CN-122004756-A - Human body part pressure partition processing method

CN122004756ACN 122004756 ACN122004756 ACN 122004756ACN-122004756-A

Abstract

The invention belongs to the technical field of human body part pressure partition signal processing, and provides a human body part pressure partition processing method which comprises the steps of obtaining human body pressure data and a preset anthropometric statistical model, determining the head position of a human body through a core interval data index according to the human body pressure data, dividing a human body part into six areas according to the gender, the height and the weight of a tested person, the preset anthropometric statistical model and the head position to obtain six human body part partitions, calculating the pressure index, the overall pressure index and the body quality index of the six human body part partitions, and evaluating by utilizing the pressure index, the overall pressure index and the body quality index of the six human body part partitions. The three-dimensional data are reduced to one-dimensional data, so that the partitioning of the human body part is realized, and meanwhile, more accurate partitioning data, pressure data and body quality index data are output by combining database preset data.

Inventors

WANG HAINING
LI GUO
WANG SHU

Assignees

湖南大学

Dates

Publication Date: 20260512
Application Date: 20260123

Claims (9)

1. A method for pressure zoning of a body part, comprising: Acquiring human body pressure data and a preset anthropometric statistical model; According to the human body pressure data, determining the head position of the human body through a core interval data index; Dividing a human body part into six areas according to the gender, the height and the weight of the tested person, and preset anthropometric statistical models and head positions to obtain six human body part areas; calculating pressure indexes, overall pressure indexes and body quality indexes of the six body part partitions; And evaluating by using the pressure index, the overall pressure index and the body quality index of the six body part partitions.
2. The method for pressure zoning according to claim 1, wherein the acquiring the pressure data and the predetermined anthropometric statistical model comprises: Acquiring a pressure sensor signal; converting the sensor signals into physical data with an X axis as a horizontal axis and a Y axis as a vertical axis; accumulating and summing physical data in the longitudinal direction of the Y axis to obtain a total pressure value of each longitudinal direction; And drawing a one-dimensional longitudinal pressure distribution curve according to the total pressure value of each longitudinal direction.
3. The method for pressure zoning according to claim 2, wherein the acquiring the pressure data and the predetermined anthropometric statistical model further comprises: Traversing the pressure sensor signal, and filtering noise points smaller than a preset pressure threshold value; and correcting the isolated bad point data in the pressure sensor signal.
4. The human body part pressure partition processing method according to claim 2, wherein determining a head position of a human body from the human body pressure data by a core interval data index comprises: Acquiring a first local minimum value in a preset first interval range according to the one-dimensional longitudinal pressure distribution curve; and marking the position corresponding to the first local minimum as a head position point.
5. The method according to claim 4, wherein dividing the human body into six regions according to the sex, height and weight of the subject, a predetermined anthropometric statistical model, and head position to obtain six human body region regions, comprises: Acquiring corresponding preset shoulder length data, preset back length data, preset waist length data, preset hip length data and preset leg length data in an anthropometric statistical model according to the gender, the height and the weight of the tested person; Adding the head position points to the preset shoulder length data to obtain shoulder position points; adding the shoulder position points to the preset back length data to obtain back position points; Adding the preset waist length data to the back position points to obtain waist position points; adding the preset hip length data to the waist position points to obtain hip position points; Adding the hip position points with the preset leg length data to obtain leg position points; wherein the region from the start point to the head position point is a head region; the area from the head position point to the shoulder position point is a shoulder area; the area from the shoulder position point to the back position point is a back area; The area from the back position point to the waist position point is a waist area; The area from the waist position point to the hip position point is a hip area; The area from the hip position point to the leg position point is a leg area.
6. A method of human body part pressure zoning according to claim 3, wherein modifying isolated bad data in the pressure sensor signal comprises: And acquiring the median of 8 sensor signals around the isolated dead pixel as the pressure sensor signal of the current isolated dead pixel data by adopting a median filtering algorithm.
7. The method according to claim 5, wherein calculating the pressure index, the overall pressure index, and the body mass index of the six body part partitions comprises: respectively calculating the maximum pressure, the average pressure, the contact area and the standard deviation of each body part partition according to the area ranges of the six areas; Calculating an overall pressure index according to the pressure data of the six body part partitions; and calculating the body quality index according to the gender, the height and the weight of the tested person.
8. The human body part pressure partition processing method according to claim 5, wherein the human body part is divided into six regions according to the sex, the height and the weight of the subject, a preset anthropometric statistical model and a head position, to obtain six human body part partitions, further comprising: acquiring a pressure value in the vicinity of the shoulder position point according to the one-dimensional longitudinal pressure distribution curve; judging whether a local significant maximum exists or not; if so, calculating the distance between the local significant maxima and the head position points in the transverse direction; and if the distance in the transverse direction is smaller than a preset threshold value, correcting the shoulder position point to be a position point corresponding to the local significant maximum value.
9. The method of pressure zoning treatment of a body part according to claim 4, comprising: The preset first interval range is defined as a preset head search interval range from the start point of the one-dimensional longitudinal pressure distribution curve.

Description

Human body part pressure partition processing method Technical Field The invention belongs to the technical field of pressure partition signal processing of human body parts, and particularly relates to a pressure partition processing method of human body parts. Background The pressure partition signal processing technology of the human body part is a key technology for connecting human body physiological state monitoring and external supporting environment, and is also a core bottom technology for realizing intelligent human body perception and self-adaptive interaction. For example, in operating tables and ambulance negative pressure stretchers, long-term lying nursing beds, ICU wards, coach lying, intelligent houses and other use scenes, the purposes of prevention, adjustment, comfort and the like are achieved by monitoring the pressure indexes, stress distribution and other data of different parts of the body in real time. In the prior art, the first method is to realize data acquisition and analysis through an image segmentation technology principle based on morphological threshold values, and the method mainly comprises the steps of setting a fixed pressure threshold value, converting a gray matrix acquired by a pressure sensor into a binary image, and then carrying out region cutting by combining with human body proportion to realize the segmentation of human body parts. Secondly, by the convolutional neural network segmentation technology principle, massive human body posture pressure images need to be acquired, a deep learning model is trained, and the partitioning of human body parts is realized. However, in the first method, the human body boundary is blurred and difficult to divide for noise data caused by the tension of thick bedding, bed sheets and the like, and in addition, the human body part is not distinguished accurately due to the physiological structure difference of different height proportions. In the second method, the learning model has strong dependence on data, and for the use condition exceeding the training data range, the accurate partition of the human body part is reduced. Disclosure of Invention The invention provides a pressure partition processing method for a human body part, and aims to solve the technical problems. The invention is realized in such a way that the pressure partition treatment method of the human body part comprises the following steps: Acquiring human body pressure data and a preset anthropometric statistical model; According to the human body pressure data, determining the head position of the human body through a core interval data index; Dividing a human body part into six areas according to the gender, the height and the weight of the tested person, and preset anthropometric statistical models and head positions to obtain six human body part areas; calculating pressure indexes, overall pressure indexes and body quality indexes of the six body part partitions; And evaluating by using the pressure index, the overall pressure index and the body quality index of the six body part partitions. Further, the acquiring the human body pressure data and the preset anthropometric statistical model includes: Acquiring a pressure sensor signal; converting the sensor signals into physical data with an X axis as a horizontal axis and a Y axis as a vertical axis; accumulating and summing the physical data in the longitudinal direction of the Y axis to obtain the total pressure value of each longitudinal direction; And drawing a one-dimensional longitudinal pressure distribution curve according to the total pressure value of each longitudinal direction. Further, the acquiring the human body pressure data and the preset anthropometric statistical model further includes: Traversing the pressure sensor signal, and filtering noise points smaller than a preset pressure threshold value; and correcting the isolated bad point data in the pressure sensor signal. Further, the determining the head position of the human body according to the human body pressure data through the core interval data index includes: Acquiring a first local minimum value in a preset first interval range according to the one-dimensional longitudinal pressure distribution curve; and marking the position corresponding to the first local minimum as a head position point. Further, the dividing the human body part into six areas according to the sex, the height and the weight of the tested person, the preset anthropometric statistical model and the head position to obtain six human body part areas, including: Acquiring corresponding preset shoulder length data, preset back length data, preset waist length data, preset hip length data and preset leg length data in an anthropometric statistical model according to the gender, the height and the weight of the tested person; Adding the head position points to the preset shoulder length data to obtain shoulder position points; adding the shoulder position points