Emotion calculation method based on in-vivo multichannel electrophysiological data

Abstract

The invention relates to the technical field of emotion recognition, in particular to an emotion calculation method based on in-vivo multichannel electrophysiological data, which comprises the following steps: acquiring distribution characteristics and association intensity of multichannel electrophysiological signals, extracting coupling intensity and fluctuation trend, analyzing energy change and interference intensity to determine a regulation boundary, and correcting a mapping instruction. And evaluating the mapping stability, and outputting an emotion mapping path stability state identifier. According to the invention, through coupling and correlation modeling of multi-channel signals, recognition of an emotional fluctuation period and correlation strength between the signals is enhanced, comprehensive capture of the signals in a complex emotional state is ensured, sensitivity of signal interference is screened, and stability and accuracy of emotional state prediction are improved. By analyzing the signal energy change trend and the external environment temperature characteristics, dynamic regulation and control of the emotional state are achieved, continuous monitoring and adjustment of the emotional state within a long time are ensured, and the reliability and stability of emotional mapping are improved.

Description

Technical Field

[0001] This invention relates to the field of emotion recognition technology, and in particular to an emotion calculation method based on in vivo multi-channel electrophysiological data. Background Technology

[0002] Emotion recognition technology is an interdisciplinary field combining artificial intelligence and biosignal analysis. It primarily studies how to collect, process, and analyze multimodal physiological signals generated by the human body under different emotional states to determine an individual's emotion category or trend. The core of this technology involves signal acquisition, feature extraction, and emotion modeling. Signal acquisition typically includes multi-channel physiological data such as electroencephalogram (EEG), electrocardiogram (ECG), electrodermal signal transduction (GSR), electromyography (EMG), and heart rate variability (HRV). Feature extraction focuses on mining emotion-related pattern features from the time and frequency domains. Emotion modeling relies on statistical models, machine learning algorithms, and neural network structures to establish a mapping relationship between emotions and physiological characteristics. Applications in this field cover multiple areas, including mental health assessment, human-computer interaction, adaptive entertainment, driving safety, and intelligent monitoring.

[0003] Traditional emotion computing methods based on in vivo multi-channel electrophysiological data refer to inferring an individual's emotional state through single or limited physiological signals, such as electroencephalography (EEG) or electrodermatology (EDS). Traditional emotion computing typically employs independent acquisition devices to sample signals, followed by feature extraction, feature selection, and classifier training to complete emotion recognition. A typical workflow includes: first, acquiring single-channel physiological signals from an individual using sensors under experimental conditions; second, preprocessing the acquired signals, such as filtering, denoising, and normalization; then, extracting effective features based on set feature indicators (such as power spectral density, signal amplitude change rate, and frequency band energy ratio); and finally, using statistical classification algorithms (such as support vector machines, k-nearest neighbors, or linear discriminant analysis) to distinguish different emotion categories. Because these methods rely on limited signal dimensions, significant individual differences, and insufficient feature coupling, they struggle to comprehensively reflect the complex physiological characteristics of emotional responses in vivo. Therefore, emotion computing methods based on in vivo multi-channel electrophysiological data have gradually developed to achieve more comprehensive emotion state recognition.

[0004] Current technologies in emotion recognition face several limitations. First, relying on single or limited physiological signals for emotion inference fails to fully reflect the complex physiological responses to emotions, thus limiting the accuracy of emotion classification. Second, signal acquisition typically relies on independent devices, making it difficult to effectively handle complex coupling relationships between signals and lacking accurate identification of signal interference and dynamic changes, resulting in unstable emotion predictions in practical applications. Furthermore, traditional emotion calculation methods fail to consider the impact of external environmental changes on emotional states, making it impossible to track and regulate emotional fluctuations over long periods, thus failing to guarantee the continuous stability of emotional states and the accuracy of feedback. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method for emotion calculation based on in vivo multi-channel electrophysiological data.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for emotion calculation based on in vivo multi-channel electrophysiological data, comprising the following steps:

[0007] S1: Obtain the distribution characteristics, correlation strength between signals, and emotional fluctuation cycle information of multi-channel electrophysiological signals; perform correlation modeling and logical verification on the three types of data; evaluate the overall signal quality; identify the dynamic correlation patterns between signals; and generate a multi-channel emotional correlation feature matrix.

[0008] S2: Based on the multi-channel emotion association feature matrix, extract the coupling strength between signals, the emotion fluctuation trend and the signal interference sensitivity, screen the key signal combination that meets the emotion calculation conditions, and determine the optimal mapping path in combination with the target emotion category to generate the optimal mapping interval model of emotion features.

[0009] S3: Based on the optimal mapping interval model of the emotional features, the energy change trend in the coupling process between signals is analyzed, the mapping balance is judged by combining the signal interference intensity, and the mapping stability is evaluated by combining the external environmental temperature characteristics. Key mapping node boundaries are screened, and dynamic control boundaries for multi-source signal mapping are generated.

[0010] S4: Call the multi-source signal mapping dynamic control boundary, compare the mapping command with the current signal state, if there is a deviation, adjust the mapping direction and weight, and combine the energy sensor feedback to correct the mapping command, so as to obtain the intelligent emotion mapping control execution cluster.

[0011] As a further aspect of the present invention, the multi-channel emotion association feature matrix includes signal distribution characteristic level, signal correlation strength index, and emotion fluctuation cycle category; the optimal mapping interval model of emotion features includes signal coupling priority, target emotion matching degree, and shortest mapping sequence; the dynamic control boundary of multi-source signal mapping includes mapping stability boundary value, energy balance tolerance interval, and temperature matching threshold; and the intelligent emotion mapping control execution cluster includes mapping direction adjustment parameters, weight correction parameters, and energy response correction coefficient.

[0012] As a further aspect of the present invention, the steps for generating the multi-channel emotion association feature matrix are as follows:

[0013] S111: Obtain the distribution characteristics, correlation strength between signals, and emotional fluctuation cycle information of multi-channel electrophysiological signals. Perform data logic verification through time synchronization markers, extract the matching relationship between signal distribution characteristics and correlation strength of each time slice, and perform state normalization processing using the logical relationship of the three types of data to generate a state registration set.

[0014] S112: Based on the matching region of signal distribution characteristics and correlation strength in the state registration set, combined with the emotional fluctuation cycle information, extract the center offset of the emotional fluctuation trajectory and the adjacent state switching angle, identify the emotional fluctuation path, and generate an emotional fluctuation parameter set.

[0015] S113: Based on the state information of each fluctuation path in the emotional fluctuation parameter group, analyze the signal configuration mapping relationship in the original state, transform the path state into a unified state system, perform signal quality and temperature verification on the channel based on path continuity and state change, integrate dynamic fluctuation segments and feasible state ranges, and generate a multi-channel emotional correlation feature matrix.

[0016] As a further aspect of the present invention, the steps for generating the optimal mapping interval model for emotion features are specifically as follows:

[0017] S211: Based on the multi-channel emotion association feature matrix, by extracting data from signal sensors, association strength detectors and temperature sensors, extracting the coupling strength between signals, emotion fluctuation trends and signal interference sensitivity, identifying key signal combinations and detecting mapping paths affected by interference, eliminating path segments that do not meet the conditions for emotion calculation, and generating priority mapping trajectories;

[0018] S212: Using the prioritized mapped trajectory and combining it with the stability detection data from the temperature sensor, identify the mapping priority segment of each trajectory segment, extract the segment information, and apply the following formula:

[0019] ;

[0020] Calculate the mapping optimization value of the segment and select the segment with the best mapping priority;

[0021] in, The mapping optimization value of the representative segment, This represents the temperature sensor measurement value for the i-th trajectory segment. This represents the average temperature across all trajectory segments. Represents the maximum value of temperature. Represents the minimum temperature. The weight coefficient represents the weight of the i-th trajectory segment. The stability detection data represents the i-th trajectory segment. Represents the total number of trajectory segments;

[0022] S213: Call the optimal mapping priority segment, combine it with the target emotion category, calculate the angular deviation between the target mapping direction and the segment direction, select the optimal path segment based on the deviation, and generate the optimal mapping interval model for emotion features.

[0023] As a further aspect of the present invention, the step of generating the dynamic control boundary of the multi-source signal mapping specifically includes:

[0024] S311: Based on the optimal mapping interval model of the emotional features, extract the energy change trend in the coupling process between signals, identify the energy value and time interval between the current frame and the previous frame, analyze the energy change trend, determine the direction and intensity of energy change, and obtain the energy offset trend value.

[0025] S312: Call the energy offset trend value, and calculate the mapping path stability quantization value by combining the mapping weight adjustment amount, the stability index of temperature feature extraction and energy boundary distance. Based on the comparison result with the energy boundary threshold, judge the mapping stability, filter the mapping node boundary, and generate the multi-source signal mapping dynamic control boundary.

[0026] As a further aspect of the present invention, the generation step of the intelligent emotion mapping and regulation execution cluster specifically includes:

[0027] S411: Call the multi-source signal mapping dynamic control boundary, extract the mapping command and the signal state calculated by the fused signal sensor and the correlation strength detector, perform logical consistency and state difference judgment, and obtain the mapping control deviation state;

[0028] S412: Based on the mapping control deviation state, combined with the energy offset and correlation strength feedback detected by the signal sensor, the mapping weight amplitude is dynamically adjusted through multi-state data fusion, the mapping path correction instruction is identified, and fed back to the mapping control to superimpose the original control quantity and update in real time, thus obtaining the intelligent emotion mapping regulation execution cluster.

[0029] As a further aspect of the present invention, the mapping instruction refers to the received control signal, which is generated through logical analysis and state determination.

[0030] The logical consistency and state difference determination refer to whether signals or data are logically consistent, and the state difference determination refers to identifying the differences between signal states and determining whether adjustments are needed.

[0031] As a further aspect of the present invention, the method further includes step S5:

[0032] S5: Based on the intelligent emotion mapping control execution cluster, it collects the continuous periodic mapping trajectory output and the signal stability length change in the energy sensor, analyzes whether the change trend is stable, determines the mapping stability period after control execution, and outputs the emotion mapping path stability status indicator.

[0033] The stability status identifier of the emotion mapping path includes the duration of mapping stability, the residual length of signal energy, and the amplitude of trajectory fluctuation.

[0034] As a further aspect of the present invention, the step of generating the emotion mapping path stability state identifier specifically includes:

[0035] S511: Based on the intelligent emotion mapping regulation execution cluster, collect the mapping trajectory output by the energy unit and the signal stability region detected by the energy unit, extract the mapping lateral offset and stability length, analyze the change amplitude of the stability length within the period, and generate a periodic stability amplitude sequence.

[0036] S512: Call the periodic stability amplitude sequence, extract the stability amplitude difference sequence within the continuous period, mark the stability trend according to the polarity of the difference change, and combine it with the signal stability threshold to determine whether the change converges into a single trend, and obtain the signal stability fusion trend value.

[0037] S513: Based on the signal stability fusion trend value, identify the lateral difference between the energy sensor state solution and the mapped positioning coordinates within the stable period, and combine it with the ratio of the signal stability region length to determine whether the mapping maintains stable path tracking, and output the emotion mapping path stability status indicator.

[0038] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0039] This invention enhances the identification of emotional fluctuation cycles and the strength of correlations between signals through multi-channel signal coupling and correlation modeling, ensuring comprehensive capture of signals under complex emotional states. Sensitivity screening for signal interference allows for more precise selection of key signal combinations, improving the stability and accuracy of emotional state prediction. By analyzing signal energy change trends and external environmental temperature characteristics, the stability of the emotional mapping process can be more effectively assessed, thereby achieving dynamic regulation of emotional states. This ensures continuous monitoring and adjustment of emotional states over long periods, reducing uncertainty caused by external factor fluctuations and improving the reliability and stability of emotional mapping. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the main steps of the present invention;

[0041] Figure 2 This is a flowchart illustrating the process of obtaining the multi-channel emotion association feature matrix in this invention.

[0042] Figure 3 This is a flowchart illustrating the process of obtaining the optimal mapping interval model for emotion features in this invention.

[0043] Figure 4 This is a flowchart illustrating the acquisition of the dynamic control boundary of multi-source signal mapping in this invention.

[0044] Figure 5 This is a flowchart illustrating the acquisition process of the intelligent emotion mapping and regulation execution cluster in this invention.

[0045] Figure 6 This is a flowchart illustrating the process of obtaining the stability status identifier of the emotion mapping path in this invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0047] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0048] All user-related information involved in this invention (including but not limited to biometric information, identity verification information, behavioral data, device information, and other data that can be used for identity verification and personalized services) is collected and processed with the user's full knowledge and voluntary consent. The collection, storage, and use of all information strictly comply with applicable national and regional laws and regulations, and meet relevant data protection standards and policy requirements. The use of data is limited to purposes necessary for providing the technical services of this invention, and reasonable technical and management measures will be taken to ensure the security and confidentiality of users' personal information in terms of information protection and privacy.

[0049] Example 1

[0050] Please see Figure 1 This invention provides a technical solution: a method for emotion calculation based on in vivo multi-channel electrophysiological data, comprising the following steps:

[0051] S1: Obtain the distribution characteristics, correlation strength between signals, and emotional fluctuation cycle information of multi-channel electrophysiological signals; perform correlation modeling and logical verification on the three types of data; evaluate the overall signal quality; identify the dynamic correlation patterns between signals; and generate a multi-channel emotional correlation feature matrix.

[0052] S2: Based on the multi-channel emotion association feature matrix, extract the coupling strength between signals, the emotion fluctuation trend and the signal interference sensitivity, screen the key signal combination that meets the emotion calculation conditions, and determine the optimal mapping path in combination with the target emotion category to generate the optimal mapping interval model of emotion features.

[0053] S3: Based on the optimal mapping interval model of emotional characteristics, the energy change trend in the coupling process between signals is analyzed, the mapping balance is judged by combining the signal interference intensity, and the mapping stability is evaluated by combining the external environmental temperature characteristics. Key mapping node boundaries are screened, and dynamic control boundaries for multi-source signal mapping are generated.

[0054] S4: Call the multi-source signal mapping dynamic control boundary, compare the mapping command with the current signal state, if there is a deviation, adjust the mapping direction and weight, and combine the energy sensor feedback to correct the mapping command, and obtain the intelligent emotion mapping control execution cluster.

[0055] S5: Based on the intelligent emotion mapping control execution cluster, it collects the continuous periodic mapping trajectory output and the signal stability length change in the energy sensor, analyzes whether the change trend is stable, determines the mapping stability period after control execution, and outputs the emotion mapping path stability status indicator.

[0056] The multi-channel emotion association feature matrix includes signal distribution characteristic level, signal correlation strength index, and emotion fluctuation cycle category. The optimal mapping interval model for emotion features includes signal coupling priority, target emotion matching degree, and shortest mapping sequence. The dynamic control boundary for multi-source signal mapping includes mapping stability boundary value, energy balance tolerance range, and temperature matching threshold. The intelligent emotion mapping control execution cluster includes mapping direction adjustment parameters, weight correction parameters, and energy response correction coefficient. The emotion mapping path stability status identifier includes mapping stability duration period, signal energy residual length, and trajectory fluctuation amplitude.

[0057] Please see Figure 2 The specific steps for generating the multi-channel sentiment association feature matrix are as follows:

[0058] S111: Obtain the distribution characteristics, correlation strength between signals, and emotional fluctuation cycle information of multi-channel electrophysiological signals. Perform data logic verification through time synchronization markers, extract the matching relationship between signal distribution characteristics and correlation strength of each time slice, and perform state normalization processing using the logical relationship of the three types of data to generate a state registration set.

[0059] First, multi-channel electrophysiological signals were acquired and processed. Specifically, signals from the Fz and Pz channels of an EEG sensor and signals from a GSR sensor were acquired. The sampling frequency was set to 256 Hz, with each processing time slice lasting 5 seconds (1280 sampling points). Data logic verification was performed by comparing the synchronization timestamps at the beginning of each time slice to ensure precise alignment of data frames on the time axis. The timestamp deviation was controlled within one sampling period (approximately 3.9 milliseconds). Subsequently, the distribution characteristics of the signals were extracted for each time slice. After applying a Hamming window to the EEG signals, a Fast Fourier Transform was performed to calculate the average power spectral density in the Alpha band (8-13 Hz), obtaining a value of 12.5 μV² / Hz within one time slice. For the GSR signals, the arithmetic mean of the skin conductance level (SCL) over the 5 seconds was calculated, obtaining a value of 3.8 μS. Next, the inter-signal inter-channel (IOL) values ​​were calculated. The correlation strength is calculated by constructing two vectors from the Alpha power sequence and SCL sequence of 10 consecutive time slices. The correlation is then calculated using the Pearson correlation coefficient method, yielding a correlation strength value of 0.78. The emotional fluctuation cycle information is represented by the duration of the time slice, which is 5 seconds. Finally, state normalization is performed according to preset logical rules. This rule base maps the signal feature combination to discrete state identifiers. For example, rule 1: if the Alpha power is greater than 10μV² / Hz, the SCL is less than 5μS, and the absolute value of the correlation strength is greater than 0.7, then the state identifier is 1 (calm). Rule 2: if the Alpha power is less than 8μV² / Hz, the SCL is greater than 7μS, and the absolute value of the correlation strength is greater than 0.7, then the state identifier is 2 (excited). Based on the calculation result of the current time slice, it satisfies rule 1, and the state is normalized to 1. By performing the above operations on the continuous data stream, a state identifier sequence arranged in chronological order is finally generated, which is the state registration set.

[0060] S112: Based on the matching region of signal distribution characteristics and correlation strength in the state registration set, combined with the emotional fluctuation cycle information, extract the center offset of the emotional fluctuation trajectory and the adjacent state switching angle, identify the emotional fluctuation path, and generate an emotional fluctuation parameter set;

[0061] Based on the state registration set, which is a sequence in this embodiment, with each state corresponding to a 5-second emotional fluctuation cycle, the following steps are taken: First, the state sequence is analyzed to extract the center offset of the emotional fluctuation trajectory. This trajectory center is obtained by calculating the arithmetic mean of all state identifiers in the sequence, i.e., (1+1+2+1) divided by 4, resulting in 1.25. The center offset is the difference between the state identifier of each time slice and this center value. For state identifier 2 in the third time slice of the sequence, its center offset is 2 minus 1.25, i.e., 0.75. For the time slice with state identifier 1, its offset is 1 minus 1.25, i.e., -0.25. Second, the switching angle between adjacent states is extracted. This process requires mapping the states to a two-dimensional space defined by signal features, where the X-axis represents Alpha power and the Y-axis represents SCL. The second time slice's state 1 corresponds to coordinates (12.2, 3.9), and the third time slice's state 2, after data acquisition and calculation, corresponds to coordinates (7.5, 7.2). The state transition in coordinate space is represented by a vector pointing from point (12.2, 3.9) to point (7.5, 7.2). The X component of this vector is 7.5 minus 12.2, resulting in -4.7, and the Y component is 7.2 minus 3.9, resulting in 3.3. The transition angle is obtained by calculating the angle between this vector and the positive X-axis direction, specifically by calculating the arctangent of the quotient of 3.3 and -4.7, which yields approximately 145.0 degrees. Finally, all the calculated center offsets and state transition angles are combined in chronological order to form a numerical sequence, such as [-0.25, 145.0°, 0.75, ...]. This sequence is the emotional fluctuation parameter set.

[0062] S113: Based on the state information of each fluctuation path in the emotional fluctuation parameter group, analyze the signal configuration mapping relationship in the original state, transform the path state into a unified state system, verify the signal quality and temperature of the channel based on the path continuity and state change, integrate the dynamic fluctuation segment and feasible state range, and generate a multi-channel emotional correlation feature matrix.

[0063] Based on the fluctuation path state information recorded in the emotion fluctuation parameter group, the path segment from state 1 to state 2 was selected for analysis. The original signal configuration mapping relationship of this path segment is that the Alpha power changes from 12.2 μV² / Hz to 7.5 μV² / Hz, and the SCL changes from 3.9 μS to 7.2 μS. In order to transform this path state to a unified state system, the signal was normalized. The effective physiological range of Alpha power was set to μV² / Hz, and the range of SCL was set to μS. The normalized coordinates of state 1 were calculated as ((12.2-5) / (20-5), (3.9-2) / (15-2)), resulting in (0.48, 0.15). The normalized coordinates of state 2 were calculated as ((7.5-5) / (20-5), (7.2-2) / (15-2)), resulting in (0.17, 0.4). Next, based on the path continuity and state change, the following was analyzed. The channel performs signal quality and temperature verification. Temperature sensor data is introduced. Within 5 seconds of state switching, the temperature reading changes from 32.5 degrees Celsius to 32.6 degrees Celsius, a change of 0.1 degrees Celsius. This value is less than the preset temperature change threshold of 0.5 degrees Celsius, so the temperature verification passes. This threshold is determined by analyzing sensor data under 1000 hours of stable environment and taking three times the standard deviation of its temperature fluctuation. Signal quality verification is performed by judging the single change amplitude of each component of normalized coordinates. The change of X component is |0.17-0.48|=0.31, and the change of Y component is |0.4-0.15|=0.25. Neither exceeds the amplitude limit of 0.5, so the quality verification passes. Finally, the dynamic fluctuation segment and feasible state range that passed the verification are integrated. The structured data including normalized coordinates, original signal values, correlation strength, temperature and verification results are integrated to generate a multi-channel emotion correlation feature matrix.

[0064] Table 1: Example of Multichannel Emotion Association Feature Matrix

[0065] Features / Time t1 t2 t3 t4 EEG channel 1 mean (μV) 4.47 8.91 15.32 16.05 EEG channel 1 standard deviation (μV) 15.01 18.23 12.45 11.98 EEG1-GSR correlation 0.78 0.85 0.65 0.63 Channel quality marking 1 1 0 1 Temperature (°C) 34.10 34.15 34.30 34.32

[0066] As shown in Table 1, this table presents a simplified multi-channel sentiment association feature matrix, where rows represent different features, columns represent consecutive time slices or state points, and in the channel quality markers, "1" represents passing the verification and "0" represents failing.

[0067] Please see Figure 3 The specific steps for generating the optimal mapping interval model for emotion features are as follows:

[0068] S211: Based on the multi-channel emotion association feature matrix, by extracting data from signal sensors, association strength detectors and temperature sensors, it extracts the coupling strength between signals, emotion fluctuation trends and signal interference sensitivity, identifies key signal combinations and detects mapping paths affected by interference, eliminates path segments that do not meet the conditions for emotion calculation, and generates priority mapping trajectories.

[0069] Data extraction and filtering are performed based on a multi-channel emotion association feature matrix. First, signal sensor data (normalized alpha power and SCL columns), association strength detector data (association strength column), and temperature sensor data (temperature column) are extracted row by row from the matrix. Then, the coupling strength between signals, emotion fluctuation trends, and signal interference sensitivity are extracted based on the data. The coupling strength is directly taken from the association strength value in the matrix, such as a value of 0.85 in a certain row. The emotion fluctuation trend is identified by analyzing the time series of the state indicator column, which reveals a fluctuation pattern from calm to excitement and then back to calm. The identification of signal interference sensitivity focuses on the stability of temperature data, setting the temperature in a single time slice (5 seconds). The threshold for the rate of change within a time slice is 0.1℃ / s. If the temperature in a certain time slice jumps from 32.5℃ to 34.0℃, the rate of change is 0.3℃ / s. This value exceeds the threshold, and the corresponding mapping path segment is marked as affected by interference. Subsequently, a path elimination operation is performed to remove all path segments marked as affected by interference, as well as path segments that do not meet other preset conditions (such as correlation strength below 0.5), from the feature matrix. The condition setting here is based on a verification experiment, which shows that when the correlation strength is below 0.5, the synchronicity between signals is insufficient to support effective emotional state inference. After screening and elimination, the remaining set of path segment data that meets the calculation conditions constitutes the priority mapping trajectory.

[0070] S212: Invoke the priority mapping trajectory, combine it with the stability detection data from the temperature sensor, identify the mapping priority segment of each trajectory segment, extract the segment information, and use the formula:

[0071] ;

[0072] Calculate the mapping optimization value of the segment and select the segment with the best mapping priority;

[0073] in, The mapping optimization value of the representative segment, This represents the temperature sensor measurement value for the i-th trajectory segment. This represents the average temperature across all trajectory segments. Represents the maximum value of temperature. Represents the minimum temperature. The weight coefficient represents the weight of the i-th trajectory segment. The stability detection data represents the i-th trajectory segment. Represents the total number of trajectory segments;

[0074] The priority mapping trajectory is invoked. This trajectory consists of a series of continuous, high-quality trajectory segments. Assuming a priority mapping trajectory of 3 segments is obtained after filtering, and combined with the stability detection data from the temperature sensor (i.e., the temperature measurement value corresponding to each trajectory segment), the mapping priority zone of each trajectory segment is identified, the zone information is extracted, and a formula is used for calculation. This formula is... The parameters and operational logic in the formula are explained below: The mapping optimization value representing the segment is used to quantify the overall quality of the trajectory. It is the index of the trajectory segment. It is the total number of trajectory segments, in this example , This represents the summation of the calculation results for all trajectory segments to obtain the total optimized value. It is the temperature sensor measurement value of the i-th trajectory segment. , , These are the average, maximum, and minimum temperatures for all trajectory segments. This term normalizes the temperature, reflecting the degree to which the temperature of each trajectory segment deviates from the overall temperature distribution, and is multiplied by a weight. This indicates a weighted consideration of this deviation. is the weighting coefficient for the i-th trajectory segment, and its setting is related to the signal quality of that trajectory segment. This is the stability detection data for the i-th trajectory segment, representing the signal stability of that trajectory segment. Taking its absolute value ensures a positive contribution. The logic of the entire formula lies in calculating a comprehensive optimization score by comprehensively considering the temperature deviation, signal quality, and signal stability of each trajectory segment, thereby selecting the optimal mapping path. Now, let's demonstrate with a calculation example: Assume the preferred mapping trajectory contains 3 trajectory segments (N=3), and the collected data is as follows:

[0075] Trajectory segment 1: Temperature =34.20℃, signal-to-noise ratio (SNR1) = 28dB, signal stability =0.95;

[0076] Track segment 2: Temperature =34.50℃, signal-to-noise ratio (SNR²) = 25dB, signal stability =-0.88 (a negative value indicates a fluctuation in a specific direction);

[0077] Track segment 3: Temperature =34.25℃, signal-to-noise ratio (SNR)3=30dB, signal stability =0.92;

[0078] Calculate temperature-related parameters: =34.50℃, =34.20℃, =(34.20+34.50+34.25) / 3=34.317℃;

[0079] Set weighting coefficients The weighting coefficients are set with reference to the signal-to-noise ratio (SNR) of the corresponding trajectory segment. The higher the SNR, the more reliable the data quality, and the greater the weight should be. The specific setting method is normalization processing. The calculation yields: ,but =28 / 83≈0.337, =25 / 83≈0.301, =30 / 83≈0.361;

[0080] Substitute the parameters into the formula to calculate. : ;

[0081] The optimized mapping value of the trajectory is calculated. This value is compared with the V values ​​calculated from other candidate priority mapping trajectories, and the trajectory with the highest V value is selected as the optimal mapping priority segment. This result shows the comprehensive quality evaluation value of this trajectory segment. If the V value of another trajectory is 0.15, then the latter is better.

[0082] The advantage of the formula is that by introducing the normalized deviation of temperature, the model can distinguish between physiological temperature changes caused by emotional fluctuations and temperature changes caused by environmental noise. At the same time, by combining the signal-to-noise ratio as a weight, the role of high-quality signals in decision-making is strengthened, the reliability of the screening results is improved, and the optimal mapping priority segment is obtained.

[0083] S213: Call the optimal mapping priority segment, combine it with the target emotion category, calculate the angle deviation between the target mapping direction and the segment direction, select the optimal path segment based on the deviation, and generate the optimal mapping interval model for emotion features;

[0084] The optimal mapping priority segment is invoked, and path matching is performed in conjunction with the target emotion category. Assuming the current target emotion category is "high focus," this category is defined in a specific direction in the feature space through extensive physiological data calibration experiments. This direction corresponds to a vector originating from the origin and pointing to the normalized coordinate point (0.2, 0.8). The angle between this direction and the positive X-axis of the coordinate system is determined to be 76.0 degrees by calculating the arctangent of the quotient of 0.8 and 0.2. Next, the direction of the optimal mapping priority segment is calculated. This segment represents the trajectory from state 1 to state 2, and its direction angle has been determined to be 145.0 degrees in the previous calculation. Then, the angular deviation between the target mapping direction and the segment direction is calculated, i.e., the absolute value of the difference between 145.0 degrees and 76.0 degrees, resulting in 6. 9.0 degrees. Subsequently, the optimal path segment is selected based on this deviation value, and an angle deviation threshold of 45 degrees is set. The establishment of this threshold is based on an experiment involving hundreds of subjects. The experimental data shows that when the deviation is less than 45 degrees, the accuracy of emotional state recognition remains above 90%, while it drops significantly when it is greater than 45 degrees. Since the currently calculated deviation of 69.0 degrees exceeds the threshold of 45 degrees, this optimal mapping priority segment is not accepted as the optimal path segment. At this time, the trajectory segment with the next lower optimization value will be evaluated. If the angle deviation of a candidate trajectory segment is calculated to be 25.3 degrees, which is less than 45 degrees, then the candidate path segment is finally determined as the optimal path segment. Based on this, information such as its signal feature range and dynamic change pattern is integrated to generate an optimal mapping interval model for emotional features.

[0085] Please see Figure 4 The specific steps for generating the dynamic control boundary of multi-source signal mapping are as follows:

[0086] S311: Based on the optimal mapping interval model of emotion features, extract the energy change trend in the coupling process between signals, identify the energy value and time interval between the current frame and the previous frame, analyze the energy change trend, determine the direction and intensity of energy change, and obtain the energy offset trend value.

[0087] Based on the optimal mapping interval model for emotional features, the energy changes during signal coupling are analyzed. Here, "energy" is quantified as the integral value of the power spectral density of the EEG signal in the 1Hz to 30Hz frequency band, with the unit being square microvolts (μV²). First, the energy values ​​of two consecutive data frames in the model and the time interval between them are identified and extracted. The length of a data frame is set to 5 seconds, i.e., the time interval is 5 seconds. Through calculation, the integral energy value of the previous frame is 150.7 μV², and the integral energy value of the current frame is 162.3 μV². Secondly, the energy change trend is analyzed. By calculating the change in energy value between two consecutive frames, i.e., 162.3μV² minus 150.7μV², the energy increment is obtained as 11.6μV². To obtain the energy change rate, this energy increment is divided by the time interval of 5 seconds, resulting in 2.32μV² / s. Finally, the direction and intensity of the energy change are determined. Since the change rate is positive, it indicates that the energy is increasing, and its intensity value is 2.32. This value + 2.32 is determined as the energy offset trend value, which is used for subsequent stability assessment calculations.

[0088] S312: Call the energy offset trend value, and combine it with the mapping weight adjustment, the stability index of temperature feature extraction, and the energy boundary distance, using the formula:

[0089] ;

[0090] Calculate the quantization value of the mapping path stability, and judge the mapping stability based on the comparison result with the energy boundary threshold, filter the mapping node boundary, and generate the dynamic control boundary of multi-source signal mapping;

[0091] in, This represents the quantization value of the mapping path stability. Represents the number of the mapped node. The total number of mapped nodes. This represents the weight adjustment amount for the k-th mapping node. The temperature feature extraction stability index represents the value of the k-th mapping node. The energy boundary distance of the k-th mapping node;

[0092] The energy shift trend value, i.e., "moderate intensity enhancement," is invoked. This is combined with the mapping weight adjustment, the stability index extracted from temperature features, and the energy boundary distance, using the formula... The parameters and calculation logic in the formula for calculating the stability quantization value of the mapping path are explained below: This is a mapping path stability quantification value, used to assess whether the current mapping path is stable. It is the number of the mapping node. A mapping path consists of multiple nodes. It is the total number of nodes. This indicates that the calculation results for all nodes along the path are summed. This is the weight adjustment amount for the k-th node, set based on the energy offset trend value. It is the stability index for temperature feature extraction at the k-th node, reflecting the temperature stability at that point. It is the energy boundary distance of the k-th node, representing the distance between the current energy value and the preset safety boundary. This represents the contribution to temperature stability after weighting, minus This contribution is used to balance the risk of energy deviation from the boundary. The absolute value and square root operations are used to obtain a non-directional, dimensionless stability metric. Now, let's demonstrate with an example: assume that the current mapping path consists of 3 nodes (M=3).

[0093] Get the parameters of each node:

[0094] Node 1 (k=1): Temperature stability index =0.98, energy boundary distance =12.5 (units can be customized);

[0095] Node 2 (k=2): Temperature stability index =0.95, energy boundary distance =15.0;

[0096] Node 3 (k=3): Temperature stability index =0.99, energy boundary distance =10.0;

[0097] Set weight adjustment amount This weight is set based on the energy offset trend value of "moderate intensity enhancement". The setting rule is: the more intense the trend, the higher the weight should be to strengthen control. "Weak" corresponds to w=0.8, "moderate" corresponds to w=1.0, and "strong" corresponds to w=1.2. Therefore, the weights of all nodes on the current path are... Set all to 1.0;

[0098] Substitute the parameters into the formula to calculate. :

[0099] ;

[0100] The quantization value of the mapping path stability is calculated. Then, based on the comparison results with the energy boundary threshold, the mapping stability is judged, and the stability threshold Sth=30 is set. This threshold is obtained by statistically analyzing the S values ​​of a large number of stable and unstable mapping paths and taking their classification boundary. Since the calculated S=34.58>Sth=30, the current mapping path is judged to be "unstable" and needs to be adjusted. Based on this judgment, the node that contributes the most to instability (e.g., node 2 with -14.05) is selected as the mapping node boundary to be adjusted first, and the dynamic control boundary of multi-source signal mapping is generated.

[0101] Please see Figure 5 The specific steps for generating the intelligent emotion mapping and regulation execution cluster are as follows:

[0102] S411: Call the multi-source signal mapping dynamic control boundary, extract the mapping command and the signal state calculated by the fused signal sensor and the correlation intensity detector, perform logical consistency and state difference judgment, and obtain the mapping control deviation state;

[0103] Mapping instructions refer to the generated commands based on received control signals through logical analysis and state determination.

[0104] Logical consistency and state difference determination: Logical consistency refers to whether signals or data are logically consistent, while state difference determination refers to identifying the differences between signal states and determining whether adjustments are needed.

[0105] The system invokes multi-source signal mapping to dynamically adjust the boundary and extracts two core types of information: mapping instructions and current signal state. The mapping instructions are target state signals issued by external control logic. For example, the instruction "switch to target state 2" represents a desired emotional state shift to "excitement." The current signal state is obtained by fusing real-time data from signal sensors (EEG, GSR) and correlation strength detectors, reflecting the immediate emotional state represented by physiological signals. Assuming that after a series of fusion calculations such as weighted averaging and nonlinear mapping, the calculated quantization value of the current state is 1.5, the system then determines the logical consistency and state difference between these two types of information. Logically, the target instruction is 2, and the current state is 1.5, indicating a discrepancy. In terms of state difference, this inconsistency is quantified by calculating the absolute value of the difference between the two, i.e., 2 minus the absolute value of 1.5, resulting in 0.5. This value of 0.5 is defined as the mapping control deviation state, directly reflecting the magnitude of the difference between the current state and the target state, and is the core basis for subsequent dynamic adjustment.

[0106] S412: Based on the mapping control deviation state, combined with the energy offset and correlation strength feedback detected by the signal sensor, the mapping weight amplitude is dynamically adjusted through multi-state data fusion, the mapping path correction command is identified, and the original control quantity is superimposed and updated in real time in the mapping control to obtain the intelligent emotion mapping regulation execution cluster.

[0107] Based on the acquired mapping control deviation state of 0.5 and combined with other real-time feedback data, the mapping weights are dynamically adjusted. The data involved in the adjustment also include the energy offset value of +2.32 detected by the signal sensor and the correlation strength value of 0.85 fed back by the correlation strength detector. The calculation logic for the weight adjustment is as follows: A base weight value, such as 0.5, is set, and then a multi-factor weighted adjustment is performed on this basis. The mapping control deviation state has a positive adjustment effect on the weights, and its adjustment amount is the product of the deviation value of 0.5 and the adjustment coefficient of 0.2, i.e., +0.1. The energy offset has a negative adjustment effect on the weights, and its adjustment amount is the product of the energy offset value of 2.32 and the adjustment coefficient of 0.05. The correlation strength is adjusted when it deviates from a certain benchmark value (e.g., 0.8). The adjustment amount is the product of (0.85-0.8) and the adjustment coefficient 0.1, which is +0.005. All adjustment amounts are added to the basic weight to obtain the final dynamic mapping weight amplitude, which is 0.5+0.1-0.116+0.005, resulting in 0.489. This weight value is identified as a mapping path correction instruction and will be fed back to the mapping control module to be superimposed on the original control amount. If the original control amount is C, the updated control amount is C+0.489. Through this closed-loop mechanism of real-time update and feedback, an intelligent emotion mapping regulation execution cluster is constructed.

[0108] Please see Figure 6 The specific steps for generating the stability state identifier of the emotion mapping path are as follows:

[0109] S511: Based on the intelligent emotion mapping regulation execution cluster, the mapping trajectory output by the energy unit and the signal stability region detected by the energy unit are collected, the mapping lateral offset and stability length are extracted, the change amplitude of the stability length within the period is analyzed, and the periodic stability amplitude sequence is generated.

[0110] Based on the data collection and analysis of the intelligent emotion mapping regulation execution cluster, the actual mapping trajectory is first collected from the output of the energy unit, and information on the signal stability region is obtained from the detection module of the energy unit. Within a standard 5-second regulation cycle, two key indicators are extracted: mapping lateral offset and stability length. Mapping lateral offset refers to the average vertical distance from a point on the actual output trajectory to the preset target trajectory (a straight line or smooth curve). After calculation, the average lateral offset within this cycle is 0.3 normalized coordinate units. Stability length refers to the total time within this 5-second cycle during which all relevant physiological signals (such as temperature and SCL) remain within the preset stable region (e.g., temperature fluctuation less than 0.1°C and SCL standard deviation less than 0.2). Statistically, the stability length within this cycle is 4.1 seconds. Next, the stability length of five consecutive regulation cycles is recorded to form a time series, such as [4.1, 3.9, 4.2, 4.0, 3.8], in seconds. This numerical sequence recording the stability performance within consecutive cycles is the final generated cycle stability amplitude sequence.

[0111] S512: Call the periodic stability amplitude sequence, extract the stability amplitude difference sequence within the continuous period, mark the stability trend according to the polarity of the difference change, and combine it with the signal stability threshold to determine whether the change converges into a single trend, and obtain the signal stability fusion trend value.

[0112] The periodic stability amplitude sequence [4.1, 3.9, 4.2, 4.0, 3.8] is retrieved and subjected to difference and trend analysis. First, by calculating the difference between adjacent terms in the sequence, the stability amplitude difference sequence within the continuous period is extracted. Specifically, the second value is subtracted from the first value (3.9 - 4.1 = -0.2), the third value is subtracted from the second value (4.2 - 3.9 = +0.3), and so on, resulting in the difference sequence [-0.2, +0.3, -0.2, -0.2]. Second, based on the positive or negative polarity of each element in this difference sequence, the trend of stability change is marked. Negative values ​​represent a decrease in stability, and positive values ​​represent an increase in stability. The corresponding polarity marking sequence is as follows: Next, based on the signal stability threshold, it is determined whether the change has converged into a single trend. The convergence condition here is: in the difference sequence, are there three or more consecutive elements with the same polarity? Observing the polarity sequence [negative, positive, negative, negative], its last three terms are [positive, negative, negative], which does not meet the condition of three consecutive terms with the same polarity. Therefore, it is determined that the stability change has not converged and is in a fluctuating state. In this case, the signal stability fusion trend value is assigned to 0. If the difference sequence is [-0.2, -0.1, -0.15, -0.12], then it satisfies the condition of four consecutive negative terms, and the trend converges to a downward trend. At this time, the signal stability fusion trend value will be assigned to -1.

[0113] S513: Based on the signal stability fusion trend value, identify the lateral difference between the energy sensor state solution and the mapped positioning coordinates within the stable period, and combine it with the ratio of the signal stability region length to determine whether the mapping maintains stable path tracking, and output the emotion mapping path stability status indicator.

[0114] A comprehensive judgment is made based on the signal stability fusion trend value. First, in this example, the value is 0, indicating that the stability is in a non-convergent fluctuation state. Second, the lateral difference between the energy sensor state solution and the mapped positioning coordinates is identified in the last stable cycle. This value has been calculated as 0.3 normalized units in the previous steps. Next, the signal stability region length ratio is evaluated. This ratio is obtained by dividing the stability length of the last cycle (3.8 seconds) by the total duration of the cycle (5 seconds), and the calculated result is 0.76. Finally, the judgment logic is executed to determine whether the mapping maintains stable path tracking. This judgment logic includes... Two conditions are required: the first condition is that the signal stability fusion trend value is not -1 (i.e., the stability has not continued to deteriorate), and the second condition is that the ratio of the signal stability region length is greater than or equal to a preset threshold, such as 0.8. This threshold is determined by statistical analysis of a large number of stable tracking samples and taking the 10th percentile of their stability length ratio. In this embodiment, the trend value of 0 satisfies the first condition, but the length ratio of 0.76 does not satisfy the second condition (0.76 < 0.8). Since both conditions are not satisfied at the same time, it is finally determined that the mapping has failed to maintain stable path tracking, and "unstable" is output as the current sentiment mapping path stability status indicator.

[0115] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for emotion calculation based on in vivo multi-channel electrophysiological data, characterized in that, Includes the following steps: S1: Obtain the distribution characteristics, correlation strength between signals, and emotional fluctuation cycle information of multi-channel electrophysiological signals; perform correlation modeling and logical verification on the three types of data; evaluate the overall signal quality; identify the dynamic correlation patterns between signals; and generate a multi-channel emotional correlation feature matrix. S2: Based on the multi-channel emotion association feature matrix, extract the coupling strength between signals, the emotion fluctuation trend and the signal interference sensitivity, screen the key signal combination that meets the emotion calculation conditions, and determine the optimal mapping path in combination with the target emotion category to generate the optimal mapping interval model of emotion features. S3: Based on the optimal mapping interval model of the emotional features, the energy change trend in the coupling process between signals is analyzed, the mapping balance is judged by combining the signal interference intensity, and the mapping stability is evaluated by combining the external environmental temperature characteristics. Key mapping node boundaries are screened, and dynamic control boundaries for multi-source signal mapping are generated. S4: Call the multi-source signal mapping dynamic control boundary, compare the mapping command with the current signal state, if there is a deviation, adjust the mapping direction and weight, and combine the energy sensor feedback to correct the mapping command, so as to obtain the intelligent emotion mapping control execution cluster.

2. The emotion calculation method based on in vivo multi-channel electrophysiological data according to claim 1, characterized in that, The multi-channel emotion association feature matrix includes signal distribution characteristic level, signal correlation strength index, and emotion fluctuation cycle category. The optimal mapping interval model of emotion features includes signal coupling priority, target emotion matching degree, and shortest mapping sequence. The dynamic control boundary of multi-source signal mapping includes mapping stability boundary value, energy balance tolerance interval, and temperature matching threshold. The intelligent emotion mapping control execution cluster includes mapping direction adjustment parameters, weight correction parameters, and energy response correction coefficient.

3. The emotion calculation method based on in vivo multi-channel electrophysiological data according to claim 1, characterized in that, The specific steps for generating the multi-channel sentiment association feature matrix are as follows: S111: Obtain the distribution characteristics, correlation strength between signals, and emotional fluctuation cycle information of multi-channel electrophysiological signals. Perform data logic verification through time synchronization markers, extract the matching relationship between signal distribution characteristics and correlation strength of each time slice, and perform state normalization processing using the logical relationship of the three types of data to generate a state registration set. S112: Based on the matching region of signal distribution characteristics and correlation strength in the state registration set, combined with the emotional fluctuation cycle information, extract the center offset of the emotional fluctuation trajectory and the adjacent state switching angle, identify the emotional fluctuation path, and generate an emotional fluctuation parameter set. S113: Based on the state information of each fluctuation path in the emotional fluctuation parameter group, analyze the signal configuration mapping relationship in the original state, transform the path state into a unified state system, perform signal quality and temperature verification on the channel based on path continuity and state change, integrate dynamic fluctuation segments and feasible state ranges, and generate a multi-channel emotional correlation feature matrix.

4. The emotion calculation method based on in vivo multi-channel electrophysiological data according to claim 3, characterized in that, The specific steps for generating the optimal mapping interval model for emotional features are as follows: S211: Based on the multi-channel emotion association feature matrix, by extracting data from signal sensors, association strength detectors and temperature sensors, extracting the coupling strength between signals, emotion fluctuation trends and signal interference sensitivity, identifying key signal combinations and detecting mapping paths affected by interference, eliminating path segments that do not meet the conditions for emotion calculation, and generating priority mapping trajectories; S212: Using the prioritized mapped trajectory and combining it with the stability detection data from the temperature sensor, identify the mapping priority segment of each trajectory segment, extract the segment information, and apply the following formula: ； Calculate the mapping optimization value of the segment and select the segment with the best mapping priority; in, The mapping optimization value of the representative segment, This represents the temperature sensor measurement value for the i-th trajectory segment. This represents the average temperature across all trajectory segments. Represents the maximum value of temperature. Represents the minimum temperature. The weight coefficient represents the weight of the i-th trajectory segment. The stability detection data represents the i-th trajectory segment. Represents the total number of trajectory segments; S213: Call the optimal mapping priority segment, combine it with the target emotion category, calculate the angular deviation between the target mapping direction and the segment direction, select the optimal path segment based on the deviation, and generate the optimal mapping interval model for emotion features.

5. The emotion calculation method based on in vivo multi-channel electrophysiological data according to claim 4, characterized in that, The specific steps for generating the dynamic control boundary of the multi-source signal mapping are as follows: S311: Based on the optimal mapping interval model of the emotional features, extract the energy change trend in the coupling process between signals, identify the energy value and time interval between the current frame and the previous frame, analyze the energy change trend, determine the direction and intensity of energy change, and obtain the energy offset trend value. S312: Call the energy offset trend value, and calculate the mapping path stability quantization value by combining the mapping weight adjustment amount, the stability index of temperature feature extraction and energy boundary distance. Based on the comparison result with the energy boundary threshold, judge the mapping stability, filter the mapping node boundary, and generate the multi-source signal mapping dynamic control boundary.

6. The emotion calculation method based on in vivo multi-channel electrophysiological data according to claim 5, characterized in that, The specific steps for generating the intelligent emotion mapping and regulation execution cluster are as follows: S411: Call the multi-source signal mapping dynamic control boundary, extract the mapping command and the signal state calculated by the fused signal sensor and the correlation strength detector, perform logical consistency and state difference judgment, and obtain the mapping control deviation state; S412: Based on the mapping control deviation state, combined with the energy offset and correlation strength feedback detected by the signal sensor, the mapping weight amplitude is dynamically adjusted through multi-state data fusion, the mapping path correction instruction is identified, and fed back to the mapping control to superimpose the original control quantity and update in real time, thus obtaining the intelligent emotion mapping regulation execution cluster.

7. The emotion calculation method based on in vivo multi-channel electrophysiological data according to claim 6, characterized in that, The mapping instruction refers to the received control signal, which is generated through logical analysis and state determination; The logical consistency and state difference determination refer to whether signals or data are logically consistent, and the state difference determination refers to identifying the differences between signal states and determining whether adjustments are needed.

8. The emotion calculation method based on in vivo multi-channel electrophysiological data according to claim 1, characterized in that, The method also includes step S5: S5: Based on the intelligent emotion mapping control execution cluster, it collects the continuous periodic mapping trajectory output and the signal stability length change in the energy sensor, analyzes whether the change trend is stable, determines the mapping stability period after control execution, and outputs the emotion mapping path stability status indicator. The stability status identifier of the emotion mapping path includes the duration of mapping stability, the residual length of signal energy, and the amplitude of trajectory fluctuation.

9. The emotion calculation method based on in vivo multi-channel electrophysiological data according to claim 8, characterized in that, The specific steps for generating the stability state identifier of the emotion mapping path are as follows: S511: Based on the intelligent emotion mapping control execution cluster, collect the mapping trajectory output by the energy unit and the signal stability region detected by the energy unit, extract the mapping lateral offset and stability length, analyze the change amplitude of the stability length within the period, and generate a periodic stability amplitude sequence. S512: Call the periodic stability amplitude sequence, extract the stability amplitude difference sequence within the continuous period, mark the stability trend according to the polarity of the difference change, and combine it with the signal stability threshold to determine whether the change converges into a single trend, and obtain the signal stability fusion trend value. S513: Based on the signal stability fusion trend value, identify the lateral difference between the energy sensor state solution and the mapped positioning coordinates within the stable period, and combine it with the ratio of the signal stability region length to determine whether the mapping maintains stable path tracking, and output the emotion mapping path stability status indicator.

Images