CN114399366B - Unmanned vending machine product selection method and system based on crowd analysis - Google Patents
Unmanned vending machine product selection method and system based on crowd analysis Download PDFInfo
- Publication number
- CN114399366B CN114399366B CN202210067171.0A CN202210067171A CN114399366B CN 114399366 B CN114399366 B CN 114399366B CN 202210067171 A CN202210067171 A CN 202210067171A CN 114399366 B CN114399366 B CN 114399366B
- Authority
- CN
- China
- Prior art keywords
- vending machine
- crowd
- characteristic
- attribute
- sample image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Control Of Vending Devices And Auxiliary Devices For Vending Devices (AREA)
Abstract
The invention discloses a method and a system for selecting an unmanned vending machine based on crowd analysis, wherein the method comprises the steps of monitoring and acquiring a first sample image and a second sample image based on people flow around the vending machine, obtaining crowd characteristics and corresponding characteristic attributes of pedestrians or consumers in the first sample image and the second sample image by adopting image analysis, further obtaining a first crowd characteristic value and a crowd characteristic value, finally carrying out similarity matching on the second crowd characteristic value and all first crowd characteristic values, and selecting a hot selling product on the existing unmanned vending machine corresponding to the closest first crowd characteristic value as a selected product recommendation of the unmanned vending machine to be selected. The invention is applied to the technical field of self-service vending machine management, recommends products matched with the hot sales of people for operators, improves the product selection efficiency of the operators, and reduces the product testing time, thereby improving the profit efficiency of the vending machine.
Description
Technical Field
The invention relates to the technical field of self-service vending machine management, in particular to a commodity selection method and system of an unmanned vending machine based on crowd analysis.
Background
Nowadays, vending machines have gradually begun to be popularized, the coverage rate and the utilization rate are rapidly improved, the use scenes of the vending machines are gradually enriched, due to the fact that the capacity of the vending machines is limited, the quantity and the types of commodities which are put in at one time are limited, a relatively long time is needed for judging which commodities are sold relatively hot, and the sales rate and the profitability of the vending machines can have relatively large influence in the time of testing products.
In the aspect of product selection of the current unmanned vending machine, currently hot-sold products are analyzed through a product selling history of the vending machine, and product selection is performed according to an analysis result. For example, a commodity optimization method mentioned in patent CN 109461027A randomly puts new commodities in a vending machine for vending and optimizes according to feedback results, which has few consideration factors, and may cause inaccurate feedback results because people near the selected machine are not target people, so that operators make wrong judgments to affect enterprise operations.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a commodity selection method and system of the unmanned vending machine based on crowd analysis, which can effectively improve the commodity selection efficiency of operators and reduce the product testing time, thereby improving the profit efficiency of the vending machine.
In order to achieve the purpose, the invention provides a goods selection method of an unmanned vending machine based on crowd analysis, which comprises the following steps:
step 1, acquiring first sample images of pedestrians or consumers in front of each existing unmanned vending machine based on a camera on the existing unmanned vending machine;
step 2, performing image analysis on all the first sample images, and extracting the crowd characteristics and the corresponding characteristic attributes of pedestrians or consumers in each first sample image;
step 3, calculating to obtain a first person group characteristic value of each existing unmanned vending machine based on the person group characteristics of pedestrians or consumers in each first sample image and the corresponding characteristic attributes;
step 4, collecting a second sample image of pedestrians or consumers in the to-be-placed area of the to-be-selected unmanned vending machine;
step 5, performing image analysis on all the second sample images, and extracting the crowd characteristics and the corresponding characteristic attributes of the pedestrians or the consumers in each second sample image;
step 6, calculating to obtain a second crowd characteristic value of the unmanned vending machine to be selected based on the crowd characteristics of the pedestrians or the consumers in each second sample image and the corresponding characteristic attributes;
and 7, calculating the feature similarity between the second crowd feature value and each first crowd feature value, and selecting a hot selling product on the existing unmanned vending machine corresponding to the first crowd feature value with the highest feature similarity as a selected product recommendation of the unmanned vending machine to be selected.
More specifically, in step 2 and step 4, the crowd characteristics of the pedestrians or consumers in the first sample image and the crowd characteristics of the pedestrians or consumers in the second sample image both include: height characteristics, clothing characteristics and age characteristics.
Further specifically, in step 2 and step 4, the characteristic attributes are specifically:
the height features have four feature attributes: the attribute 1 is that the height is less than 120cm, the attribute 2 is that the height is within the interval of 120cm-140cm, the attribute 3 is that the height is within the interval of 140cm-160cm, the attribute 4 is that the height is above 160cm;
the garment features have four characteristic attributes: attribute 1 is school uniform, attribute 2 is casual uniform, attribute 3 is work uniform, attribute 4 is sportswear;
the age feature has six feature attributes: the attribute 1 is an age less than 10 years, the attribute 2 is an age in the range of 10 years to 18 years, the attribute 3 is an age in the range of 18 years to 35 years, the attribute 4 is an age in the range of 35 years to 45 years, the attribute 5 is an age in the range of 45 years to 46 years, and the attribute 6 is an age over 60 years.
More specifically, in step 3, the obtaining process of the first person group feature value is as follows:
for the existing vending machine n, the number of the first sample images corresponding to the vending machine n is acquired to be P n Counting the number of samples of each characteristic attribute in a first sample image corresponding to the vending machine n;
calculating the crowd characteristic value of each characteristic attribute in each crowd characteristic in the first sample image corresponding to the vending machine n, wherein the crowd characteristic value is as follows:
V nij ＝Q nij /P n
in the formula, V nij A demographic value, Q, representing the jth characteristic attribute of the ith personal demographic in the vending machine's n captured image nij A sample number of jth feature attributes representing ith person group feature in the image captured by the vending machine n, i representing the ith person group feature, and j representing the jth feature attribute in the ith person group feature.
More specifically, in step 6, the process of obtaining the second population characteristic value is:
acquiring the number P of second sample images corresponding to the unmanned vending machine for the to-be-selected products, and counting the number of samples of each characteristic attribute in the second sample images corresponding to the unmanned vending machine for the to-be-selected products;
calculating the crowd characteristic value of each characteristic attribute in each crowd characteristic in the second sample image corresponding to the to-be-selected-product unmanned vending machine, wherein the crowd characteristic value is as follows:
V ij ＝Q ij /P
in the formula, V ij A crowd characteristic value Q of the jth characteristic attribute of the ith personal crowd characteristic in the acquired image of the unmanned vending machine for the selected product ij And the number of samples of the j-th characteristic attribute representing the ith personal group characteristic in the picked image of the unmanned vending machine to be selected.
More specifically, in step 7, the process of calculating the feature similarity between the second population feature value and each first population feature value is as follows:
calculating the similarity of the single features:
in the formula, S ni Representing the feature similarity of the existing vending machine n and the vending machine to be selected on the ith personal group feature, and J representing the total number of feature attributes in the ith personal group feature;
calculating the feature similarity as follows:
in the formula, F n Representing the feature similarity of the existing vending machine n and the vending machine to be selected, I representing the total number of the population features, A ni A feature weight representing the characteristic of the ith personal group.
Further specifically, it is characterized in that in step 2 and step 4, the image analysis is performed based on the trained convolutional neural network.
In order to achieve the purpose, the invention also provides a goods selection system of the unmanned vending machine based on crowd analysis, and the goods selection method of the unmanned vending machine is adopted.
Further specifically, the vending machine selection system comprises:
the image acquisition unit is used for acquiring a first sample image acquired by the existing unmanned vending machine and a second sample image of a to-be-selected product unmanned vending machine placement area;
the image analysis unit is used for analyzing the first sample image and the second sample image and extracting the crowd characteristics and the corresponding characteristic attributes of pedestrians or consumers in each first sample image and each second sample image;
the crowd similarity calculation unit is used for calculating a first crowd characteristic value and a second crowd characteristic value according to the crowd characteristics of pedestrians or consumers and the corresponding characteristic attributes;
and the similarity calculation unit is used for screening out the first crowd characteristic value closest to the second crowd characteristic value and taking the hot-sold product corresponding to the first crowd characteristic value on the existing unmanned vending machine as the selected item recommendation of the unmanned vending machine to be selected.
Compared with the prior art, the method and the system for selecting the products of the unmanned vending machine based on the crowd analysis are based on the crowd monitoring around the vending machine, matching is carried out according to the crowd characteristics and the characteristic attributes, the crowd around the existing unmanned vending machine is matched, the products matched with the hot sales of the crowd are recommended for the operator, the product selection efficiency of the operator is improved, the product testing time is shortened, and the profit efficiency of the vending machine is improved.
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, 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 the structures shown in the drawings without creative efforts.
FIG. 1 is a flow chart of a vending machine selection method in an embodiment of the present invention;
fig. 2 is a block diagram of an automatic vending machine selection system according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that all directional indicators (such as up, down, left, right, front, back ...) in the embodiments of the present invention are only used to explain the relative positional relationship between the components, the motion situation, etc. in a specific posture (as shown in the attached drawings), and if the specific posture is changed, the directional indicator is changed accordingly.
In addition, the descriptions related to "first", "second", etc. in the present invention are only for descriptive purposes 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" or "second" may explicitly or implicitly include at least one of the feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
In the present invention, unless otherwise expressly stated or limited, the terms "connected," "secured," and the like are to be construed broadly, and for example, "secured" may be a fixed connection, a removable connection, or an integral part; the connection can be mechanical connection, electrical connection, physical connection or wireless communication connection; they may be directly connected or indirectly connected through intervening media, or they may be connected internally or in any other suitable relationship, unless expressly stated otherwise. The specific meanings of the above terms in the present invention can be understood according to specific situations by those of ordinary skill in the art.
In addition, the technical solutions in the embodiments of the present invention may be combined with each other, but it must be based on the realization of those skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination of technical solutions should not be considered to exist, and is not within the protection scope of the present invention.
The embodiment discloses a commodity selection method of an unmanned vending machine based on crowd analysis, which is based on people flow monitoring around the vending machine, matches people around the existing unmanned vending machine according to the crowd characteristics and the characteristic attributes, recommends a product matched with the hot sales of the people for an operator, improves the commodity selection efficiency of the operator, reduces the product testing time, and improves the profit efficiency of the vending machine. Referring to fig. 1, the method for selecting a product by an unmanned vending machine in the embodiment specifically includes the following steps:
step 1, collecting a first sample image of a pedestrian or a consumer before passing through each existing unmanned vending machine based on a camera on the existing unmanned vending machine;
step 2, performing image analysis on all the first sample images, and extracting the crowd characteristics and the corresponding characteristic attributes of pedestrians or consumers in each first sample image;
step 3, calculating to obtain a first person group characteristic value of each existing unmanned vending machine based on the person group characteristics of pedestrians or consumers in each first sample image and the corresponding characteristic attributes;
step 4, collecting a second sample image of pedestrians or consumers in the to-be-placed area of the to-be-selected unmanned vending machine;
step 5, performing image analysis on all the second sample images, and extracting the crowd characteristics and the corresponding characteristic attributes of the pedestrians or the consumers in each second sample image;
step 6, calculating to obtain a second crowd characteristic value of the unmanned vending machine to be selected based on the crowd characteristics of the pedestrians or the consumers in each second sample image and the corresponding characteristic attributes;
and 7, calculating the feature similarity between the second crowd feature value and each first crowd feature value, and selecting a hot selling product on the existing unmanned vending machine corresponding to the first crowd feature value with the highest feature similarity as a selection recommendation of the unmanned vending machine to be selected.
In this embodiment, the crowd characteristics of the pedestrians or the consumers in the first sample image and the crowd characteristics of the pedestrians or the consumers in the second sample image both include: height characteristics, clothing characteristics and age characteristics. And attributes for each demographic are divided into:
the height features have four feature attributes: the attribute 1 is that the height is less than 120cm, the attribute 2 is that the height is within an interval of 120cm-140cm, the attribute 3 is that the height is within an interval of 140cm-160cm, and the attribute 4 is that the height is above 160cm, wherein 100, 120cm belongs to the attribute 2, 140cm belongs to the attribute 3, and 160cm belongs to the attribute 4;
the garment features have four characteristic attributes: attribute 1 is school uniform, attribute 2 is sportswear, attribute 3 is work uniform, attribute 4 is sportswear;
the age feature has six feature attributes: the attribute 1 is an age less than 10 years, the attribute 2 is an age in a range from 10 years to 18 years, the attribute 3 is an age in a range from 18 years to 35 years, the attribute 4 is an age in a range from 35 years to 45 years, the attribute 5 is an age in a range from 45 years to 60 years, and the attribute 6 is an age above 46 years, wherein 10 years belongs to the attribute 2, 18 years belongs to the attribute 3, 35 years belongs to the attribute 4, 45 years belongs to the attribute 5, and 60 years belongs to the attribute 6.
That is, for any one of the first sample image or the second sample image, it has three population characteristics, and each population characteristic has a certain attribute in the first sample image or the second sample image. In a specific implementation process, the crowd characteristics and the corresponding characteristic attributes obtained based on the image analysis can be obtained by adopting a convolutional neural network, and the convolutional neural network is trained after the crowd characteristics and the corresponding characteristic attributes in the crowd images are marked by collecting a large number of crowd images, so that the convolutional neural network can be used for the image analysis in the embodiment.
In a specific implementation process, a characteristic similarity calculation process between the existing unmanned vending machine and the unmanned vending machine for the to-be-selected products is as follows:
preferably, the first person group characteristic value corresponding to each existing computer is calculated, and the calculation process is as follows:
for the existing vending machine n, the number of the first sample images corresponding to the vending machine n is acquired to be P n Counting the number of samples of each characteristic attribute in a first sample image corresponding to the vending machine n;
calculating the crowd characteristic value of each characteristic attribute in each crowd characteristic in the first sample image corresponding to the vending machine n, wherein the crowd characteristic value is as follows:
V nij ＝Q nij /P n
in the formula, V nij A demographic value, Q, representing the jth characteristic attribute of the ith personal demographic in the vending machine's n captured image nij The sample number of jth characteristic attributes of ith personal group characteristics in the acquired image of the vending machine n is represented, i represents ith personal group characteristics, and j represents jth characteristic attributes in the ith personal group characteristics;
secondly, calculating a second crowd characteristic value of the unmanned vending machine for the to-be-selected goods, wherein the calculating process is as follows:
acquiring the number P of second sample images corresponding to the to-be-selected unmanned vending machine, and counting the number of samples of each characteristic attribute in the second sample images corresponding to the to-be-selected unmanned vending machine;
calculating the crowd characteristic value of each characteristic attribute in each crowd characteristic in the second sample image corresponding to the to-be-selected-product unmanned vending machine, wherein the crowd characteristic value is as follows:
V ij ＝Q ij /P
in the formula, V ij A crowd characteristic value Q of a jth characteristic attribute representing the ith personal crowd characteristic in the acquired image of the unmanned vending machine for the selected product ij The sample number of the jth characteristic attribute of the ith personal group characteristic in the acquired image of the unmanned vending machine to be selected is represented;
then, calculating the single feature similarity between the second population characteristic value and each first population characteristic value, wherein the single feature similarity is as follows:
in the formula, S ni Representing the feature similarity of the existing vending machine n and the vending machine to be selected on the ith personal group feature, and J representing the total number of feature attributes in the ith personal group feature;
and finally, the total feature similarity between the second population feature value and each first population feature value, namely the feature similarity between the existing unmanned vending machine and the unmanned vending machine for the to-be-selected goods, is as follows:
in the formula, F n Representing the feature similarity of the existing vending machine n and the vending machine to be selected, I representing the total number of the crowd features, A ni A feature weight representing the feature of the ith personal group. Namely, according to the size of the feature similarity value, the hot sales product of the unmanned vending machine with the highest similarity value is taken as a recommendation.
Referring to fig. 2, based on the above-mentioned unmanned vending machine product selection method, this embodiment also discloses an unmanned vending machine product selection system based on crowd analysis, and the unmanned vending machine product selection system adopts the above-mentioned unmanned vending machine product selection method to recommend a newly added unmanned vending machine. Specifically, the vending machine selection system comprises an image acquisition unit, an image analysis unit, a crowd similarity calculation unit and a similarity matching unit. The image acquisition unit is used for acquiring a first sample image acquired by the existing unmanned vending machine and a second sample image of a to-be-placed area of the to-be-selected unmanned vending machine; the image analysis unit is a training convolutional neural network and is used for analyzing the first sample image and the second sample image and extracting the crowd characteristics and the corresponding characteristic attributes of pedestrians or consumers in each first sample image and each second sample image; the crowd similarity calculation unit is used for calculating a first crowd characteristic value and a second crowd characteristic value according to the crowd characteristics of pedestrians or consumers and the corresponding characteristic attributes, and the calculation process is the same as that of the method, so that the calculation process is not repeated; and the similarity calculation unit is used for screening out the first crowd characteristic value closest to the second crowd characteristic value and taking the hot-sold product corresponding to the first crowd characteristic value on the existing unmanned vending machine as the selected item recommendation of the unmanned vending machine to be selected.
The vending machine selection method in this embodiment is further described below with reference to specific examples.
Supposing that 3 unmanned vending machines M1, M2 and M3 are provided, wherein M1 and M2 are the existing unmanned vending machines, and M3 is the unmanned vending machine for the to-be-selected goods;
the number of sampling persons of each vending machine is 10000, namely P1= P2= P3= 10000;
has 3 population characteristics, which are: height characteristic C1, clothing characteristic C2 and age characteristic C3; wherein, the height characteristic weight A1=1, the clothing characteristic weight A2=1, and the age characteristic weight A3=1.2.
The height characteristics have four attributes, which are: attribute 1T 11 <120cm; attribute 2T 12 =120cm-140cm; attribute 3T 13 =140cm-160cm; attribute 4T 14 >＝160cm。
The clothing features four attributes, which are: attribute 1T 21 = school uniform; attribute 2T 22 = casual wear; attribute 3T 23 = work clothes; attribute 4T 24 = sportswear.
The age characteristic has six attributes, which are: attribute 1T 31 <Age 10; attribute 2T 32 =10-18 years old; attribute 3T 33 =18-35 years old; attribute 4T 34 = 35-45 years old; attribute 5T 35 = 45-60 years old; attribute 6T 36 >=60 years of age.
If the existing unmanned vending machine M1 is near middle school, and the number of the collected characteristic attributes is set as follows:
height characteristic attribute quantity: q 111 ＝1000,Q 112 ＝5000，Q 113 ＝3000，Q 114 ＝1000；
Number of clothing feature attributes: q 121 ＝4000,Q 122 ＝4000，Q 123 ＝500，Q 124 ＝1500；
Age characteristic attribute number: q 131 ＝1000,Q 132 ＝6000，Q 133 ＝1000，Q 134 ＝1000，Q 135 ＝800，Q 136 ＝200。
The population characteristic values of the vending machine M1 are:
height characteristic population characteristic value: v 111 ＝0.1,V 112 ＝0.5,V 113 ＝0.3,V 114 ＝0.1；
Clothing characteristic population characteristic value: v 121 ＝0.4,V 122 ＝0.4,V 123 ＝0.05,V 124 ＝0.15；
Age characteristic population characteristic value: v 131 ＝0.1,V 132 ＝0.6,V 133 ＝0.1,V 134 ＝0.1,V 135 ＝0.08,V 136 ＝0.02。
If the existing vending machine M2 is in a commercial area, and the number of the collected characteristic attributes is set as follows:
height characteristic attribute quantity: q 211 ＝1000,Q 212 ＝1000，Q 213 ＝2000，Q 214 ＝6000；
Number of clothing feature attributes: q 221 ＝1000,Q 222 ＝4000，Q 223 ＝1000，Q 224 ＝4000；
Number of age characteristic attributes: q 231 ＝1000,Q 232 ＝1500，Q 233 ＝2000，Q 234 ＝2000，Q 135 ＝1500，Q 136 ＝1000。
The crowd characteristic value of the vending machine M2 is:
height characteristic population characteristic value: v 211 ＝0.1,V 212 ＝0.1,V 213 ＝0.2,V 214 ＝0.6；
Clothing characteristic population characteristic value: v 221 ＝0.1,V 222 ＝0.4,V 223 ＝0.1,V 224 ＝0.4；
Age characteristic population characteristic value: v 231 ＝0.1,V 232 ＝0.15,V 233 ＝0.2,V 234 ＝0.2,V 235 ＝0.15,V 236 ＝0.1。
If the to-be-placed area of the to-be-selected unmanned vending machine M3 is also near middle school, and the collected characteristic attribute quantity is set as:
height characteristic attribute quantity: q 311 ＝1200,Q 312 ＝4500，Q 313 ＝3300，Q 114 ＝1000；
Number of clothing feature attributes: q 321 ＝4200,Q 322 ＝3800，Q 323 ＝600，Q 124 ＝1400；
Age characteristic attribute number: q 331 ＝1200,Q 332 ＝6200，Q 333 ＝800，Q 134 ＝800，Q 135 ＝800，Q 136 =200. The crowd characteristic value of the unmanned vending machine M3 for the to-be-selected product is as follows:
height characteristic population characteristic value: v 311 ＝0.12,V 312 ＝0.45,V 313 ＝0.33,V 314 ＝0.1；
Clothing characteristic population characteristic value: v 321 ＝0.42,V 322 ＝0.38,V 323 ＝0.06,V 324 ＝0.14；
Age characteristic population characteristic value: v 331 ＝0.12,V 332 ＝0.62,V 333 ＝0.08,V 334 ＝0.08,V 335 ＝0.08,V 336 ＝0.02。
The single-feature similarity between the existing unmanned vending machine M1 and the unmanned vending machine M3 for the to-be-selected goods is as follows:
the height single feature similarity is as follows:
the similarity of the single characteristics of the clothes is as follows:
age chart feature similarity is:
calculating to obtain the feature similarity between the unmanned vending machine M1 and the unmanned vending machine M3 to be selected as follows:
F 1 ＝S 11 +S 12 +1.2×S 13 ＝0.9951+0.9985+0.952×1.2＝3.136。
the single-feature similarity between the existing unmanned vending machine M2 and the unmanned vending machine M3 to be selected is as follows:
the height single feature similarity is as follows:
the clothing single characteristic similarity is as follows:
age chart feature similarity is:
calculating to obtain the feature similarity between the unmanned vending machine M2 and the unmanned vending machine M3 to be selected as follows:
F 2 ＝S 21 +S 22 +1.2×S 23 ＝0.4873+0.7486+0.6131×1.2＝1.972。
in summary, F 1 >F 2 Therefore, the hot-sold product on the existing unmanned vending machine M1 is selected as the selection recommendation of the unmanned vending machine to be selected.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention, and all equivalent structural changes made by using the contents of the present specification and the drawings, or any other related technical fields, which are directly or indirectly applied to the present invention, are included in the scope of the present invention.
Claims (2)
1. A vending machine product selection method based on crowd analysis is characterized by comprising the following steps:
step 1, acquiring first sample images of pedestrians or consumers in front of each existing unmanned vending machine based on a camera on the existing unmanned vending machine;
step 2, performing image analysis on all the first sample images, and extracting the crowd characteristics and the corresponding characteristic attributes of pedestrians or consumers in each first sample image;
step 3, calculating to obtain a first crowd characteristic value of each existing unmanned vending machine based on the crowd characteristics of pedestrians or consumers in each first sample image and the corresponding characteristic attributes;
step 4, collecting a second sample image of pedestrians or consumers in the to-be-placed area of the to-be-selected unmanned vending machine;
step 5, performing image analysis on all the second sample images, and extracting the crowd characteristics and the corresponding characteristic attributes of the pedestrians or the consumers in each second sample image;
step 6, calculating to obtain a second crowd characteristic value of the unmanned vending machine for the to-be-selected products based on the crowd characteristics of the pedestrians or the consumers in each second sample image and the corresponding characteristic attributes;
step 7, calculating the feature similarity between the second crowd feature value and each first crowd feature value, and selecting hot sales products on the existing unmanned vending machine corresponding to the first crowd feature value with the highest feature similarity as selection recommendation of the unmanned vending machine to be selected;
in step 2 and step 4, the crowd characteristics of the pedestrians or the consumers in the first sample image and the crowd characteristics of the pedestrians or the consumers in the second sample image both include: height characteristics, clothing characteristics and age characteristics;
in step 2 and step 4, the characteristic attributes are specifically:
the height features have four feature attributes: the attribute 1 is that the height is less than 120cm, the attribute 2 is that the height is within the interval of 120cm-140cm, the attribute 3 is that the height is within the interval of 140cm-160cm, the attribute 4 is that the height is above 160cm;
the garment features have four characteristic attributes: attribute 1 is school uniform, attribute 2 is casual uniform, attribute 3 is work uniform, attribute 4 is sportswear;
the age characteristic has six characteristic attributes: the attribute 1 is an age less than 10 years, the attribute 2 is an age in a range of 10 years to 18 years, the attribute 3 is an age in a range of 18 years to 35 years, the attribute 4 is an age in a range of 35 years to 45 years, the attribute 5 is an age in a range of 45 years to 46 years, and the attribute 6 is an age over 60 years;
in step 2 and step 4, image analysis is performed based on the trained convolutional neural network;
any one of the first sample image or the second sample image having three demographic characteristics, each of which has a certain attribute in the first sample image or the second sample image;
the acquisition process of the first person group characteristic value comprises the following steps:
for the existing vending machine n, the number of the first sample images corresponding to the vending machine n is P n Counting the number of samples of each characteristic attribute in a first sample image corresponding to the vending machine n;
calculating the crowd characteristic value of each characteristic attribute in each crowd characteristic in the first sample image corresponding to the vending machine n, wherein the crowd characteristic value is as follows:
V nij ＝Q nij /P n
in the formula, V nij A demographic value, Q, representing the jth characteristic attribute of the ith personal demographic in the vending machine's n captured image nij A sample number of jth feature attributes representing ith personal group feature in the acquired image of the vending machine n;
the acquisition process of the second population characteristic value is as follows:
acquiring the number P of second sample images corresponding to the to-be-selected unmanned vending machine, and counting the number of samples of each characteristic attribute in the second sample images corresponding to the to-be-selected unmanned vending machine;
calculating the crowd characteristic value of each characteristic attribute in each crowd characteristic in the second sample image corresponding to the unmanned vending machine for the to-be-selected product, wherein the crowd characteristic value is as follows:
V ij ＝Q ij /P
in the formula, V ij A crowd characteristic value Q of a jth characteristic attribute representing the ith personal crowd characteristic in the acquired image of the unmanned vending machine for the selected product ij The sample number of the jth characteristic attribute of the ith personal group characteristic in the acquired image of the unmanned vending machine to be selected is represented;
the process of calculating the feature similarity between the second population feature value and each first population feature value is as follows:
calculating the similarity of the single features:
in the formula, S ni Representing the feature similarity of the existing vending machine n and the vending machine to be selected on the ith personal group feature, and J representing the total number of feature attributes in the ith personal group feature;
calculating the feature similarity as follows:
in the formula, F n Representing the feature similarity of the existing vending machine n and the vending machine to be selected, I representing the total number of the population features, A ni A feature weight representing the characteristic of the ith personal group.
2. An automatic vending machine product selection system based on crowd analysis is characterized in that the automatic vending machine product selection method in claim 1 is adopted;
the vending machine selection system comprises:
the image acquisition unit is used for acquiring a first sample image acquired by the existing unmanned vending machine and a second sample image of a to-be-placed area of the unmanned vending machine to be selected;
the image analysis unit is used for analyzing the first sample image and the second sample image and extracting the crowd characteristics and the corresponding characteristic attributes of pedestrians or consumers in each first sample image and each second sample image;
the crowd similarity calculation unit is used for calculating a first crowd characteristic value and a second crowd characteristic value according to the crowd characteristics of pedestrians or consumers and the corresponding characteristic attributes;
and the similarity calculation unit is used for screening out the first crowd characteristic value closest to the second crowd characteristic value and taking the hot-sold product corresponding to the first crowd characteristic value on the existing unmanned vending machine as the selection recommendation of the unmanned vending machine to be selected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210067171.0A CN114399366B (en) | 2022-01-20 | 2022-01-20 | Unmanned vending machine product selection method and system based on crowd analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210067171.0A CN114399366B (en) | 2022-01-20 | 2022-01-20 | Unmanned vending machine product selection method and system based on crowd analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114399366A CN114399366A (en) | 2022-04-26 |
| CN114399366B true CN114399366B (en) | 2023-04-18 |
Family
ID=81233482
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210067171.0A Active CN114399366B (en) | 2022-01-20 | 2022-01-20 | Unmanned vending machine product selection method and system based on crowd analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114399366B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292667A (en) * | 2017-06-23 | 2017-10-24 | 北京小度信息科技有限公司 | Information-pushing method and device |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI615813B (en) * | 2016-12-12 | 2018-02-21 | 尚茂智能科技股份有限公司 | Intelligent vending machine |
| CN108230086B (en) * | 2017-11-30 | 2022-05-20 | 广东数相智能科技有限公司 | Commodity selling adjustment method and storage medium |
| CN108596735A (en) * | 2018-04-28 | 2018-09-28 | 北京旷视科技有限公司 | Information-pushing method, apparatus and system |
| CN109493177A (en) * | 2018-11-12 | 2019-03-19 | 深圳云行智能科技有限公司 | A kind of intelligent Matching SKU selection method and its system and storage medium |
| CN109658207A (en) * | 2019-01-15 | 2019-04-19 | 深圳友朋智能商业科技有限公司 | Method of Commodity Recommendation, system and the device of automatic vending machine |
| CN110378770A (en) * | 2019-07-25 | 2019-10-25 | 上海尊溢商务信息咨询有限公司 | A kind of method and device that e-commerce operation selects money to recommend |
| CN111626790A (en) * | 2020-06-01 | 2020-09-04 | 浪潮软件股份有限公司 | Consumer group feature identification method for retail terminal |
| CN112967109A (en) * | 2021-03-15 | 2021-06-15 | 杭州邻汇网络科技有限公司 | Intelligent flash shop product selection method |
-
2022
- 2022-01-20 CN CN202210067171.0A patent/CN114399366B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292667A (en) * | 2017-06-23 | 2017-10-24 | 北京小度信息科技有限公司 | Information-pushing method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114399366A (en) | 2022-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106384219B (en) | Storage sub-warehouse auxiliary analysis method and device | |
| CN109711890B (en) | User data processing method and system | |
| CN110070067A (en) | The training method of video classification methods and its model, device and electronic equipment | |
| CN109978918A (en) | A kind of trajectory track method, apparatus and storage medium | |
| CN110175549A (en) | Face image processing process, device, equipment and storage medium | |
| CN109145921A (en) | A kind of image partition method based on improved intuitionistic fuzzy C mean cluster | |
| WO2017138406A1 (en) | Information processing device, information processing method, and storage medium | |
| EP2894577B1 (en) | Retrieving system, retrieving method, and security inspection device based on contents of fluoroscopic images | |
| CN108537166B (en) | Method and device for determining shelf browsing amount and analyzing browsing amount | |
| CN110458195A (en) | A kind of abnormal data discriminating method based on the fusion of more criterions | |
| CN106557728B (en) | Query image processing and image search method and device and monitoring system | |
| CN108664986B (en) | Based on lpNorm regularized multi-task learning image classification method and system | |
| CN109902644A (en) | Face identification method, device, equipment and computer-readable medium | |
| CN110188796A (en) | User identification method, device, storage medium and processor | |
| CN110704659B (en) | Image list ordering method and device, storage medium and electronic device | |
| JP2003085194A (en) | Process analyzing method and information system | |
| CN112288375B (en) | Big data-based intelligent supply chain coordination management system | |
| CN109712324A (en) | A kind of automatic vending machine image-recognizing method, good selling method and vending equipment | |
| JP2019192082A (en) | Server for learning, image collection assisting system for insufficient learning, and image estimation program for insufficient learning | |
| CN107944487B (en) | Crop breeding variety recommendation method based on mixed collaborative filtering algorithm | |
| CN110490486B (en) | Enterprise big data management system | |
| CN114399366B (en) | Unmanned vending machine product selection method and system based on crowd analysis | |
| CN112131477A (en) | Library book recommendation system and method based on user portrait | |
| JP6684777B2 (en) | Good product / defective determination system and good product / defective determination method | |
| CN110378299A (en) | A kind of clothing identifying system under the conditions of indoor light |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |