阅读说明：本技术 一种基于活体重力感应与边缘智能识别技术的出门检测系统 (Detection system of going out based on living body gravity induction and edge intelligent recognition technology ) 是由 刘甘霖 蔡慧敏 马利峰 于 2021-07-28 设计创作，主要内容包括：本发明公开了一种基于活体重力感应与边缘智能识别技术的出门检测系统,包括边缘计算设备和称重系统,所述称重系统通过温度补偿系数和/或加速度补偿系数对测得的重量进行修,然后测出温度补偿系数和加速度补偿系数,再通过重量精确计算公式获得精准重量。本发明的有益效果是通过对影响称重因素的温度和加速度做标定,能够有效消除温度与速度对重量称量带来的影响,并通过对拟合后做修正,得到准确性较高的温度补偿系数和速度补偿系数,统一了基准,消除了温度和速度等相关因素带来的影响,可靠性高,测量过程简单,特别适合对活体中午重量的测定。(The invention discloses a detection system for going out of a door based on living body gravity sensing and edge intelligent identification technology, which comprises edge calculation equipment and a weighing system, wherein the weighing system corrects the measured weight through a temperature compensation coefficient and/or an acceleration compensation coefficient, then measures the temperature compensation coefficient and the acceleration compensation coefficient, and finally obtains the accurate weight through a weight accurate calculation formula. The method has the advantages that the influence of temperature and speed on weight weighing can be effectively eliminated by calibrating the temperature and the acceleration which influence weighing factors, the temperature compensation coefficient and the speed compensation coefficient with higher accuracy are obtained by correcting after fitting, the reference is unified, the influence of temperature, speed and other related factors is eliminated, the reliability is high, the measuring process is simple, and the method is particularly suitable for measuring the midday weight of the living body.)

1. The utility model provides a detecting system that comes out based on live body gravity-feed tank and edge intelligent recognition technique which characterized in that: comprising an edge calculation device and a weighing system which corrects the measured weight by means of a temperature compensation coefficient and/or an acceleration compensation coefficient.

2. The system according to claim 1, wherein the system comprises: the temperature compensation coefficient measurement process is as follows,

s1, weighing the weight with fixed weight in a certain temperature interval by using standard weight measuring equipment, and recording the weight of each liter of weight with high fixed temperature;

s2 fitting the data by moving least square method, selecting a group of functions r₁(x),r₂(x),r₃(x),...r_m(x) Let f (x) be a₁r₁(x)+a₂r₂(x)+...+a_mr_m(x) Wherein a is₁、a₂、...a_mIs the undetermined coefficient; requiring coefficients to be fitted to the experimental data, a least squares criterion is followed: i.e. n points (x)_i,y_i) Distance δ from curve y ═ f (x)_iThe sum of squares is minimal, i.e.:

determining a₁,a₂,.. making J (a)₁,a₂...,a_m) The value of (d) is minimal;

the curve equation after function fitting is a cubic function T ═ k₁t³+k₂t²+k₃t+d

Wherein T is a temperature compensation coefficient, T is temperature, k₁、k₂、k₃And d is a constant.

3. The system according to claim 1, wherein the system comprises: the speed compensation coefficient measurement process is as follows,

s1, measuring the weight when the weight with fixed weight slides across the weight sensor at different speeds, and recording the weight measurement weight at different speeds;

s2, calculating the accelerations corresponding to different weights according to the speed calculation formula a, namely delta v/delta t;

s3, obtaining a function curve most closely influenced by the acceleration on the weight sensor by using a curve fitting mode, wherein the curve equation after the function fitting is a linear function A-k₁a + d, wherein A is an acceleration compensation coefficient, a is acceleration, and k is₁And d is a constant.

4. The system according to claim 2, wherein the system comprises: when fitting the data using the moving least squares method, the temperature is on the x-axis and the weight at this temperature is on the y-axis, and a function curve y ═ f (x) is sought such that f (x) is the closest to all measured data points.

5. The system for detecting the exit based on the living body gravity sensing and the edge intelligent identification technology as claimed in claims 2 and 3, wherein: according to the weight value W that weight inductive transducer acquireed again, can know actual weight value W, accurate weight computational formula is:

W＝w×T×A 。

Technical Field

The invention belongs to the technical field of intelligent retail, and particularly relates to a detection system for going out based on living body gravity sensing and edge intelligent identification technology.

Background

Today, as the smart retail industry becomes more mature and more widely used, the stability of smart retail systems and the experience of customers becomes more important. Retail merchants demand intelligent retail system service providers to provide more accurate and experienced systems. The exit detection system is the most important ring of the intelligent retail system and is a safety valve of the whole set of system. Whether the detection result of the exit detection system is accurate or not and whether the experience of the customer when the customer exits is smooth or not are determined, and whether the whole scheme is complete or not is determined. Therefore, both aspects need to be solved and continuously optimized for system service providers, and weight-sensing-based exit detection systems are currently available in the market and applied to the field of intelligent retail sales. Assuming that m weight sensors are arranged, a group of weight data is output every 100ms, after the system receives the data, the system averages n groups of data, namely the weight value output in the time period, and the specific result is as follows

The calculated W (weight value) is an effective weight output value, and is set as W₁. The second 100ms likewise outputs a value of W, W₁The average value of W and W is the second effective value, W₂＝(W₁+ W)/2; in this analogy, the nth group of data is Wn ═ W (W)₁+W₂+.. + W)/n. The method is easy to obtain, and the more n is, the more accurate the obtained effective value Wn is. The weight value of W is used as the y axis, the sampling times are used as the x axis, and a coordinate graph is drawn to enable the discussion to be more visual. As shown in fig. 3. Essentially around an axis, is a gradually converging wave pattern. The undulations of the first half of the waveform are large because there are too few samples to sample. When the sample is sufficient, it substantially converges to the true weight value.

From the above calculations, it can be readily seen that such a system is currently deficient. Such a detection system for going out has the following drawbacks

1. Acceleration effects, since the larger the n value is, the more accurate it is, the system manufacturer can only increase the number of sets of samples as much as possible in order to obtain a relatively accurate value. In actual sampling, the value of n is usually 100 groups, i.e. 10s (each sampling interval is 100ms) as shown in fig. 2.

2. Temperature affects the accuracy problem, which is manifested in the ambient temperature at which the weight-sensing sensor operates.

The weight sensor operates on the principle of the piezoelectric effect. The so-called piezoelectric effect is that for a heteropolar crystal without a symmetric center, an external force applied to the crystal will change the polarization state of the crystal in addition to deforming the crystal, and an electric field is established inside the crystal. "according to this principle, the mechanical force signal is converted into an electrical signal for output. The signal conversion process is susceptible to temperature and weight acceleration, and the error can be solved without increasing the sampling rate. To solve this problem, only the sensor itself can be used to find a solution to the problem. For the above reasons, the general accuracy error of the target system is about 100g, and in order that a customer can normally enter and exit the door, the method adopted by a system manufacturer is often to increase the error threshold. Therefore, with such a practical scenario, if a customer holds less than 100g of a product, such as a small bag of paper towels, the exit detection system cannot detect the product because the product falls within the allowable error range of the system manufacturer. Therefore, in an unattended scene, it is not appropriate to prevent a commodity having too small weight.

Therefore, a system for detecting the going-out based on the living body gravity sensing and the edge intelligent identification technology needs to be designed to solve the problems.

Disclosure of Invention

The invention aims to provide a detection system for going out based on living body gravity sensing and edge intelligent identification technology.

In order to solve the technical problems, the invention adopts the technical scheme that:

the exit detection system based on the living body gravity sensing and edge intelligent identification technology comprises edge computing equipment and a weighing system, wherein the weighing system corrects the measured weight through a temperature compensation coefficient and/or an acceleration compensation coefficient.

Preferably, the temperature compensation coefficient measuring process is as follows,

s1, weighing the weight with fixed weight in a certain temperature interval by using standard weight measuring equipment, and recording the weight of each liter of weight with high fixed temperature;

s2 fitting the data by moving least square method, selecting a group of functions r₁(x),r₂(x),r₃(x),...r_m(x) Let f (x) be a₁r₁(x)+a₂r₂(x)+...+a_mr_m(x) Wherein a is₁、 a₂、...a_mIs the undetermined coefficient; requiring coefficients to be fitted to the experimental data, a least squares criterion is followed: i.e. n points (x)_i,y_i) Distance δ from curve y ═ f (x)_iThe sum of squares is minimal, i.e.:

determining a₁,a₂,.. making J (a)₁,a₂...,a_m) The value of (d) is minimal;

the curve equation after function fitting is a cubic function T ═ k₁t³+k₂t²+k₃t+d

Wherein T is a temperature compensation coefficient, T is temperature, k₁、k₂、k₃And d is a constant.

Preferably, the speed compensation coefficient measuring process is as follows,

s1, measuring the weight when the weight with fixed weight slides across the weight sensor at different speeds, and recording the weight measurement weight at different speeds;

s2, calculating the accelerations corresponding to different weights according to the speed calculation formula a, namely delta v/delta t;

s3, obtaining a function curve most closely influenced by the acceleration on the weight sensor by using a curve fitting mode, and fitting the function curveThe equation is a linear function A ═ k₁a + d, wherein A is an acceleration compensation coefficient, a is acceleration, and k is₁D is a constant;

preferably, when fitting the data using the moving least squares method, the temperature is on the x-axis and the weight at this temperature is on the y-axis, and a function curve y ═ f (x) is sought such that f (x) is the closest to all measured data points.

Preferably, according to the weight value W that weight-sensitive sensor acquireed, can know actual weight value W, accurate weight computational formula is:

W＝w×T×A

the invention has the advantages and positive effects that:

the invention can effectively eliminate the influence of temperature and speed on the weight weighing by calibrating the temperature and the acceleration which influence the weighing factors, obtains the temperature compensation coefficient and the speed compensation coefficient with higher accuracy by correcting the fitted temperature and the speed compensation coefficient, unifies the reference, eliminates the influence of the temperature, the speed and other related factors, has high reliability and simple measurement process, and is particularly suitable for measuring the midday weight of the living body.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

FIG. 1 is a logic diagram of weight detection of a exit detection system based on living body gravity sensing and edge intelligent identification technology according to the present invention;

FIG. 2 is a waveform of time versus measurement for a weight sensor only state;

FIG. 3 is a trend graph of the weight transmission process within the multi-dimensional sensor;

FIG. 4 is a temperature influence graph trend graph;

FIG. 5 is an acceleration influence trend graph;

FIG. 6 is a formula for calculating the accurate weight of the exit detection system based on the living body gravity sensing and the edge intelligent identification technology.

Detailed Description

In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on those shown in the drawings, and are used only for convenience in describing the present invention and for simplicity in description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and thus, are not to be construed as limiting the present invention. Furthermore, the terms "first", "second", etc. are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first," "second," etc. may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless otherwise specified.

In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art through specific situations.

The invention will be further described with reference to the accompanying drawings in which:

example 1

As shown in fig. 1, an exit detection system based on living body gravity sensing and edge intelligent recognition technology includes an edge computing device and a weighing system, wherein the weighing system corrects the measured weight through a temperature compensation coefficient and/or an acceleration compensation coefficient.

Preferably, the temperature compensation coefficient measurement process is as follows,

s1, weighing the weight with fixed weight in a certain temperature interval by using standard weight measuring equipment, and recording the weight of each liter of weight with high fixed temperature;

s2 fitting the data by moving least square method, selecting a group of functions r₁(x),r₂(x),r₃(x),...r_m(x) Let f (x) be a₁r₁(x)+a₂r₂(x)+...+a_mr_m(x) Wherein a is₁、 a₂、...a_mIs the undetermined coefficient; requiring coefficients to be fitted to the experimental data, a least squares criterion is followed: i.e. n points (x)_i,y_i) Distance δ from curve y ═ f (x)_iThe sum of squares is minimal, i.e.:

determining a₁,a₂,.. making J (a)₁,a₂...,a_m) The value of (d) is minimal;

the curve equation after function fitting is a cubic function T ═ k₁t³+k₂t²+k₃t+d

Wherein T is a temperature compensation coefficient, T is temperature, k₁、k₂、k₃D is a constant;

the speed compensation coefficient measurement process is as follows,

s1, measuring the weight when the weight with fixed weight slides across the weight sensor at different speeds, and recording the weight measurement weight at different speeds;

s2, calculating the accelerations corresponding to different weights according to the speed calculation formula a, namely delta v/delta t;

s3 obtaining the function curve of the influence of the most close acceleration on the weight sensor by using a curve fitting mode, and performing function fitting to obtain the curve equationAs a function of the first order a ═ k₁a + d, wherein A is an acceleration compensation coefficient, a is acceleration, and k is₁And d is a constant.

Preferably, when fitting the data using the moving least squares method, the temperature is on the x-axis and the weight at this temperature is on the y-axis, and a function curve y ═ f (x) is sought such that f (x) is the closest to all measured data points.

Preferably, according to the weight value W that weight-sensitive sensor acquireed, can know actual weight value W, accurate weight computational formula is:

W＝w×T×A

the working process of the embodiment: firstly, a temperature compensation coefficient and an acceleration compensation coefficient are determined,

1. determining temperature compensation coefficient

Obtaining experimental data

According to the calibration procedure of the electronic balance, the weight standard of 10g is selected, the range of-10 ℃ to 40 ℃ is selected as the loading point, one loading point is taken at every 5 ℃, and the analysis experiment of the corresponding temperature range is carried out. The results obtained are shown in table one. (the smaller the interval value, the higher the accuracy according to the accuracy requirement.) the trend graph is shown in FIG. 4.

Watch 1

According to the basic principle of curve fitting, all data points of the curve reflect the change trend of the data as a whole.

Selecting a set of functions r₁(x)，r₂(x)，r₃(x)，...r_m(x) Order:

f(x)＝a₁r₁(x)+a₂r₂(x)+...+a_mr_m(x)，

wherein a is₁、a₂、...a_mIs the undetermined coefficient.

Requiring coefficients to be fitted to the experimental data, a least squares criterion is followed: i.e. n points (x)_i，y_i) Distance δ from curve y ═ f (x)_iThe sum of squares is minimal, i.e.:

determining a₁，a₂,.. making J (a)₁，a₂...，a_m) The value of (d) is minimal;

the curve equation after function fitting is a cubic function T ═ k₁t³+k₂t²+k₃t+d

Wherein T is a temperature compensation coefficient, T is temperature, k₁、k₂、k₃And d is a constant.

Calculating a temperature compensation coefficient T according to the ambient temperature T measured by the temperature sensor during actual weighing

Also, an acceleration sensor is introduced to compensate for the gravity signal.

2. Determining an acceleration compensation factor

The acceleration sensor and the gravity sensing sensor are integrated together to measure the acceleration.

The experimental data are obtained, the acceleration is not continuously obtained, and the experimental data are obtained by changing the speed of applying force. As shown in Table two

Table two, experimental data of gravity change under different accelerations, and a trend chart is shown in fig. 5.

And similarly, acquiring a function curve most closely influenced by the acceleration on the weight sensor by using a curve fitting mode.

The curve equation after function fitting is a linear function A ═ k₁a+d。

Wherein A is an acceleration compensation coefficient, a is an acceleration, and k₁And d is a constant.

And calculating an acceleration compensation coefficient A according to the acceleration a measured by the acceleration sensor and the calculation model.

The accurate temperature compensation coefficient T and the accurate acceleration compensation coefficient A are obtained through the above mode, and then according to the gravity value W obtained by the gravity induction sensor, the actual gravity value W can be known as follows:

W＝w×T×A

the influence of temperature and acceleration on weight measurement can be effectively eliminated through calculation, and accurate weight data can be obtained.

While one embodiment of the present invention has been described in detail, the description is only a preferred embodiment of the present invention and should not be taken as limiting the scope of the invention. All equivalent changes and modifications made within the scope of the present invention shall fall within the scope of the present invention.