Human body part pressure partition processing method
By determining the head position using human pressure data and a preset model, and dividing it into six zones based on gender, height, and weight, the system solves the problems of noise interference and physiological structure differences, achieving high-precision human body part zoning, which is suitable for low-cost embedded chips.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUNAN UNIV
- Filing Date
- 2026-01-23
- Publication Date
- 2026-05-12
AI Technical Summary
Existing technologies suffer from noise data interference in pressure zoning of human body parts, leading to blurred boundaries and difficulty in segmentation. Furthermore, physiological structural differences in different height proportions result in inaccurate zoning.
By acquiring human pressure data, using a pre-set anthropometry statistical model and core interval data index, the head position is determined, and the human body is divided into six zones according to gender, height, and weight. Pressure and quality indicators are calculated, and the zones are processed in conjunction with a one-dimensional longitudinal pressure distribution curve.
It achieves high-precision human body part partitioning on low-cost embedded chips, reduces dependence on complex models, improves the accuracy and robustness of partitioning, and adapts to physiological structures with different height proportions.
Smart Images
Figure CN122004756A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of human body part pressure zone signal processing technology, and particularly relates to a human body part pressure zone processing method. Background Technology
[0002] Human body pressure zoning signal processing technology is a key technology connecting human physiological state monitoring with the external support environment, and also a core underlying technology for realizing intelligent human perception and adaptive interaction. For example, in usage scenarios such as operating tables and ambulance negative pressure stretchers, long-term lying nursing beds, ICU wards, long-distance passenger cabins, and smart homes, it is necessary to monitor the pressure index and force distribution of different parts of the body in real time to achieve the purposes of prevention, regulation, and comfort.
[0003] In existing technologies, the first common approach involves data acquisition and analysis using image segmentation techniques based on morphological thresholding. This method primarily involves setting a fixed pressure threshold, converting the grayscale matrix collected by the pressure sensor into a binary image, and then combining this with human body proportions to segment regions and achieve body part partitioning. The second approach utilizes convolutional neural network segmentation techniques, which requires acquiring massive amounts of human posture and pressure images and training a deep learning model to achieve body part partitioning.
[0004] However, in the first method, noise data caused by thick blankets, sheet tension, etc., makes human body boundaries blurred and difficult to segment. Additionally, physiological differences in body structure due to varying heights lead to inaccurate differentiation of body parts. In the second method, the learning model is highly dependent on the data; using data beyond the training data range results in a decrease in the accuracy of body part segmentation. Summary of the Invention
[0005] This invention provides a method for pressure zoning of human body parts, aiming to solve the above-mentioned technical problems.
[0006] This invention is implemented as follows: a method for pressure zoning of human body parts, comprising:
[0007] Acquire human stress data and pre-set anthropometry statistical models;
[0008] Based on the human body pressure data, the head position of the human body is determined through the core region data index;
[0009] Based on the subject's gender, height and weight, a pre-set anthropometry statistical model and head position, the human body is divided into six regions, resulting in six human body region partitions.
[0010] Calculate the pressure index, overall pressure index, and body mass index for six body part zones;
[0011] The assessment is conducted using the stress index, overall stress index, and body mass index of the six body part zones.
[0012] Furthermore, the acquisition of human body pressure data and the preset anthropometry statistical model includes:
[0013] Acquire pressure sensor signals;
[0014] The sensor signals are converted into physical data with the X-axis as the horizontal axis and the Y-axis as the vertical axis;
[0015] The total pressure value for each longitudinal direction is obtained by summing up the physical data along the Y-axis.
[0016] A one-dimensional longitudinal pressure distribution curve is plotted based on the total pressure value for each longitudinal direction.
[0017] Furthermore, the acquisition of human body pressure data and the preset anthropometry statistical model also includes:
[0018] The pressure sensor signals are iterated through, and noise points below a preset pressure threshold are filtered out.
[0019] Correct isolated bad pixel data in the pressure sensor signal.
[0020] Furthermore, determining the head position of the human body based on the human pressure data and through core region data indexing includes:
[0021] Based on the one-dimensional longitudinal pressure distribution curve, obtain the first local minimum value within a preset first interval range;
[0022] Mark the position corresponding to the first local minimum as the head position point.
[0023] Furthermore, based on the subject's gender, height, weight, a pre-set anthropometry statistical model, and head position, the human body is divided into six regions, resulting in six human body region partitions, including:
[0024] Based on the subject's gender, height, and weight, obtain the corresponding preset shoulder length data, preset back length data, preset waist length data, preset hip length data, and preset leg length data in the anthropometry statistical model;
[0025] Add the preset shoulder length data to the head position point to obtain the shoulder position point;
[0026] Add the preset back length data to the shoulder position point to obtain the back position point;
[0027] Add the preset waist length data to the back position point to obtain the waist position point;
[0028] Add the preset hip length data to the waist position point to obtain the hip position point;
[0029] Add the preset leg length data to the hip position point to obtain the leg position point;
[0030] The area from the starting point to the head position point is the head region.
[0031] The area from the head position point to the shoulder position point is the shoulder area;
[0032] The area from the shoulder position point to the back position point is the back area;
[0033] The area from the back position point to the waist position point is the waist area;
[0034] The area from the waist position point to the hip position point is the hip area;
[0035] The area from the hip position point to the leg position point is the leg area.
[0036] Furthermore, correcting isolated bad pixel data in the pressure sensor signal includes:
[0037] The median filtering algorithm is used to obtain the median of the eight sensor signals surrounding the isolated defect as the pressure sensor signal of the current isolated defect data.
[0038] Furthermore, the calculation of pressure indices for six body part zones, overall pressure indices, and body mass indices includes:
[0039] Based on the area range of the six zones, calculate the maximum pressure, average pressure, contact area, and standard deviation for each human body part zone;
[0040] At the same time, the overall pressure index is calculated based on the pressure data of six body part zones;
[0041] Body mass indexes were calculated based on the subject's gender, height, and weight.
[0042] Furthermore, based on the subject's gender, height, weight, a pre-set anthropometry statistical model, and head position, the human body is divided into six regions, resulting in six human body region partitions, which also include:
[0043] Based on the one-dimensional longitudinal pressure distribution curve, the pressure value in the neighborhood of the shoulder location point is obtained;
[0044] Determine if a local significant maximum exists;
[0045] If it exists, calculate the distance between the local salient maximum and the head position point in the lateral direction;
[0046] If the distance in the lateral direction is less than a preset threshold, the shoulder position point is corrected to the position point corresponding to a local significant maximum.
[0047] Furthermore, including:
[0048] The preset first interval range is defined as the preset head search interval range starting from the starting point of the one-dimensional longitudinal pressure distribution curve.
[0049] This invention provides a method for processing pressure zoning of human body parts. It involves reducing the acquired pressure blanket sensor signal data from three dimensions to easily processed one-dimensional physical data. The physical data in the Y-axis direction are then summed to obtain the total longitudinal pressure value, and a one-dimensional longitudinal pressure distribution curve is plotted. Based on this curve, the head is located using local minima, and the positions of other body parts are anchored by adding the length of the head to the length of other pre-defined body parts, thereby determining the human body part zoning. The algorithm has extremely low complexity and can be directly deployed on low-cost embedded chips. Attached Figure Description
[0050] Figure 1 Schematic diagram of operations for acquiring human pressure data in different postures;
[0051] Figure 2 The length values of each body part and the number of squares in the pressure blanket are corresponding to the percentage of height of a thousand people;
[0052] Figure 3 This is a schematic diagram of a one-dimensional longitudinal pressure distribution curve;
[0053] Figure 4 A diagram showing the division of six human body parts. Detailed Implementation
[0054] 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.
[0055] This application provides a method for pressure zoning of human body parts. The core of this method is how to divide human body parts into zones and evaluate the zones to determine whether there is uneven pressure distribution in each zone, thereby achieving the functions of prevention, adjustment and ensuring comfort.
[0056] The embodiments disclosed in this application introduce a three-dimensional human body database with thousands of users, construct a mapping mechanism, utilize one-dimensional dimensionality reduction processing to achieve millisecond-level partitioning, and output objective feature data. Combined with multi-dimensional stress indicators, a medical-grade health assessment system is constructed. Specific embodiments are as follows:
[0057] In step S100, the subject's gender, height and weight, human pressure data, and a preset anthropometry statistical model are obtained.
[0058] In this embodiment, human body pressure data is acquired through the sensors of a flexible pressure sensing matrix (collectively referred to as a pressure pad or pressure blanket), and the acquired data is converted into data that can be directly used and calculated in the method of this application embodiment.
[0059] See Figure 1 , Figure 1 A schematic diagram illustrating the operation of acquiring human pressure data in different postures.
[0060] according to Figure 1 It is known that the pressure pad is laid flat on the mattress, with the top edge of the pillow aligned with the top edge of the pressure point. The subject lies supine with their body horizontal and their body axis in a straight line, head resting on the pillow, shoulders outside the bottom edge of the pillow, and arms and legs naturally extended.
[0061] After determining the posture, data collection begins, and the results are as follows: Figure 1 The pressure heatmap shown can macroscopically represent the stress on the human body through color. However, it cannot quantify the stress on individual body regions, therefore, the pressure heatmap needs to be quantified.
[0062] Therefore, based on the pressure heatmap, complete human body pressure data can be output, specifically, the data can be output through an Excel spreadsheet.
[0063] The pre-set anthropometric statistical model is based on a 3D human body scan database constructed from a sample of thousands of people, obtained through experiments. (See [link to relevant documentation]). Figure 2 , Figure 2 The values represent the lengths of various body parts and the number of pressure blanket squares for each percentage of the population's height.
[0064] The average lengths of the head, shoulders, back, waist, hips, and legs of 1000 people were calculated by dividing them into human percentiles (P0 / P10 / P20 / P40 / P50 / P60 / P80 / P90 / P100) for both men and women. These average lengths were then converted to the total number of segments on the pressure pad. For example, the average head length for men (P50) is 247.6857mm. Based on the 28x42 squares (rectangular) of the experimental pressure pad, with each segment measuring 45mm x 30mm, the head length for men (P50) corresponds to 6 segments on the pressure pad.
[0065] Specifically, in S101, the pressure sensor signal is acquired.
[0066] In step S102, the sensor signal is converted into physical data with the X-axis as the horizontal axis and the Y-axis as the vertical axis.
[0067] In step S103, the physical data in the longitudinal direction of the Y-axis are summed to obtain the total pressure value in each longitudinal direction.
[0068] In step S104, a one-dimensional longitudinal pressure distribution curve is plotted based on the total pressure value for each longitudinal direction.
[0069] like Figure 3 As shown, Figure 3 This is a schematic diagram of a one-dimensional longitudinal pressure distribution curve.
[0070] exist Figure 3 In this process, the acquired pressure sensor signals are converted into physical data. With the length of the human body as the X-axis and the width as the Y-axis, the physical data in the longitudinal direction of the Y-axis are summed to obtain the total pressure value in each longitudinal direction. The total pressure value is then plotted as a one-dimensional longitudinal pressure distribution curve.
[0071] Specifically, system parameter reading: The physical size of the sensor used in this embodiment is 2000mm×900mm, and the effective sensing area is 1890mm×840mm.
[0072] The matrix specifications are as follows. Adapting to the human lying posture, the coordinate system is defined as follows:
[0073] X-axis (vertical axis / length direction): corresponds to the direction of the human head, feet, and eyes, including N. X Each sensor array has a physical sampling interval defined as Px = 45 mm due to the anisotropic design of the array.
[0074] Y-axis (horizontal axis / width direction): corresponds to the left and right width direction of the human body, including N. Y =28 sensor rows, with a physical sampling interval defined as Py=30mm.
[0075] In addition, when acquiring pressure sensor signals, pressure data of objects such as blankets and quilts are also acquired. Therefore, in this embodiment, noise points below a preset pressure threshold can be filtered out by traversing the pressure sensor signal data, and isolated bad point data in the pressure sensor signal can be corrected.
[0076] Specifically, set an adaptive pressure threshold T. noise(In this embodiment, 0.2 kPa is used). This threshold is not arbitrarily set, but based on the signal-to-noise ratio (SNR) principle. This ensures that while eliminating noise, the shoulder signal in a micro-contact state can be identified. Traversing the original matrix, if P(x,y)... <T noise Then, the sensor is forced to zero to eliminate sheet tension and sensor zero drift under no-load conditions.
[0077] To suppress isolated bad data, a 3×3 median filter is used. For any non-zero point in the matrix, the median of its 8-neighbor values is taken to replace the current value, thus eliminating spike noise caused by poor hardware contact.
[0078] Collect N consecutive frames of data, such as 5 frames within 1 second. Calculate the arithmetic mean matrix when the variance of the pressure center displacement converges. steady As a subsequent input, the effects of respiratory micromovements are eliminated.
[0079] After eliminating high-frequency spatial noise introduced by uncontrolled environments, such as bedding wrinkles and sheet tension, robust human axial morphological features are extracted.
[0080] By employing longitudinal projection technology, the preprocessed two-dimensional pressure matrix is accumulated along the sensor's Y-axis, thereby mapping the two-dimensional discrete pressure field into a one-dimensional longitudinal pressure distribution curve V distributed along the X-axis. x :
[0081]
[0082] Where, N Y =28. This operation uses the principle of integration to eliminate the degree of freedom in the width direction, smooth out local noise, and effectively enhance the overall anatomical morphological characteristics of the human body along the longitudinal axis.
[0083] Furthermore, the following implementation methods are provided for correcting pressure data such as ineffective contact of the skin within the human body pressure zone:
[0084] An effective support threshold of 1.33 kPa (approximately 10 mmHg) is defined to filter out ineffective skin surface contact, focusing only on the effective pressure-bearing area. The corrected pressure variable P̂(x, y) is defined as:
[0085] P̂ = .
[0086] In the above steps, the three-dimensional data is converted into two-dimensional physical data, and then the dimensionality is reduced to obtain a one-dimensional longitudinal pressure distribution curve. This step is the basis for subsequent calculations and evaluations, reduces the performance support of inference calculations, and also eliminates the need for training complex models, thus reducing the dependence on model training.
[0087] In step S200, the head position of the human body is determined based on the human body pressure data and through the core region data index.
[0088] In this embodiment, the head position of the human body is specifically the boundary anchor point between the head and neck. For ease of understanding and subsequent position point calculation, the decomposed anchor point of the head and neck is defined as the head position.
[0089] After acquiring human pressure data, the head position is determined using the core interval data index. Here, the core interval data is set to the first 1 / 3 of the search interval, specifically including:
[0090] Based on the one-dimensional longitudinal pressure distribution curve, the first local minimum value within the preset first interval is obtained, and the position corresponding to the first local minimum value is marked as the head position point.
[0091] The first preset interval range is the preset head search interval range of the one-dimensional longitudinal pressure distribution curve. This area is the head region. Within this area, the first local minimum value can be found quickly and accurately. This minimum value is the location point where the neck is under the least force, so it can be marked as the head location point P1.
[0092] In step S300, the human body is divided into six regions based on the subject's gender, height and weight, a preset anthropometry statistical model and head position, resulting in six human body region partitions.
[0093] Once the head position is determined, it can be determined according to a pre-set anthropometry statistical model, that is... Figure 2 As shown, other locations are determined, and finally, six human body part zones are defined.
[0094] Specifically, this includes: obtaining preset shoulder length data, preset back length data, preset waist length data, preset hip length data, and preset leg length data in the anthropometry statistical model based on the subject's gender, height, and weight;
[0095] from Figure 2 As shown, based on the subject's or patient's gender, height, and weight, corresponding length data for each part can be obtained. Figure 4 As shown, Figure 4 A diagram showing the division of six human body parts.
[0096] Add the preset shoulder length data to the head position point P1 to obtain the shoulder position point P2. Since the head position point P is the dividing anchor point between the head and the neck, the shoulder position point P2 can be obtained by adding the preset shoulder length data to the head position point P1. Similarly, the preset back length data is added to the shoulder position point P2 to obtain the back position point P3; the preset waist length data is added to the back position point P3 to obtain the waist position point P4; the preset hip length data is added to the waist position point P4 to obtain the hip position point P5; and the preset leg length data is added to the hip position point P5 to obtain the leg position point P6. The area from the starting point to the head position point is the head area; the area from the head position point P1 to the shoulder position point P2 is the shoulder area; the area from the shoulder position point P2 to the back position point P3 is the back area; the area from the back position point P3 to the waist position point P4 is the waist area; the area from the waist position point P4 to the hip position point P5 is the hip area; and the area from the hip position point P5 to the leg position point P6 is the leg area.
[0097] Specific implementation method: The system uses a database containing a built-in anthropometric statistical model constructed from 3D human body scan data of thousands of people; or, the database system contains a built-in Chinese adult anthropometric model conforming to national standards. This model stores the average longitudinal projection length Lphy of six key anatomical segments in a supine position for people of different genders and height percentiles, including preset shoulder length data, preset back length data, preset waist length data, preset hip length data, and preset leg length data.
[0098] Based on the user's input of gender, height, and weight, if no information is provided, the P50 mean is used by default, and the corresponding L value is retrieved. phy， Convert it to the number of grid cells N in the sensor coordinate system using the formula. grid。
[0099]
[0100] Example: For a P50 male, the projected head length is 247.7 mm. Calculations show that 247.7 / 45 is approximately 5.5, rounded down to the nearest integer, resulting in 6 grid cells. This discretized value provides a precise search window for subsequent partitioning.
[0101] Shoulder position point P2 is locked as follows:
[0102] Based on prior theoretical positioning and following the logic of the above steps, the preset shoulder length data N is directly called. shoulder Starting from the head position point P1, the coordinates P of the theoretical shoulder position point P2 are derived. 2-theory : + .
[0103] The coordinate derivation of subsequent parts will take the finalized P2 as the new starting point, and based on the length grid number of each part in the database, the subsequent coordinates will be deduced sequentially:
[0104] Back position point P3: P3 = P2 + N bac k;
[0105] Waist position point P4: P4 = P3 + N waist ;
[0106] Hip position point P5: P5 = P4 + N hip ;
[0107] Leg position point P6: P6 = P5 + N leg ;
[0108] Connect the region from the starting point to point P1 to obtain the head region; connect the regions P1 and P2 to obtain the shoulder region; connect the region from P2 to P3 to obtain the back region; connect the region from P3 to P4 to obtain the waist region; connect the region from P4 to P5 to obtain the hip region; and connect the region from P5 to P6 to obtain the leg region.
[0109] Furthermore, for the theoretical value of shoulder position point P2, another embodiment is provided to repair the theoretical value of shoulder position point P2.
[0110] Specifically, this includes: obtaining the pressure value in the neighborhood of the shoulder position point based on the one-dimensional longitudinal pressure distribution curve; determining whether there is a local significant maximum; if so, calculating the distance between the local significant maximum and the head position point in the lateral direction; if the distance in the lateral direction is less than a preset threshold, correcting the shoulder position point to the position point corresponding to the local significant maximum.
[0111] To accommodate individual differences in experience, such as variations in neck length, in theoretical coordinate P2 -theroy Within the smallest neighborhood, detect whether there are local significant maxima, that is, significant pressure peaks.
[0112] The decision logic is to calculate the distance Δv in the lateral direction between the local maximum value in the neighborhood of the shoulder location point and the neck trough P1.
[0113] If Δv is less than or equal to the threshold, and the threshold is set within a reasonable range of 50mm-80mm, it indicates that the distance between the user's shoulder and neck is relatively close. In this case, the local salient maximum is more suitable as the shoulder position point. Therefore, the shoulder position point can be corrected to the position point corresponding to the local salient maximum.
[0114] In step S400, the pressure index, overall pressure index, and body mass index of the six body part zones are calculated.
[0115] After determining the six body part zones, the maximum pressure, average pressure, contact area, and standard deviation for each zone are calculated based on their respective area ranges. Simultaneously, the overall pressure index is calculated based on the pressure data from the six body part zones.
[0116] The overall pressure index PI, a composite index of mean deviation and dispersion, is used to evaluate the uniformity of the overall load distribution.
[0117]
[0118] Among them, P avy SD represents the normalized average pressure, and SD represents the standard deviation of the pressure distribution.
[0119] Body Mass Index (BMI), also known as Body Mass Index, can reflect the density of the human body to a certain extent.
[0120] In step S500, the pressure index, overall pressure index, and body mass index of the six human body parts are used for evaluation.
[0121] Maximum pressure refers to the highest pressure value among all contact points, reflecting the maximum pressure a local area can withstand. The higher the value, the greater the local pressure and the greater the risk of tissue damage.
[0122] Average pressure refers to the average pressure value of all contact points, reflecting the overall pressure distribution level of that part. A value that is too high may indicate insufficient support, while a value that is too low may indicate insufficient contact area.
[0123] The contact area is simply the pressure contact area between the body and the wound; the larger the value, the larger the contact area.
[0124] Standard deviation measures the dispersion of pressure values at different locations on a pressure blanket, reflecting the uniformity or volatility of pressure distribution. A low value indicates highly concentrated and uniform pressure values, while a high value indicates large pressure fluctuations and uneven distribution, potentially indicating the presence of localized high or low pressure areas.
[0125] A lower PI value indicates a more uniform load distribution and better ergonomic support. It comprehensively reflects the risk of pressure distribution, as well as the intensity and uniformity of the pressure distribution. A lower pressure index indicates better comfort and support, better pressure relief, and the potential presence of localized high or low pressure areas.
[0126] During algorithm execution, the subject's BMI value can be derived from their height, gender, and weight. BMI provides a necessary physiological reference for judging the reasonableness of the stress distribution. Finally, by combining the BMI value with the stress data under mechanical conditions, the subject's physiological and mechanical characteristics are evaluated.
[0127] As can be seen from the above embodiments, the human body pressure zoning method provided in this application introduces a three-dimensional human body database of thousands of people, i.e., an anthropometric statistical model, and uses one-dimensional dimensionality reduction processing to achieve human body part zoning. It outputs data including maximum pressure, average pressure, contact area, standard deviation, overall pressure index, and body mass index, constructing a medical-grade health assessment system. The introduced three-dimensional human body database of thousands of people can output more accurate zoning data for physiological structures with different height proportions. Furthermore, converting three-dimensional data into one-dimensional data reduces the learning model's dependence on data while simultaneously ensuring accurate results.
[0128] Furthermore, by reducing 3D data to 1D signal processing and handling only a single frame of data, the algorithm has extremely low complexity and can be directly deployed on low-cost embedded chips. It also exhibits high robustness and consistency; actual measurements show that under longitudinal displacement disturbances in different supine positions, the pass rate for a one-grid deviation reaches 94.4%.
[0129] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for pressure zoning of human body parts, characterized in that, include: Acquire human stress data and pre-set anthropometry statistical models; Based on the human body pressure data, the head position of the human body is determined through the core region data index; Based on the subject's gender, height and weight, a pre-set anthropometry statistical model and head position, the human body is divided into six regions, resulting in six human body region partitions. Calculate the pressure index, overall pressure index, and body mass index for six body part zones; The assessment is conducted using the stress index, overall stress index, and body mass index of the six body part zones.
2. The method for processing pressure zones in human body parts according to claim 1, characterized in that, The acquisition of human body pressure data and the preset anthropometry statistical model include: Acquire pressure sensor signals; The sensor signals are converted into physical data with the X-axis as the horizontal axis and the Y-axis as the vertical axis; The physical data in the longitudinal direction of the Y-axis are summed to obtain the total pressure value in each longitudinal direction; A one-dimensional longitudinal pressure distribution curve is plotted based on the total pressure value for each longitudinal direction.
3. The method for processing pressure zones in human body parts according to claim 2, characterized in that, The acquisition of human body pressure data and the preset anthropometry statistical model also include: The pressure sensor signals are iterated through, and noise points below a preset pressure threshold are filtered out. Correct isolated bad pixel data in the pressure sensor signal.
4. The method for processing pressure zones in human body parts according to claim 2, characterized in that, Based on the aforementioned human pressure data, the head position is determined using a core region data index, including: Based on the one-dimensional longitudinal pressure distribution curve, obtain the first local minimum value within a preset first interval range; Mark the position corresponding to the first local minimum as the head position point.
5. The method for processing pressure zones in human body parts according to claim 4, characterized in that, Based on the subjects' gender, height, weight, a pre-set anthropometry statistical model, and head position, the human body is divided into six regions, resulting in six human body region partitions, including: Based on the subject's gender, height, and weight, obtain the corresponding preset shoulder length data, preset back length data, preset waist length data, preset hip length data, and preset leg length data in the anthropometry statistical model; Add the preset shoulder length data to the head position point to obtain the shoulder position point; Add the preset back length data to the shoulder position point to obtain the back position point; Add the preset waist length data to the back position point to obtain the waist position point; Add the preset hip length data to the waist position point to obtain the hip position point; Add the preset leg length data to the hip position point to obtain the leg position point; The area from the starting point to the head position point is the head region. The area from the head position point to the shoulder position point is the shoulder area; The area from the shoulder position point to the back position point is the back area; The area from the back position point to the waist position point is the waist area; The area from the waist position point to the hip position point is the hip area; The area from the hip position point to the leg position point is the leg area.
6. The method for processing pressure zones in human body parts according to claim 3, characterized in that, Correcting isolated bad pixel data in the pressure sensor signal includes: The median filtering algorithm is used to obtain the median of the eight sensor signals surrounding the isolated defect as the pressure sensor signal of the current isolated defect data.
7. The method for processing pressure zones in human body parts according to claim 5, characterized in that, The calculation of pressure indices for six body part zones, overall pressure indices, and body mass indices includes: Based on the area range of the six zones, calculate the maximum pressure, average pressure, contact area, and standard deviation for each human body part zone; The overall pressure index is calculated based on the pressure data of six body parts. Body mass indexes were calculated based on the subject's gender, height, and weight.
8. The method for processing pressure zones in human body parts according to claim 5, characterized in that, Based on the subject's gender, height, weight, a pre-set anthropometry statistical model, and head position, the human body is divided into six regions, resulting in six human body region partitions, which also include: Based on the one-dimensional longitudinal pressure distribution curve, the pressure value in the neighborhood of the shoulder location point is obtained; Determine if a local significant maximum exists; If it exists, calculate the distance between the local salient maximum and the head position point in the lateral direction; If the distance in the lateral direction is less than a preset threshold, the shoulder position point is corrected to the position point corresponding to a local significant maximum.
9. The method for processing pressure zones in human body parts according to claim 4, characterized in that, include: The preset first interval range is defined as: the preset head search interval range starting from the starting point of the one-dimensional longitudinal pressure distribution curve.