Mattress spring deformation detection method and device based on image recognition

Abstract

The present invention discloses a method and device for detecting spring deformation for a mattress based on image recognition, and relates to the field of image recognition technology. The method comprises: reading a predetermined physical sign index to obtain target physical sign index information; dividing the target mattress spring into blocks, extracting the first block from multiple blocks, obtaining a first evaluation result, obtaining a first spring deformation graph sequence, comparing it with a predetermined spring image to obtain a first comparison result, obtaining a first deformation value, generating a first deformation detection curve, and judging whether the first supporting force meets the required threshold; if it meets, the deformation detection passes, and if it does not meet, the detection fails. The present invention solves the technical problem in the prior art that the accuracy of mattress spring deformation detection is low and cannot meet personalized needs, and achieves the technical effect of improving the accuracy of mattress spring deformation detection, meeting personalized needs, and improving the comfort and support performance of the mattress.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and in particular to a method and device for detecting spring deformation of a mattress based on image recognition. Background Art

[0002] As people's living standards improve and they pay more attention to sleep quality, the comfort and support performance of mattresses have become crucial. One of the core components of a mattress is the spring, whose deformation characteristics directly affect the support effect of the mattress and the user's sleeping experience. Traditional mattress spring testing methods often rely on manual observation and simple physical testing, which is not only inefficient, but also difficult to guarantee accuracy and reliability. Since it is impossible to accurately quantify and analyze the deformation of the spring in multiple dimensions, it is difficult to meet the increasingly diverse and personalized needs of consumers.

[0003] The existing technology has the technical problem of low accuracy in detecting the deformation of mattress springs and being unable to meet personalized needs. Summary of the invention

[0004] The present application provides a mattress spring deformation detection method and device based on image recognition, which is used to solve the technical problem that the existing technology has low accuracy in mattress spring deformation detection and cannot meet personalized needs.

[0005] In view of the above problems, the present application provides a method and device for detecting spring deformation of a mattress based on image recognition.

[0006] In a first aspect of the present application, a mattress spring deformation detection method based on image recognition is provided, the method comprising:

[0007] Reading a predetermined physical sign index, and performing multi-dimensional feature collection on a target user based on the predetermined physical sign index to obtain target physical sign index information; dividing a target mattress spring into blocks to obtain a target division result, wherein the target division result includes a plurality of blocks; extracting a first block from the plurality of blocks, and evaluating the first block in combination with the target physical sign index information to obtain a first evaluation result, wherein the first evaluation result includes a first pressure evaluation threshold and a first support force requirement threshold; obtaining a first spring deformation image sequence, wherein the first spring deformation image sequence refers to a spring deformation image sequence dynamically collected by an image acquisition device when performing deformation detection on the first spring based on the first pressure evaluation threshold, wherein the first spring refers to a spring at a position corresponding to the first block. spring; comparing the first spring image in the first spring deformation graph sequence with the predetermined spring image to obtain a first comparison result, and analyzing the first comparison result to obtain a first deformation value, wherein the first spring image is the spring deformation image under the first pressure in the first pressure assessment threshold, and the first deformation value and the first pressure have a first corresponding relationship; generating a first deformation detection curve of the first spring according to the first corresponding relationship between the first deformation value and the first pressure; analyzing the first deformation detection curve to obtain a first supporting force of the first spring, and judging whether the first supporting force meets the first supporting force requirement threshold; if so, the deformation detection of the first spring passes, and if not, the deformation detection of the first spring fails.

[0008] The second aspect of the present application provides a spring deformation detection device for a mattress based on image recognition, the device comprising:

[0009] A target vital sign indicator information acquisition module, the target vital sign indicator information acquisition module is used to read predetermined vital sign indicators, and based on the predetermined vital sign indicators, perform multi-dimensional feature collection on the target user to obtain target vital sign indicator information; a target segmentation result acquisition module, the target segmentation result acquisition module is used to perform block segmentation on the target mattress spring to obtain a target segmentation result, and the target segmentation result includes multiple blocks; a first block extraction module, the first block extraction module is used to extract a first block from the multiple blocks, and evaluate the first block in combination with the target vital sign indicator information to obtain a first evaluation result, and the first evaluation result includes a first pressure evaluation threshold and a first support force requirement threshold; a first spring deformation image sequence acquisition module, the first spring deformation image sequence acquisition module is used to acquire a first spring deformation image sequence, the first spring deformation image sequence refers to a spring deformation image sequence dynamically collected by an image acquisition device when a deformation detection is performed on the first spring based on the first pressure evaluation threshold, and the first spring refers to a corresponding a spring at a position in the first block; a first comparison result acquisition module, the first comparison result acquisition module is used to compare the first spring image in the first spring deformation sequence with the predetermined spring image to obtain a first comparison result, and analyze the first comparison result to obtain a first deformation value, wherein the first spring image is the spring deformation image under the first pressure in the first pressure evaluation threshold, and the first deformation value and the first pressure have a first corresponding relationship; a first deformation detection curve generation module, the first deformation detection curve generation module is used to generate a first deformation detection curve of the first spring according to the first corresponding relationship between the first deformation value and the first pressure; a first support force judgment module, the first support force judgment module is used to analyze the first deformation detection curve to obtain the first support force of the first spring, and judge whether the first support force meets the first support force requirement threshold; if so, the deformation detection of the first spring passes, and if not, the deformation detection of the first spring fails.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] Read the predetermined physical sign index to obtain the target physical sign index information; divide the target mattress spring into blocks to obtain the target division result, and the target division result includes multiple blocks; extract the first block from the multiple blocks, and evaluate the first block in combination with the target physical sign index information to obtain the first evaluation result; obtain the first spring deformation graph sequence, and the first spring refers to the spring at the position corresponding to the first block; compare the first spring image in the first spring deformation graph sequence with the predetermined spring image to obtain the first comparison result, and analyze the first comparison result to obtain the first deformation value, and generate the first deformation detection curve of the first spring according to the first corresponding relationship between the first deformation value and the first pressure; analyze the first deformation detection curve to obtain the first support force of the first spring, and judge whether the first support force meets the first support force requirement threshold; if it meets, the deformation detection of the first spring passes, and if it does not meet, the deformation detection of the first spring fails. The technical effect of improving the accuracy of mattress spring deformation detection, meeting personalized needs, and improving the comfort and support performance of the mattress is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 A schematic diagram of a mattress spring deformation detection method based on image recognition provided in an embodiment of the present application;

[0014] Figure 2 A schematic structural diagram of a mattress spring deformation detection device based on image recognition provided in an embodiment of the present application.

[0015] Explanation of the accompanying drawings: target vital sign index information acquisition module 10, target division result acquisition module 20, first block extraction module 30, first spring deformation pattern sequence acquisition module 40, first comparison result acquisition module 50, first deformation detection curve generation module 60, first support force judgment module 70, first spring deformation detection module 80. DETAILED DESCRIPTION

[0016] The present application provides a mattress spring deformation detection method and device based on image recognition, which is used to solve the technical problem that the mattress spring deformation detection accuracy in the prior art is low and cannot meet personalized needs.

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] Embodiment 1

[0019] like Figure 1 As shown, the present application provides a mattress spring deformation detection method based on image recognition, the method comprising:

[0020] Step S100: Reading predetermined physical sign indicators, and collecting multi-dimensional features of a target user based on the predetermined physical sign indicators to obtain target physical sign indicator information.

[0021] Specifically, first obtain pre-set physical sign indicators, which are determined according to the general standards of mattress design or specific user needs. Then, collect characteristics of the target users from multiple different dimensions, including but not limited to obtaining the user's weight data using professional measuring equipment, obtaining height information through height measuring instruments, obtaining sleeping habits (such as side-lying frequency, number of turning over, etc.) through questionnaires or user behavior monitoring, and obtaining the pressure distribution characteristics of users when lying down through pressure sensors. The data collected from these multiple dimensions will be comprehensively analyzed and processed to finally obtain comprehensive, detailed and accurate target physical sign indicator information about the target users. This information will provide important basis and reference for the subsequent evaluation and testing of mattress springs, provide users with mattress spring designs and evaluations that are more in line with their individual needs, and realize personalized product customization.

[0022] Step S200: dividing the target mattress spring into blocks to obtain a target division result, wherein the target division result includes a plurality of blocks.

[0023] Specifically, the springs of the target mattress are divided into blocks. This division is based on certain rules and considerations, including the size of the mattress, the layout of the springs, and the differences in the support requirements of different parts of the human body during sleep. By dividing the mattress springs into multiple blocks, each block can be regarded as a relatively independent unit for subsequent evaluation and analysis, which can enable a more detailed and targeted evaluation of the spring performance in different areas of the mattress. When a problem with the mattress springs is detected, the specific block can be more accurately located, thereby improving the overall comfort and support performance of the mattress.

[0024] Step S300: extracting a first block from the multiple blocks, and evaluating the first block in combination with the target vital sign index information to obtain a first evaluation result, wherein the first evaluation result includes a first pressure evaluation threshold and a first support force requirement threshold.

[0025] Specifically, the first block is selected from the multiple blocks divided previously for key processing, and the first block is evaluated in combination with the target user's physical sign indicator information collected previously. This evaluation is carried out by comprehensively considering the user's physical characteristics (such as weight, height, sleeping habits, etc.) and the characteristics of the block. The first evaluation result obtained by the evaluation includes a first pressure evaluation threshold and a first support force requirement threshold. For the first block, the first pressure evaluation threshold obtained by the evaluation is determined according to the reasonable pressure range that the body part corresponding to the block can withstand. For example, if the first block corresponds to the shoulder position, since the shoulder is relatively sensitive and bears less pressure, its pressure evaluation threshold may be lower; and if it corresponds to the hip position, since the hip can withstand greater pressure, the pressure evaluation threshold is higher. The first support force requirement threshold is also determined according to the body part corresponding to the block. For example, the waist, as an important supporting part of the body, usually has a greater demand for support, so the support force requirement threshold of the corresponding block will be higher; while the calf part has a relatively small demand for support, and the support force requirement threshold of the corresponding block is relatively low. This detailed and personalized evaluation can provide precise guidance for the subsequent design, adjustment and testing of the mattress springs to ensure that the mattress can provide users with the best sleeping support and comfort.

[0026] Step S400: Obtain a first spring deformation image sequence, where the first spring deformation image sequence refers to a spring deformation image sequence dynamically acquired by an image acquisition device when deformation detection of the first spring is performed based on the first pressure assessment threshold, and the first spring refers to a spring at a position corresponding to the first block.

[0027] Specifically, it is clear that the object to be detected is the first spring located at the first block position. This first block is a specific area determined when the mattress spring is divided into blocks. The deformation detection of the first spring is carried out according to the first pressure evaluation threshold obtained in the previous step. This threshold specifies the pressure range and change law applied to the first spring. During the detection process, an image acquisition device, such as a high-speed high-definition camera, is started. The device will continuously shoot the state of the first spring under pressure at an extremely high frame rate and accuracy. As the pressure is gradually applied to the first spring in a predetermined manner, the image acquisition device will capture the instantaneous shape of the spring under different pressure values ​​in sequence. These spring deformation images obtained in chronological order constitute an orderly image sequence, namely the first spring deformation image sequence. This image sequence records in detail the entire process of the first spring from the initial state to the gradual deformation under different pressures. Each image reflects the specific shape and characteristic changes of the spring at a specific pressure moment. By obtaining such a complete and detailed first spring deformation image sequence, an intuitive basis is provided for the design improvement of the mattress spring, thereby continuously improving the quality and comfort of the mattress.

[0028] Step S500: Compare the first spring image in the first spring deformation sequence with the predetermined spring image to obtain a first comparison result, and analyze the first comparison result to obtain a first deformation value, wherein the first spring image is a spring deformation image under the first pressure in the first pressure assessment threshold, and the first deformation value has a first corresponding relationship with the first pressure.

[0029] Specifically, the specific spring deformation image under the first pressure in the first pressure evaluation threshold is extracted from the first spring deformation image sequence, which is called the first spring image. At the same time, a predetermined spring image is prepared, which is an image of the spring state when it is not under pressure. Then, the first spring image and the predetermined spring image are compared one by one in terms of details, for example, the differences in their spring bending degree, stretching length, shape change, etc. are compared. Through this detailed comparison operation, a first comparison result is obtained. Based on the degree of deformation difference reflected by the image comparison, a specific value is obtained according to the spring deformation evaluation function, and this value is the first deformation value. It should be noted that there is a clear and specific correspondence between this first deformation value and the first pressure applied. This correspondence plays a very important role in evaluating the performance of the spring, judging whether it meets the design requirements, and in the overall quality control of the mattress. Through comparison and analysis, the deformation of the first spring under a specific pressure can be accurately quantified, and the spring performance can be accurately evaluated.

[0030] Step S600: generating a first deformation detection curve of the first spring according to the first corresponding relationship between the first deformation value and the first pressure.

[0031] Specifically, the deformation detection curve of the first spring is generated by using the first corresponding relationship between the first deformation value and the first pressure determined previously. A plurality of different first pressure values ​​and their corresponding first deformation values ​​are used as data points, with the pressure value as the horizontal coordinate and the deformation value as the vertical coordinate. These points are plotted in the coordinate system, and these points are connected by appropriate mathematical methods (such as fitting curves) to form a continuous curve. This curve intuitively shows the changing trend and law of the deformation of the first spring under different pressures. The curve can be used to predict the possible deformation of the spring under other pressure values ​​and predict the spring performance.

[0032] Step S700: Analyze the first deformation detection curve to obtain a first supporting force of the first spring, and determine whether the first supporting force meets the first supporting force requirement threshold.

[0033] Specifically, the first deformation detection curve is analyzed in detail. By observing the trend of the curve, the change in slope, the degree of curvature and other characteristics, and combining the material properties, geometric dimensions and other parameters of the spring, a large number of experiments are carried out to measure the support force of the spring under different deformations, and then a relationship model between deformation and support force is established through mathematical methods (such as linear regression, polynomial fitting, etc.), and the support force is predicted by the model according to the new deformation data, and the support force that the first spring can generate under specific conditions is calculated, which is defined as the first support force. Next, the calculated first support force is compared with the first support force requirement threshold determined in advance. This first support force requirement threshold is a reasonable range set by comprehensively considering multiple factors such as the user's physical condition, sleeping habits, and mattress design requirements. It helps to screen out springs that do not meet the support force threshold, thereby improving the overall quality and stability of the mattress and providing users with reliable sleep support.

[0034] Step S800: If it is in compliance, the deformation test of the first spring passes; if it is not in compliance, the deformation test of the first spring fails.

[0035] Specifically, if the first support force meets the first support force requirement threshold, it means that the first spring has reached the expected standards and requirements in terms of deformation and support force, and the deformation test of the first spring is determined to have passed. On the contrary, if the first support force does not meet the first support force requirement threshold, that is, it does not fall within the specified range, then it means that the performance of the first spring does not meet the design and use standards, and the deformation test of the first spring is determined to have failed, which effectively screens out qualified and unqualified springs, helps to find problems in a timely manner, and improves the consistency and stability of products.

[0036] In a possible implementation, step S100 further includes:

[0037] Step S110: The predetermined physical sign indicators at least include height, weight and body shape.

[0038] Specifically, the predetermined physical signs mentioned cover at least three important aspects: height, weight, and body shape. The height factor affects the size of the mattress and the stretching space of the human body during sleep. Taller people may need a longer mattress to ensure that all parts of the body are fully supported. Weight is directly related to the amount of pressure on the mattress. Heavier people will exert greater pressure on the mattress, so the support structure of the mattress needs to be able to withstand and provide corresponding support. Body shape comprehensively considers the width, fatness, and proportions of various parts of the human body. Different body shapes will have different distributions of support requirements for mattresses during sleep. For example, fatter people may need a wider support area, or specific parts (such as shoulders and hips) need stronger support. Comprehensive consideration of these predetermined physical signs can provide important basic data for the subsequent design, evaluation, and testing of mattress springs, so as to achieve personalized customization and optimized design of mattresses and improve the comfort and applicability of mattresses.

[0039] In a possible implementation, step S300 further includes:

[0040] Step S310: Matching the first body part corresponding to the first block.

[0041] Step S320: Estimate the gravity of the first body part in combination with the target physical sign indicator information, and determine a first gravity threshold.

[0042] Step S330: reading a first predetermined correlation coefficient of the first body part, and weighting the first gravity threshold to obtain the first pressure assessment threshold.

[0043] Specifically, the first block into which the mattress is divided corresponds to a specific part of the human body. This requires a pre-set mapping rule or model between mattress blocks and body parts. For example, according to the layout of the mattress and the distribution of common contact parts of the human body during sleep, the block at the front of the mattress is matched to the area corresponding to the feet, and the block in the middle and upper part is matched to the area corresponding to the waist, etc. Through this matching, the position of the first block in the human body structure can be clarified, providing an accurate basis for subsequent gravity estimation, pressure assessment and other operations for this part. This helps to achieve accurate evaluation and customized design of the performance of different areas of the mattress to better meet the support and comfort needs of different parts of the human body.

[0044] First, the target physical sign indicator information is obtained, including but not limited to height, weight and body shape. Then, the principles of physics and ergonomics are used to estimate the gravity generated by the first body part that matches the first block. For example, people with heavier weight usually generate greater gravity on the same body part; people with larger body shape will also have different distribution areas and weight distribution of body parts, which will affect the estimation of gravity. After estimating the gravity, a reasonable range is further determined, namely the first gravity threshold. This threshold is not a fixed value, but an interval. When determining the first gravity threshold, many factors are considered, such as the use scenario of the mattress (whether it is for home use, hotel use or medical use, etc.), the special needs of the user (such as preference for harder or softer mattresses), and the design standards and safety margins of the mattress. By determining the first gravity threshold, an important reference basis can be provided for the subsequent evaluation of the support capacity and comfort of the mattress for the body part.

[0045] A first predetermined correlation coefficient specifically for a first body part is read from a preset database or parameter set. This coefficient is associated with the contact area between the body part and the mattress. Since the contact area between different body parts and the mattress varies, for example, the buttocks usually have a larger contact area, while the shoulders are relatively smaller, the size of the contact area will affect the distribution and size of the pressure. The read correlation coefficient is used to perform a weighted operation on the first gravity threshold obtained previously. The weighted method is to multiply the first gravity threshold by the correlation coefficient. Through this weighted processing, the influence of the contact area between the body part and the mattress on the pressure is fully considered, thereby obtaining a more accurate and actual first pressure assessment threshold.

[0046] In a possible implementation, step S300 further includes:

[0047] Step S340: reading a first predetermined support coefficient of the first body part, and weighting the first pressure assessment threshold to obtain the first support force requirement threshold.

[0048] Specifically, a first predetermined support coefficient corresponding to a first body part is read from a pre-set and stored database or parameter set. This coefficient is obtained through in-depth scientific research and evaluation, and is intended to accurately quantify the support strength required for different body parts. Due to their physiological structure, the force applied during sleep, and different requirements for comfort, different body parts require significantly different support strengths. For example, the waist, as an important load-bearing part of the human body, usually requires a greater support strength to maintain the normal physiological curvature of the spine and reduce the burden on the waist muscles; while the legs may require less support strength, relatively speaking. Next, a weighted operation is performed. This is achieved through a specific weighted formula. Assuming that the first pressure assessment threshold is P and the first predetermined support coefficient is C, the weighted formula is: the first support force requirement threshold = C × P. Through this weighted calculation, the support requirement of a specific body part represented by the first predetermined support coefficient is combined with the first pressure assessment threshold, thereby obtaining a first support force requirement threshold that can accurately reflect the actual support force requirement of the first body part. The determination of this threshold provides a critical and precise quantitative standard for the design, production and quality assessment of mattresses, helping to ensure that the mattress can provide appropriate support for each part of the body and improve the user's sleep quality and comfort.

[0049] In a possible implementation, step S500 further includes:

[0050] Step S510: Reading a predetermined spring deformation characteristic.

[0051] Step S520: using the predetermined deformation feature parameters of the predetermined spring image extracted based on the predetermined spring deformation feature as a comparison reference.

[0052] Step S530: comparing the first deformation feature parameter of the first spring image extracted based on the predetermined spring deformation feature with the comparison reference to obtain a first parameter comparison deviation, and using the first parameter comparison deviation as the first comparison result.

[0053] Step S540: introducing a spring deformation evaluation function to evaluate and analyze the first comparison result to obtain the first deformation value.

[0054] Specifically, information about predetermined spring deformation characteristics is obtained from a pre-set storage location (such as a specific database, configuration file or register). These predetermined spring deformation characteristics are the deformation characteristics that ideal or standard springs that have been carefully studied and determined should exhibit when subjected to specific external forces. These characteristics may include, but are not limited to, the following aspects: the amplitude of deformation, such as the length range of compression or extension of the spring when subjected to a certain pressure. The speed of deformation, how fast the spring deforms during the force process. The mode of deformation, such as whether the spring is uniformly compressed or the local deformation is more obvious. The recovery characteristics, the rebound speed and degree of recovery of the spring after the external force is removed. By reading these predetermined deformation characteristics, a clear standard and reference basis are provided for subsequent comparison and evaluation with the actual spring deformation.

[0055] Based on the read predetermined spring deformation characteristics, specific parameters related to deformation are extracted from the image of the predetermined spring by using image recognition, data analysis and other technical means. These parameters accurately reflect the deformation characteristics of the predetermined spring under specific conditions. Then, the extracted predetermined deformation characteristic parameters are set as the benchmark for subsequent comparison, and the predetermined deformation characteristic parameters in the predetermined spring image are extracted and set as the benchmark for comparison. This benchmark represents the spring deformation state that is considered to be ideal or normal.

[0056] Similarly, according to the method and standard adopted by the predetermined spring deformation characteristics, the image of the first spring is processed and analyzed to extract the first deformation characteristic parameters. Subsequently, these first deformation characteristic parameters are compared one by one or comprehensively with the previously set comparison benchmark (i.e., the predetermined deformation characteristic parameters of the predetermined spring image). This comparison involves the calculation of numerical differences, the evaluation of the matching degree of features, etc. Through the comparison, the difference between the deformation characteristic parameters of the first spring and the benchmark is obtained, and this difference is the first parameter comparison deviation. Finally, this first parameter comparison deviation is regarded as the first comparison result. This result intuitively reflects the degree of deviation of the deformation of the first spring from the predetermined standard, and provides a key basis for subsequent evaluation and decision-making.

[0057] A specially constructed spring deformation evaluation function is introduced. This function is designed based on the relevant theories, experimental data and engineering experience of spring deformation. The first comparison result (i.e., the first parameter comparison deviation) obtained in the previous step is passed as input to this evaluation function. The evaluation function will perform complex calculations and analyses on the input comparison results. It will comprehensively consider multiple factors, such as the size and direction of the deviation, the material properties of the spring, the use environment, etc. Through the processing of the function, a quantitative value is finally output, i.e., the first deformation value, which can more comprehensively and accurately reflect the deformation degree and nature of the first spring.

[0058] In a possible implementation, step S510 further includes:

[0059] Step S511: The predetermined spring deformation characteristics at least include spring length, spring curvature and spring spacing.

[0060] Specifically, the predetermined spring deformation characteristics mentioned in this step at least cover three important aspects: spring length, spring curvature and spring spacing. Spring length refers to the overall length change of the spring under unstressed and stressed conditions. This feature directly reflects the expansion and contraction of the spring when subjected to external force, which is crucial for evaluating the elastic properties and load-bearing capacity of the spring. Spring curvature describes whether the spring has abnormal bending or twisting during the force process. Under normal circumstances, the spring should expand and contract evenly along its axial direction. If there is obvious bending, it may mean that there is a problem with the structural stability of the spring or the force distribution is uneven. The spring spacing focuses on the change in the distance between adjacent springs. In some structures with multiple spring combinations, the consistency and change law of the spring spacing will affect the support effect and stability of the overall structure. Comprehensively considering the predetermined spring deformation characteristics of these three aspects can more comprehensively and accurately evaluate the deformation performance and quality of the spring, providing an important basis for subsequent analysis and judgment.

[0061] In a possible implementation, step S540 further includes:

[0062] Step S541:

[0063] Wherein, de(x) refers to the first deformation value of the spring in the first spring image x, x n It refers to the nth parameter comparison deviation in the first comparison result, and the first comparison result includes N parameter comparison deviations in total, N is an integer greater than or equal to 3, ε(x n ) refers to the weight coefficient of the nth parameter comparison deviation, cor(x n ) refers to the correlation coefficient between the nth parameter contrast deviation and the first deformation value de(x), α and β are the first deformation coefficient and the second deformation coefficient respectively, and α+β=1.

[0064] Specifically, this formula is a weighted summation of multiple parameter comparison deviations and adjustment combined with two deformation coefficients to obtain the final first deformation value de(x), which refers to the first deformation value of the spring in the first spring image, x n It refers to the nth parameter comparison deviation in the first comparison result, and the first comparison result includes N parameter comparison deviations in total, where N is an integer greater than or equal to 3. n) refers to the weight coefficient of the nth parameter comparison deviation, and different parameter comparison deviations are given different weights due to their different importance. n ) refers to the correlation coefficient between the nth parameter comparison deviation and the first deformation value de(x), reflecting the influence of each parameter comparison deviation on the final deformation value. α and β are the first deformation coefficient and the second deformation coefficient respectively, and α+β=1, which is a constant determined based on experience, experimental data or theoretical analysis, and is used to further adjust and optimize the calculation results. Through this formula, multiple parameter comparison deviations and their weights, correlations and other factors can be comprehensively considered to more accurately calculate the first deformation value, thereby making a comprehensive and accurate assessment of the deformation of the first spring.

[0065] In a possible implementation, step S600 further includes:

[0066] Step S610: generating a first deformation detection scatter plot according to the first corresponding relationship.

[0067] Step S620: obtaining a first sampling scatter point set in the first deformation detection scatter point map based on the random sampling consistency principle.

[0068] Step S630: performing fitting analysis on the first sampling scattered point set to obtain a first curve.

[0069] Step S640: obtaining a first spatial distance from a first scatter point to the first curve, wherein the first scatter point refers to any scatter point in the first deformation detection scatter point graph after excluding the first sampling scatter point set.

[0070] Step S650: When the first spatial distance meets the spatial distance limit, the first scattered point is recorded as a first inner point; when the first spatial distance does not meet the spatial distance limit, the first scattered point is recorded as a first outer point.

[0071] Step S660: If the ratio of the number of the first inner points to the number of the first outer points reaches a ratio limit, the first curve is used as the first deformation detection curve.

[0072] Specifically, the first correspondence refers to the correlation between certain characteristics of the spring (such as force magnitude, material properties, etc.) and the degree of deformation that have been determined. By processing and converting the data involved in these correlations, each data pair (such as the deformation value under a specific force) is represented as a point on a two-dimensional plane. The collection of all these points constitutes the first deformation detection scatter plot. In this scatter plot, the horizontal axis and the vertical axis respectively represent the two variables involved in the first correspondence (for example, the horizontal axis can be the force magnitude, and the vertical axis can be the degree of deformation). By observing the distribution of the scatter points, the general trend and relationship between these data can be understood, providing an intuitive basis for subsequent analysis and processing.

[0073] The Random Sampling Consensus (RANSAC) principle is a method for estimating model parameters from data containing outliers. Based on this principle, the first sampling scatter point set is obtained from the first deformation detection scatter plot, and some scatter points are randomly selected as the initial sample set. Then, a model is fitted based on these sample points. The distance from all scatter points to the fitted model is calculated. If the distance is less than a preset threshold, the scatter point is considered to be an inlier, that is, it is consistent with the fitted model; otherwise, it is an outlier. By repeating the above random selection and fitting process many times, the sample set containing the most inliers is finally selected as the first sampling scatter point set. The first sampling scatter point set obtained in this way is more likely to represent the main trend of the data, while reducing the influence of outliers or outliers, laying the foundation for more accurate fitting and analysis later.

[0074] The first sampling scattered point set obtained by the principle of random sampling consistency is fitted and analyzed. The fitting process is to find a curve that can best describe the distribution trend of these scattered points. Common fitting methods include linear fitting, polynomial fitting, exponential fitting, etc. The specific method to be selected depends on the characteristics of the data and the needs of the analysis. For example, if the data presents an approximately linear relationship, linear fitting will be selected to determine the equation of the line by calculating the slope and intercept; if the data changes are more complex, polynomial fitting may be used to determine the coefficients of the polynomial. When fitting the first sampling scattered point set, the error between the fitting curve and the scattered points will be calculated. By continuously adjusting the fitting parameters, the error is minimized, thereby obtaining the first curve that can better reflect the distribution law of the scattered point set. The first curve obtained can be used to further analyze the deformation characteristics of the spring, predict unknown data, or compare with other curves.

[0075] In this step, we must first clarify the definition of the first scatter point, which is any point remaining in the first deformation detection scatter plot after removing the first sampling scatter point set obtained by the random sampling consistency principle. Then, we need to calculate the first spatial distance from this first scatter point to the first curve. The calculation of this distance is determined according to the specific coordinate system and curve equation.

[0076] According to the first spatial distance from the first scatter point to the first curve that has been calculated, there is a pre-set spatial distance limit, which is a standard for judging whether the proximity between the first scatter point and the first curve is within an acceptable range. If the first spatial distance from the first scatter point to the first curve is less than or equal to the spatial distance limit, the first scatter point is considered to be close enough to the first curve and is marked as the first inner point, which means that the scatter point is considered to conform to the law or pattern represented by the curve. On the contrary, if the first spatial distance is greater than the spatial distance limit, it means that the first scatter point deviates greatly from the first curve and exceeds the acceptable range, and it is marked as the first outer point. Through such marking, the scatter points can be divided into two categories: inner points and outer points, which is helpful for further analysis and processing of the data, such as judging the fitting effect of the curve, screening out abnormal data, etc., to improve the accuracy of deformation detection.

[0077] The number of the first inner point and the first outer point marked above is counted and compared. The ratio of the number of the first inner point to the first outer point is calculated. At the same time, there is a pre-set ratio limit, which is determined based on the requirements for detection accuracy and reliability. If the ratio of the number of the first inner point to the first outer point reaches or exceeds this ratio limit, it means that the first curve obtained by fitting can well represent the distribution law of most data points and has a high goodness of fit and reliability. In this case, this first curve is determined as the first deformation detection curve. This means that it is recognized as a mathematical model that can effectively describe the deformation of the spring and can be used for subsequent deformation analysis, prediction or quality assessment. On the contrary, if the ratio of the number does not reach the ratio limit, it is necessary to re-sample, fit or adjust the relevant parameters to obtain a curve that better meets the requirements.

[0078] In a possible implementation, step S660 further includes:

[0079] If the ratio of the number of the first inner points to the first outer points does not reach the ratio limit, a second sampling scatter point set in the first deformation detection scatter point map is obtained based on the principle of random sampling consistency, and an iterative fitting test is performed on the second sampling scatter point set.

[0080] Specifically, if the ratio of the number of the first inner point to the first outer point does not reach the preset ratio limit, this means that the current fitting result is not ideal and cannot represent the distribution law of the data well. At this time, based on the principle of random sampling consistency, a new set of sampling scatter points is obtained from the first deformation detection scatter plot again, which is called the second sampling scatter point set. Then, an iterative fitting test is performed on this second sampling scatter point set. The process of iterative fitting test includes fitting the second sampling scatter point set using a similar fitting method as before to obtain a new curve, calculating the new inner points and outer points again, and comparing the ratio of their numbers with the ratio limit. This process will be repeated until the ratio of the number of inner points to outer points reaches the ratio limit, or after a certain number of iterations, it is still not reached but the best result is selected as the final deformation detection curve. Through such an iterative process, it is intended to continuously optimize the fitting results to obtain a more accurate detection curve that better reflects the deformation law of the spring.

[0081] Embodiment 2

[0082] Based on the same inventive concept as the mattress spring deformation detection method based on image recognition in the aforementioned embodiment, Figure 2 As shown, the present application provides a spring deformation detection device for a mattress based on image recognition, and the system and method embodiments in the present application are based on the same inventive concept. The device includes:

[0083] The target vital sign indicator information acquisition module 10 is used to read the predetermined vital sign indicators and perform multi-dimensional feature collection on the target user based on the predetermined vital sign indicators to obtain the target vital sign indicator information.

[0084] The target division result acquisition module 20 is used to divide the target mattress spring into blocks to obtain a target division result, and the target division result includes a plurality of blocks.

[0085] The first block extraction module 30 is used to extract the first block from the multiple blocks, and evaluate the first block in combination with the target vital sign indicator information to obtain a first evaluation result, wherein the first evaluation result includes a first pressure evaluation threshold and a first support force requirement threshold.

[0086] The first spring deformation pattern sequence acquisition module 40 is used to acquire a first spring deformation pattern sequence, wherein the first spring deformation pattern sequence refers to a spring deformation image sequence dynamically acquired by an image acquisition device when deformation detection of the first spring is performed based on the first pressure assessment threshold, and the first spring refers to a spring at a position corresponding to the first block.

[0087] The first comparison result acquisition module 50 is used to compare the first spring image in the first spring deformation sequence with the predetermined spring image to obtain a first comparison result, and analyze the first comparison result to obtain a first deformation value, wherein the first spring image is a spring deformation image under the first pressure in the first pressure assessment threshold, and the first deformation value has a first corresponding relationship with the first pressure.

[0088] The first deformation detection curve generating module 60 is configured to generate a first deformation detection curve of the first spring according to the first corresponding relationship between the first deformation value and the first pressure.

[0089] The first supporting force judging module 70 is used for analyzing the first deformation detection curve to obtain the first supporting force of the first spring, and judging whether the first supporting force meets the first supporting force requirement threshold.

[0090] The first spring deformation detection module 80 is used to determine if the deformation of the first spring passes, or if the deformation of the first spring does not pass, the deformation of the first spring fails.

[0091] Furthermore, the target vital sign index information acquisition module 10 also includes:

[0092] The predetermined physical sign indicators at least include height, weight and body shape.

[0093] Furthermore, the first block extraction module 30 further includes:

[0094] A first body part matching unit is used to match a first body part corresponding to the first block.

[0095] A first gravity threshold determination unit is used to estimate the gravity of the first body part in combination with the target vital sign index information and determine a first gravity threshold.

[0096] A first pressure assessment threshold acquisition unit is used to read a first predetermined correlation coefficient of the first body part and weight the first gravity threshold to obtain the first pressure assessment threshold.

[0097] A first support force requirement threshold value acquisition unit is used to read a first predetermined support coefficient of the first body part and weight the first pressure assessment threshold value to obtain the first support force requirement threshold value.

[0098] Furthermore, the first comparison result acquisition module 50 further includes:

[0099] A spring deformation characteristic reading unit is used to read a predetermined spring deformation characteristic.

[0100] A comparison reference extraction unit is used to use the predetermined deformation feature parameters of the predetermined spring image extracted based on the predetermined spring deformation feature as a comparison reference.

[0101] A first comparison result acquisition unit is used to compare the first deformation feature parameter of the first spring image extracted based on the predetermined spring deformation feature with the comparison reference to obtain a first parameter comparison deviation, and use the first parameter comparison deviation as the first comparison result.

[0102] A first deformation value acquisition unit is used to introduce a spring deformation evaluation function to evaluate and analyze the first comparison result to obtain the first deformation value.

[0103] Wherein, the predetermined spring deformation characteristics at least include spring length, spring curvature and spring spacing.

[0104] Furthermore, the first deformation value obtaining unit further includes:

[0105]

[0106] Wherein, de(x) refers to the first deformation value of the spring in the first spring image x, x n It refers to the nth parameter comparison deviation in the first comparison result, and the first comparison result includes N parameter comparison deviations in total, N is an integer greater than or equal to 3, ε(x n ) refers to the weight coefficient of the nth parameter comparison deviation, cor(x n ) refers to the correlation coefficient between the nth parameter contrast deviation and the first deformation value de(x), α and β are the first deformation coefficient and the second deformation coefficient respectively, and α+β=1.

[0107] Furthermore, the first deformation detection curve generating module 60 further includes:

[0108] A first deformation detection scatter plot generating unit is configured to generate a first deformation detection scatter plot according to the first corresponding relationship.

[0109] A first sampling scatter point set acquisition unit is configured to acquire a first sampling scatter point set in the first deformation detection scatter point map based on a random sampling consistency principle.

[0110] A first curve acquisition unit is used to perform fitting analysis on the first sampling scattered point set to obtain a first curve.

[0111] A first spatial distance acquisition unit is used to acquire a first spatial distance from a first scatter point to the first curve, wherein the first scatter point refers to any scatter point in the first deformation detection scatter point graph after the first sampling scatter point set is eliminated.

[0112] The first scatter point judgment unit is used to record the first scatter point as a first inner point when the first spatial distance meets the spatial distance limit, and record the first scatter point as a first outer point when the first spatial distance does not meet the spatial distance limit.

[0113] A first deformation detection curve acquisition unit is configured to use the first curve as the first deformation detection curve if the ratio of the number of the first inner points to the first outer points reaches a ratio limit.

[0114] Furthermore, the first deformation detection curve generating module 60 further includes:

[0115] An iterative fitting verification unit is used to obtain a second sampling scatter point set in the first deformation detection scatter point diagram based on a random sampling consistency principle if the ratio of the number of the first inner points to the first outer points does not reach the ratio limit, and perform an iterative fitting test on the second sampling scatter point set.

[0116] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0117] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0118] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A mattress spring deformation detection method based on image recognition, characterized in that: include: Reading predetermined physical sign indicators, and performing multi-dimensional feature collection on a target user based on the predetermined physical sign indicators to obtain target physical sign indicator information; Performing block division on the target mattress spring to obtain a target division result, wherein the target division result includes a plurality of blocks; Extracting a first block from the multiple blocks, and evaluating the first block in combination with the target vital sign indicator information to obtain a first evaluation result, wherein the first evaluation result includes a first pressure evaluation threshold and a first support force requirement threshold; Acquire a first spring deformation image sequence, where the first spring deformation image sequence refers to a spring deformation image sequence dynamically acquired by an image acquisition device when deformation detection of the first spring is performed based on the first pressure assessment threshold, and the first spring refers to a spring at a position corresponding to the first block; Comparing a first spring image in the first spring deformation sequence with a predetermined spring image to obtain a first comparison result, and analyzing the first comparison result to obtain a first deformation value, wherein the first spring image is a spring deformation image under a first pressure in the first pressure assessment threshold, and the first deformation value has a first corresponding relationship with the first pressure; generating a first deformation detection curve of the first spring according to the first corresponding relationship between the first deformation value and the first pressure; Analyzing the first deformation detection curve to obtain a first supporting force of the first spring, and determining whether the first supporting force meets the first supporting force requirement threshold; If it is in compliance, the deformation test of the first spring passes; if it is not in compliance, the deformation test of the first spring fails.

2. The mattress spring deformation detection method based on image recognition according to claim 1 is characterized in that: The predetermined physical sign indicators at least include height, weight and body shape.

3. The mattress spring deformation detection method based on image recognition according to claim 2, characterized in that: include: matching a first body part corresponding to the first block; estimating the gravity of the first body part in combination with the target physical sign indicator information, and determining a first gravity threshold; A first predetermined correlation coefficient of the first body part is read, and the first gravity threshold is weighted to obtain the first pressure assessment threshold.

4. The mattress spring deformation detection method based on image recognition according to claim 3 is characterized in that: include: A first predetermined support coefficient of the first body part is read, and the first pressure assessment threshold is weighted to obtain the first support force requirement threshold.

5. The mattress spring deformation detection method based on image recognition according to claim 1, characterized in that: include: Reading predetermined spring deformation characteristics; using predetermined deformation feature parameters of the predetermined spring image extracted based on the predetermined spring deformation feature as a comparison reference; Comparing the first deformation feature parameter of the first spring image extracted based on the predetermined spring deformation feature with the comparison reference to obtain a first parameter comparison deviation, and using the first parameter comparison deviation as the first comparison result; A spring deformation evaluation function is introduced to evaluate and analyze the first comparison result to obtain the first deformation value.

6. The mattress spring deformation detection method based on image recognition according to claim 5, characterized in that: The predetermined spring deformation characteristics include at least spring length, spring curvature and spring spacing.

7. The mattress spring deformation detection method based on image recognition according to claim 5, characterized in that: The expression of the spring deformation evaluation function is as follows: Wherein, de(x) refers to the first deformation value of the spring in the first spring image x, x n It refers to the nth parameter comparison deviation in the first comparison result, and the first comparison result includes N parameter comparison deviations in total, N is an integer greater than or equal to 3, ε(x n ) refers to the weight coefficient of the nth parameter comparison deviation, cor(x n ) refers to the correlation coefficient between the nth parameter contrast deviation and the first deformation value de(x), α and β are the first deformation coefficient and the second deformation coefficient respectively, and α+β=1.

8. The mattress spring deformation detection method based on image recognition according to claim 1, characterized in that: include: generating a first deformation detection scatter plot according to the first corresponding relationship; Acquire a first sampling scatter point set in the first deformation detection scatter point graph based on a random sampling consistency principle; Performing fitting analysis on the first sampling scattered point set to obtain a first curve; Acquire a first spatial distance from a first scatter point to the first curve, wherein the first scatter point refers to any scatter point in the first deformation detection scatter point graph after excluding the first sampling scatter point set; When the first spatial distance meets the spatial distance limit, the first scattered point is recorded as a first inner point; when the first spatial distance does not meet the spatial distance limit, the first scattered point is recorded as a first outer point; If the ratio of the number of the first inner points to the number of the first outer points reaches a ratio limit, the first curve is used as the first deformation detection curve.

9. The mattress spring deformation detection method based on image recognition according to claim 8, characterized in that: If the ratio of the number of the first inner points to the first outer points does not reach the ratio limit, a second sampling scatter point set in the first deformation detection scatter point map is obtained based on the principle of random sampling consistency, and an iterative fitting test is performed on the second sampling scatter point set.

10. A mattress spring deformation detection device based on image recognition, characterized in that: The device comprises: A target physical sign indicator information acquisition module, which is used to read predetermined physical sign indicators and perform multi-dimensional feature collection on the target user based on the predetermined physical sign indicators to obtain target physical sign indicator information; A target division result acquisition module, wherein the target division result acquisition module is used to divide the target mattress spring into blocks to obtain a target division result, wherein the target division result includes a plurality of blocks; A first block extraction module, the first block extraction module is used to extract a first block from the multiple blocks, and evaluate the first block in combination with the target vital sign indicator information to obtain a first evaluation result, wherein the first evaluation result includes a first pressure evaluation threshold and a first support force requirement threshold; A first spring deformation image sequence acquisition module, the first spring deformation image sequence acquisition module is used to acquire a first spring deformation image sequence, the first spring deformation image sequence refers to a spring deformation image sequence dynamically acquired by an image acquisition device when deformation detection of the first spring is performed based on the first pressure assessment threshold, and the first spring refers to a spring at a position corresponding to the first block; a first comparison result acquisition module, the first comparison result acquisition module being used to compare the first spring image in the first spring deformation graph sequence with a predetermined spring image to obtain a first comparison result, and to analyze the first comparison result to obtain a first deformation value, wherein the first spring image is a spring deformation image under a first pressure in the first pressure assessment threshold, and the first deformation value has a first corresponding relationship with the first pressure; a first deformation detection curve generating module, the first deformation detection curve generating module being used to generate a first deformation detection curve of the first spring according to the first corresponding relationship between the first deformation value and the first pressure; a first supporting force judging module, the first supporting force judging module being used for analyzing the first deformation detection curve to obtain a first supporting force of the first spring, and judging whether the first supporting force meets the first supporting force requirement threshold; The first spring deformation detection module is used for passing the deformation detection of the first spring if the conditions are met, and failing the deformation detection of the first spring if the conditions are not met.

Images