CN109567317B - A Distributed Method of Barefoot Information Collecting Sensors - Google Patents

Abstract

The invention discloses a method for distributing barefoot information collecting sensors, comprising: S1: determining the number and distribution mode of the sole sensors; S2: determining the number and distribution mode of the foot sensors; the present application also provides a method based on the position of the sensors. Installation steps: The method of collecting barefoot information has been improved, so that the collected barefoot information is more comprehensive and complete, which can meet the quantitative collection of various footprint information, and lay a more solid foundation for barefoot 3D reconstruction.

Description

A Distributed Method of Barefoot Information Collecting Sensors

technical field

The invention relates to a method for distributing sensors, in particular to a method for distributing sensors for barefoot information collection.

Background technique

As we all know, wearable devices are regarded by the industry as a new industry growth point after smartphones and tablet computers. Now the representative smart wearable products on the market include smart bracelets, smart watches and smart glasses. These wearable devices are used in For the upper body of the human body, in order to improve the use of smart wearable devices, smart shoes are an effective supplement. At the same time, wearable plantar pressure testing technology is a new type of medical monitoring technology that has developed rapidly in recent years. It breaks through the limitation of the space scope of monitoring and diagnosis in medical institutions in the past, and can easily realize the daily monitoring of dynamic human plantar pressure. Foot medical care, sports posture correction and scientific shoemaking provide the basis; therefore, the data collection of plantar pressure is extremely important. In the prior art, there is only research on the plantar pressure sensor, but the research on the distribution of the plantar pressure sensor is still blank.

SUMMARY OF THE INVENTION

The present application provides a method for distributing barefoot information collection sensors, which can satisfy the quantitative collection of various footprint information.

A first technical solution of the present application is: a method for distributing barefoot information collection sensors, comprising:

S1: Determine the number and distribution of sole sensors;

S2: Determine the number and distribution of foot sensors.

Further, the following formula is used to determine the number of sole sensors:

N represents the number of sensors, S represents the average footprint area of an adult, in mm ² ; d is the average diameter of the smallest collection unit of the footprint, in mm.

Further, the distribution method of the sole sensor is as follows:

A. According to the distribution of acupoints, the soles of the feet are divided into regions;

B. Construct the optimal method of regional distribution with the objective function as the reference;

C. The sensors are distributed in each acupoint area by step-by-step calculation of the area skeleton of the binary map.

Further, the sole is divided into three types of areas: large acupoint area, small acupoint area and blank acupoint area, the number of sensors in the large acupoint area is x, the number of sensors in the small acupoint area is y, and the number of sensors in the blank acupoint area is z.

Further, the objective function includes the amount of information, the quantification degree of acupoint attention, and the uniformity of distribution; the solution method based on the amount of information is:

计算信息量；小穴位区域包括Q个穴位区域，某个区域的面积为S_Cj，面积最小S_Cmin，面积最大为S_Cmax，面积越小，信息量越大，依据公式

计算信息量，这里构建目标函数H：The amount of information per square millimeter in the blank acupoint area is h _A , h _A =v, and the overall area is S _A ; the large acupoint area includes P acupoint areas, the area of a certain area is S _Bi , the smallest area is S _Bmin , and the largest area is S _Bmax , the smaller the area, the greater the amount of information, according to the formula

Calculate the amount of information; the small acupoint area includes Q acupoint areas, the area of a certain area is S _Cj , the smallest area is S _Cmin , the largest area is S _Cmax , the smaller the area, the greater the amount of information, according to the formula

To calculate the amount of information, here the objective function H is constructed:

1≤i≤P

1≤j≤Q

另外，每个区域的传感器密度不得少于平均d个/平方毫米，即x,y_i,z_j＞d，以上两点认为是约束条件c₁，c₂，则结合目标函数，得到如下关系：When the number of sensors N is given,

In addition, the density of sensors in each area should not be less than the average d/mm2, that is, x, y _i , z _j > d, the above two points are considered as constraints c ₁ , c ₂ , then combined with the objective function, the following relationship is obtained :

Satisfy the c ₁ , c ₂ constraints, and there are infinitely many sets of solutions in this state. Here, the nearest neighbor method or the insertion method or the genetic algorithm or the neural network algorithm is used to obtain the optimal solution.

Further, each acupoint area distributes the sensors by gradually calculating the skeleton of the binary image area, as follows:

Assuming that the number of sensors in the acupoint area is K, to perform at least T sensor arrangements (the skeleton can be obtained by performing T corrosions), K/T sensors need to be distributed each time, and the specific steps are as follows:

i. Define the binary image of a certain acupoint area as I, and the corrosion count is initially 0;

ii. Extract edges for I, and evenly distribute K/T sensors on the edges of I;

iii. Corrode I and add 1 to the corrosion count;

iv. Determine whether the number of corrosion has reached T, if so, complete the sensor distribution;

If not, proceed to ii.

Further, the following formula is used to determine the number of foot sensors:

m=f(C,Δ)

Divide the bare feet into n circles, where Δ is the error of the polygon perimeter approaching the circumference, C is the perimeter of the circle, m is the minimum number of sides satisfying the error condition, L is the reference circumference of the metatarsal toe, L _i is the circumference of each circle, and M is the number of sensors.

Furthermore, the distribution mode of the foot surface sensors is a uniform distribution mode.

As a further step, there are two prototypes carried by the above sensor, one of which is the insole; the installation method of the sensor on the insole is as follows:

a. Construct an average foot model by acquiring a large number of barefoot footprints. The methods of constructing an average foot include: average of key points after size normalization, average of pressure footprint, etc.;

b. Coincide the foot bisector of the insole with the average foot bisector, and confirm that there is a 5mm interval between the back edge of the insole heel and the average heel back edge;

c. Areas other than the average foot follow:

i. Distribute sensors evenly in each area;

ii. Distributing sensors in such a way that the number of sensors on the average foot edge gradually decreases (morphological skeleton processing);

d. In the average foot, the sensors are distributed according to the number of sensors provided by different partitions and different optimization methods.

As a further, there are two kinds of prototypes that the above-mentioned sensors carry, and the other is shoes; the installation method of the sensor on the sole part of the shoe is to install the above-mentioned insole as the load-bearing installation method, and the installation method of the sensor on the upper is: constructing a coordinate system , where the z-axis is the ground normal direction, the y-axis is the centerline direction of the footsteps, and the x-axis is the direction perpendicular to the zy plane;

a. According to the upper of the fixed size, starting from the heel, measure the maximum perimeter L of the upper, and calculate the step length D of the sensor distribution through L/minimum number of sensors; the vertical distance between the intersection of the shoe center line and the heel is D As the reference point O ₁ , where the upper intersects the xy plane of O ₁ , a sensor is installed every distance D;

b. The vertical distance between the intersection of the shoe center line and the heel is 2D as the reference point O ₂ , the position where the vamp intersects the O ₂ xy plane, install a sensor every distance D, and so on, the vamp and O ₂ until there is no intersection in the xy plane.

The beneficial effects of the present invention are as follows: the present invention provides a sensor distribution calculation method for the collection of barefoot information and an installation step based on the position of the sensor, so that the method for collection of barefoot information is improved, so that the collected The barefoot information is more comprehensive and complete, which can meet the quantitative collection of various footprint information, and lay a more solid foundation for the barefoot 3D reconstruction.

Description of drawings

The present invention has 3 accompanying drawings:

Fig. 1 is a schematic diagram of acupoint division;

Fig. 2 is the schematic diagram that barefoot is divided into 4 circles in the embodiment;

FIG. 3 is a schematic diagram of the coordinate system constructed in the embodiment.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

Example 1

This embodiment provides a method for distributing sensors for barefoot information collection, and the sensor requirements are as follows:

1) Sensors include but are not limited to pressure-sensitive resistors, capacitors or power supplies, with high pressure-sensing sensitivity, small size, low power consumption, and low pressure-sensing noise levels;

2) The sensor has a long service life, wear resistance and corrosion resistance;

3) The noise level of analog-to-digital conversion is low, and the number of voltage or current quantization digits is more than 8;

4) The bearing prototype of the overall sensor is the shoe, and the bearing prototype of the sole sensor is the insole.

1. Sole of foot:

1) To solve the problem:

(1) The amount of information collected (resolution, number of sensors, number of pressure-sensitive element points) is inconsistent with feet of different sizes;

(2) It is meaningless to ensure the point position definition of the sensor.

2) Distribution mode (point distribution):

(1) Confirmation of the number of sensors:

N is the number of sensors (in megapixels), S is the average footprint area (in mm ² ) of an adult male, and d is the average diameter (in mm) of the acquisition unit with the smallest footprint.

According to the particle size reflected by the lesions of traditional Chinese medicine (the average diameter of millet grains is about 1mm), the information points collected on the sole of the foot in the smallest area collect at least 2 points per 1mm, which is converted into a resolution of 66PPI. Therefore, the distribution density of the sensors of the entire sole collection device At least 4/mm2, the average footprint size of an adult male is about ^23400mm2 (foot length 25cm, foot width about 9cm), at least 963600 points are required, considering that low-resolution footprint information is not enough to collect part of the foot Large male, and the sensor distribution is not a pure two-dimensional simple arrangement, and additional points are required to ensure that the acquisition area can completely collect footprints, this application increases the resolution to more than twice the theoretical value - 2 million points ( The foot area of an adult female is smaller than that of a male foot, so it can be determined according to the male footprint resolution requirement), that is, regardless of gender and foot condition, the sole information of each person is collected using 2 million points. This is the minimum requirement for the number of sensors to collect information per foot.

(2) Distribution of sensors:

According to the theory of traditional Chinese medicine, there are 66 important acupoints on each foot of the human body, 33 on the left and right feet, and the reflected areas of the lesions are very different in terms of area. The more points and the smaller the area, the more data actually needs to be collected (based on the amount of information). A sensor distribution method based on the acupoint area is given, as follows:

A. According to the distribution of acupoints, the soles of the feet are divided into three types of areas: large acupoint area, small acupoint area, and blank acupoint area. Refer to Figure 1 for the schematic diagram of acupoint division. Light gray is the small acupoint area, and dark gray is the large acupoint area. Black is the blank acupoint area.

B. The number of sensors per square millimeter in the blank acupoint area is set as x, and the number of sensors per square millimeter in the large acupoint area is set as y (the sensor distribution of different acupoints can be different or the same, so it is represented by a vector. There are no acupoints in the area, so a single variable can be represented), the number of sensors per square millimeter in the small acupoint area is set as z, and an optimal method for regional distribution is constructed with the objective function as a reference. The objective function includes but is not limited to the amount of information, acupoint attention The degree of quantification, distribution uniformity, etc., here is the solution method based on the amount of information: define the amount of information carried by each:

计算信息量，小穴位区域(共Q个穴位)，某个区域的面积为S_Cj，面积最小S_Cmin，面积最大为S_Cmax，面积越小，信息量越大，依据公式

计算信息量，这里构建目标函数H：The amount of information per square millimeter in the blank acupoint area is h _A , h _A =v, the overall area is S _A , the area included in the large acupoint area (a total of P acupoints), the area of a certain area is S _Bi , and the minimum area is S _Bmin , the maximum area is S _Bmax , the smaller the area, the greater the amount of information, according to the formula

Calculate the amount of information, small acupoint area (a total of Q acupoints), the area of a certain area is S _Cj , the smallest area is S _Cmin , the largest area is S _Cmax , the smaller the area, the greater the amount of information, according to the formula

To calculate the amount of information, here the objective function H is constructed:

1≤i≤P

1≤j≤Q

另外，每个区域的传感器密度不得少于平均d个/平方毫米，即x,y_i,z_j＞d，以上两点认为是约束条件c₁，c₂，则结合目标函数，可以得到如下关系：When the number of sensors N is given,

In addition, the density of sensors in each area should not be less than the average d/mm2, that is, x, y _i , z _j > d, the above two points are considered as constraints c ₁ , c ₂ , then combined with the objective function, the following can be obtained relation:

Satisfying the c ₁ and c ₂ constraints, there are infinitely many sets of solutions in this state, which is an NP-hard problem. To solve such problems, the nearest neighbor method, the insertion method, the genetic algorithm or the neural network algorithm can be used;

C. The sensors are distributed in each acupoint area by gradually calculating the skeleton of the binary map area. Assuming that the number of sensors in the acupoint area is K, at least T times of sensor arrangement (the skeleton can be obtained by performing T times of corrosion), then each The second need to distribute K/T sensors, the specific way is:

i. Define the binary image of a certain acupoint area as, the corrosion count is initially 0;

ii. For the extraction edge, the sensors are evenly distributed on the edge;

iii. Corrode the pair and increment the corrosion count by 1;

iv. Determine whether the number of corrosion has been reached, and if so, complete the sensor distribution;

If not, proceed to ii.

2. Foot surface:

1) Solve the problem: The error of the information collected by the sensor is too large, and it cannot be detailed to the shoe size and fat and thin type.

2) Distribution method:

(1) Confirmation of the number of sensors:

m=f(C,Δ)

Δ is the error of approximating the circumference of the polygon by the circumference of the polygon, C is the circumference of the circle, m is the minimum number of sides satisfying the error condition, L is the circumference of the reference _metatarsal toe, and Li is the circumference of each circle in Figure 2 , M is the number of sensors.

Since the barefoot surface of each person is very similar in a static environment, the shape difference will not be very obvious, and the degree of change is limited, but considering the stability characteristics of each person during walking (metatarsophalangeal circumference, anterior tarsus circumference, Pocket, heel circumference, etc.) measurement, it is necessary to have a sufficient number of sensors to ensure the error between the digital features and the real features. When making shoes, every 3.5mm of the metatarsal toe circumference is a shoe size (distinguishing point), so the number of sensors on the foot surface , It must be ensured that the measurement error of each circumference feature should not exceed 3.5mm. Based on the analysis of the 25th foot type of an adult male, the circumference of the metatarsal toe is about 246.5mm ("Footwear Designer"). If the measurement accuracy is less than 3.5mm, each At least 9 points are required for the measurement of each surrounding feature (the error is calculated by the method of approximating a circle by a polygon), and a maximum of 12 points are required, that is, one point is collected every 27.38mm. Similarly, this indicator is also applicable to the vertical direction of the surrounding feature. Divide the bare feet into 4 circles (as shown in Figure 2), and the radius of each circle is 45/90/70/60mm, then the collection points in the vertical direction are 11/21/17/14 respectively, considering that For the common part of each circle, the overall number of acquisitions is about 35. Based on this, it is determined that for people with feet 25, the number of sensors required for the acquisition of foot information is at least 35*12=420. Considering the low-resolution The footprint information is not enough to collect some men with big feet. Here, the number of sensors in the girth measurement direction and the vertical girth measurement direction are doubled, 70*24 respectively, and there are 1680 sensors in total.

(2) The distribution method of the sensor: the uniform distribution method can be used, and the distribution is based on one sensor point per 13.69mm.

Example 2

This embodiment provides a sensor installation method, as follows:

Here are the prototypes carried by the two sensors:

1) Insole:

(1) The insole can only be used for sensor installation on the sole of the foot. According to the fixed shoe size insole provided, the known quantities here are:

a. The overall minimum number of sensors to be distributed;

b. The minimum number of sensors in critical areas of the fixed footprint;

c. Minimum number of sensors per zone.

(2) Installation method:

a. Construct an average foot model by acquiring a large number of barefoot footprints. The methods of constructing an average foot include: average of key points after size normalization, average of pressure footprint, etc.;

b. Coincide the foot bisector of the insole with the average foot bisector, and confirm that there is a 5mm interval between the heel edge of the insole and the average heel rear edge;

c. Areas other than the average foot are as follows:

i. The minimum number of sensors in each area is evenly distributed sensors;

ii. Distributing sensors in such a way that the number of sensors on the average foot edge gradually decreases (morphological skeleton processing);

d. In the average foot, the sensors are distributed according to the number of sensors provided by different partitions and different optimization methods.

2) Shoes:

(1) The installation method of the sensor of the sole part of the shoe can refer to the method of installing the insole as the load-bearing method. The upper sensor is based on the provided fixed shoe size shoes. The known quantities before installation are:

a. The minimum number of overall sensors that need to be distributed;

b. Sensor distribution requirements.

(2) Installation method:

a. According to the upper of the shoe with the fixed shoe size, starting from the heel, measure the maximum perimeter L of the upper, and calculate the step length D of the sensor distribution through the L/minimum number of sensors. The intersection of the shoe center line and the heel is perpendicular The distance is D as the reference point O ₁ , and the coordinate system is constructed in the direction shown in Figure 3, z is the ground normal direction, y is the direction of the center line of the footprint foot, x is the direction perpendicular to the zy plane _; At the intersection of the planes, a sensor is installed every distance D;

b. Take the vertical distance of the intersection of the shoe center line and the heel as 2D as the reference point O ₂ , the intersection of the vamp and the O ₂ xy plane, install a sensor every distance D, and so on, the vamp and O ₂ until the xy plane has no intersection.

The present application provides a method for calculating the distribution of sensors for the collection of barefoot information and an installation procedure based on the position of the sensors, which improves the method for collecting the barefoot information and makes the collected barefoot information more comprehensive and complete. It can meet the quantitative collection of various footprint information, and lay a more solid foundation for barefoot 3D reconstruction.

The above is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited to this. The equivalent replacement or change of the inventive concept thereof shall be included within the protection scope of the present invention.

Claims (9)

1. A distribution method of barefoot information acquisition sensors is characterized by comprising the following steps:

s1: determining the number and the distribution mode of the sole sensors;

s2: determining the number and distribution mode of the foot surface sensors;

the following formula is adopted for determining the number of the sole sensors:

n denotes the number of sensors and S denotes the average footprint area in mm of an adult²(ii) a d is the average diameter of the collection unit with the smallest footprint, in mm.

2. The distribution method of the barefoot information acquisition sensors according to claim 1, wherein the distribution mode of the sole sensors is specifically as follows:

A. dividing the region of the sole according to the distribution condition of the acupuncture points;

B. constructing a region distribution optimal method with the objective function as a reference to solve;

C. each acupoint area distributes sensors by a method of gradually calculating a binary image area skeleton.

3. The distribution method of the barefoot information acquisition sensors as claimed in claim 2, wherein the sole of the foot is divided into three types of regions: the acupuncture point detection device comprises a large acupuncture point area, a small acupuncture point area and a blank acupuncture point area, wherein the number of sensors in the large acupuncture point area is x, the number of sensors in the small acupuncture point area is y, and the number of sensors in the blank acupuncture point area is z.

4. The distribution method of the barefoot information acquisition sensors as claimed in claim 3, wherein the objective function comprises information amount, acupoint importance quantification degree and distribution uniformity; the solving mode based on the information quantity is as follows:

the information amount per square millimeter of the blank hole region is h_A，h_AV, total area S_A(ii) a The large acupoint region comprises P acupoint regions with an area of S_BiMinimum area S_BminMaximum area is S_BmaxThe smaller the area, the larger the information quantity, according to the formula

Calculating the information quantity; the small acupoint region comprises Q acupoint regions, and the area of one region is S_CjMinimum area S_CminMaximum area is S_CmaxThe smaller the area, the larger the information quantity, according to the formula

Calculating the information amount, here constructing an objective function H:

when given the number of sensors N,

in addition, the sensor density per area must not be less than an average of d per square millimeter, i.e.

x,y_i,z_j＞d；

The above two points are considered as constraint conditions c₁，c₂Then, combining with the objective function, the following relationship is obtained:

satisfy c₁，c₂There are infinite groups of solutions under the constraint condition, and the optimal solution is obtained by adopting a neighbor method, an insertion method, a genetic algorithm or a neural network algorithm.

5. The distribution method of the barefoot information acquisition sensors as claimed in claim 4, wherein each acupoint area distributes the sensors by gradually calculating a binary image area skeleton, specifically as follows:

assuming that the number of sensors in the acupoint area is K, K/T sensors are required to be distributed for the least T times, and the method comprises the following specific steps:

i. defining a binary image of a certain cave region as I, and the initial corrosion frequency is 0;

extracting edges of the I, and uniformly distributing K/T sensors on the edges of the I;

etching I, and adding 1 to the etching times;

iv, judging whether the corrosion times reach T, if so, finishing the distribution of the sensors; if not, continuing to step ii.

6. The distribution method of the barefoot information acquisition sensors as claimed in claim 1, wherein the following formula is adopted for determining the number of the sensors on the foot surface:

m＝f(C,Δ)；

dividing the barefoot into n circles, wherein delta is the error of the polygon circumference approaching the circumference, C is the circumference of the circle, m is the minimum number of sides meeting the error condition, L is the reference metatarsophalangeal circumference, and L_iFor the perimeter of each circle, M is the number of sensors.

7. The method as claimed in claim 6, wherein the distribution of the sensors is uniform.

8. The method of claim 4, 6 or 7, wherein the sensor carries two types of prototypes, one of which is an insole; the installation method of the sensor on the insole comprises the following steps:

a. constructing an average foot model by acquiring a large number of barefoot footprints;

b. coinciding the bisector of the insole with the average bisector and confirming that the heel back edge of the insole has a certain distance from the heel back edge of the average heel;

c. the areas outside the average foot were treated as follows:

i. each area is uniformly distributed with sensors;

distributing the sensors in such a way that the number of average foot edge sensors decreases gradually;

d. and in the average foot, the sensors are distributed according to the number of the sensors provided by different partitions and different optimizing modes.

9. The method of claim 8, wherein the sensor carries two types of prototypes, the other type being a shoe; the method for installing the sensor on the sole part of the shoe is implemented by the method for installing the sensor on the insole, and the method for installing the sensor on the vamp comprises the following steps: constructing a coordinate system, wherein the z axis is the direction of a ground normal, the y axis is the direction of the center line of the footprint foot, and the x axis is the direction vertical to the zy plane;

a. measuring the maximum perimeter L of the vamp from the heel, and calculating the step length D of the distribution of the sensors through L/the minimum number of the sensors; the positive vertical distance D of the intersection point of the center line of the shoe and the heel is used as a reference point O₁Shoe upper and O₁Mounting a sensor at every distance D at the intersection position of the xy planes;

b. the positive vertical distance of the intersection point of the shoe center line and the heel is 2D and is used as a reference point O₂Shoe upper and O₂Mounting a sensor at every distance D at the position where the xy planes intersect, and so on until the vamp and the O₂The xy plane has no point of intersection.

Images