Intelligent identification system and identification method for confiscated tobacco warehouse based on QR code

By collecting and analyzing warehouse data, a neural network model is constructed to determine the best factors for QR code identification, which solves the problem of factors affected by the recognition success rate and speed in the prior art, and achieves efficient identification of items entering and leaving the warehouse.

CN118446235BActive Publication Date: 2024-12-06SHENZHEN HONGTUWULIAN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410493514.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-23
Publication Date
2024-12-06
Estimated Expiration
2044-04-23

AI Technical Summary

Technical Problem

During the identification process, the existing QR code recognition system is affected by factors such as the distance, angle, relative position and identification environment brightness between the scanning device and the QR code, resulting in a decrease in the recognition success rate and speed, affecting the entry and exit of the item. process.

Method used

By collecting warehouse historical data, randomly selecting samples for QR code scanning and identification, determining the identifiable threshold of influencing factors, and building a neural network model to obtain the best value of the scanning time and influencing factors, adjusting the scanning device to achieve the best recognition state .

Benefits of technology

It improves the recognition efficiency of items entering and leaving the warehouse, ensures the optimal duration of the identification process, and improves the efficiency of the item process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118446235B_ABST
    Figure CN118446235B_ABST
Patent Text Reader

Abstract

The present invention discloses a confiscated tobacco warehouse intelligent identification system and identification method based on QR codes, and relates to the field of intelligent identification technology. By collecting historical in-and-out data in the warehouse, information on tobacco product types is obtained, at least one number of samples are randomly selected from each type of tobacco product as test samples, and the QR code positions of each type of test sample are scanned and identified by a scanning device, and identification result information of the test sample in different influencing factors and test sample scanning time data are obtained, and the scanning time is divided into a high-speed identification area and a low-speed identification area according to a threshold value, and the influencing factor data of each type of tobacco product test sample when the scanning time is in the high-speed identification area is obtained, and a neural network model is constructed, and a data set of optimal values ​​of scanning time and influencing factors is obtained according to the model, and the scanning device is adjusted to the optimal value state of various influencing factors in the process of identifying the tobacco product.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of intelligent identification technology, and in particular to an intelligent identification system for a confiscated tobacco warehouse based on a two-dimensional code and an identification method thereof. Background Art

[0002] Confiscated tobacco refers to cigarette products purchased or confiscated by tobacco management departments in accordance with the law. According to relevant management regulations, confiscated tobacco needs to be stored in designated warehouses. Generally, the outer packaging of finished confiscated tobacco products is printed with a QR code or barcode that stores information about the product. When items enter or leave the warehouse, the type of item can be determined by identifying the identification code, so as to formulate storage or outbound strategies.

[0003] When the existing QR code recognition system is performing recognition operations, the success rate of the recognition result is directly related to the speed of recognition, which is affected by factors including the distance, angle, relative position between the scanning device and the QR code, and the brightness of the recognition environment. When any of the influencing factors deviates from the optimal recognition state, the success rate and speed of recognition will be reduced, affecting the process of goods entering and leaving the warehouse. To this end, we propose a QR code-based intelligent recognition system for confiscated tobacco goods warehouses and its recognition method. Summary of the invention

[0004] The main purpose of the present invention is to provide a confiscated tobacco goods warehouse intelligent identification system and identification method based on QR codes, which can effectively solve the problems in the background technology.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] The intelligent identification method of the warehouse of confiscated tobacco items based on QR code includes:

[0007] Collect historical warehouse in-and-out data, obtain tobacco product category information, and randomly select at least one number of samples from each type of tobacco product as test samples;

[0008] Scan and identify the QR code positions of various test samples by scanning the scanning device to obtain the recognition result information of the test samples under different influencing factors, wherein the influencing factors include recognition distance, recognition angle, recognition environment brightness, and cleanliness of the scanning device, and the recognition result includes recognition failure status and recognition success status;

[0009] Determine the identifiable threshold of the influencing factor according to the obtained identification result information [S ijmin , S ijmax ], where S ijmin is the minimum value that the scanning device can identify for the jth influencing factor of the i-th type of test sample, S ijmaxis the maximum value that the scanning device can identify for the jth influencing factor of the i-th type of test sample;

[0010] Obtaining scanning time data of various tobacco article test samples when the recognition result is a recognition success state, and dividing the scanning time into two categories of a high-speed recognition area and a low-speed recognition area according to a threshold value;

[0011] Obtaining the influencing factor data of various tobacco article test samples when the scanning time is in the high-speed recognition area, taking the obtained normalized data of the influencing factors as input, and taking the corresponding normalized data of the scanning time as output to build a neural network model, and obtaining a data set of optimal values ​​of the scanning time and the influencing factors according to the model;

[0012] According to the acquired data set, the scanning device is adjusted to the optimal value state for various influencing factors in the tobacco article identification process.

[0013] An intelligent identification system for confiscated tobacco product warehouses based on QR codes, the system comprising a tobacco product type information collection module, a scanning device, a data collection module, a data analysis module, a data classification module, a central control module, and an equipment control module;

[0014] The tobacco article type information collection module is used to collect historical in-and-out warehouse data in the warehouse, obtain the in-and-out tobacco article type information, and randomly select at least one number of samples from each type of tobacco article as test samples;

[0015] The data acquisition module is connected to the scanning device and is used to obtain the recognition result information of the test sample under different influencing factors, wherein the influencing factors include recognition distance, recognition angle, recognition environment brightness, and scanning device cleanliness, and the recognition result includes recognition failure status and recognition success status;

[0016] The data analysis module is connected to the data acquisition module and is used to determine the identifiable threshold of the influencing factor according to the acquired identification result information;

[0017] The data classification module is connected to the data analysis module and the data acquisition module, and is used to divide the scanning time data of various tobacco article test samples into two categories, high-speed recognition area and low-speed recognition area, according to the threshold value when the recognition result is a recognition success state;

[0018] The central control module is connected to the data classification module and the data acquisition module, and is used to obtain the influencing factor data of various tobacco product test samples when the scanning time is in the high-speed recognition area, and to construct a neural network model with the obtained normalized data of the influencing factors as input and the corresponding normalized data of the scanning time as output, and to obtain a data set of optimal values ​​of the scanning time and the influencing factors according to the model;

[0019] The device control module is connected to the central control module and the scanning device, and is used to generate a control instruction for driving the scanning device according to the data set of the scanning duration and the optimal value of the influencing factor acquired by the central control module;

[0020] The scanning device comprises a scanning unit, a distance adjustment unit, an angle adjustment unit, an ambient brightness adjustment unit, a cleanliness adjustment unit, and a signal transceiver unit. The scanning unit is used to scan the QR code position of the tobacco article. The signal transceiver unit is used to receive the control instruction of the device control module. In response to the control instruction, the distance adjustment unit, the angle adjustment unit, the ambient brightness adjustment unit, and the cleanliness adjustment unit respectively adjust the influencing factors in the tobacco article identification process to the optimal value state;

[0021] The system further comprises a data storage module, which is connected to the tobacco article type information collection module and the data collection module and is used to store the acquired tobacco article type information and influencing factors and identification result information data;

[0022] The system also includes a memory, a processor, and an electronic program stored in the memory and capable of running on the processor.

[0023] The present invention has the following beneficial effects:

[0024] Compared with the prior art, the technical solution of the present invention obtains tobacco product category information by collecting historical in-and-out warehouse data in the warehouse, randomly selects at least one number of samples from each type of tobacco product as test samples, scans and identifies the QR code positions of each type of test sample through a scanning device, and obtains recognition result information of the test sample under different influencing factors, wherein the influencing factors include recognition distance, recognition angle, recognition environment brightness, and scanning device cleanliness; the recognition result includes a recognition failure state and a recognition success state; the identifiable threshold of the influencing factor is determined according to the obtained recognition result information, and the recognition result when the recognition result is a recognition success state is obtained. The scanning time data of various tobacco product test samples are obtained, the scanning time is divided into two categories, high-speed recognition area and low-speed recognition area, according to the threshold, and the influencing factor data of various tobacco product test samples when the scanning time is in the high-speed recognition area is obtained, and the normalized data of the influencing factors is used as input, and the corresponding normalized data of the scanning time is used as output to build a neural network model, and a data set of optimal values ​​of scanning time and influencing factors is obtained according to the model, and the scanning device is adjusted to the optimal value state for various influencing factors in the process of tobacco product identification according to the obtained data set, so as to achieve the optimal scanning time state for scanning tobacco products in and out of the warehouse, thereby improving the efficiency of goods in and out of the warehouse. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flow chart of the intelligent identification method of the warehouse of confiscated tobacco articles based on QR codes of the present invention;

[0026] Figure 2 The present invention is a block diagram of the overall structure of the intelligent identification system for confiscated tobacco warehouses based on QR codes. DETAILED DESCRIPTION

[0027] The present invention will be further described below in conjunction with specific implementation methods, wherein the accompanying drawings are only used for exemplary descriptions and represent only schematic diagrams rather than actual drawings, and should not be understood as limiting the present invention. In order to better illustrate the specific implementation methods of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product.

[0028] Example 1

[0029] like Figure 1 The flowchart of the intelligent identification method for confiscated tobacco warehouse based on QR code and Figure 2 The overall structural block diagram of the confiscated tobacco items warehouse intelligent identification system based on QR code is shown.

[0030] The specific implementation process of the technical solution of the present invention includes the following steps:

[0031] Step 1: Collect historical in-and-out warehouse data in the warehouse through the tobacco article type information collection module, obtain the in-and-out tobacco article type information, and randomly select at least one number of samples from each type of tobacco article as test samples.

[0032] Step two, use a scanning device to scan and identify the QR code position of each test sample under different influencing factors, and obtain the recognition result information of the test sample under different influencing factors through the data acquisition module, where the influencing factors include recognition distance, recognition angle, recognition environment brightness, and scanning equipment cleanliness. The recognition results include recognition failure status and recognition success status.

[0033] Step 3: Determine the identifiable threshold of the influencing factor based on the obtained identification result information through the data analysis module [S ijmin , S ijmax ], where S ijmin is the minimum value that the scanning device can identify for the jth influencing factor of the i-th type of test sample, S ijmax The maximum value that the scanning device can identify for the jth influencing factor of the i-th type of test sample is as follows:

[0034] Step 31, obtaining all influencing factor data and recognition result information of the test sample during the scanning and recognition process;

[0035] Step 32, classifying the acquired data into recognition failure state data and recognition success state data according to the state of the recognition result;

[0036] Step 33, using the control variable method to respectively filter the data of any influencing factor in the recognition failure state data and the recognition success state data in different states, and construct a threshold analysis data set;

[0037] Step 34, arranging the influencing factor data of different states in the threshold analysis data set in order of size, and selecting the data points where the recognition results change as the value points of the identifiable threshold;

[0038] Step 35, repeat the above steps 33 and 34 until the identifiable thresholds of all influencing factors are determined.

[0039] Step 4: When the recognition result is a successful recognition state, the scanning time data of each type of tobacco product test sample is divided into two categories: a high-speed recognition area and a low-speed recognition area according to the threshold value through the data classification module. The threshold values ​​of the high-speed recognition area and the low-speed recognition area are respectively Among them, v imax is the maximum scanning time when the recognition result in the i-th test sample is a successful recognition state, v imin It is the minimum scanning time when the recognition result of the i-th test sample is a successful recognition state.

[0040] Step 5: The influencing factor data of various tobacco product test samples when the scanning time is in the high-speed recognition area is obtained through the central control module, and the normalized data of the influencing factors is used as input, and the normalized data of the corresponding scanning time is used as output to build a neural network model, and a data set of the optimal value of the scanning time and the influencing factor is obtained according to the model. Taking the construction of the BP neural network model using MATLAB as an example, the specific steps are:

[0041] Step 51, obtaining the influencing factor data of various tobacco article test samples when the scanning time is within the high-speed recognition area;

[0042] Step 52, normalize the acquired data, and the normalization formula is:

[0043]

[0044] In the formula, X is the normalized data; x is the acquired source data; x min is the minimum value of the source data obtained; x max is the maximum value of the source data obtained;

[0045] Step 53, select 85% of the data as training data and 15% of the data as test data, and train the model with the selected training data to obtain a neural network model, wherein the calculation formula for the number of hidden layer nodes of the model is:

[0046]

[0047] Where s is the number of nodes in the hidden layer; m is the number of nodes in the input layer, and n is the number of nodes in the output layer. In this model, m is equal to the number of influencing factor types, n = 1; α is an integer from 1 to 9;

[0048] Step 54, set the target error to 0.01, the maximum number of cycles to 1000, select the S-type tangent function tans ig as the transfer function of the hidden layer neurons, select the S-type logarithmic function logs ig as the transfer function of the output layer neurons, and select the traindx function as the training function;

[0049] Step 54, by analyzing the error, determine the final structure of the model, and output the model to predict the scanning time and the optimal value of the influencing factors, and use the model output data to build a data set (d i' , a i' , l i' , r i' , v i' ), where v i' , d i' , a i' , l i' , r i' They are the scanning time value, optimal recognition distance value, optimal recognition angle value, optimal recognition environment brightness value, and optimal scanning equipment cleanliness value obtained by the model. Among them, the scanning time value obtained by the model is within the high-speed recognition area threshold. within the range.

[0050] Step 6: When tobacco products need to be put in and taken out of the warehouse, the data set of the best value of the scanning time and the influencing factors (d i' , a i' , l i' , r i' , v i' ) is used as the basis for adjustment to determine the optimal value of the recognition distance, recognition angle, recognition environment brightness, and scanning equipment cleanliness of this type of tobacco products. The equipment control module generates control instructions for driving the scanning equipment based on the data set of the optimal values ​​of each influencing factor.

[0051] Step seven, the signal transceiver unit receives the control instruction of the device control module, and in response to the control instruction, the distance adjustment unit, angle adjustment unit, ambient brightness adjustment unit, and cleanliness adjustment unit of the scanning device respectively adjust the parameters in the tobacco article identification process, including the distance, angle, and ambient brightness between the scanning device and the tobacco article to be scanned, and the cleanliness of the scanning camera of the scanning device, so as to adjust various influencing factors to the optimal value state, so as to achieve the optimal scanning time state for scanning tobacco articles in and out of the warehouse, thereby improving the efficiency of article in and out of the warehouse.

[0052] In summary, the QR code-based intelligent identification system and identification method for confiscated tobacco warehouses of the present invention is to obtain tobacco product category information by collecting historical in-and-out data in the warehouse, randomly select at least one number of samples from each type of tobacco product as test samples, and use a scanning device to scan and identify the QR code positions of each type of test sample to obtain the identification result information of the test sample under different influencing factors, wherein the influencing factors include identification distance, identification angle, identification environment brightness, and scanning device cleanliness, and the identification result includes an identification failure state and an identification success state, and the identifiable threshold of the influencing factor is determined according to the obtained identification result information, and the identification result of the test sample under different influencing factors is obtained. The recognition result is the scanning time data of various tobacco product test samples when the recognition is successful. The scanning time is divided into two categories: high-speed recognition area and low-speed recognition area according to the threshold. The influencing factor data of various tobacco product test samples when the scanning time is in the high-speed recognition area is obtained, and the normalized data of the influencing factors is used as input, and the corresponding normalized data of the scanning time is used as output to build a neural network model. According to the model, a data set of optimal values ​​of scanning time and influencing factors is obtained, and according to the obtained data set, the scanning device is adjusted to the optimal value state for various influencing factors in the tobacco product recognition process, so as to achieve the optimal scanning time state for scanning tobacco products in and out of the warehouse, thereby improving the efficiency of goods in and out of the warehouse.

[0053] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. Intelligent identification method of confiscated tobacco warehouse based on QR code, It is characterized in that include: Collect historical warehouse in-and-out data, obtain tobacco product category information, and randomly select at least one number of samples from each type of tobacco product as test samples; Scan and identify the QR code positions of various test samples by scanning the scanning device to obtain the recognition result information of the test samples under different influencing factors, wherein the influencing factors include recognition distance, recognition angle, recognition environment brightness, and cleanliness of the scanning device, and the recognition result includes recognition failure status and recognition success status; Determine the identifiable threshold of the influencing factor based on the obtained identification result information Among them, S ijmin is the minimum value that the scanning device can identify for the jth influencing factor of the i-th type of test sample, S ijmax is the maximum value that the scanning device can identify for the jth influencing factor of the i-th type of test sample; Obtaining scanning time data of various tobacco article test samples when the recognition result is a recognition success state, and dividing the scanning time into two categories of a high-speed recognition area and a low-speed recognition area according to a threshold value; Obtaining the influencing factor data of various tobacco article test samples when the scanning time is in the high-speed recognition area, taking the obtained normalized data of the influencing factors as input, and taking the corresponding normalized data of the scanning time as output to build a neural network model, and obtaining a data set of optimal values ​​of the scanning time and the influencing factors according to the model; According to the acquired data set, the scanning device is adjusted to the optimal value state for various influencing factors in the tobacco article recognition process, and the thresholds of the high-speed recognition area and the low-speed recognition area are respectively Among them, v imax is the maximum scanning time when the recognition result in the i-th test sample is a successful recognition state, v imin is the minimum scanning time when the recognition result in the i-th test sample is a successful recognition state. The method for determining the identifiable threshold of the influencing factor is: Step s31, obtaining all influencing factor data and recognition result information of the test sample during the scanning and recognition process; Step s32, classifying the acquired data into recognition failure state data and recognition success state data according to the state of the recognition result; Step s33, using a control variable method to respectively filter the data of the influencing factors in any one of the recognition failure state data and the recognition success state data as data of different states, and construct a threshold analysis data set; Step s34, arranging the influencing factor data of different states in the threshold analysis data set in order of size, and selecting the data points where the recognition results change as the value points of the identifiable threshold; Step s35, repeating the above steps s33 and s34 until the identifiable thresholds of all influencing factors are determined.

2. The intelligent identification method for confiscated tobacco warehouses based on QR codes according to claim 1, It is characterized in that The calculation formula for the number of hidden layer nodes of the neural network model is: Where s is the number of nodes in the hidden layer; m is the number of nodes in the input layer, and n is the number of nodes in the output layer. In this model, m is equal to the number of influencing factor types, n = 1; α is an integer from 1 to 9.

3. The method for intelligently identifying warehouses of confiscated tobacco items based on QR codes according to claim 1, It is characterized in that The normalization formula for the input and output data of the neural network model is: In the formula, X is the normalized data; x is the acquired source data; x min is the minimum value of the source data obtained; x max The maximum value in the source data obtained.

4. The method for intelligently identifying warehouses of confiscated tobacco items based on QR codes according to claim 1, It is characterized in that The data set of scanning time and optimal value of influencing factors is recorded as (d i' , a i' , l i' , r i' , v i' ), where v i' , d i' , a i' , l i' , r i' They are respectively the scanning time value, the optimal recognition distance value, the optimal recognition angle value, the optimal recognition environment brightness value, and the optimal scanning equipment cleanliness value obtained through the model.

5. The intelligent identification method for confiscated tobacco warehouses based on QR codes according to claim 1, It is characterized in that The scanning time value obtained by the model is within the high-speed recognition area threshold within the range.

6. Intelligent identification system for confiscated tobacco warehouses based on QR codes, It is characterized in that The system includes a tobacco article type information collection module, a scanning device, a data collection module, a data analysis module, a data classification module, a central control module, and an equipment control module; The tobacco article type information collection module is used to collect historical in-and-out warehouse data in the warehouse, obtain the in-and-out tobacco article type information, and randomly select at least one number of samples from each type of tobacco article as test samples; The data acquisition module is connected to the scanning device and is used to obtain the recognition result information of the test sample under different influencing factors, wherein the influencing factors include recognition distance, recognition angle, recognition environment brightness, and scanning device cleanliness, and the recognition result includes recognition failure status and recognition success status; The data analysis module is connected to the data acquisition module and is used to determine the identifiable threshold of the influencing factor according to the acquired identification result information; The data classification module is connected to the data analysis module and the data acquisition module, and is used to divide the scanning time data of various tobacco article test samples into two categories, high-speed recognition area and low-speed recognition area, according to the threshold value when the recognition result is a recognition success state; The central control module is connected to the data classification module and the data acquisition module, and is used to obtain the influencing factor data of various tobacco product test samples when the scanning time is in the high-speed recognition area, and to construct a neural network model with the obtained normalized data of the influencing factors as input and the corresponding normalized data of the scanning time as output, and to obtain a data set of optimal values ​​of the scanning time and the influencing factors according to the model; The device control module is connected to the central control module and the scanning device, and is used to generate a control instruction for driving the scanning device according to the data set of the scanning duration and the optimal value of the influencing factor acquired by the central control module; The scanning device comprises a scanning unit, a distance adjustment unit, an angle adjustment unit, an ambient brightness adjustment unit, a cleanliness adjustment unit, and a signal transceiver unit. The scanning unit is used to scan the QR code position of the tobacco article. The signal transceiver unit is used to receive the control instruction of the device control module. In response to the control instruction, the distance adjustment unit, the angle adjustment unit, the ambient brightness adjustment unit, and the cleanliness adjustment unit respectively adjust the influencing factors in the tobacco article identification process to the optimal value state. The method for determining the identifiable threshold of the influencing factors is: Step s31, obtaining all influencing factor data and recognition result information of the test sample during the scanning and recognition process; Step s32, classifying the acquired data into recognition failure state data and recognition success state data according to the state of the recognition result; Step s33, using a control variable method to respectively filter the data of the influencing factors in any one of the recognition failure state data and the recognition success state data as data of different states, and construct a threshold analysis data set; Step s34, arranging the influencing factor data of different states in the threshold analysis data set in order of size, and selecting the data points where the recognition results change as the value points of the identifiable threshold; Step s35, repeating the above steps s33 and s34 until the identifiable thresholds of all influencing factors are determined.

7. The QR code-based intelligent identification system for confiscated tobacco warehouses according to claim 6, It is characterized in that The system further comprises a data storage module, which is connected to the tobacco article type information collection module and the data collection module and is used to store the acquired tobacco article type information and influencing factors and identification result information data; The system further comprises a memory, a processor and an electronic program stored in the memory and capable of running on the processor, wherein the processor can implement the steps of the method according to any one of claims 1 to 5 when running the electronic program.