be heart damage, depression, anger, fear, etc. And the negative factors can include natural factors and human social factors, wherein the natural factors include: at least one of overcast and rainy weather, rainstorm weather and flood weather; social factors are believed to include: the lost object is at least one of a search, divorce, car accident, death, and attack. After the original images are collected in advance, a group of original scene image databases can be formed. Of course, the image data and the video data may be merged into the same database, or a set of original scene video databases may be formed separately for the video data.

Further, for classifying the anxiety disorder co-morbidity depressive disorder and the non-anxiety induced depressive disorder, image library screening is carried out on clinical samples, physiological evidence related to the mental disorder, clinical psychology expert judgment evidence, screening of the emotional disorder including the anxiety disorder and the depressive disorder and multi-angle evidence of auxiliary diagnostic tool evidence are taken as standards, stimulation effectiveness of the original scene image database is reevaluated, and image library deletion is carried out according to the stimulation effectiveness.

The method comprises the steps of firstly, determining brain function evidence, wherein the brain science diagnosis evidence of the depressive disorder is a non-reward attractor network state with an external orbital frontal cortex as a core according to a neuroscience principle of the depressive disorder proposed by Rolls, E.T. of brain science researchers of Oxford university; the brain scientific diagnostic evidence for anxiety disorders is the continuous activity of the amygdala-thalamus-adrenal circuit (HPA) axis, and the two neuroscience evidences were taken as the first part of the clinical sample for distinguishing depressive disorders from anxiety disorders.

The clinical psychiatric medical experts judge the evidence, a plurality of clinical psychiatric medical experts of the psychiatry department clinic are selected to diagnose the clinical samples, and the plurality of experts obtain the consensus, so that the evidence can be used as an effective grouping evidence.

The clinical samples are filled with diagnostic tools, mainly completing the filling of four tools of MMPI, SDS, SAS and SCL-90, and grouping according to scores of anxiety disorder co-morbid depressive disorder and non-anxiety induced depressive disorder. The evidence points to the samples with consistent directions from the three angles, and the samples can be divided into an anxiety disorder co-morbid depressive disorder group and a non-anxiety induced depressive disorder group, and the processes are repeatedly iterated to finally form two groups of 200 clinical samples.

And secondly, starting the collection of screening standard data. Requiring a clinical sample to lie on an MRI observation cabin or an EEG monitoring bed, completing the observation of the image library, and synchronously completing the MRI and EEG data acquisition; meanwhile, the 5 clinical psychiatric experts carry out sample emotional state observation on site and evaluate whether emotional abnormal reactions occur. When the change of the MRI data or the EEG data of the brain wave data is consistent with the judgment of the expert, the image corresponding to the abnormal reaction is marked and marked as an effective image. And repeating the iteration of the process to finish the labeling of the image library.

And thirdly, screening image materials capable of generating emotional stimulation to the anxiety disorder co-morbid depressive disorder and the non-anxiety induced depressive disorder respectively in the image library according to the evaluation result of the previous step, and meanwhile, meeting the requirements of generating stimulation to the anxiety disorder co-morbid depressive disorder but not generating stimulation to the non-anxiety induced depressive disorder, and generating stimulation to the non-anxiety induced depressive disorder but not generating stimulation to the anxiety disorder co-morbid depressive disorder.

S102: and acquiring multi-dimensional physiological indexes of the user when the user watches the negative image data.

Wherein, the multi-dimensional physiological indexes comprise: at least a plurality of dimensions of eye tracking data, facial expression information, heart rate data, heart rate variability data. The facial expression information (which may include facial expressions FE: active/passive, high arousal/low arousal) and the heart rate data (RR) and Heart Rate Variability (HRV) data captured by the sensors may be captured by the user's eye tracking data (which may include saccade patterns SM, attention transfer durations ATT) and camera while observing the relevant stimulus images.

S103: and determining to generate a multi-dimensional quantization description vector according to the negative image data in advance, and determining the classification of the negative image data and the grade under the classification according to the multi-dimensional quantization description vector.

Specifically, for each negative image data, answer data of multiple persons to the negative image data according to a plurality of questions set in advance is acquired by a multiple person evaluation method (for example, a 5 person evaluation method). Presetting a plurality of questions may include:

problem 1: whether a scene in an image/video has been lost or will be dangerous. Problem 2: scenes in the image/video feel how uncomfortable you feel. (the answer included four options: no discomfort-slight discomfort-relatively uncomfortable-very uncomfortable) question 3: please summarize the main content of the image/video using a dialog.

Forming a group of multidimensional quantization description vectors { LT, VS, CT } according to the answer data, wherein the LT vector is a binary variable and represents binary classification judgment on loss and threat; the VS vector is a continuous variable, represents a valence index, and is provided with a maximum score of 4 and a minimum score of 1; the CT vector is a sentence content vector consisting of natural language vocabularies and describes the content of the negative image data.

Further, for the CT vector, a description subject word corresponding to the negative image data is obtained by analyzing through natural language processing (for example, LDA natural language processing), where the description subject word is used to indicate a category corresponding to the negative image data. Wherein, the processing procedure of LDA natural language processing mode can be as formula

As shown.

And aiming at the VS vector, pre-scoring the negative image data by a plurality of users for an average term, then further performing cluster analysis on the negative image data belonging to the same description subject word by taking the average term as a basis, and solving a Minkowski distance between every two clusters in the process of cluster analysis so as to enable the intra-cluster distance to be smaller than the inter-cluster distance, and obtaining the optimized cluster number under the condition of minimizing the intra-cluster distance and maximizing the inter-cluster distance, wherein the optimized cluster number is used for representing the grade of the negative image data. The process of cluster analysis can be as formula

As shown.

The negative Image data after the calculation can be edited into a knowledge graph according to the body structure of the Image/Video-LT-CT-VS, so as to be convenient for subsequent use.

Wherein the classifying may include: suffer from at least one of loss, suffer from slight stain, dangerous scene, and social separation.

Specifically, the loss, the slight stain, and the social separation each include a plurality of levels, and the dangerous scene does not distinguish the levels.

For example, the suffered loss may include four levels, where level 1 (lightest) is a slight loss of a avatar in the scene losing a few change, losing an umbrella, etc.; level 2 (lighter) is that the avatar in the scene has lost the identity card or bank card; level 3 (heavy) is that a avatar in the scene encounters a divorce event; level 4 (severe) is that the avatar in the scene encountered a car accident and the avatar in the scene died.

The libel suffered from the pollution can comprise three levels, wherein the level 1 (lightest) is that the virtual human in the scene refuses certain requests by the leader and certain requests by the family; level 2 (medium) is the avatar in the scene being abused by other avatars; level 3 (severe) is when the avatar in the scene is discarded or expelled from the house after being abused by the house, relatives.

Social separation may include three levels, level 1 (lightest) being that the avatar in the scene is rejected by individual colleagues, classmates, who are not close friends of the avatar; level 2 (medium) is that the virtual person in the scene is rejected by relatives, friends; level 3 (severe) is that avatars in the scene are rejected by close relatives, ancestors, lovers.

The danger scenario is then not graded, which may include: the scene where the virtual person is located is about to have storm, the scene where the virtual person is located has criminals in dark places, and the like.

In the image library, at least 10 image combinations with the same content are covered in each classification so as to ensure the diversity when the images are displayed for users.

S104: generating a judgment result according to the multi-dimensional physiological indexes acquired when the user observes negative image data of different classifications and grades and a preset judgment rule so as to distinguish the categories of the psychological state of the user, wherein the categories of the psychological state at least comprise: anxiety disorders co-morbid depressive disorders, non-anxiety induced depressive disorders.

The mental states may include: health status, sub-health status, depressive disorder status, etc., and the depressive disorder status can be further subdivided into anxiety disorder co-morbid depressive disorder, non-anxiety induced depressive disorder, etc. In these categories of mental states, the differentiation of depressive disorders is more difficult to do in traditional solutions. Several conventional schemes are provided, such as using artificial intelligence analysis of physiological signals to infer whether a user has depressive disorder, using an artificial intelligence analysis method of electroencephalogram signals to distinguish healthy users from depressed users, and using audio-video and natural language processing techniques to perform multi-modal characterization analysis of users to infer whether a user has a depressive disorder-related depression.

However, the above conventional solutions mainly aim at the problem of "whether there is a depressive disorder" or "whether there is an affective characterization of a depressive disorder", and provide an automatic determination function by analyzing related performance information using an artificial intelligence technique. However, in clinical practice, whether or not a depressive disorder is suffered is not a core problem of practice, but it is a practical problem to distinguish a non-anxiety-causing depressive disorder from an anxiety disorder co-morbid with a depressive disorder.

According to theoretical research of psychiatry and cognitive neuroscience, the main cause of depression is the excessive sensitivity of an individual to non-reward stimulation signals and punishment signals, so that the duration of the lateral orbit frontal cortex after activation is longer, and stable signal communication is formed between the lateral orbit frontal cortex and a speech path, an attention path and the like, so that a non-reward network attractor effect is formed, the symptom of depression is generated, the symptoms are mainly clinically expressed as low mood and depression, and the situation of loss is relatively high in sensitivity; the main cause of anxiety is the increased activation of the thalamic-adrenal circuits by an individual's hypersensitivity to signals of threatening stimuli, with greater sensitivity to the threatening (not actually occurring) situation that is expected to occur, combining the feature that depression and anxiety have different sensitivities to the actual and expected losses in brain function and cognitive function, the scheme in this application can differentiate the category of depressive disorder suffered by the user based on this feature.

Specifically, if the user observes negative image data classified as suffering from loss and having the highest rank (4 ranks) and observes negative image data classified as social separation, and the multidimensional physiological indicators of the user all meet preset criteria, it is determined that the psychological state of the user belongs to non-anxiety-induced depressive disorder.

And if the multi-dimensional physiological indexes of the user meet the preset standard when the user observes that the user is classified as suffering from the slight, judging that the psychological state of the user belongs to non-anxiety induced depressive disorder.

If the user observes the negative image data classified as suffering from the loss and the multi-dimensional physiological indexes of the user accord with preset standards when observing the negative image data classified as suffering from the slight loss, judging that the psychological state of the user belongs to anxiety disorder comorbid depressive disorder.

And if the multi-dimensional physiological indexes of the user meet the preset standard when the user observes the negative image data classified into the dangerous scene, judging that the psychological state of the user belongs to the anxiety disorder comorbid depressive disorder.

Wherein, the preset standard comprises: the saccade mode is fine processing saccade, the time length of attention transfer is longer than the preset time length, facial expression information is negative, the heart rate is increased, and the heart rate variability is reduced.

After the preset standard is obtained, the body structure of the Image/Video-LT-CT-VS-SM-ATT-FE-RR-HRV is used as a data structure of a user and is edited into a knowledge graph, and the two types of assistant discrimination knowledge graphs of the non-anxiety induced depressive disorder and the anxiety comorbid depressive disorder are edited so as to automatically give an analysis conclusion according to the knowledge graph subsequently.

As shown in fig. 3, an embodiment of the present application further provides a mental state distinguishing apparatus based on a quantization description vector, including:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to cause the at least one processor to perform such as:

determining to generate a multidimensional quantization description vector according to the negative image data in advance, and determining the classification of the negative image data obtained according to the multidimensional quantization description vector and the grade under the classification;

An embodiment of the present application further provides a non-volatile computer storage medium storing computer-executable instructions, where the computer-executable instructions are configured to:

The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on differences from other embodiments. In particular, for the device and media embodiments, the description is relatively simple as it is substantially similar to the method embodiments, and reference may be made to some descriptions of the method embodiments for relevant points.

The device and the medium provided by the embodiment of the application correspond to the method one to one, so the device and the medium also have the similar beneficial technical effects as the corresponding method, and the beneficial technical effects of the method are explained in detail above, so the beneficial technical effects of the device and the medium are not repeated herein.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, 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 disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of other like elements in a process, method, article, or apparatus comprising the element.

The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement or the like made within the spirit and principle of the present application shall be included in the scope of the claims of the present application.

Claims

1. A method for distinguishing a psychological state based on a quantized description vector, comprising:

2. The method according to claim 1, wherein generating a multidimensional quantization description vector from the negative image data specifically comprises:

for each negative image data, acquiring answer data of multiple persons to the negative image data according to multiple preset questions by a multi-person evaluation mode, and forming a group of multidimensional quantization description vectors { LT, VS, CT } according to the answer data, wherein the LT vector is a binary variable and represents two-classification judgment on loss and threat; the VS vector is a continuous variable, represents a valence index, and is provided with a highest score and a lowest score; the CT vector is a sentence content vector composed of natural language vocabularies and describes the content of the negative image data.

3. The method according to claim 2, wherein the obtaining of the classification of the negative image data and the grade under the classification according to the multi-dimensional quantization description vector comprises:

analyzing the CT vector through natural language processing to obtain a description subject term corresponding to the negative image data, wherein the description subject term is used for representing the category corresponding to the negative image data;

4. The method according to claim 2, wherein the presetting of the plurality of questions comprises: whether the negative image data has been lost or will be dangerous, the level of discomfort to you for the scene in the negative image data, a summary of the main content of the negative image data using a dialog.

5. The method according to claim 1, wherein the multi-dimensional physiological index comprises: at least a plurality of dimensions in eye tracking data, facial expression information, heart rate data, heart rate variability data;

6. The method according to claim 5, wherein the step of generating a determination result according to the multi-dimensional physiological index collected when the user observes negative image data of different classifications and levels and a preset determination rule specifically comprises:

If the user observes that the user suffers from the libel, the multi-dimensional physiological indexes of the user all accord with the preset standard, and then the psychological state of the user is judged to belong to the non-anxiety induced depressive disorder; and/or

7. The method according to claim 6, wherein the eye tracking data comprises a saccade mode, an attention transfer duration;

the preset standard comprises the following steps: the glance mode is a fine-process glance, the attention transfer duration is higher than a preset duration, the facial expression information is negative, the heart rate is increased, and the heart rate variability is decreased.

8. The method according to claim 1, wherein the negative image data comprises image data, video data; the negative factors comprise natural factors and human social factors, wherein the natural factors comprise: at least one of overcast and rainy weather, rainstorm weather and flood weather; the human social factors include: the lost object is at least one of a search, divorce, car accident, death, and attack.

9. A psychological state discrimination apparatus based on a quantization description vector, comprising:

at least one processor; and the number of the first and second groups,

10. A non-transitory computer storage medium storing computer-executable instructions, the computer-executable instructions configured to: