CN114022944A - Intelligent monitoring system - Google Patents
Intelligent monitoring system Download PDFInfo
- Publication number
- CN114022944A CN114022944A CN202210003345.7A CN202210003345A CN114022944A CN 114022944 A CN114022944 A CN 114022944A CN 202210003345 A CN202210003345 A CN 202210003345A CN 114022944 A CN114022944 A CN 114022944A
- Authority
- CN
- China
- Prior art keywords
- risk
- image
- analysis module
- analysis
- face image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/70—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mental therapies, e.g. psychological therapy or autogenous training
Landscapes
- Health & Medical Sciences (AREA)
- Child & Adolescent Psychology (AREA)
- Developmental Disabilities (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Engineering & Computer Science (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to an intelligent monitoring system, comprising: the acquisition module is used for shooting the face image information within a preset time interval T0; the recognition module is used for recognizing the characteristics of the face image; the analysis module is used for determining the actual risk level of the monitored person according to the micro-expression in the plurality of frames of face images which accord with the analysis standard, determining and sending early warning information of the corresponding level according to the actual risk level, and sending an early warning instruction of the corresponding level to the early warning module by the analysis module when the early warning information of the corresponding level is determined and sent; the early warning module is used for receiving the early warning instruction sent by the analysis module and sending a corresponding early warning signal according to the early warning instruction; the storage module is used for storing data generated when the system runs, and the micro expression analysis method and the system can accurately analyze the micro expression of the monitored person, improve the accuracy of micro expression analysis and further accurately master the mental health condition of the monitored person.
Description
Technical Field
The invention relates to the technical field of intelligent monitoring, in particular to an intelligent monitoring system.
Background
Among the daily management of prison, the video monitor system of prison has realized the coverage of total area basically, but prior art's monitored control system only is used for the monitoring at the scope of activity of escorting personnel, can't carry out at the mental health analysis of escorting personnel according to the facial image of monitoring image, and unable accurate screening facial image carries out the micro expression analysis with accurate to the person of escorting to accurate grasp at the mental health condition of escorting personnel.
Disclosure of Invention
Therefore, the invention provides an intelligent monitoring system which is used for solving the problem that in the prior art, a face image cannot be accurately screened so as to accurately analyze the micro expression of escort personnel.
In order to achieve the above object, the present invention provides an intelligent monitoring system, comprising:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module comprises at least one acquisition unit and is used for shooting face image information in a preset time interval T0, and the face image information comprises a plurality of frames of face images with a preset time sequence acquired according to a preset central standard;
the recognition module is connected with the acquisition module and comprises at least one recognition unit used for recognizing the micro-expression characteristics of the face image according to a preset recognition center;
the analysis module is connected with the recognition module and comprises at least one analysis unit, the analysis unit is used for comparing the micro expressions with a preset risk image model according to the micro expressions in a plurality of frames of human face images which accord with analysis standards so as to determine the actual risk level of the monitored person, the analysis module determines to send early warning information of corresponding level according to the actual risk level, and when the early warning information of corresponding level is determined to be sent, the analysis module sends an early warning instruction of corresponding level to the early warning module; the analysis module calculates the actual image contact ratio of each frame of facial image in sequence, compares the actual contact ratio with a preset value to select the facial image which accords with the micro-expression analysis standard, calculates the actual risk score of the monitored person according to the facial image which accords with the micro-expression analysis standard, sends an early warning instruction of a corresponding grade according to the actual risk score, and corrects the actual risk grade according to the time interval of the adjacent risk image when sending the early warning instruction;
the early warning module is connected with the analysis module and used for receiving the early warning instruction sent by the analysis module and sending a corresponding early warning signal according to the early warning instruction;
and the storage module is respectively connected with the acquisition module, the identification module, the analysis module and the early warning module and is used for storing data generated during the operation of the system.
Further, when the analysis module determines a face image meeting an analysis standard, the analysis module obtains a first frame of face image shot by the acquisition module, and the first frame of face image is marked as a standard face image, the analysis module compares the first frame of face image with the remaining other frame images in sequence to calculate an actual image coincidence degree a, the analysis module compares the actual image coincidence degree with a preset image coincidence degree a0, and judges whether the remaining other frame face images meet the analysis standard according to a comparison result;
the preset image overlap ratio A0 comprises a first preset image overlap ratio A1 and a second preset image overlap ratio A2, wherein A1 < A2;
when A is larger than A2, the analysis module judges that the frame of face image does not meet the analysis standard and does not use the frame of face image to carry out micro-expression analysis;
when A is not less than A1 and not more than A2, the analysis module judges that the frame of face image meets the analysis standard and uses the frame of face image to perform micro-expression analysis;
when A is less than A2, the analysis module judges that the frame of face image does not meet the analysis standard and does not use the frame of face image to carry out micro expression analysis.
Further, when the analysis module judges that the face image meets the analysis standard, the analysis calculates an actual image overlap ratio difference value delta A, the actual image overlap ratio difference value delta A is compared with a preset image overlap ratio difference value delta A0, a micro-expression analysis position is judged according to a comparison result, and delta A = | A- (A1 + A2)/2 | is set;
the preset image overlap ratio difference Δ a0 comprises a first preset image overlap ratio difference Δ a1 and a second preset image overlap ratio difference Δ a2, wherein Δ a1 < Δa 2;
when delta A > -delta A2, setting the position of the facial image for micro expression analysis as a lip by the analysis module;
when the delta A1 is less than the delta A and less than the delta A2, the analysis module sets the position of the facial image for micro expression analysis as the eyes;
when the delta A is less than or equal to the delta A1, the analysis module sets the position of the facial image for micro expression analysis as the eyebrow.
Further, when the analysis is completed at the position for performing the micro expression analysis, the analysis module compares each frame of facial image with a preset risk image model in sequence to calculate the actual similarity B of the facial image, and when the actual similarity of the facial image is greater than 93%, the analysis module judges that the frame of facial image is a risk image.
Further, when the analysis module judges the face image as a risk image, the analysis module calculates an actual similarity difference value delta B, compares the actual image similarity difference value with a preset similarity difference value, determines the risk grade of the risk image according to the comparison result, and sets delta B = B-93%;
the preset similarity difference comprises a first similarity difference delta B1, a second similarity difference delta B2 and a third similarity difference delta B3, wherein delta B1 < [ delta ] B2 < [ delta ] B3;
when the delta B is less than the delta B1, the analysis module judges the risk level of the frame of the human face image as a primary risk image;
when the delta B is not less than delta B1 and less than delta B2, the analysis module judges the risk grade of the frame of the human face image as a secondary risk image;
when the delta B2 is more than or equal to the delta B and less than the delta B3, the analysis module judges the risk grade of the frame of the human face image as a three-level risk image;
when the delta B is not less than the delta B3, the analysis module judges the risk level of the frame of the face image as a four-level risk image;
wherein the order of risk levels is: the four-level risk image > the three-level risk image > the two-level risk image > the one-level risk image.
Further, when the analysis module finishes the judgment of the risk level of the risk image, the analysis module calculates an actual risk score Q of the face image information, compares the actual risk score Q with a preset risk score Q0, and judges the face image information level according to the comparison result;
the preset risk scores Q0 include a first preset risk score Q1, a second preset risk score Q2, and a third preset risk score Q3, wherein Q1 < Q2 < Q3;
when Q is less than Q1, the analysis module judges the risk of the face image information as primary risk image information, the analysis module judges that a preset time interval T0 needs to be adjusted and does not need to send an early warning instruction to the early warning module, the analysis module marks the adjusted preset time interval as T1 and sets T1= T0 x (1- (Q1-Q)/Q1);
when Q1 is not less than Q < Q2, the analysis module judges the risk of the face image information as secondary risk image information and sends a secondary early warning instruction to the early warning module;
when Q2 is not less than Q < Q3, the analysis module judges the risk of the face image information as three-level risk image information and sends a three-level early warning instruction to the early warning module;
when Q is less than Q3, the analysis module judges the risk of the face image information as four-level risk image information and sends a four-level early warning instruction to the early warning module;
wherein, the early warning instruction rank order is: the fourth-level early warning instruction is larger than the third-level early warning instruction, the second-level early warning instruction is larger than the first-level early warning instruction.
Further, when the analysis module calculates an actual risk score Q of the face image information, the analysis module obtains a time interval Δ T of the two adjacent frames of risk images, Δ T = Ta-Tb, where Ta is an adjacent first frame of risk image capturing time, and Tb is an adjacent second frame of risk image capturing time, the analysis module corrects the actual risk score Q according to Δ T, and the analysis module records the corrected actual risk score as Qa, and sets Qa = Q x (1 + (Δt1+ (+ Δ T2+ … … +/Δ Tn)/0.4 × T0), where Δ T1 represents the time interval of the first two adjacent frames of risk images, and Δ Tn represents the time interval of the nth two adjacent frames of risk images.
Further, the actual risk score Q is calculated using the following formula,
Q=（F1+F2）ⅹK+F3ⅹ2+F4ⅹ3；
wherein F1 is the number of first-level risk images, F2 is the number of second-level risk images, F3 is the number of third-level risk images, F4 is the number of fourth-level risk images, and K is the risk score calculation weight.
Further, the risk score calculation weight K is calculated using the following formula,
K=1.15ⅹ（F2/F1）；
wherein the value range of K is 0.98-1.33.
Further, when the analysis module determines a face image meeting an analysis standard and calculates the coincidence degree A of an actual image, the analysis module acquires a second frame of face image shot by the acquisition module, the second frame of face image is marked as a standard face image, and the analysis module compares the second frame of face image with the rest of other frame images in sequence;
the analysis module acquires the j frame of face image shot by the acquisition module, the j frame of face image is marked as a standard face image, and the analysis module compares the j frame of face image with the rest of other frame images in sequence.
Compared with the prior art, the method has the advantages that the information of the facial image shot by the acquisition module in the preset time interval is acquired in real time through the analysis module, the acquired first frame image is used as a standard image, the definition of the standard image and the rest of other frame images is calculated, the facial image which meets the micro-expression analysis standard can be accurately screened, on one hand, when the coincidence degree of the actual image is too large, the facial image indicates that the expression of the monitored person is not changed or the expression is slightly changed, the analysis module further eliminates the frame of facial image, when the coincidence degree is too small, the expression of the monitored person is obviously changed and the real mental health condition of the monitored person cannot be displayed, so that the analysis module does not use the frame of image to perform micro-expression analysis, and the coincidence degree of the actual image is compared in real time through the analysis module, rejection that can be accurate is not conform to and carries out the facial image of micro expression analysis standard, and then analysis module can be accurate carries out the micro expression analysis to being monitored personnel, micro expression analysis's rate of accuracy has been improved, and then accurate grasp is by monitoring personnel's mental health condition, another aspect thereof, analysis module is when carrying out the micro expression analysis, analysis module compares actual facial image with predetermineeing the risk image model, and then when improving micro expression analysis's rate of accuracy, can accurately send early warning information, the staff can be according to different early warning signal, carry out psychological tutor to being monitored personnel, thereby effectual assurance is at escort personnel's mental health.
Furthermore, the preset image overlap ratio is set as two standards, the face image meeting the analysis standard is determined by the analysis module, the analysis module can compare the image overlap ratio in real time and screen the face image information accurately according to the comparison result, so that the face image meeting the micro-expression analysis and marking can be selected accurately, the accuracy of micro-expression analysis can be effectively improved through the accurate selection of the analysis module, the psychological health condition of the escorting personnel can be mastered more accurately, the monitored personnel can be subjected to psychological guidance according to the actual condition, and the psychological health of the escorting personnel is effectively guaranteed.
Furthermore, after the analysis module screens the images which meet the micro-expression analysis standard, the analysis module calculates the difference value of the coincidence degree of the actual images in real time and compares the difference value of the coincidence degree of the actual images with a preset value, when the difference value of the coincidence degree of the actual images is larger, the analysis module judges that the micro-expression change of the monitored person is larger, and further judges that the position of the micro-expression change is in the lips, so as to perform key analysis on the lip parts, further accurately master the mental health condition of the monitored person through the micro-expression change of the lips, when the difference value of the coincidence degree of the actual images is moderate, the analysis module judges that the micro-expression change of the monitored person is moderate, and further judges that the position of the micro-expression change is in the eyes, so as to perform key analysis on the empirical parts, further accurately master the mental health condition of the monitored person through the micro-expression change of the eyes, when the difference value of the coincidence degree of actual image is less, the analysis module judges that the change of micro expression of the monitored person is less, and then the position of judging that the change of micro expression occurs is at the eyebrow, thereby carrying out key analysis on the eyebrow part, and then the change of micro expression through the eyebrow, the mental health condition of the monitored person is mastered accurately, thereby the analysis module can pass through multiple angles and multiple dimensions, the analysis of micro expression is carried out from different parts in the face image, the accuracy of the analysis of micro expression is further improved, and the mental health of the monitored person is further effectively ensured.
Furthermore, the analysis module of the invention is also provided with standard image similarity, when the analysis module finishes the part for carrying out micro expression analysis on the face image, the face image is compared with the preset risk image model, when the actual image similarity reaches the preset image similarity standard, the analysis module marks the frame image as a risk image, and the analysis module accurately judges the risk image, thereby further improving the accuracy of micro expression analysis and further more effectively ensuring the mental health of the monitored personnel.
Furthermore, the analysis module is also provided with a plurality of standard similarity difference values, when the analysis module evaluates the risk level of the risk image, the analysis module calculates the actual similarity difference values in real time, when the actual similarity difference values are larger, the analysis module judges that the frame of risk image has larger similarity with the preset sharing image, and further indicates that the probability of the monitored person having the mental health problem is larger, so that the larger the level of the frame of risk image is, the analysis module can accurately master the mental health condition of the monitored person according to the risk level of each frame of image, further improves the accuracy of micro-expression analysis, and further effectively ensures the mental health of the monitored person.
Furthermore, the analysis module of the invention can also calculate the actual risk score of the face information according to the risk grade of each frame of risk image, the psychological condition of the monitored person can be reflected more accurately by calculating the actual risk score in real time, when the analysis module judges that the face image information is the primary risk image information, the analysis module judges that the probability of the monitored person having the psychological monitoring problem is smaller, the psychological health of the monitored person can be analyzed secondarily by correcting the shooting time interval, thereby the information health condition of the monitored person can be mastered more accurately, when the analysis module judges that the risk grade of the face image information is two-level or more, the analysis module judges that the probability of the monitored person having the psychological monitoring problem is larger, and further the psychological health condition of the escort person can be mastered more accurately by sending the early warning instruction, and psychological counseling can be carried out on the monitored personnel according to the grade of the early warning signal, so that the psychological health of the escort personnel is effectively ensured.
Furthermore, when the analysis module calculates the actual risk score, the analysis module obtains the time interval between two actual adjacent frames of risk images and corrects the actual risk score, when the time interval between two adjacent frames of risk images is smaller, the analysis module judges that the probability of the monitored personnel suffering from the mental health problem is higher, when the time interval between two adjacent frames of risk images is larger, the analysis module judges that the probability of the monitored personnel suffering from the mental health problem is smaller, the risk score is corrected accurately through the analysis module, the mental health condition of the escorting personnel can be mastered more accurately, and the mental health of the monitored personnel can be assisted according to the early warning signal level, so that the mental health of the escorting personnel is effectively ensured.
Furthermore, when the analysis module calculates the actual risk score Q, the analysis module counts the number of the primary risk images and the number of the secondary risk images, when the primary risk images account for a larger proportion, the analysis module judges that the probability of the monitored personnel suffering from the mental health problem is smaller, the analysis module judges that the actual risk score is lower and selects a smaller calculation weight to accurately calculate the actual risk score, when the number of the secondary risk images is larger, the analysis module judges that the probability of the monitored personnel suffering from the mental health problem is larger, the analysis module judges that the actual risk score is higher and selects a larger calculation weight, and the risk score calculation weight is accurately selected by the analysis module, so that the risk grade of the image information of the personnel can be accurately mastered, and the mental health condition of the escort personnel can be accurately mastered, and psychological counseling can be carried out on the monitored personnel according to the grade of the early warning signal, so that the psychological health of the escort personnel is effectively ensured.
Furthermore, when the analysis module screens the face images meeting the micro-expression analysis standard, different frames are selected as standard images, the face images meeting the micro-expression analysis standard can be screened more comprehensively, missing and wrong selection can be effectively prevented, the face images not meeting the micro-expression analysis standard can be accurately eliminated, the analysis module can accurately perform micro-expression analysis on the monitored personnel, the accuracy of micro-expression analysis is improved, and the mental health condition of the monitored personnel can be accurately mastered.
Drawings
Fig. 1 is a schematic structural diagram of an intelligent monitoring system according to the present invention.
Detailed Description
In order that the objects and advantages of the invention will be more clearly understood, the invention is further described below with reference to examples; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and do not limit the scope of the present invention.
It should be noted that in the description of the present invention, the terms of direction or positional relationship indicated by the terms "upper", "lower", "left", "right", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, which are only for convenience of description, and do not indicate or imply that the device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.
Furthermore, it should be noted that, in the description of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
Referring to fig. 1, a schematic structural diagram of the intelligent monitoring system according to an embodiment of the present invention is shown, including:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module comprises at least one acquisition unit and is used for shooting face image information in a preset time interval T0, and the face image information comprises a plurality of frames of face images with a preset time sequence acquired according to a preset central standard;
the recognition module is connected with the acquisition module and comprises at least one recognition unit used for recognizing the micro-expression characteristics of the face image according to a preset recognition center;
the analysis module is connected with the recognition module and comprises at least one analysis unit, the analysis unit is used for comparing the micro expressions with a preset risk image model according to the micro expressions in a plurality of frames of human face images which accord with analysis standards so as to determine the actual risk level of the monitored person, the analysis module determines to send early warning information of corresponding level according to the actual risk level, and when the early warning information of corresponding level is determined to be sent, the analysis module sends an early warning instruction of corresponding level to the early warning module; the analysis module calculates the actual image contact ratio of each frame of facial image in sequence, compares the actual contact ratio with a preset value to select the facial image which accords with the micro-expression analysis standard, calculates the actual risk score of the monitored person according to the facial image which accords with the micro-expression analysis standard, sends an early warning instruction of a corresponding grade according to the actual risk score, and corrects the actual risk grade according to the time interval of the adjacent risk image when sending the early warning instruction;
and the storage module is respectively connected with the acquisition module, the identification module, the analysis module and the early warning module and is used for storing data generated during the operation of the system.
Specifically, in the embodiment of the present invention, the analysis module acquires the information of the facial image captured by the acquisition module in real time within a preset time interval, and uses the acquired first frame image as the standard image, and the definition of the standard image and the remaining other frame images is calculated, so as to precisely screen the facial image meeting the criteria for performing the micro-expression analysis, wherein on one hand, when the coincidence degree of the actual image is too large, it indicates that the expression of the monitored person is not changed or the expression is slightly changed, and then the analysis module eliminates the facial image of the frame, when the coincidence degree is too small, it indicates that the expression of the monitored person is obviously changed and the real mental health condition of the monitored person cannot be displayed, so that the analysis module does not use the frame image to perform the micro-expression analysis, and the analysis module compares the coincidence degree of the actual image in real time, and can precisely eliminate the facial image not meeting the criteria for performing the micro-expression analysis, and then the analysis module can be accurate carries out the micro expression analysis to being monitored personnel, the rate of accuracy of micro expression analysis has been improved, and then accurate grasp is by the mental health condition of monitoring personnel, another aspect thereof, the analysis module is when carrying out the micro expression analysis, the analysis module compares actual facial image with predetermineeing the risk image model, and then when improving the rate of accuracy of micro expression analysis, can accurately send early warning information, the staff can be according to different early warning signals, carry out psychological tutor to being monitored personnel, thereby effectual assurance is at the psychological health of escort personnel.
Specifically, when the analysis module determines a face image meeting an analysis standard, the analysis module obtains a first frame of face image shot by the acquisition module, and the first frame of face image is marked as a standard face image, the analysis module compares the first frame of face image with the rest of other frame images in sequence to calculate an actual image coincidence degree a, the analysis module compares the actual image coincidence degree with a preset image coincidence degree a0, and judges whether the rest of other frame face images meet the analysis standard according to a comparison result;
the preset image overlap ratio A0 comprises a first preset image overlap ratio A1 and a second preset image overlap ratio A2, wherein A1 < A2;
when A is larger than A2, the analysis module judges that the frame of face image does not meet the analysis standard and does not use the frame of face image to carry out micro-expression analysis;
when A is not less than A1 and not more than A2, the analysis module judges that the frame of face image meets the analysis standard and uses the frame of face image to perform micro-expression analysis;
when A is less than A2, the analysis module judges that the frame of face image does not meet the analysis standard and does not use the frame of face image to carry out micro expression analysis.
Specifically, the embodiment of the invention sets the preset image overlap ratio as two standards, determines the face image meeting the analysis standard in the analysis module, and the analysis module can compare the image overlap ratio in real time and screen the face image information accurately according to the comparison result, so that the face image meeting the micro-expression analysis and marking can be selected accurately.
Specifically, when the analysis module judges that the face image meets the analysis standard, the analysis calculates an actual image overlap ratio difference value delta A, the actual image overlap ratio difference value delta A is compared with a preset image overlap ratio difference value delta A0, a micro-expression analysis position is judged according to a comparison result, and delta A = | A- (A1 + A2)/2 | is set;
the preset image overlap ratio difference Δ a0 comprises a first preset image overlap ratio difference Δ a1 and a second preset image overlap ratio difference Δ a2, wherein Δ a1 < Δa 2;
when delta A > -delta A2, setting the position of the facial image for micro expression analysis as a lip by the analysis module;
when the delta A1 is less than the delta A and less than the delta A2, the analysis module sets the position of the facial image for micro expression analysis as the eyes;
when the delta A is less than or equal to the delta A1, the analysis module sets the position of the facial image for micro expression analysis as the eyebrow.
Specifically, after the analysis module of the embodiment of the present invention screens the images meeting the micro-expression analysis standard, the analysis module calculates the difference of the coincidence degree of the actual images in real time, and compares the difference of the coincidence degree of the actual images with a preset value, when the difference of the coincidence degree of the actual images is larger, the analysis module determines that the change of the micro-expression of the monitored person is larger, and further determines that the position of the change of the micro-expression is in the lips, so as to perform the key analysis on the lips, and further accurately grasp the mental health condition of the monitored person through the change of the micro-expression of the lips, when the difference of the coincidence degree of the actual images is moderate, the analysis module determines that the change of the micro-expression of the monitored person is moderate, and further determines that the position of the change of the micro-expression is in the eyes, so as to perform the key analysis on the experienced position, and further accurately grasp the mental health condition of the monitored person through the change of the micro-expression of the eyes, when the difference value of the coincidence degree of actual image is less, the analysis module judges that the change of micro expression of the monitored person is less, and then the position of judging that the change of micro expression occurs is at the eyebrow, thereby carrying out key analysis on the eyebrow part, and then the change of micro expression through the eyebrow, the mental health condition of the monitored person is mastered accurately, thereby the analysis module can pass through multiple angles and multiple dimensions, the analysis of micro expression is carried out from different parts in the face image, the accuracy of the analysis of micro expression is further improved, and the mental health of the monitored person is further effectively ensured.
Specifically, when the analysis is completed at the position where the micro expression analysis is performed, the analysis module compares each frame of facial image with a preset risk image model in sequence to calculate the actual similarity B of the facial image, and when the actual similarity of the facial image is greater than 93%, the analysis module judges that the frame of facial image is a risk image.
Specifically, the analysis module in the embodiment of the invention is further provided with the standard image similarity, when the analysis module completes the micro expression analysis of the human face image, the human face image is compared with the preset risk image model, when the actual image similarity reaches the preset image similarity standard, the analysis module marks the frame image as a risk image, and the analysis module accurately judges the risk image, so that the accuracy of micro expression analysis is further improved, and the mental health of the monitored personnel is further effectively ensured.
Specifically, when the analysis module judges the face image as a risk image, the analysis module calculates an actual similarity difference value delta B, compares the actual image similarity difference value with a preset similarity difference value, determines the risk grade of the risk image according to the comparison result, and sets delta B = B-93%;
the preset similarity difference comprises a first similarity difference delta B1, a second similarity difference delta B2 and a third similarity difference delta B3, wherein delta B1 < [ delta ] B2 < [ delta ] B3;
when the delta B is less than the delta B1, the analysis module judges the risk level of the frame of the human face image as a primary risk image;
when the delta B is not less than delta B1 and less than delta B2, the analysis module judges the risk grade of the frame of the human face image as a secondary risk image;
when the delta B2 is more than or equal to the delta B and less than the delta B3, the analysis module judges the risk grade of the frame of the human face image as a three-level risk image;
when the delta B is not less than the delta B3, the analysis module judges the risk level of the frame of the face image as a four-level risk image;
wherein the order of risk levels is: the four-level risk image > the three-level risk image > the two-level risk image > the one-level risk image.
Specifically, the analysis module in the embodiment of the present invention is further provided with a plurality of standard similarity difference values, and when the analysis module performs risk level evaluation on a risk image, the analysis module calculates an actual similarity difference value in real time, and when the actual similarity difference value is larger, the analysis module determines that the frame of risk image has a larger similarity to a preset shared image, thereby indicating that the probability of the monitored person suffering from a mental health problem is larger, so that the larger the level of the frame of risk image is, the analysis module can perform analysis according to the risk level of each frame of image, thereby accurately grasping the mental health condition of the monitored person, further improving the accuracy of micro-expression analysis, and further effectively ensuring the mental health of the monitored person.
Specifically, when the analysis module finishes the judgment of the risk level of the risk image, the analysis module calculates an actual risk score Q of the face image information, compares the actual risk score Q with a preset risk score Q0, and judges the face image information level according to the comparison result;
the preset risk scores Q0 include a first preset risk score Q1, a second preset risk score Q2, and a third preset risk score Q3, wherein Q1 < Q2 < Q3;
when Q is less than Q1, the analysis module judges the risk of the face image information as primary risk image information, the analysis module judges that a preset time interval T0 needs to be adjusted and does not need to send an early warning instruction to the early warning module, the analysis module marks the adjusted preset time interval as T1 and sets T1= T0 x (1- (Q1-Q)/Q1);
when Q1 is not less than Q < Q2, the analysis module judges the risk of the face image information as secondary risk image information and sends a secondary early warning instruction to the early warning module;
when Q2 is not less than Q < Q3, the analysis module judges the risk of the face image information as three-level risk image information and sends a three-level early warning instruction to the early warning module;
when Q is less than Q3, the analysis module judges the risk of the face image information as four-level risk image information and sends a four-level early warning instruction to the early warning module;
wherein, the early warning instruction rank order is: the fourth-level early warning instruction is larger than the third-level early warning instruction, the second-level early warning instruction is larger than the first-level early warning instruction.
Specifically, the analysis module of the embodiment of the invention can also calculate the actual risk score of the face information according to the risk level of each frame of risk image, can more accurately reflect the psychological condition of the monitored person by calculating the actual risk score in real time, when the analysis module judges that the face image information is the primary risk image information, the analysis module judges that the probability of the psychological monitoring problem of the monitored person is smaller, and can secondarily analyze the psychological health of the monitored person by correcting the shooting time interval, thereby more accurately mastering the information health condition of the monitored person, when the analysis module judges that the risk level of the face image information is two-level or more, the analysis module judges that the probability of the psychological monitoring problem of the monitored person is larger, and further more accurately mastering the psychological health condition of the escort person by sending the early warning instruction, and psychological counseling can be carried out on the monitored personnel according to the grade of the early warning signal, so that the psychological health of the escort personnel is effectively ensured.
Specifically, when the analysis module calculates an actual risk score Q of the face image information, the analysis module acquires a time interval Δ T,. DELTA.T = Ta-Tb of the two adjacent frames of risk images, where Ta is an adjacent first frame of risk image capturing time and Tb is an adjacent second frame of risk image capturing time, the analysis module corrects the actual risk score Q based on Δ T, and the analysis module records the corrected actual risk score as Qa, and sets Qa = Q x (1 + (. DELTA.T 1 +. DELTA.T 2+ … … +. DELTA.Tn)/0.4 x T0), where Δ T1 represents the time interval of the first two adjacent frames of risk images and Δ Tn represents the time interval of the nth two adjacent frames of risk images.
Specifically, when the analysis module calculates the actual risk score, the analysis module obtains the time interval between two actual adjacent frames of risk images and corrects the actual risk score, when the time interval between two adjacent frames of risk images is smaller, the analysis module judges that the probability of the monitored person suffering from the mental health problem is greater, when the time interval between two adjacent frames of risk images is larger, the analysis module judges that the probability of the monitored person suffering from the mental health problem is smaller, the risk score is corrected accurately through the analysis module, the mental health condition of the escorting person can be mastered more accurately, and the monitored person can be psychologically guided according to the early warning signal level, so that the mental health of the escorting person is effectively guaranteed.
Specifically, the actual risk score Q is calculated using the following formula,
Q=（F1+F2）ⅹK+F3ⅹ2+F4ⅹ3；
wherein F1 is the number of first-level risk images, F2 is the number of second-level risk images, F3 is the number of third-level risk images, F4 is the number of fourth-level risk images, and K is the risk score calculation weight.
Specifically, the risk score calculation weight K is calculated using the following formula,
K=1.15ⅹ（F2/F1）；
wherein the value range of K is 0.98-1.33.
Specifically, when the analysis module calculates the actual risk score Q, the analysis module calculates the number of the primary risk images and the number of the secondary risk images, when the primary risk images account for a larger proportion, the analysis module determines that the probability of the monitored person suffering from the mental health problem is smaller, the analysis module determines that the actual risk score is lower, and selects a smaller calculation weight to accurately calculate the actual risk score, when the number of the secondary risk images is larger, the analysis module determines that the probability of the monitored person suffering from the mental health problem is larger, the analysis module determines that the actual risk score is higher, and selects a larger calculation weight, and the risk score calculation weight is accurately selected by the analysis module, so that the risk level of the image information of the person can be accurately mastered, and the mental health condition of the escort person can be accurately mastered, and psychological counseling can be carried out on the monitored personnel according to the grade of the early warning signal, so that the psychological health of the escort personnel is effectively ensured.
Specifically, when the analysis module determines a face image meeting an analysis standard and calculates the coincidence degree A of an actual image, the analysis module acquires a second frame of face image shot by the acquisition module, the second frame of face image is marked as a standard face image, and the analysis module compares the second frame of face image with the rest of other frame images in sequence;
the analysis module acquires the j frame of face image shot by the acquisition module, the j frame of face image is marked as a standard face image, and the analysis module compares the j frame of face image with the rest of other frame images in sequence.
Specifically, when the analysis module screens the face images meeting the micro-expression analysis standard, different frames are selected as standard images, the face images meeting the micro-expression analysis standard can be screened more comprehensively, missing and wrong selection can be effectively prevented, the face images not meeting the micro-expression analysis standard can be accurately removed, the analysis module can accurately perform micro-expression analysis on the monitored person, the accuracy of micro-expression analysis is improved, and the mental health condition of the monitored person can be accurately mastered.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention; various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An intelligent monitoring system, comprising:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module comprises at least one acquisition unit and is used for shooting face image information in a preset time interval T0, and the face image information comprises a plurality of frames of face images with a preset time sequence acquired according to a preset central standard;
the recognition module is connected with the acquisition module and comprises at least one recognition unit used for recognizing the micro-expression characteristics of the face image according to a preset recognition center;
the analysis module is connected with the recognition module and comprises at least one analysis unit, the analysis unit is used for comparing the micro expressions with a preset risk image model according to the micro expressions in a plurality of frames of human face images which accord with analysis standards so as to determine the actual risk level of the monitored person, the analysis module determines to send early warning information of corresponding level according to the actual risk level, and when the early warning information of corresponding level is determined to be sent, the analysis module sends an early warning instruction of corresponding level to the early warning module; the analysis module calculates the actual image contact ratio of each frame of facial image in sequence, compares the actual contact ratio with a preset value to select the facial image which accords with the micro-expression analysis standard, calculates the actual risk score of the monitored person according to the facial image which accords with the micro-expression analysis standard, sends an early warning instruction of a corresponding grade according to the actual risk score, and corrects the actual risk grade according to the time interval of the adjacent risk image when sending the early warning instruction;
the early warning module is connected with the analysis module and used for receiving the early warning instruction sent by the analysis module and sending a corresponding early warning signal according to the early warning instruction;
and the storage module is respectively connected with the acquisition module, the identification module, the analysis module and the early warning module and is used for storing data generated during the operation of the system.
2. The intelligent monitoring system according to claim 1, wherein when the analysis module determines that the face image meets the criteria for performing the micro-expression analysis, the analysis module obtains a first frame of face image captured by the obtaining module and marks the first frame of face image as a standard face image, the analysis module compares the first frame of face image with the remaining other frame images in sequence to calculate an actual image overlap ratio a, the analysis module compares the actual image overlap ratio with a preset image overlap ratio a0, and determines whether the remaining other frame face images meet the analysis criteria according to the comparison result;
the preset image overlap ratio A0 comprises a first preset image overlap ratio A1 and a second preset image overlap ratio A2, wherein A1 < A2;
when A is larger than A2, the analysis module judges that the frame of face image does not meet the analysis standard and does not use the frame of face image to carry out micro-expression analysis;
when A is not less than A1 and not more than A2, the analysis module judges that the frame of face image meets the analysis standard and uses the frame of face image to perform micro-expression analysis;
when A is less than A2, the analysis module judges that the frame of face image does not meet the analysis standard and does not use the frame of face image to carry out micro expression analysis.
3. The intelligent monitoring system according to claim 2, wherein when the analysis module determines that the face image meets the analysis standard, the analysis module analyzes and calculates an actual image overlap ratio difference Δ a, compares the actual image overlap ratio difference Δ a with a preset image overlap ratio difference Δ a0, and determines a micro-expression analysis position according to a comparison result, and sets Δ a = | a- (a 1+ a 2)/2 |;
the preset image overlap ratio difference Δ a0 comprises a first preset image overlap ratio difference Δ a1 and a second preset image overlap ratio difference Δ a2, wherein Δ a1 < Δa 2;
when delta A > -delta A2, setting the position of the facial image for micro expression analysis as a lip by the analysis module;
when the delta A1 is less than the delta A and less than the delta A2, the analysis module sets the position of the facial image for micro expression analysis as the eyes;
when the delta A is less than or equal to the delta A1, the analysis module sets the position of the facial image for micro expression analysis as the eyebrow.
4. The intelligent monitoring system according to claim 3, wherein when the analysis is completed at the position where the micro expression analysis is performed, the analysis module compares each frame of the face image with a preset risk image model in sequence to calculate the actual similarity B of the face image, and when the actual similarity of the face image is greater than 93%, the analysis module determines that the frame of the face image is a risk image.
5. The intelligent monitoring system according to claim 4, wherein when the analysis module determines the face image as a risk image, the analysis module calculates an actual similarity difference Δ B, compares the actual image similarity difference with a preset similarity difference, determines a risk level of the risk image according to a comparison result, and sets Δ B = B-93%;
the preset similarity difference comprises a first similarity difference delta B1, a second similarity difference delta B2 and a third similarity difference delta B3, wherein delta B1 < [ delta ] B2 < [ delta ] B3;
when the delta B is less than the delta B1, the analysis module judges the risk level of the frame of the human face image as a primary risk image;
when the delta B is not less than delta B1 and less than delta B2, the analysis module judges the risk grade of the frame of the human face image as a secondary risk image;
when the delta B2 is more than or equal to the delta B and less than the delta B3, the analysis module judges the risk grade of the frame of the human face image as a three-level risk image;
when the delta B is not less than the delta B3, the analysis module judges the risk level of the frame of the face image as a four-level risk image;
wherein the order of risk levels is: the four-level risk image > the three-level risk image > the two-level risk image > the one-level risk image.
6. The intelligent monitoring system according to claim 5, wherein when the analysis module completes the determination of the risk level of the risk image, the analysis module calculates an actual risk score Q of the face image information, compares the actual risk score Q with a preset risk score Q0, and determines the face image information level according to the comparison result;
the preset risk scores Q0 include a first preset risk score Q1, a second preset risk score Q2, and a third preset risk score Q3, wherein Q1 < Q2 < Q3;
when Q is less than Q1, the analysis module judges the risk of the face image information as primary risk image information, the analysis module judges that a preset time interval T0 needs to be adjusted and does not need to send an early warning instruction to the early warning module, the analysis module marks the adjusted preset time interval as T1 and sets T1= T0 x (1- (Q1-Q)/Q1);
when Q1 is not less than Q < Q2, the analysis module judges the risk of the face image information as secondary risk image information and sends a secondary early warning instruction to the early warning module;
when Q2 is not less than Q < Q3, the analysis module judges the risk of the face image information as three-level risk image information and sends a three-level early warning instruction to the early warning module;
when Q is less than Q3, the analysis module judges the risk of the face image information as four-level risk image information and sends a four-level early warning instruction to the early warning module;
wherein, the early warning instruction rank order is: the fourth-level early warning instruction is larger than the third-level early warning instruction, the second-level early warning instruction is larger than the first-level early warning instruction.
7. The intelligent monitoring system according to claim 6, wherein when the analysis module calculates the actual risk score Q of the face image information, the analysis module obtains a time interval DeltaT of two adjacent frames of risk images, DeltaT = Ta-Tb, wherein Ta is an adjacent first frame of risk image capturing time and Tb is an adjacent second frame of risk image capturing time, the analysis module revises the actual risk score Q according to DeltaT, the analysis module revises the revised actual risk score as Qa and sets Qa = Qx (1 + (DeltaT 1 +. DELTA T2+ … … +. DELTA Tn)/0.4 xT 0), wherein DeltaT 1 represents the time interval of the first two adjacent frames of risk images and DeltaTn represents the time interval of the nth two adjacent frames of risk images.
8. The intelligent monitoring system according to claim 7, wherein the actual risk score Q is calculated using the following formula,
Q=（F1+F2）ⅹK+F3ⅹ2+F4ⅹ3；
wherein F1 is the number of first-level risk images, F2 is the number of second-level risk images, F3 is the number of third-level risk images, F4 is the number of fourth-level risk images, and K is the risk score calculation weight.
9. The intelligent monitoring system according to claim 8, wherein the risk score calculation weight K is calculated using the following formula,
K=1.15ⅹ（F2/F1）；
wherein the value range of K is 0.98-1.33.
10. The intelligent monitoring system according to claim 2, wherein when the analysis module determines a face image meeting an analysis standard and calculates an actual image contact ratio a, the analysis module obtains a second frame of face image captured by the acquisition module, and the second frame of face image is marked as a standard face image, and the analysis module compares the second frame of face image with the remaining other frame images in sequence;
the analysis module acquires the j frame of face image shot by the acquisition module, the j frame of face image is marked as a standard face image, and the analysis module compares the j frame of face image with the rest of other frame images in sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210003345.7A CN114022944A (en) | 2022-01-05 | 2022-01-05 | Intelligent monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210003345.7A CN114022944A (en) | 2022-01-05 | 2022-01-05 | Intelligent monitoring system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114022944A true CN114022944A (en) | 2022-02-08 |
Family
ID=80069657
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210003345.7A Pending CN114022944A (en) | 2022-01-05 | 2022-01-05 | Intelligent monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114022944A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114154172A (en) * | 2022-02-10 | 2022-03-08 | 北京国信网联科技有限公司 | Network security monitoring system based on situation awareness |
| CN114612828A (en) * | 2022-03-10 | 2022-06-10 | 中化学建设投资集团有限公司 | Construction site fire monitoring and early warning method based on image analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109584050A (en) * | 2018-12-14 | 2019-04-05 | 深圳壹账通智能科技有限公司 | Consumer's risk degree analyzing method and device based on micro- Expression Recognition |
| CN109766859A (en) * | 2019-01-17 | 2019-05-17 | 平安科技（深圳）有限公司 | Campus monitoring method, device, equipment and storage medium based on micro- expression |
| CN109766766A (en) * | 2018-12-18 | 2019-05-17 | 深圳壹账通智能科技有限公司 | Employee work condition monitoring method, device, computer equipment and storage medium |
| CN112820072A (en) * | 2020-12-28 | 2021-05-18 | 深圳壹账通智能科技有限公司 | Dangerous driving early warning method and device, computer equipment and storage medium |
| CN113051573A (en) * | 2021-02-19 | 2021-06-29 | 广州银汉科技有限公司 | Host safety real-time monitoring alarm system based on big data |
-
2022
- 2022-01-05 CN CN202210003345.7A patent/CN114022944A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109584050A (en) * | 2018-12-14 | 2019-04-05 | 深圳壹账通智能科技有限公司 | Consumer's risk degree analyzing method and device based on micro- Expression Recognition |
| CN109766766A (en) * | 2018-12-18 | 2019-05-17 | 深圳壹账通智能科技有限公司 | Employee work condition monitoring method, device, computer equipment and storage medium |
| CN109766859A (en) * | 2019-01-17 | 2019-05-17 | 平安科技（深圳）有限公司 | Campus monitoring method, device, equipment and storage medium based on micro- expression |
| CN112820072A (en) * | 2020-12-28 | 2021-05-18 | 深圳壹账通智能科技有限公司 | Dangerous driving early warning method and device, computer equipment and storage medium |
| CN113051573A (en) * | 2021-02-19 | 2021-06-29 | 广州银汉科技有限公司 | Host safety real-time monitoring alarm system based on big data |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114154172A (en) * | 2022-02-10 | 2022-03-08 | 北京国信网联科技有限公司 | Network security monitoring system based on situation awareness |
| CN114612828A (en) * | 2022-03-10 | 2022-06-10 | 中化学建设投资集团有限公司 | Construction site fire monitoring and early warning method based on image analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| RU2292839C2 (en) | Method and device for analyzing human behavior | |
| CN114022944A (en) | Intelligent monitoring system | |
| CN109215806A (en) | A kind of public place health monitoring systems and method based on recognition of face | |
| CN106169071B (en) | A kind of Work attendance method and system based on dynamic human face and chest card recognition | |
| CN209543514U (en) | Monitoring and alarm system based on recognition of face | |
| CN105791299A (en) | Unattended monitoring type intelligent on-line examination system | |
| CN111144262A (en) | Process anomaly detection method based on monitoring video | |
| CN202130312U (en) | Driver fatigue driving monitoring device | |
| CN112957042B (en) | Non-contact target emotion recognition method and system | |
| CN113080855B (en) | Facial pain expression recognition method and system based on depth information | |
| CN116563797B (en) | Monitoring management system for intelligent campus | |
| CN115546899A (en) | Examination room abnormal behavior analysis method, system and terminal based on deep learning | |
| CN112102530B (en) | Campus Internet of things intelligent cloud lock management system | |
| CN117292836A (en) | Body measurement score monitoring method and system | |
| CN112528767A (en) | Machine vision-based construction machinery operator fatigue operation detection system and method | |
| CN104063689B (en) | Face image identification method based on binocular stereoscopic vision | |
| CN117218680A (en) | Scenic spot abnormity monitoring data confirmation method and system | |
| CN113035329A (en) | Medical image quality control system | |
| CN114783097A (en) | Hospital epidemic prevention management system and method | |
| CN114388143A (en) | Method and device for acquiring facial data of Alzheimer's disease based on game interaction | |
| CN109145684A (en) | Head state monitoring method based on region most matching characteristic point | |
| CN113284320A (en) | Method and system for predicting fatigue state of train driver in advance | |
| CN113901866A (en) | Fatigue driving early warning method based on machine vision | |
| CN113327404A (en) | Post fatigue state monitoring and warning system for air traffic controller | |
| CN112597842A (en) | Movement detection facial paralysis degree evaluation system based on artificial intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220208 |