Visual technology-based wearable oversleeve system for psychiatric patient

Abstract

The invention discloses a wearable oversleeve system for a psychiatric patient based on a visual technology, and relates to the field of intelligent medical treatment, and the system comprises an expression recognition module which is used for obtaining a real-time facial image of the patient, building a patient facial expression recognition model, and carrying out the emotion recognition of the real-time facial image of the patient, and obtaining the initial analysis data of an emotion state; the physiological data module is used for acquiring a patient heart rate fluctuation difference, a patient skin electrical activity numerical value fluctuation difference and a patient blood pressure numerical value fluctuation difference corresponding to the patient suffering from the expression according to the emotional state preliminary analysis data to obtain patient suffering physiological data; the emotion release module is used for releasing the emotion of the patient suffering from the pain emotion according to the physiological data of the pain of the patient, the original self-injury behavior of the patient is replaced, the patient can relieve the internal pain and nervous emotion through visual and tactile feedback, and the risk of body injury of the patient can be effectively reduced.

Description

Technical Field

[0001] The present invention belongs to the field of smart medical care and relates to visual technology, in particular to a wearable sleeve system for psychiatric patients based on visual technology. Background Art

[0002] Self-harm in psychiatric patients refers to the behavior of individuals intentionally causing harm or damage to their own bodies, including cutting, burning, hitting, squeezing, etc., with the purpose of relieving inner emotional pain or tension through physical pain. Self-harm is common in patients with mental illnesses such as mood disorders, post-traumatic stress disorder, and personality disorders, reflecting a coping mechanism when they face emotional distress. This behavior requires comprehensive psychiatric and psychological assessments to understand the motivations behind it and its impact, and to provide appropriate treatment and intervention support;

[0003] The existing treatment and nursing methods for self-harm behavior of psychiatric patients have the following deficiencies: 1. Relying on drugs to control patients' mental symptoms is prone to a series of side effects and requires a long medication period; 2. Psychotherapy / skill intervention requires the active participation and cooperation of patients, but some patients may not be able to fully devote themselves to psychotherapy / skill training due to the influence of their illness or personal reasons, which affects the treatment effect.

[0004] To this end, we propose a wearable sleeve system for psychiatric patients based on visual technology to replace patients' self-harming behaviors under the influence of painful emotions. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a wearable sleeve system for psychiatric patients based on visual technology. The present invention is based on acquiring real-time facial images of patients, establishing a facial expression recognition model library for patients, performing emotion recognition on real-time facial images of patients, and obtaining preliminary analysis data of emotional states. According to the preliminary analysis data of emotional states, the patient's heart rate fluctuation difference, the patient's skin electrical activity value fluctuation difference and the patient's blood pressure value fluctuation difference corresponding to the patient's painful expression are obtained to obtain the patient's painful physiological data, and the patient's painful emotions are emotionally released according to the patient's painful physiological data.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: The specific working process of each module of a wearable sleeve system for psychiatric patients based on visual technology is as follows: Expression recognition module: used to obtain the patient's real-time facial image, establish the patient's facial expression recognition model to perform emotion recognition on the patient's real-time facial image, and obtain preliminary analysis data of the emotional state; Physiological data module: used to obtain the patient's heart rate fluctuation difference, skin electrical activity value fluctuation difference and blood pressure value fluctuation difference corresponding to the patient's painful expression according to the preliminary analysis data of the emotional state, and obtain the patient's painful physiological data; Emotional release module: used to release the patient's painful emotions based on the patient's painful physiological data.

[0007] Furthermore, the expression recognition module obtains preliminary analysis data of the emotional state, as follows: Acquire the patient's facial image in real time through the first sleeve system sub-device to obtain the patient's real-time facial image; The big data crawler technology is used to crawl images with facial expressions of psychiatric patients as keywords to obtain multiple facial image data of psychiatric patients, and the obtained images are named as j1th to jjth facial images of psychiatric patients respectively; Performing image preprocessing on the j1th psychiatric patient's facial image to obtain the j1th processed facial image; Repeat the process of acquiring the j1th processed facial image, pre-process the j2th to jjth psychiatric patient facial images respectively, and obtain the j2th to jjth processed facial images; Establishing a patient facial expression recognition model according to the processing of the j1-th to jj-th facial images; Use the tested patient facial expression recognition model to perform expression recognition on the patient's real-time facial image to obtain preliminary analysis data of the emotional state; If the real-time facial image of the patient is a first expression image, it is determined that the patient is in a first emotional state interval; If the real-time facial image of the patient is a second expression image, it is determined that the patient is in the second emotional state interval.

[0008] Furthermore, the expression recognition module obtains the j1th processed facial image as follows: In the j1st psychiatric patient's facial image, the patient's facial edge contour is filled with lines to obtain a facial contour line circle, with the patient's nose tip as the center point, and the longest distance between the facial contour line circle and the center point as the characteristic length; In the j1th psychiatric patient's facial image, the center point is taken as the intersection point of the diagonal lines of the cropping rectangle, the feature length is half the diagonal length of the cropping rectangle, and the cropping rectangle corresponding to the j1th psychiatric patient's facial image is obtained. The j1th psychiatric patient's facial image is cropped by the cropping rectangle to obtain the j1th cropped facial image; The image brightness corresponding to the j1-th cropped facial image is adjusted to the preset target brightness of the image to obtain the j1-th processed facial image.

[0009] Furthermore, the expression recognition module creates and tests a patient facial expression recognition model, as follows: Dividing the j1-th to jj-th processed facial images into a facial image test set and a facial image test set according to a preset division ratio; Each processed facial image in the facial image test set is divided into a first expression image and a second expression image by artificial division, and the division result corresponding to each processed facial image is marked in the image to obtain a facial image annotation test set; Repeat the process of dividing the facial image annotation test set, divide the facial image test set into images, and obtain the facial image annotation test set; Use the existing artificial intelligence model to establish a platform to create a patient facial expression recognition model, use the facial image annotation test set to perform a recognition test on the patient facial expression recognition model, wait for the facial image annotation test set to complete a round of testing on the patient facial expression recognition model, use the facial image annotation test set to test the patient facial expression recognition model, and obtain the test results to obtain a reference value for the accuracy of the test results; Obtain the test result accuracy target value. If the test result accuracy reference value is greater than or equal to the test result accuracy target value, the test of the patient's facial expression recognition model is completed. If the test result accuracy reference value is less than the test result accuracy target value, continue to use the facial image annotation test set to repeatedly test the patient's facial expression recognition model until the test result accuracy reference value is greater than or equal to the test result accuracy target value.

[0010] Furthermore, the physiological data module acquires the physiological data of the patient's pain, as follows: Obtain preliminary analysis data of emotional state; According to the preliminary analysis data of emotional state, the heart rate data of the patients in the first emotional state interval are analyzed to obtain the heart rate fluctuation difference of the patients; According to the preliminary analysis data of emotional state, skin electrical activity data of patients in the first emotional state interval are analyzed to obtain the fluctuation difference of the patients' skin electrical activity values; According to the preliminary analysis data of emotional state, blood pressure data of patients in the first emotional state interval are analyzed to obtain the fluctuation difference of blood pressure values ​​of patients; The patient's heart rate fluctuation difference, the patient's skin electrical activity value fluctuation difference and the patient's blood pressure value fluctuation difference are defined as the patient's painful physiological data.

[0011] Furthermore, the physiological data module obtains the patient's heart rate fluctuation difference as follows: When the patient is in the first emotional state interval, the patient's real-time heart rate value is obtained through the wearable cuff to obtain the patient's first heart rate value; Acquire a heart rate value corresponding to when the patient is in a second emotional state interval to obtain a second heart rate value of the patient; The difference between the first heart rate value of the patient and the second heart rate value of the patient is calculated to obtain the patient's heart rate fluctuation difference.

[0012] Furthermore, the physiological data module obtains the fluctuation difference of the patient's skin electrical activity value, as follows: When the patient is in the first emotional state interval, the patient's real-time skin electrical activity value is obtained through the wearable sleeve to obtain a first skin electrical activity value; Acquire the patient's skin electrical activity value corresponding to when the patient is in the second emotional state interval to obtain a second skin electrical activity value; The difference between the first skin electrical activity value and the second skin electrical activity value is calculated to obtain the patient's skin electrical activity value fluctuation difference.

[0013] Furthermore, the physiological data module obtains the fluctuation difference of the patient's blood pressure value, as follows: When the patient is in the first emotional state interval, the real-time diastolic blood pressure value and the real-time systolic blood pressure value of the patient are obtained through the wearable cuff, and they are named as the first diastolic blood pressure value and the first systolic blood pressure value respectively; Obtaining a diastolic blood pressure value and a systolic blood pressure value corresponding to the patient in the second emotional state interval, and naming them a second diastolic blood pressure value and a second systolic blood pressure value, respectively; The first diastolic pressure value, the second diastolic pressure value, the first systolic pressure value and the second systolic pressure value are calculated to obtain the patient's blood pressure value fluctuation difference corresponding to the first emotional state interval.

[0014] Furthermore, the emotion release module obtains the patient's emotion release coefficient, as follows: Acquire the patient's painful physiological data, and acquire the patient's heart rate fluctuation difference, the patient's skin electrical activity value fluctuation difference, and the patient's blood pressure value fluctuation difference based on the patient's painful physiological data; The patient's emotional release coefficient is obtained by calculating the patient's heart rate fluctuation difference, the patient's skin electrical activity value fluctuation difference and the patient's blood pressure value fluctuation difference; Obtain the patient's emotion release coefficient threshold, compare the patient's emotion release coefficient with the patient's emotion release coefficient threshold, and release the patient's emotions according to the comparison result.

[0015] Furthermore, the emotion release module performs emotion release on the patient, specifically as follows: Obtaining the patient's heart rate baseline fluctuation difference, the patient's skin electrical activity value baseline fluctuation difference, and the patient's blood pressure value baseline fluctuation difference; The patient's emotional release coefficient threshold is obtained by calculating the patient's heart rate baseline fluctuation difference, the patient's skin electrical activity value baseline fluctuation difference and the patient's blood pressure value baseline fluctuation difference; When the patient's emotional release coefficient is greater than or equal to the patient's emotional release coefficient threshold, the wearable sleeve produces red lines and a blurring effect, and releases electromagnetic pulses, causing the patient to feel pain and present a visual effect of self-injury; When the patient's emotional release coefficient is less than the patient's emotional release coefficient threshold, the wearable sleeve releases a slight electromagnetic pulse, causing the patient to feel slight pain.

[0016] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. The present invention can replace the patient's original self-harm behavior, allowing the patient to relieve the inner pain and tension through visual and tactile feedback, and can effectively reduce the occurrence of physical injuries; 2. The present invention monitors the emotional state of psychiatric patients through facial recognition and physiological data, timely discovers the precursors of self-harm and provides effective intervention, which can effectively reduce the risk of physical injury; 3. The present invention determines the intensity of the patient's painful emotions by calculating the patient's emotion release coefficient, which can effectively ensure the pertinence of emotion relief measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0018] Figure 1 is the overall system block diagram of the present invention; Figure 2 is the first expression image of the present invention; Figure 3 It is the second expression image of the present invention. DETAILED DESCRIPTION

[0019] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Embodiment 1 See also Figure 1 , the present invention provides a technical solution: a wearable sleeve system for psychiatric patients based on visual technology, comprising an expression recognition module, a physiological data module, an emotion release module and a server, wherein the expression recognition module, the physiological data module and the emotion release module are respectively connected to the server; The expression recognition module obtains the patient's facial expression image to judge the painful expression and obtains preliminary analysis data of the emotional state; The patient's facial image is acquired in real time through the first sleeve system sub-device to obtain the patient's real-time facial image.

[0021] It should be noted here that: In this application, the first sleeve system sub-device is a camera; By using big data crawler technology to crawl images with facial expressions of psychiatric patients as keywords, multiple facial image data of psychiatric patients are obtained, and the obtained images are named as facial images of psychiatric patients from j1 to jj, respectively.

[0022] It should be noted here that: jj is the corresponding number of facial images of psychiatric patients crawled in this application; The j1th psychiatric patient's facial image is preprocessed to obtain the j1th processed facial image.

[0023] The details are as follows: In the j1st psychiatric patient's facial image, the patient's facial edge contour is filled with lines to obtain a facial contour line circle, with the patient's nose tip as the center point, and the longest distance between the facial contour line circle and the center point as the characteristic length; In the j1th psychiatric patient's facial image, the center point is taken as the intersection point of the diagonal lines of the cropping rectangle, the feature length is half the diagonal length of the cropping rectangle, and the cropping rectangle corresponding to the j1th psychiatric patient's facial image is obtained. The j1th psychiatric patient's facial image is cropped by the cropping rectangle to obtain the j1th cropped facial image; The image brightness corresponding to the j1-th cropped facial image is adjusted to the preset target brightness of the image to obtain the j1-th processed facial image; Repeat the acquisition process of processing the j1th facial image, pre-process the facial images of the j2th to jjth psychiatric patients respectively, and obtain the j2th to jjth processed facial images; The facial images processed from j1 to jj are divided into a facial image test set and a facial image test set according to a preset division ratio.

[0024] It should be noted here that: In the present application, the preset division ratio is specifically 3:7, that is, the ratio of the number of the facial image test set to the number of the facial image test set is 3:7; Each processed facial image in the facial image test set is divided into a first expression image and a second expression image by manual division, and the division result corresponding to each processed facial image is marked in the image to obtain a facial image annotation test set.

[0025] It should be noted here that: See also Figure 2 , the first expression image is a painful expression image; See also Figure 3 , the second expression image is a non-painful expression image; Repeat the process of dividing the facial image annotation test set, divide the facial image test set into images, and obtain the facial image annotation test set; Use the existing artificial intelligence model to establish a platform to create a patient facial expression recognition model, use the facial image annotation test set to perform a recognition test on the patient facial expression recognition model, wait for the facial image annotation test set to complete a round of testing on the patient facial expression recognition model, use the facial image annotation test set to test the patient facial expression recognition model, and obtain the test results to obtain a reference value for the accuracy of the test results; Obtain the test result accuracy target value. If the test result accuracy reference value is greater than or equal to the test result accuracy target value, the test of the patient's facial expression recognition model is completed. If the test result accuracy reference value is less than the test result accuracy target value, continue to use the facial image annotation test set to repeatedly test the patient's facial expression recognition model until the test result accuracy reference value is greater than or equal to the test result accuracy target value. Use the tested patient facial expression recognition model to perform expression recognition on the patient's real-time facial image to obtain preliminary analysis data of the emotional state; If the real-time facial image of the patient is a first expression image, it is determined that the patient is in a first emotional state interval; If the real-time facial image of the patient is a second expression image, determining that the patient is in a second emotional state interval; The expression recognition module obtains preliminary analysis data of the emotional state and transmits it to the physiological data module; The physiological data module obtains the physiological data corresponding to the patient with painful expression according to the preliminary analysis data of the emotional state, and obtains the physiological data of the patient's pain; Obtain preliminary analysis data of emotional state; According to the preliminary analysis data of the emotional state, the heart rate data of the patients in the first emotional state interval are analyzed to obtain the heart rate fluctuation difference of the patients.

[0026] The details are as follows: When the patient is in the first emotional state interval, the patient's real-time heart rate value is obtained through the wearable cuff to obtain the patient's first heart rate value; Acquire a heart rate value corresponding to when the patient is in a second emotional state interval to obtain a second heart rate value of the patient; The difference between the first heart rate value of the patient and the second heart rate value of the patient is calculated to obtain the patient's heart rate fluctuation difference.

[0027] It should be noted here that: The patient's heart rate value involved here is obtained by monitoring the patient's pulse through a wearable cuff; According to the preliminary analysis data of the emotional state, the skin electrical activity data of the patients in the first emotional state interval are analyzed to obtain the fluctuation difference of the skin electrical activity values ​​of the patients.

[0028] The details are as follows: When the patient is in the first emotional state interval, the patient's real-time skin electrical activity value is obtained through the wearable sleeve to obtain a first skin electrical activity value; Acquire the patient's skin electrical activity value corresponding to when the patient is in the second emotional state interval to obtain a second skin electrical activity value; Calculating the difference between the first skin electrical activity value and the second skin electrical activity value to obtain the patient's skin electrical activity value fluctuation difference; According to the preliminary analysis data of emotional state, blood pressure data of patients in the first emotional state interval are analyzed to obtain the fluctuation difference of blood pressure values ​​of patients; When the patient is in the first emotional state interval, the real-time diastolic blood pressure value and the real-time systolic blood pressure value of the patient are obtained through the wearable cuff, and they are named as the first diastolic blood pressure value and the first systolic blood pressure value respectively; Obtaining a diastolic blood pressure value and a systolic blood pressure value corresponding to the patient in the second emotional state interval, and naming them a second diastolic blood pressure value and a second systolic blood pressure value, respectively; The first diastolic pressure value, the second diastolic pressure value, the first systolic pressure value and the second systolic pressure value are calculated to obtain the patient's blood pressure value fluctuation difference corresponding to the first emotional state interval; Calculate the fluctuation difference of the patient's blood pressure value. The specific formula is as follows: ; Wherein, Hxb is the fluctuation difference of the patient's blood pressure value, Sy1 is the first diastolic pressure value, Sy2 is the second diastolic pressure value, Sz1 is the first systolic pressure value, and Sz2 is the second systolic pressure value; The patient's heart rate fluctuation difference, the patient's skin electrical activity value fluctuation difference and the patient's blood pressure value fluctuation difference are defined as the patient's painful physiological data; The physiological data module acquires the patient's painful physiological data and transmits it to the emotional release module; The emotional release module provides emotional release for patients with painful emotions based on their painful physiological data; Acquire the patient's painful physiological data, and acquire the patient's heart rate fluctuation difference, the patient's skin electrical activity value fluctuation difference, and the patient's blood pressure value fluctuation difference based on the patient's painful physiological data; The patient's emotional release coefficient is obtained by calculating the patient's heart rate fluctuation difference, the patient's skin electrical activity value fluctuation difference and the patient's blood pressure value fluctuation difference; The patient's emotional release coefficient is calculated, and the specific formula configuration is as follows: ; Among them, Qhs is the patient's emotional release coefficient, Hxb is the patient's blood pressure value fluctuation difference, Pfd is the patient's skin electrical activity value fluctuation difference, Xlb is the patient's blood pressure value fluctuation difference, a1 and a2 are both set proportional coefficients and a1 and a2 are both greater than 0; Obtain the patient's emotion release coefficient threshold, compare the patient's emotion release coefficient with the patient's emotion release coefficient threshold, and release the patient's emotions according to the comparison result.

[0029] The details are as follows: Obtaining the patient's heart rate baseline fluctuation difference, the patient's skin electrical activity value baseline fluctuation difference, and the patient's blood pressure value baseline fluctuation difference; The patient's emotional release coefficient threshold is obtained by calculating the patient's heart rate baseline fluctuation difference, the patient's skin electrical activity value baseline fluctuation difference and the patient's blood pressure value baseline fluctuation difference; The patient's emotional release coefficient threshold is calculated, and the specific formula configuration is as follows: ; Among them, Qhy is the threshold value of the patient's emotional release coefficient, Hxj is the patient's blood pressure value baseline fluctuation difference, Pfj is the patient's skin electrical activity value baseline fluctuation difference, Xlj is the patient's blood pressure value baseline fluctuation difference, a1 and a2 are both set proportional coefficients and a1 and a2 are both greater than 0; When the patient's emotional release coefficient is greater than or equal to the patient's emotional release coefficient threshold, the wearable sleeve produces red lines and a blurring effect, and releases electromagnetic pulses, causing the patient to feel pain and present a visual effect of self-injury; When the patient's emotional release coefficient is less than the patient's emotional release coefficient threshold, the wearable sleeve releases a slight electromagnetic pulse, causing the patient to feel slight pain.

[0030] In this application, if corresponding calculation formulas appear, the above calculation formulas are all dimensionless and take their numerical calculations. The weight coefficients, proportional coefficients and other coefficients in the formulas are set to a result value obtained by quantifying each parameter. The size of the weight coefficient and the proportional coefficient can be determined as long as it does not affect the proportional relationship between the parameter and the result value.

[0031] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A wearable sleeve system for psychiatric patients based on visual technology, characterized in that: include: Expression recognition module: used to obtain the patient's real-time facial image, establish the patient's facial expression recognition model to perform emotion recognition on the patient's real-time facial image, and obtain preliminary analysis data of the emotional state; Physiological data module: used to obtain the patient's heart rate fluctuation difference, skin electrical activity value fluctuation difference and blood pressure value fluctuation difference corresponding to the painful expression according to the preliminary analysis data of the emotional state, and obtain the patient's painful physiological data; Emotional release module: used to release the patient's painful emotions based on the patient's physiological data.

2. A wearable sleeve system for psychiatric patients based on visual technology according to claim 1, characterized in that: The expression recognition module obtains preliminary analysis data of the emotional state, as follows: Acquire the patient's facial image in real time through the first sleeve system sub-device to obtain the patient's real-time facial image; The big data crawler technology is used to crawl images with facial expressions of psychiatric patients as keywords to obtain multiple facial image data of psychiatric patients, and the obtained images are named as j1th to jjth facial images of psychiatric patients respectively; Performing image preprocessing on the j1th psychiatric patient's facial image to obtain the j1th processed facial image; Repeat the process of processing and acquiring the j1th facial image, and pre-process the j2th to jjth facial images of psychiatric patients respectively to obtain the j2th to jjth facial images; Establishing a patient facial expression recognition model according to the processing of the j1-th to jj-th facial images; Use the tested patient facial expression recognition model to perform expression recognition on the patient's real-time facial image to obtain preliminary analysis data on the patient's emotional state; If the real-time facial image of the patient is a first expression image, it is determined that the patient is in a first emotional state interval; If the real-time facial image of the patient is a second expression image, it is determined that the patient is in the second emotional state interval.

3. A wearable sleeve system for psychiatric patients based on visual technology according to claim 2, characterized in that: The expression recognition module obtains the j1th processed facial image as follows: In the j1st psychiatric patient's facial image, the patient's facial edge contour is filled with lines to obtain a facial contour line circle, with the patient's nose tip as the center point, and the longest distance between the facial contour line circle and the center point as the characteristic length; In the j1th psychiatric patient's facial image, the center point is taken as the intersection point of the diagonal lines of the cropping rectangle, the feature length is half the diagonal length of the cropping rectangle, and the cropping rectangle corresponding to the j1th psychiatric patient's facial image is obtained. The j1th psychiatric patient's facial image is cropped by the cropping rectangle to obtain the j1th cropped facial image; The image brightness corresponding to the j1-th cropped facial image is adjusted to the preset target brightness of the image to obtain the j1-th processed facial image.

4. A wearable sleeve system for psychiatric patients based on visual technology according to claim 2, characterized in that: The expression recognition module creates and tests the patient's facial expression recognition model as follows: Dividing the j1-th to jj-th processed facial images into a facial image test set and a facial image test set according to a preset division ratio; Each processed facial image in the facial image test set is divided into a first expression image and a second expression image by artificial division, and the division result corresponding to each processed facial image is marked in the image to obtain a facial image annotation test set; Repeat the process of dividing the facial image annotation test set, divide the facial image test set into images, and obtain the facial image annotation test set; Use the existing artificial intelligence model to establish a platform to create a patient facial expression recognition model, use the facial image annotation test set to perform a recognition test on the patient facial expression recognition model, wait for the facial image annotation test set to complete the test on the patient facial expression recognition model, use the facial image annotation test set to test the patient facial expression recognition model, and obtain the test results to obtain a reference value for the accuracy of the test results; Obtain the test result accuracy target value. If the test result accuracy reference value is greater than or equal to the test result accuracy target value, the test of the patient's facial expression recognition model is completed. If the test result accuracy reference value is less than the test result accuracy target value, continue to use the facial image annotation test set to repeatedly test the patient's facial expression recognition model until the test result accuracy reference value is greater than or equal to the test result accuracy target value.

5. A wearable sleeve system for psychiatric patients based on visual technology according to claim 1, characterized in that: The physiological data module acquires the patient's painful physiological data, as follows: Obtain preliminary analysis data of emotional state; According to the preliminary analysis data of emotional state, the heart rate data of the patients in the first emotional state interval are analyzed to obtain the heart rate fluctuation difference of the patients; According to the preliminary analysis data of emotional state, skin electrical activity data of patients in the first emotional state interval are analyzed to obtain the fluctuation difference of the patients' skin electrical activity values; According to the preliminary analysis data of emotional state, blood pressure data of patients in the first emotional state interval are analyzed to obtain the fluctuation difference of blood pressure values ​​of patients; The patient's heart rate fluctuation difference, skin electrode activity value fluctuation difference and blood pressure value fluctuation difference are defined as the patient's painful physiological data.

6. A wearable sleeve system for psychiatric patients based on visual technology according to claim 5, characterized in that: The physiological data module obtains the patient's heart rate fluctuation difference, specifically as follows: When the patient is in the first emotional state interval, obtaining the patient's real-time heart rate value to obtain the patient's first heart rate value; Acquire a heart rate value corresponding to when the patient is in a second emotional state interval to obtain a second heart rate value of the patient; The difference between the first heart rate value of the patient and the second heart rate value of the patient is calculated to obtain the patient's heart rate fluctuation difference.

7. A wearable sleeve system for psychiatric patients based on vision technology according to claim 5, characterized in that: The physiological data module obtains the fluctuation difference of the patient's skin electrical activity value, as follows: When the patient is in the first emotional state interval, obtaining the patient's real-time skin electrical activity value to obtain a first skin electrical activity value; Acquire the patient's skin electrical activity value corresponding to when the patient is in the second emotional state interval to obtain a second skin electrical activity value; The difference between the first skin electrical activity value and the second skin electrical activity value is calculated to obtain the patient's skin electrical activity value fluctuation difference.

8. A wearable sleeve system for psychiatric patients based on vision technology according to claim 5, characterized in that: The physiological data module obtains the fluctuation difference of the patient's blood pressure value, as follows: When the patient is in the first emotional state interval, obtaining the patient's real-time diastolic blood pressure value and real-time systolic blood pressure value, and naming them as a first diastolic blood pressure value and a first systolic blood pressure value, respectively; Obtaining a diastolic blood pressure value and a systolic blood pressure value corresponding to the patient in the second emotional state interval, and naming them a second diastolic blood pressure value and a second systolic blood pressure value, respectively; The first diastolic pressure value, the second diastolic pressure value, the first systolic pressure value and the second systolic pressure value are calculated to obtain the patient's blood pressure value fluctuation difference corresponding to the first emotional state interval.

9. A wearable sleeve system for psychiatric patients based on vision technology according to claim 1, characterized in that: The emotion release module obtains the patient's emotion release coefficient, as follows: Obtaining the patient's painful physiological data, and obtaining the patient's heart rate fluctuation difference, skin electrical activity value fluctuation difference, and blood pressure value fluctuation difference based on the patient's painful physiological data; The patient's emotional release coefficient is obtained by calculating the patient's heart rate fluctuation difference, skin electrical activity value fluctuation difference and blood pressure value fluctuation difference; Obtain the patient's emotion release coefficient threshold, compare the patient's emotion release coefficient with the patient's emotion release coefficient threshold, and release the patient's emotions according to the comparison result.

10. A wearable sleeve system for psychiatric patients based on vision technology according to claim 9, characterized in that: The emotion release module performs emotion release on the patient, specifically as follows: Obtain the patient's heart rate baseline fluctuation difference, skin electrical activity value baseline fluctuation difference, and blood pressure value baseline fluctuation difference; The patient's emotional release coefficient threshold is obtained by calculating the patient's heart rate baseline fluctuation difference, skin electrical activity value baseline fluctuation difference and blood pressure value baseline fluctuation difference; When the patient's emotional release coefficient is greater than or equal to the patient's emotional release coefficient threshold, the wearable sleeve produces red lines and a blurring effect, and releases electromagnetic pulses, causing the patient to feel pain and present a visual effect of self-injury; When the patient's emotional release coefficient is less than the patient's emotional release coefficient threshold, the wearable sleeve releases a slight electromagnetic pulse, causing the patient to feel slight pain.

Images