TW202113751A - An on-site supervision method and system thereof - Google Patents

Abstract

The present invention provides a method for on-site supervision, comprising: identifying an emotional value of a user's emotion by image or voice; determining whether the emotion is lower than a threshold; performing feedback or supervision based on the determining; and storing the identifying and feedback as history information.

Description

On-site supervision method and system

The present invention relates to a field supervision system, in particular to a medical field supervision system.

In the existing service industry, such as the medical system, in order to provide better services, medical personnel must observe the patient’s state or emotions by themselves. However, when the medical personnel themselves have other work conditions, even if the patients receive poor services If you experience bad mood or complain, it is easy for medical staff to fail to notice. Therefore, how to pay attention to the user's emotions and improve the service will be a problem that we desperately need to solve.

In view of the purpose of the present invention, the present invention provides an on-site supervision method, which is characterized in that it includes: recognizing user emotions by images or sounds to generate an emotional value; judging whether the emotion is lower than a threshold value; giving feedback or supervision based on the judgment; and Store the identification and feedback as historical data. The server can recognize the user's emotions by at least one of the video or the sound to provide more timely feedback and improve the service, reducing the situation where the user's bad emotions are not noticed and cannot be improved. Since the database is stored as historical data, the system can For training or learning based on the data, the judgment of emotions will be more accurate, and users with poor emotions can be found more accurately.

In view of the purpose of the present invention, the present invention provides an on-site supervision method, which is characterized in that it includes: recognizing the user's emotional characteristics by video or sound; judging whether the emotional characteristics meet the preset emotional characteristics; giving feedback based on the judgment; and storing The identification and the judgment are regarded as historical data. The server can recognize the user's emotions by at least one of the video or the sound to provide more timely feedback and improve the service, reducing the situation where the user's bad emotions are not noticed and cannot be improved. Since the database is stored as historical data, the system can perform training or learning based on the received data, so that the judgment of emotions will be more accurate, and users with poor emotions can be found more accurately.

In view of the purpose of the present invention, the present invention provides an on-site supervision system, which is characterized by comprising: a recognition module, which receives images or voices to recognize the user’s emotion and generates an emotional value; a judgment module, which determines whether the emotional value is lower than a threshold value ; The feedback module, which provides feedback based on the judgment; and the memory, which stores the identification and the judgment as historical data. The server can recognize the user's emotions by at least one of the video or the sound to provide more timely feedback and improve the service, reducing the situation where the user's bad emotions are not noticed and cannot be improved. Since the database is stored as historical data, the system can perform training or learning based on the received data, so that the judgment of emotions will be more accurate, and users with poor emotions can be found more accurately.

In view of the purpose of the present invention, the present invention provides an on-site supervision system, which is characterized by comprising: an emotional characteristic confirmation module, which receives images or voices to recognize the user’s emotional characteristic; a judgment module, which judges whether the emotional characteristic matches a preset mood Features; feedback module, Feedback based on the judgment; and a memory, storing the identification and the feedback as historical data. The server can recognize the user's emotions by at least one of the video or the sound to provide more timely feedback and improve the service, reducing the situation where the user's bad emotions are not noticed and cannot be improved.

100‧‧‧On-site supervision method

101~105‧‧‧Step

201~210‧‧‧Step

300‧‧‧On-site supervision method

301~304‧‧‧Step

400‧‧‧Methods of generating preset emotional characteristics

Steps 401~405‧‧‧

500‧‧‧The process of proactively identifying dissatisfied users

501~505‧‧‧Step

600‧‧‧Characteristic Data Server

1100‧‧‧Site Supervision System

1101‧‧‧Identification Module

1102‧‧‧Input Module

1103‧‧‧Judgment Module

1104‧‧‧Feedback Module

1105‧‧‧Memory

1201‧‧‧Search Module

1202‧‧‧First capture module

1203‧‧‧Comparison Module

1204‧‧‧Screening Module

1205‧‧‧Capture Module

1206‧‧‧Memory

1207‧‧‧Satisfaction Feedback Module

1208‧‧‧Second Capture Module

1209‧‧‧Corresponding module

1210‧‧‧Feature Generation Module

1300‧‧‧Site Supervision System

1301‧‧‧Emotional characteristics confirmation module

1302‧‧‧Judgment Module

1303‧‧‧Feedback Module

1304‧‧‧Memory

1400‧‧‧Establishing a system of pre-determined emotional characteristics

1401‧‧‧Third capture module

1402‧‧‧User Information Deletion Module

1403‧‧‧Fourth capture module

1404‧‧‧Keyword Extraction Module

1405‧‧‧Characteristic data creation module

The embodiments of the present application are shown in the accompanying drawings by way of example rather than limitation, and similar reference numerals indicate similar elements.

Fig. 1 shows a block diagram of an on-site supervision method according to an embodiment of the present invention.

Fig. 2 shows a block diagram for generating a threshold value according to another embodiment of the present invention.

Fig. 3 shows a block diagram of an on-site supervision method according to an embodiment of the present invention.

Fig. 4 shows a block diagram of generating the preset emotional feature according to another embodiment of the present invention.

FIG. 5 shows a schematic diagram of a process of actively identifying dissatisfied users according to an embodiment of the present invention.

FIG. 6 shows a schematic diagram of a feature data server according to an embodiment of the invention.

FIG. 7 shows a schematic diagram of a user feedback screen according to an embodiment of the present invention.

FIG. 8 shows a schematic diagram of a user feedback screen according to another embodiment of the present invention.

Fig. 9 is a schematic diagram of a management background system of the system according to an embodiment of the present invention.

FIG. 10 is a schematic diagram of a background system screen of the management software according to an embodiment of the present invention.

Fig. 11 shows a block diagram of an on-site supervision system according to an embodiment of the present invention.

Fig. 12 shows a block diagram of a system for generating a threshold value according to an embodiment of the present invention.

Fig. 13 shows a block diagram of an on-site supervision system according to an embodiment of the present invention.

Fig. 14 shows a block diagram of a system for establishing preset emotional characteristics according to an embodiment of the present invention.

The principle of the present invention will be described below with reference to several exemplary embodiments shown in the drawings. It should be understood that these embodiments are described only to enable those skilled in the art to better understand and then implement the present invention, but not to limit the scope of the present invention in any way.

Specific examples are given below to illustrate the content of the case in detail, and the drawings are used as an auxiliary description. The symbols mentioned in the description refer to the schematic symbols. It is worth noting that the "include" mentioned in the description is a one-off term and should be interpreted as "includes but is not limited to." In addition, those with general knowledge in the field should understand that the same component/name usually has different addressing methods in different situations, such as "user" or "consumer" or "patient", or "data" or "patient". Data collection", therefore, components/products in the same field and with similar functions mentioned in this description should of course be included in the scope of this description.

The present invention can be applied to many situations. For example, in a medical system, patients are equivalent to users or consumers. Similarly, the method and system of the present invention can also be applied to other industries without departing from the spirit of the present invention.

First, please refer to FIG. 1, which is a block diagram of an on-site supervision method 100 according to an embodiment of the present invention, including: recognizing user emotions by images or sounds to generate an emotion value 101; dissatisfied with the user’s active feedback 102; determining whether the emotion is low At the threshold 103; feedback or supervise 104 based on the judgment; store the identification and feedback as historical data 105; and store the identification and feedback as historical data 105.

For example, in a clinic, the monitoring system can identify at least one of the patient’s images or sounds, that is, when different patients’ emotions respond to expressions, different patient’s emotional values can be generated for different expressions, and then according to the disease The patient’s emotions generate emotional values. If the patient’s emotions are not good, for example, the emotion value is lower than the threshold, it will be judged that the patient’s emotions are not good. At this time, the system can notify the relevant personnel or supervisors of the clinic by SMS to go there Caring for the patient or eliminating the situation that caused the patient's unhealthy mood at the scene. Furthermore, the system can notify the relevant staff of the clinic or the supervisor to link the scene-related video or sound, for example, the system uses e-mail, text message or communication software Send the Internet link to the relevant personnel or supervisors in the form of messages, etc., click on the link to view the relevant video or sound, or directly send the video or sound, so that the relevant personnel or supervisors in the clinic can be Directly confirm the video or sound to immediately grasp the situation on the spot and eliminate or improve it, for example, you can directly care for the patient. In one embodiment, the patient can directly report the on-site situation to the system or clinic through a mobile device or a computer, and can also notify the on-site related personnel or supervisors of the clinic of dissatisfaction. The system can store the above-mentioned content, such as the recognition result or the feedback situation, in the server as historical reference data.

The threshold value is obtained through preprocessing of images or sound. Please refer to FIG. 2, which is a block diagram of generating the threshold value according to another embodiment of the present invention.

The generation threshold can be divided into two situations: the user experiences the service for the first time and the user experiences the service for the second time. The method for generating the threshold value when the user experiences the service for the first time includes: retrieving the user's public image 201; capturing the user image 202; performing comparison 203; screening the public image corresponding to the user image as the first image 204; And the feature of the first image is captured as the feature data 205 whose user satisfaction is above the threshold; and stored in the database 206. The method for generating the threshold value when the user experiences the service for the second time later includes: user feedback satisfaction 207; capturing the user's image 208; corresponding to the satisfaction data to the corresponding image 209; capturing the corresponding difference The characteristics of the image of satisfaction generate the first data 210; and store it in the database 206.

Similarly, taking the clinic as an example, it is divided into two situations: the initial diagnosis of the patient and the second diagnosis of the patient.

First of all, when the patient is first diagnosed, after the patient registers or registers his personal information, the clinic’s system receives the patient’s personal information. The system can then use the patient’s personal information such as name to retrieve relevant information on the Internet, public information, Images are like personal photos. At the same time, the clinic's camera will also record the patient's on-site images, capture the patient's image and upload it to the system. Then, the system compares the patient’s public images with the patient images captured on-site, and selects all public images that match the patient as the first image set, and then captures the features in the first image set as the patient’s satisfaction Characteristic data above medium degree. For example, satisfaction is 1 to 5 points, the most dissatisfied is 0 points, the most satisfied is 5 points, and the satisfaction level is above 3 points. Finally, the characteristic data with the satisfaction level above the middle level is stored in the database as the characteristic data with the satisfaction level above the threshold value.

In addition, when a patient returns to the clinic, such as when the clinic patient is notified during registration, the clinic’s camera will record the patient’s on-site image, capture the patient’s image and upload it to the system to become The second image collection, and when the patient ends the consultation, the satisfaction level of the patient can be reported immediately or afterwards. The satisfaction data returned by the patient corresponds to the patient images captured at the time, and the second image collection is captured The characteristics can generate the characteristic data corresponding to the patients' feedback of different satisfactions, and store them in the database as the characteristic data corresponding to the different satisfactions. The system can train or learn according to the received data, so that the judgment of emotion will be more accurate, and it can also further adjust the threshold value to obtain characteristic data corresponding to different threshold values. For example, if a clinic wants to improve the quality of service, it can lower the threshold. When the patient's emotional characteristics exceed the lower threshold, the system will determine that feedback or supervision is needed to improve the quality of service.

Next, please refer to FIG. 3, which is a block diagram of an on-site supervision method 300 according to an embodiment of the present invention. The method includes: recognizing the user's emotional characteristic 301 by image or sound; judging whether the emotional characteristic meets the preset emotional characteristic 302; giving feedback 303 based on the judgment; and storing the recognition and the judgment as historical data 304.

For example, in a clinic, the monitoring system can recognize at least one of the patient's image or sound, that is, the system can recognize the patient's emotional characteristics. Then determine whether the patient’s emotional characteristics are in line with the preset emotional characteristics. Assuming that the system’s default emotional characteristics are not good, if the patient’s emotional response is not good, the system will determine whether the patient’s emotional characteristics are in line with the system preset Emotional characteristics. If the conditions are met, the system can notify the relevant personnel or supervisors of the clinic by text message to care about the patients or rule out the situation that caused the patients to be emotionally unhealthy. Furthermore, the system can notify the clinic On-site related personnel or supervisors connect to site-related images or sounds, for example, the system sends Internet links to on-site related personnel or supervisors by e-mail, short message or communication software messages, and then click the link View related images or sounds, or directly send images or sounds Sound, so that the relevant personnel or supervisors of the clinic can directly confirm the video or sound to grasp the situation of the scene in real time and eliminate or improve it. For example, they can directly care for the patient. In one embodiment, the patient can directly report the on-site situation to the system or clinic through a mobile device or a computer, and can also notify the on-site related personnel or supervisors of the clinic of dissatisfaction. The system can store the above-mentioned content, such as the recognition result or the feedback situation, in the server as historical reference data.

The preset emotion feature is obtained through preprocessing of images or sound. Please refer to FIG. 4, which is a block diagram of generating the preset emotion feature according to another embodiment of the present invention. The method 400 for generating a preset emotion feature includes: capturing the user audio 401; deleting the information about the user in the audio to generate the first data 402; capturing the first information in which the user's emotion value is lower than a threshold The content of the value becomes the second data 403; the keywords in the second data are retrieved to become the third data 404; and the preset emotion characteristic database 405 is established based on the third data.

For example, more than one clinic has recording equipment that can record on-site patients and even their family members, and then the system deletes the personal data in the recording to generate the first data set, which can also be referred to as the first data, the first data The collection can also be to directly filter out the segments in which the patient is not satisfied with the clinic’s services, and then retrieve the keywords in the first data. In one embodiment, the keywords belonging to poor mood are retrieved. In another implementation, the keywords are retrieved. Take the content of the patient’s emotional value below the threshold in the first data as the second data, and then extract the keywords in the second data to become the third data, and then based on the third data, create information about poor mood or emotions Dissatisfied keyword feature database or preset emotional feature database.

Next, please refer to FIG. 5, which is a schematic diagram of a process of actively identifying dissatisfied users according to an embodiment of the present invention. The process 500 for proactively identifying dissatisfied users includes: image analysis 501; confirm the dissatisfaction feature 502; notify the scene 503; view the image 504; communicate and process 505.

For example, the clinic is equipped with a photographing device. The images taken by the photographing device are analyzed by the system or software to confirm whether the content of the image meets the dissatisfaction characteristics or the emotional value is lower than the preset value. The parts of the dissatisfaction characteristics are confirmed as described above , So the description will not be repeated. If no emotional dissatisfaction characteristics are found in the image, no action is required. However, if it is confirmed that the image content meets the dissatisfaction characteristics or the emotional value is lower than the preset value, in one embodiment, the system can notify the clinic by SMS Relevant personnel or supervisors at the scene will be explained here with supervisors as representatives. In one embodiment, after receiving the short message, the supervisor can click on the link in the short message to confirm the image related to the patient’s emotional dissatisfaction. In another embodiment, the system can directly send the relevant video directly to the supervisor for confirmation , After the supervisor receives the text message or confirms the image, he can immediately deal with it on the spot to eliminate the situation that caused the patient's dissatisfaction, or communicate with the patient to eliminate the patient's dissatisfaction.

Please refer to FIG. 6, according to an embodiment of the present invention, a schematic diagram of a feature data server. The feature data server 600 may include: a dissatisfied body motion picture feature database; a dissatisfied facial expression picture feature database; dissatisfied speech, voice and intonation feature data Database; a database of dissatisfaction speech keyword characteristics. Different databases can be integrated into the same database, or different databases can be processed. In one embodiment, the database can select at least one of them for implementation. The various feature databases mentioned above are only examples, and the present invention is not limited thereto.

Please refer to FIG. 7, which is a schematic diagram of a user feedback screen according to an embodiment of the present invention. For example, in the previous embodiment, when dissatisfied with the user’s active feedback 102 in Figure 1, the system can provide an interface as shown in Figure 7. After the patient has completed the consultation, he can use a mobile device or a desktop computer to directly feedback the current consultation. Experience. Please refer to FIG. 8, which is a schematic diagram of a user feedback screen according to another embodiment of the present invention. For example, disease The system provides further detailed items for patients to fill in and give feedback on the situation of the patient. In this embodiment, the option method is used to make it easy for patients to fill in, and the willingness to give back will not be reduced due to the difficulty of the way of giving back. For example, the person who caused dissatisfaction can be selected as at least one of physicians, counter staff, follow-up staff, or consultants. In another embodiment, the system can also provide patients with direct input of dissatisfaction opinions (not shown). The patient can describe the opinions in more detail so that the clinic understands the problem of the service. Next, please refer to FIG. 9, which is a schematic diagram of the management background system of the system according to an embodiment of the present invention. In one embodiment, after the patient uses the interface as shown in Figure 7 to give feedback, the control background of the clinic can directly monitor the patient's feedback status (such as satisfied, normal, or unsatisfied), and can also learn about services at different times Quality, so that the clinic can be further improved. In one embodiment, the system can also aggregate the satisfaction status and present it in a chart form, so that the supervisors of the clinic can clearly understand the satisfaction status of the service, such as the percentage of patient satisfaction. Further refer to FIG. 10, which is a schematic diagram of the background system screen of the management software according to an embodiment of the present invention. In one embodiment, the system can reflect the percentage of satisfaction for different dates. In another embodiment, the satisfaction level can also be set with a specific warning standard. If the satisfaction level is lower than the standard, the system can further mark it or directly notify the relevant personnel.

Please refer to FIG. 11, which is a block diagram of an on-site supervision system according to an embodiment of the present invention.

The on-site supervision system 1100 includes: a recognition module 1101, which receives images or sounds to recognize user emotions and generates an emotional value; an input module 1102, which is dissatisfied with the user’s active feedback; a judgment module 1103, which determines whether the emotional value is lower than a threshold; The feedback module 1104 provides feedback based on the judgment; and the memory 1105 stores the identification and the judgment as historical data.

For example, in a clinic, at least one of the images or sounds of the patient is captured by the monitoring system and the recognition module 1101 is used to identify the patient’s emotions and generate emotional values, that is, different diseases When the patient’s emotion is reacted to an expression, the recognition module 1101 can generate different emotion values of the patient for different expressions, and then the judgment module 1103 is used to determine the patient’s emotion. If the patient’s emotional response is not good, for example, the emotion value is low At the threshold, it is judged that the patient's mood is not good. At this time, the system can notify the relevant staff or supervisors of the clinic by SMS through the feedback module 1104, to go to care about the patient or eliminate the mood of the patient at the scene. Furthermore, the system can notify the relevant staff in the clinic or the supervisor to link the scene-related video or sound through the feedback module 1104. For example, the system sends an Internet link to the relevant staff in the scene through a text message or a message. Supervisors can view the relevant video or sound after clicking on the link, and can also directly transmit the video or sound, so that the relevant personnel or supervisors in the clinic can directly confirm the video or sound in real time to grasp the situation of the scene and eliminate or improve it , For example, you can go ahead and care about the patient directly. In one embodiment, the patient can directly report the on-site situation to the system or clinic through the mobile device or the input module 1102 of the computer, and can also notify the on-site related personnel or supervisors of the clinic of dissatisfaction. The system can store the aforementioned content, such as the recognition result or the feedback situation, in the memory 1105 of the server as historical reference data.

In particular, part or all of the identification module 1101, the input module 1102, the judgment module 1103, the feedback module 1104, and the memory 1105 included in the on-site supervision system 1100 can be located in different servers or systems. It can be coupled through a wired or wireless network, and can also be integrated in the same server or system. Other embodiments can also be implemented in this manner, and the description will not be repeated.

The threshold value is obtained through preprocessing of images or sound. Please refer to FIG. 12, which is a block diagram of a system for generating a threshold value according to another embodiment of the present invention.

The generation threshold can be divided into two situations: the user experiences the service for the first time and the user experiences the service for the second time. The system that generates a threshold value when the user experiences the service for the first time includes: a retrieval module 1201, which retrieves the user’s public images; a first capture module 1202, which captures user images; and a comparison module 1203, which performs comparison Yes; the screening module 1204, which screens the public image corresponding to the user image as the first image; and the capture module 1205, which captures the features of the first image as the feature data whose user satisfaction is above the threshold; and memory Body 1206, save data. The system that generates a threshold value when the user experiences the service for the second time later includes: a satisfaction feedback module 1207, which feedbacks the satisfaction of the user; a second capture module 1208, which captures an image of the user; a corresponding module 1209 , The data corresponding to the satisfaction degree is sent to the corresponding image; the feature generation module 1210 captures the characteristics of the image corresponding to different satisfaction degrees to generate the first data; and the memory 1206 stores the data.

Similarly, taking the clinic as an example, it is divided into two situations: the initial diagnosis of the patient and the second diagnosis of the patient.

First of all, when the patient is first diagnosed, after the patient registers or registers his personal data, the clinic’s system receives the patient’s personal data, and the system can use the patient’s personal data such as name to retrieve relevant information on the Internet through the retrieval module 1201 Information, public information, and images such as personal photos. At the same time, the camera in the clinic will also record the patient's on-site images, and capture the patient's image through the first capture module 1202 and upload it to the system. Then, the system compares the patient’s public images with the patient’s images captured on-site through the comparison module 1203, then selects all public images that match the patient as the first image set through the screening module 1204, and then through the capture The fetching module 1205 fetches the features in the first image set as feature data with a patient satisfaction level of medium or above. For example, satisfaction is 1~5 points, the most dissatisfied is 0 points, and the most satisfied is 5 points, then A score of 3 or more is average for satisfaction. Finally, store the characteristic data with the satisfaction level above the middle level in the memory 1206 of the database as the characteristic data with the satisfaction level above the threshold value.

Furthermore, when the patient returns to the clinic, such as when the clinic patient is notified during registration, the clinic’s camera will record the patient’s on-site image, and the patient’s image will be captured through the second capture module 1208 and uploaded to the system Become the second image collection, and when the patient ends the consultation, the satisfaction feedback module 1207 can be used to feedback the satisfaction at the time of the consultation immediately or afterwards. The system can use the corresponding module 1209 to respond to the satisfaction data received from the patient at the time. The acquired patient images, through the feature generation module 1210 to capture the features of the second image set, can generate the feature data corresponding to the patient's feedback of different satisfaction levels, and store it in the database as the feature data corresponding to different satisfaction levels . The system can train or learn according to the received data, so that the judgment of emotion will be more accurate, and it can also further adjust the threshold value to obtain the characteristic data corresponding to different threshold values. For example, if the clinic wants to improve the quality of service, it can lower the threshold. When the patient's emotional characteristics exceed the lower threshold, the system will determine that feedback or supervision is needed to improve the quality of service.

Next, please refer to FIG. 13, which is a block diagram of an on-site supervision system according to an embodiment of the present invention. The on-site supervision system 1300 includes: an emotional characteristic confirmation module 1301, which receives images or voices to recognize the user’s emotional characteristic; a judgment module 1302, which judges whether the emotional characteristic meets a preset emotional characteristic; a feedback module 1303, which provides feedback based on the judgment ; And the memory 1304, storing the identification and the feedback as historical data.

For example, in a clinic, the monitoring system can capture at least one of images or sounds of the patient, and the system can recognize the patient's emotional characteristics through the emotional characteristic confirmation module 1301. Then, through the judgment module 1302, it is judged whether the patient’s emotional characteristics meet the preset emotional characteristics. It is assumed that the default emotional characteristics of the system is a situation of poor mood. If the patient’s emotional response is not good Good, the system will determine whether the patient’s emotional characteristics are in line with the preset emotional characteristics of the system. If the conditions are met, the system can notify the relevant personnel or supervisors at the clinic by SMS through the feedback module 1303 to care for the patient. Or to rule out the situation that caused the patient's unhealthy mood at the scene. Furthermore, the system can notify the relevant staff of the clinic or the supervisor to link the scene-related video or sound through the feedback module 1303, for example, the system can send text messages or messages. Internet links to relevant personnel or supervisors at the scene. After clicking on the link, you can view the relevant video or sound, and you can also directly send the video or sound, so that the relevant personnel or supervisors in the clinic can directly confirm the video or sound And immediately grasp the situation on the spot and eliminate or improve it, for example, you can directly care for the patient. In one embodiment, the patient can directly report the on-site situation to the system or clinic through a mobile device or a computer, and can also notify the on-site related personnel or supervisors of the clinic of dissatisfaction. The system can store the above-mentioned content, such as the recognition result or the feedback situation, in the memory 1304 of the server as historical reference data.

The preset emotion features are obtained through preprocessing of images or sounds. Please refer to FIG. 14, which is a block diagram of a system for generating preset emotion features according to another embodiment of the present invention. The system 1400 for creating preset emotional characteristics includes: a third capturing module 1401 to capture the user's audio; a user information deletion module 1402 to delete the information about the user in the audio to generate the first data; The fourth capturing module 1403 captures the content of the first information whose emotional value is lower than the threshold value as the second data; the keyword capturing module 1404 captures the keywords in the second data; and The feature data creation module 1405 creates a preset emotional feature database based on the keywords.

For example, more than one clinic uses the third capture module 1401, such as a video recorder with a radio device, or a recording device with a microphone. As long as it can capture audio, this Ming does not appear here. The third capturing module 1401 can record the patients and even their family members at the scene, and then the system deletes the personal data in the recording through the user information deletion module 1402 to generate the first data set, which can also be referred to as the first data. The first data set can also be used to directly filter out segments in which the patient is not satisfied with the clinic service through the fourth capturing module 1403, and then use the keyword capturing module 1404 to capture the keywords in the first data. In one embodiment In the first data, the keyword extraction module 1404 captures keywords that belong to poor emotions. In another implementation, the fourth capture module 1403 captures the content of the patient's emotional value below the threshold in the first data. The second data, then the keywords in the second data are retrieved by the keyword retrieving module 1404 to become the third data, and then based on the third data, the characteristic data creation module 1405 is used to create information about poor mood or emotional dissatisfaction Keyword feature database or preset emotional feature database.

Now, examples of configurations related to the present invention will be added below. Note that the present invention is not limited to the following configuration.

Configuration 1 may include an on-site supervision method, which is characterized in that it includes: recognizing user emotions by images or sounds to generate an emotional value; determining whether the emotional value is lower than a threshold value; giving feedback based on the determination; and storing the recognition and the Judgment as historical data.

Configuration 2 can be based on the on-site supervision method of configuration 1, wherein the threshold is obtained through preprocessing of the image or sound.

Configuration 3 can be based on the on-site supervision method of configuration 1, wherein the method for generating the threshold includes: retrieving the user’s public image; capturing the user’s image; comparing the public image with the user’s image; screening the corresponding user The public image of the image is the first image; and the feature of the first image is captured as the feature data whose user satisfaction is above the threshold.

Configuration 4 can be based on the on-site supervision method of configuration 1, wherein the method for generating the threshold includes: user feedback satisfaction; capturing the user's image; corresponding to the satisfaction data to the corresponding image; capturing The first data is generated corresponding to the characteristics of the image with different satisfaction levels.

Configuration 5 includes an on-site supervision method, which is characterized by: recognizing the user’s emotional characteristics by video or sound; judging whether the emotional characteristics meets the preset emotional characteristics; giving feedback based on the judgment; and storing the recognition and the judgment as Historical data.

Configuration 6 can be based on the on-site supervision method of configuration 5, wherein the preset emotional feature is obtained through preprocessing of the image or sound.

Configuration 7 can be based on the on-site supervision method of configuration 5, wherein the method for generating the preset emotional characteristics includes: capturing the user's audio; deleting the information about the user in the audio to generate the first data; and capturing the first data. The content of a data whose emotional value of the user is lower than the threshold becomes the second data; the keywords in the second data are retrieved to become the third data; and a preset emotional characteristic database is established based on the third data.

Configuration 8 includes an on-site supervision system, which is characterized in that it includes: a recognition module, which receives images or sounds to recognize the user’s emotions and generates an emotional value; a judgment module, which determines whether the emotional value is lower than a threshold; a feedback module, based on The judgment is fed back; and the memory stores the identification and the judgment as historical data.

Configuration 9 includes an on-site supervision system, which is characterized by including: an emotional feature confirmation module, which receives images or voices to recognize the user’s emotional characteristics; a judgment module, which judges the Whether the emotional characteristics meet the preset emotional characteristics; the feedback module provides feedback based on the judgment; and the memory stores the recognition and the feedback as historical data.

Configuration 10 can be based on the on-site supervision method of configuration 9, in which the system that generates the preset emotional characteristics includes; a third capturing module to capture the user’s audio; a user information deletion module to delete the information about the usage in the audio The first data is generated by the information of the person; the fourth capturing module captures the content of the first information whose emotional value of the user is lower than the threshold value as the second data; the keyword capturing module captures the first data 2. Keywords in the data; and a feature data creation module, based on the keywords to create a preset emotional feature database.

The above descriptions are only optional embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

100‧‧‧On-site supervision method

101~105‧‧‧Step

Claims (10)

An on-site supervision method, characterized in that it includes: recognizing user emotions by images or sounds to generate emotional values; judging whether the emotional value is lower than a threshold value; giving feedback based on the judgment; and storing the recognition and the judgment as historical data . The on-site supervision method according to claim 1, wherein the threshold value is obtained through preprocessing of the image or sound. The on-site supervision method according to claim 1, wherein the method for generating the threshold includes: retrieving the user’s public image; capturing the user’s image; comparing the public image with the user’s image; filtering the corresponding use The public image of the user image is the first image; and the feature of the first image is captured as the feature data whose user satisfaction is above the threshold. The on-site supervision method according to claim 1, wherein the method for generating the threshold value includes: the user feedbacks satisfaction; captures the user's image; the data corresponding to the satisfaction is sent to the corresponding image; The first data is generated by taking the characteristics of the image corresponding to different satisfaction levels. An on-site supervision method, which is characterized in that it includes: recognizing the emotional characteristics of the user by video or sound; Determine whether the emotional feature meets the preset emotional feature; provide feedback based on the judgment; and store the identification and the judgment as historical data. The on-site supervision method according to claim 5, wherein the predetermined emotional characteristic is obtained through preprocessing of the image or sound. The on-site supervision method according to claim 5, wherein the method for generating the preset emotional feature includes: capturing the user's audio; deleting the information about the user in the audio to generate the first data; and capturing the first data The content of a data whose emotional value of the user is lower than the threshold becomes the second data; the keywords in the second data are retrieved to become the third data; and a preset emotional characteristic database is established based on the third data. An on-site supervision system, which is characterized by comprising: an identification module, which receives images or voices to recognize user emotions and generates an emotional value; a judgment module, which judges whether the emotional value is lower than a threshold; and a feedback module, based on the judgment Feedback; and memory, storing the identification and the judgment as historical data. An on-site supervision system, which is characterized by comprising: an emotional characteristic confirmation module, which receives images or voices to recognize the user’s emotional characteristic; a judgment module, which judges whether the emotional characteristic meets a preset emotional characteristic; and a feedback module, based on the judgment Feedback; and memory, storing the identification and the feedback as historical data. The on-site supervision system according to claim 9, wherein the system that generates the preset emotional characteristics includes; a third capturing module to capture the user's audio; a user information deletion module to delete the information about the usage in the audio The first data is generated by the information of the person; the fourth capturing module captures the content of the first information whose emotional value of the user is lower than the threshold value as the second data; the keyword capturing module captures the first data 2. Keywords in the data; and a feature data creation module, based on the keywords to create a preset emotional feature database.

Images