A method and system for checking accuracy of a barcode scanner

Abstract

The present invention relates to the technical field of statistical analysis, and specifically discloses a method and system for checking the accuracy of a barcode scanner, comprising the following steps: step S1: setting external factors and their levels, obtaining a verification database of the barcode scanner, and selecting a sample database whose difference external factors belong to an actual level range; step S2: when each external factor is at a conventional level, the barcode scanner is recorded as being in a conventional scene, and an average recognition accuracy rate and an average recognition time are obtained; step S3: selecting an unconventional scene; obtaining a recognition quality index according to the verification result, and obtaining a recognition curve; selecting influencing factors according to the recognition curve; step S4: selecting the error-prone level of the influencing factors, and when each influencing factor is at an error-prone level, the barcode scanner is recorded as being in an error-prone scene; step S5: when the barcode scanner is verified, if the barcode scanner successfully recognizes a barcode image, the barcode scanner is recorded as qualified.

Description

Technical Field

[0001] The present invention relates to the technical field of statistical analysis, and in particular to an accuracy verification method and system for a barcode scanning gun. Background Art

[0002] A barcode scanner is a device used to scan barcodes or QR codes. It is usually used in stores, warehouses, libraries and other places to quickly identify and record goods or items. It uses technologies such as lasers or cameras to quickly and accurately read the information on the barcode or QR code and convert it into digital or text form for further processing and identification by computers or other systems.

[0003] Accuracy verification is an important step to ensure that the scanner can accurately read and identify barcodes, including scanning test barcodes, checking scanning results, multi-angle scanning and large-scale scanning tests; the above methods ensure that the scanner can accurately and reliably read and identify barcodes in actual application scenarios.

[0004] In the prior art, the recognition accuracy of a barcode scanner varies in different usage scenarios; however, the accuracy test scenario for the barcode scanner at the factory is relatively simple and cannot accurately reflect the accuracy of the barcode scanner in actual application scenarios; and the actual application scenarios are complex and changeable, making it difficult to determine a suitable test scenario for verifying the accuracy of the barcode scanner. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for verifying the accuracy of a barcode scanner to solve the above technical problems.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A method for checking the accuracy of a barcode scanner includes the following steps:

[0008] Step S1: setting external factors and their levels, and dividing the external factors into difference external factors and common external factors; obtaining a verification database of the barcode scanner, the verification database including real-time verification results and levels of external factors, the verification results including recognition results and recognition time, and the recognition results are recognition success and recognition failure;

[0009] Acquire the actual level range of the difference external factors in the actual application scenario, and select data in the verification database that the difference external factors belong to the actual level range, and record it as a sample database;

[0010] Step S2: In the sample database, select the mode of the level of each external factor and record it as the normal level of the external factor; when each external factor is at the normal level, the barcode scanner is in a normal scene, and obtain the average recognition accuracy R of the barcode scanner at this time ave and the average recognition time T ave ;

[0011] Step S3: In the sample database, select data in which one external factor is at an unconventional level and the other external factors are at a conventional level, and record it as an unconventional scene; obtain the recognition quality index according to the verification result Where R represents the current recognition accuracy, and T represents the average value of the current recognition time; according to the recognition quality index and the total number of data in the unconventional scene selected in real time, the recognition curve corresponding to the unconventional scene is constructed;

[0012] Set the standard deviation threshold St, select several monitoring points on the identification curve, and obtain the standard deviation S of the identification curve according to each monitoring point; select the external factors at the unconventional level in the unconventional scene with S ≥ St in the identification curve and record them as influencing factors;

[0013] Step S4: Obtain the average value QI of the identification quality index when the influencing factors are at different levels in the unconventional scenario ave ; Set the index threshold QIt and select QI ave The level of the influencing factors of <QIt is recorded as the error-prone level; when each influencing factor is at the error-prone level, the barcode scanner is recorded as being in an error-prone scenario;

[0014] Step S5: Setting shooting points at the scanning end of the barcode scanner to respectively obtain barcode images facing the scanning end in error-prone scenarios and conventional scenarios; when the barcode scanner is calibrated, if the barcode scanner successfully identifies the barcode image, the barcode scanner is considered qualified.

[0015] As a further solution of the present invention: in step S1, the external factors include angle, distance, light intensity and barcode quality, the difference external factors include light intensity, and the common external factors include angle, distance and barcode quality.

[0016] As a further solution of the present invention: in step S1, the process of setting the level of the external factor includes:

[0017] Obtain the degree range of the external factors, select several nodes within the degree range, and the nodes include the starting point and the end point of the range; each two adjacent nodes constitute an interval, and mark the levels of each interval as A1, A2, ..., A according to the order of the intervals. n , where n is the number of intervals.

[0018] As a further solution of the present invention: in step S3, the process of constructing the identification curve includes taking the total number of data as the horizontal axis and the identification quality index as the vertical axis.

[0019] As a further solution of the present invention: in step S3, the process of obtaining the standard deviation of the identification curve includes:

[0020] The ordinates {y1, y2, ..., y j}, y1 represents the ordinate of the first monitoring point, j is the number of monitoring points; then the average ordinate of each monitoring point is y ave , and get the standard deviation S of the identification curve:

[0021]

[0022] As a further solution of the present invention: in step S4, the process of obtaining the average value of the recognition quality index includes:

[0023] When the influencing factor is at one of the levels, the recognition quality index of each verification result is {QI1, QI2, ..., QI m}, where QI1 represents the recognition quality index of the first verification result, and m is the number of verification results when the influencing factor is at one of the levels; the average value of the recognition quality index is obtained Where i∈[1,m] and i is a positive integer.

[0024] As a further solution of the present invention: in step S5, the process of verifying the barcode scanner includes:

[0025] The barcode image obtained by the shooting point in an error-prone scene is recorded as an error-prone barcode image, and the barcode image obtained by the shooting point in a normal scene is recorded as a normal barcode image;

[0026] Set two determination coefficients k1 and k2, and set the recognition time threshold range to (T ave -ε,T ave +ε), where ε is a preset constant and ε>0;

[0027] When the barcode scanner correctly identifies the barcode in the error-prone barcode image and the recognition time falls within the recognition time threshold range, record k1=1; otherwise, record k1=0; when the barcode scanner correctly identifies the barcode in the conventional barcode image and the recognition time falls within the recognition time threshold range, record k2=1; otherwise, record k2=0; if and only if k1k2=1, the barcode scanner is qualified.

[0028] As a further solution of the present invention: a barcode scanner accuracy verification system, comprising:

[0029] Sample database acquisition module: set external factors and their levels, and divide the external factors into difference external factors and common external factors; obtain the verification database of the barcode scanner, the verification database includes real-time verification results and levels of external factors, the verification results include recognition results and recognition time, and the recognition results are recognition success and recognition failure;

[0030] Acquire the actual level range of the difference external factors in the actual application scenario, and select data in the verification database that the difference external factors belong to the actual level range, and record it as a sample database;

[0031] Conventional scene determination module: In the sample database, select the mode of the levels of each external factor and record it as the conventional level of the external factor; when each external factor is at the conventional level, the barcode scanner is recorded in the conventional scene, and the average recognition accuracy R of the barcode scanner at this time is obtained. ave and the average recognition time T ave ;

[0032] Influencing factor determination module: In the sample database, select data where one external factor is at an unconventional level and the other external factors are at a conventional level, and record it as an unconventional scenario; obtain the recognition quality index based on the verification results Where R represents the current recognition accuracy, and T represents the average value of the current recognition time; according to the recognition quality index and the total number of data in the unconventional scene selected in real time, the recognition curve corresponding to the unconventional scene is constructed;

[0033] Set the standard deviation threshold St, select several monitoring points on the identification curve, and obtain the standard deviation S of the identification curve according to each monitoring point; select the external factors at the unconventional level in the unconventional scene with S ≥ St in the identification curve and record them as influencing factors;

[0034] Error-prone scenario judgment module: In unconventional scenarios, the average value of the quality index (QI) is obtained when the influencing factors are at different levels. ave ; Set the index threshold QIt and select QI ave The level of the influencing factors of <QIt is recorded as the error-prone level; when each influencing factor is at the error-prone level, the barcode scanner is recorded as being in an error-prone scenario;

[0035] Verification module: a shooting point is set at the scanning end of the barcode scanner, and the shooting point obtains the barcode images facing the scanning end in error-prone scenarios and normal scenarios respectively; when the barcode scanner is verified, if the barcode scanner successfully recognizes the barcode image, the barcode scanner is recorded as qualified.

[0036] Beneficial effects of the present invention:

[0037] In the prior art, when the usage scenarios of the barcode scanner are different, the recognition accuracy of the barcode scanner is different in different usage environments; however, the scenario for the accuracy test of the barcode scanner at the factory is relatively single, which cannot accurately reflect the accuracy of the barcode scanner in the actual application scenario; and the actual application scenario is complex and changeable, and it is difficult to determine a suitable test scenario for verifying the accuracy of the barcode scanner; compared with the prior art, the present invention establishes a batch of test data sets, namely the verification database, in which the verification results of the barcode scanner under various levels of external factors are included; the test data is recorded in real time, and each time a verification is performed, a verification result and an external factor and its level are recorded, and a A sample database; the test data only needs to be established once, and then a sample database is selected from the sample database for analysis according to different actual usage scenarios; this method can obtain a sample database flexibly according to different actual usage scenarios of the barcode scanner; then a conventional scenario is selected according to the sample database, and the conventional scenario is the scenario that appears most frequently in the actual application scenario, and the average recognition accuracy and average recognition time of the barcode scanner in the conventional scenario are obtained according to the sample database, and the average recognition accuracy and average recognition time in the conventional scenario are used as the standard to calculate the recognition quality index of the barcode scanner in the unconventional scenario; the recognition quality index increases with the increase of the recognition accuracy, and increases with the average recognition time. The value increases and decreases, indicating the recognition quality of the barcode scanner during each verification; unconventional scenes are selected from the sample database, and the unconventional scenes are used to remove the influence of other external factors, and judge the recognition quality of the barcode scanner when there is only one external factor, and then establish a recognition curve under unconventional scenes, and the recognition curve indicates the difference in recognition quality index during each verification as the number of selected data increases; therefore, when the fluctuation of the recognition curve is large, that is, the larger the standard deviation of the recognition curve, the greater the influence of external factors at the unconventional level; therefore, when the standard deviation of the recognition curve under an unconventional scene exceeds the standard deviation threshold, the influence of external factors at the unconventional level in the unconventional scene is large , recorded as the influencing factor; then obtain the average value of the recognition quality index at each verification in the sample database under different levels of influencing factors. The higher the average value of the recognition quality index, the smaller the impact of the influencing factor at this level on the accuracy of recognition. Therefore, the level of the influencing factor whose average value of the recognition quality index is less than the index threshold is selected and recorded as the error-prone level of the influencing factor; in this way, the analysis of whether the public external factor is an influencing factor only needs to be performed once, and the analysis result can be reused in different subsequent practical application scenarios; then when the influencing factor is at the error-prone level, the barcode scanner is recorded as being in an error-prone scenario; obtain the barcode images faced by the barcode scanner in the conventional scenario and the error-prone scenario respectively as the test sample images;At a fixed angle and distance, use the barcode scanner to be verified to identify the test sample image and detect the accuracy of recognition; this method greatly improves the verification efficiency and is in line with actual application scenarios. ; BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The present invention will be further described below in conjunction with the accompanying drawings.

[0039] Figure 1 The invention is a schematic structural diagram of a method and system for checking the accuracy of a barcode scanner. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] See also Figure 1 As shown, the present invention is a method for checking the accuracy of a barcode scanner, comprising the following steps:

[0042] Step S1: setting external factors and their levels, and dividing the external factors into difference external factors and common external factors; obtaining a verification database of the barcode scanner, the verification database including real-time verification results and levels of external factors, the verification results including recognition results and recognition time, and the recognition results are recognition success and recognition failure;

[0043] Acquire the actual level range of the difference external factors in the actual application scenario, and select data in the verification database that the difference external factors belong to the actual level range, and record it as a sample database;

[0044] It should be further explained that the above process establishes a batch of test data, namely the verification database, in which the verification results of the barcode scanner under various levels of external factors are included; the test data is recorded in real time, and the verification result and the external factors and their levels are recorded once during each verification, and a sample database is obtained; the test data is established once, and a sample database is selected from the sample database for analysis according to different actual usage scenarios; this method can select a sample database flexibly according to different actual usage scenarios of the barcode scanner;

[0045] The external factors include angle, distance, light intensity and barcode quality, the difference external factors include light intensity, and the common external factors include angle, distance and barcode quality;

[0046] It can be understood that the common external factors are external factors that exist in different usage scenarios and have the same degree range, while the different external factors are external factors that have different degree ranges in different usage scenarios;

[0047] The process of setting the level of the external factors includes:

[0048] Obtain the degree range of the external factors, select several nodes within the degree range, and the nodes include the starting point and the end point of the range; each two adjacent nodes constitute an interval, and mark the levels of each interval as A1, A2, ..., A according to the order of the intervals. n , where n is the number of intervals;

[0049] It should be further explained that, for example, the range of the angle is generally (0°, 180°), and the range of indoor light intensity is generally (100lux, 1000lux); and for example, the range of the barcode quality, several different blur levels are set, and different blur levels are marked as different levels;

[0050] Step S2: In the sample database, select the mode of the level of each external factor and record it as the normal level of the external factor; when each external factor is at the normal level, the barcode scanner is in a normal scene, and obtain the average recognition accuracy R of the barcode scanner at this time ave and the average recognition time T ave ;

[0051] It can be understood that conventional scenarios are selected based on the sample database, and the conventional scenarios are the scenarios that appear most frequently in actual application scenarios. The average recognition accuracy and average recognition time of the barcode scanner in the conventional scenarios are obtained based on the sample database, and the recognition quality index of the barcode scanner in the unconventional scenarios is calculated based on the average recognition accuracy and average recognition time in the conventional scenarios;

[0052] The process of obtaining the average recognition accuracy and the average recognition time includes:

[0053] Obtain the total number of data at this time N, and the number of data with successful recognition results at this time N1, then the recognition accuracy at this time R = N1 / N*100%; then obtain the average value of the recognition accuracy of all data and record it as the average recognition accuracy; and obtain the average value of the recognition time of all data and record it as the average recognition time;

[0054] It can be understood that all data in the average recognition accuracy and average recognition time are all data in the sample database in normal scenarios;

[0055] Step S3: In the sample database, select data in which one external factor is at an unconventional level and the other external factors are at a conventional level, and record it as an unconventional scene; obtain the recognition quality index according to the verification result Where R represents the current recognition accuracy, and T represents the average value of the current recognition time; according to the recognition quality index and the total number of data in the unconventional scene selected in real time, the recognition curve corresponding to the unconventional scene is constructed;

[0056] Set the standard deviation threshold St, select several monitoring points on the identification curve, and obtain the standard deviation S of the identification curve according to each monitoring point; select the external factors at the unconventional level in the unconventional scene with S ≥ St in the identification curve and record them as influencing factors;

[0057] It is worth noting that the recognition quality index increases with the increase of recognition accuracy and decreases with the increase of the average value of recognition time, indicating the recognition quality of the barcode scanner during each verification; the selection process of the monitoring point includes setting a selection interval threshold, and selecting a monitoring point every other selection interval threshold in the recognition curve; the selection of the monitoring points is kept as dispersed and dense as possible to accurately reflect the fluctuation degree of the recognition curve; the average value of the recognition time is obtained based on the recognition time in the data selected each time and the total number of data currently selected; the recognition curve is generated in real time during the real-time selection process from the sample database;

[0058] The construction process of the identification curve includes taking the total number of data as the horizontal axis and the identification quality index as the vertical axis;

[0059] The process of obtaining the standard deviation of the identification curve includes:

[0060] The ordinates {y1, y2, ..., y j}, y1 represents the ordinate of the first monitoring point, j is the number of monitoring points; then the average ordinate of each monitoring point is y ave , and get the standard deviation S of the identification curve:

[0061]

[0062] It can be understood that the unconventional scenario is used to remove the influence of other external factors, determine the recognition quality of the barcode scanner when there is only one external factor, and then establish a recognition curve in the unconventional scenario, and the recognition curve represents the difference in the recognition quality index at each verification as the number of selected data increases; therefore, when the fluctuation of the recognition curve is large, that is, the larger the standard deviation of the recognition curve, the greater the influence of the external factors at the unconventional level; therefore, when the standard deviation of the recognition curve in the unconventional scenario exceeds the standard deviation threshold, the influence of the external factors at the unconventional level in the unconventional scenario is large, and is recorded as an influencing factor;

[0063] Step S4: Obtain the average value QI of the identification quality index when the influencing factors are at different levels in the unconventional scenario ave ; Set the index threshold QIt and select QI ave The level of the influencing factors of <QIt is recorded as the error-prone level; when each influencing factor is at the error-prone level, the barcode scanner is recorded as being in an error-prone scenario;

[0064] It should be further explained that the process of obtaining the data of each influencing factor at different levels is to select from the sample data in real time, and obtain the recognition quality index while selecting in real time; when all the data are acquired, the larger the average value of the recognition quality index is, the larger the recognition quality index of the overall data is, the higher the recognition quality is, and the smaller the influence of the influencing factors at this level on the recognition accuracy is, so the level of the influencing factor whose average value of the recognition quality index is less than the index threshold is selected and recorded as the error-prone level of the influencing factor;

[0065] The process of obtaining the average value of the recognition quality index includes:

[0066] When the influencing factor is at one of the levels, the recognition quality index of each verification result is {QI1, QI2, ..., QI m}, where QI1 represents the recognition quality index of the first verification result, and m is the number of verification results when the influencing factor is at one of the levels; the average value of the recognition quality index is obtained Where i∈[1,m] and i is a positive integer;

[0067] Step S5: setting shooting points at the scanning end of the barcode scanner, and the shooting points respectively obtain the barcode images facing the scanning end in the error-prone scene and the normal scene; when the barcode scanner is calibrated, if the barcode scanner successfully recognizes the barcode image, the barcode scanner is recorded as qualified;

[0068] It can be understood that the barcode images faced by the barcode scanner in the conventional scenario and the error-prone scenario are obtained as the test sample images respectively; the test sample images are recognized by the barcode scanner to be verified at a fixed angle and distance to detect the recognition accuracy;

[0069] It should be further explained that the process of obtaining the barcode image at the shooting point includes:

[0070] The barcode scanner is fixed, and external factors are adjusted so that the fixed barcode scanner is placed in an error-prone scene or a regular scene. The shooting point obtains image data directly in front of the scanning end of the barcode scanner, which is recorded as a barcode image.

[0071] The process of verifying the barcode scanner includes:

[0072] The barcode image obtained by the shooting point in an error-prone scene is recorded as an error-prone barcode image, and the barcode image obtained by the shooting point in a normal scene is recorded as a normal barcode image;

[0073] Set two determination coefficients k1 and k2, and set the recognition time threshold range to (T ave -ε,T ave +ε), where ε is a preset constant and ε>0;

[0074] When the barcode scanner correctly identifies the barcode in the error-prone barcode image and the recognition time falls within the recognition time threshold range, record k1=1; otherwise, record k1=0; when the barcode scanner correctly identifies the barcode in the conventional barcode image and the recognition time falls within the recognition time threshold range, record k2=1; otherwise, record k2=0; if and only if k1k2=1, the barcode scanner is qualified.

[0075] An accuracy verification system for a barcode scanner, comprising:

[0076] Sample database acquisition module: set external factors and their levels, and divide the external factors into difference external factors and common external factors; obtain the verification database of the barcode scanner, the verification database includes real-time verification results and the levels of external factors, and the verification results include recognition accuracy, recognition time and total number of data;

[0077] Acquire the actual level range of the difference external factors in the actual application scenario, and select data in the verification database that the difference external factors belong to the actual level range, and record it as a sample database;

[0078] Conventional scene determination module: In the sample database, select the mode of the levels of each external factor and record it as the conventional level of the external factor; when each external factor is at the conventional level, the barcode scanner is recorded in the conventional scene, and the average recognition accuracy R of the barcode scanner at this time is obtained.ave and the average recognition time T ave ;

[0079] Influencing factor determination module: In the sample database, select data where one external factor is at an unconventional level and the other external factors are at a conventional level, and record it as an unconventional scenario; obtain the recognition quality index based on the verification results Where R represents the current recognition accuracy, and T represents the average value of the current recognition time; according to the real-time recognition quality index and the total number of data in the unconventional scene, the recognition curve corresponding to the unconventional scene is constructed;

[0080] Set the standard deviation threshold St, select several monitoring points on the identification curve, and obtain the standard deviation S of the identification curve according to each monitoring point; select the external factors at the unconventional level in the unconventional scene with S ≥ St in the identification curve and record them as influencing factors;

[0081] Error-prone scenario judgment module: In unconventional scenarios, the average value of the quality index (QI) is obtained when the influencing factors are at different levels. ave ; Set the index threshold QIt and select QI ave The level of the influencing factors ≥QIt is recorded as the error-prone level; when each influencing factor is at the error-prone level, the barcode scanner is recorded as being in an error-prone scenario;

[0082] Verification module: a shooting point is set at the scanning end of the barcode scanner, and the shooting point obtains the barcode images facing the scanning end in error-prone scenarios and normal scenarios respectively; when the barcode scanner is verified, if the barcode scanner successfully recognizes the barcode image, the barcode scanner is recorded as qualified.

[0083] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for verifying the accuracy of a barcode scanner, characterized in that: The following steps are involved: Step S1: setting external factors and their levels, and dividing the external factors into differential external factors and common external factors; obtaining a verification database of the barcode scanner, the verification database including real-time verification results and levels of external factors, the verification results including recognition results and recognition time, and the recognition results including recognition success and recognition failure; Obtaining the actual level range of the difference external factors in the actual application scenario, and selecting data in the verification database whose difference external factors fall within the actual level range, and recording the data as a sample database; Step S2: In the sample database, select the mode of the level of each external factor and record it as the normal level of the external factor; when each external factor is at the normal level, the barcode scanner is considered to be in a normal scene, and the average recognition accuracy R of the barcode scanner at this time is obtained. ave and the average recognition time T ave ; Step S3: In the sample database, select data in which one external factor is at an unconventional level and the other external factors are at a conventional level, and record it as an unconventional scenario; obtain the recognition quality index based on the verification result , where R represents the current recognition accuracy, and T represents the average value of the current recognition time; based on the recognition quality index and the total number of data in the unconventional scene selected in real time, the recognition curve corresponding to the unconventional scene is constructed; Set the standard deviation threshold St, select several monitoring points on the recognition curve, and obtain the standard deviation S of the recognition curve based on each monitoring point; Select the external factors at the unconventional level in the unconventional scenario of S≥St in the identification curve and record them as influencing factors; Step S4: Obtain the average value of the quality index QI when the influencing factors are at different levels in the unconventional scenario ave ; Set the index threshold QIt and select QI ave The level of the factors affecting QIt is recorded as the error-prone level. When each influencing factor is at the error-prone level, the barcode scanner is recorded as being in an error-prone scenario. Step S5: Setting shooting points at the scanning end of the barcode scanner to respectively capture barcode images of the scanning end in error-prone scenarios and normal scenarios; when the barcode scanner is calibrated, if the barcode scanner successfully recognizes the barcode image, the barcode scanner is considered qualified; The external factors include angle, distance, light intensity and barcode quality, the difference external factors include light intensity, and the common external factors include angle, distance and barcode quality.

2. The accuracy verification method for a barcode scanner according to claim 1, characterized in that: In step S1, the process of setting the level of the external factors includes: Obtain the degree range of the external factors, select several nodes within the degree range, and the nodes include the starting point and the end point of the range; each two adjacent nodes form an interval, and according to the order of the intervals, mark the levels of each interval as A1, A2, ..., A n , where n is the number of intervals.

3. The accuracy verification method for a barcode scanner according to claim 1, characterized in that: In step S2, the process of obtaining the average recognition accuracy and the average recognition time includes: Get the total number of data at this time N, and the number of data with successful recognition results N1 at this time, then the current recognition accuracy R=N1 / N*100%; then get the average value of the recognition accuracy of all data and record it as the average recognition accuracy; and get the average value of the recognition time of all data and record it as the average recognition time.

4. The accuracy verification method for a barcode scanner according to claim 1, characterized in that: In step S3, the process of constructing the recognition curve includes taking the total number of data as the horizontal axis and the recognition quality index as the vertical axis.

5. The accuracy verification method for a barcode scanner according to claim 4, characterized in that: In step S3, the process of obtaining the standard deviation of the identification curve includes: Obtain the vertical coordinates {y1, y2, ..., y j }, y1 represents the vertical coordinate of the first monitoring point, j is the number of monitoring points; the average value of the vertical coordinates of each monitoring point is y ave , and obtain the standard deviation S of the identification curve: 。 6. The accuracy verification method for a barcode scanner according to claim 1, characterized in that: In step S4, the process of obtaining the average value of the recognition quality index includes: When the influencing factor is at one of the levels, the recognition quality index of each verification result is {QI1, QI2, ..., QI m }, where QI1 represents the recognition quality index of the first verification result, and m is the number of verification results when the influencing factor is at one level; the average value of the recognition quality index is obtained , where i∈[1,m] and i is a positive integer.

7. The accuracy verification method for a barcode scanner according to claim 1, characterized in that: In step S5, the process of verifying the barcode scanner includes: The barcode image obtained by the shooting point in an error-prone scene is recorded as an error-prone barcode image, and the barcode image obtained by the shooting point in a normal scene is recorded as a normal barcode image; Set two determination coefficients k1 and k2, and set the recognition time threshold range to (T ave -ε,T ave +ε), where ε is a preset constant and ε>0; When the barcode scanner correctly identifies the barcode in the error-prone barcode image and the recognition time falls within the recognition time threshold range, record k1=1; otherwise, record k1=0; when the barcode scanner correctly identifies the barcode in the regular barcode image and the recognition time falls within the recognition time threshold range, record k2=1; otherwise, record k2=0; if and only if k1k2=1, the barcode scanner is qualified.

8. The system for verifying the accuracy of a barcode scanner according to claim 1, comprising: Sample database collection module: setting external factors and their levels, and dividing the external factors into differential external factors and common external factors; Obtaining a verification database of the barcode scanner, the verification database including real-time verification results and levels of external factors, the verification results including recognition results and recognition time, and the recognition results including recognition success and recognition failure; Obtaining the actual level range of the difference external factors in the actual application scenario, and selecting data in the verification database whose difference external factors fall within the actual level range, and recording the data as a sample database; Conventional scene determination module: In the sample database, select the mode of the level of each external factor and record it as the conventional level of the external factor; when each external factor is at the conventional level, the barcode scanner is recorded in the conventional scene, and the average recognition accuracy R of the barcode scanner at this time is obtained. ave and the average recognition time T ave ; Influencing factor determination module: In the sample database, select the data where one external factor is at an unconventional level and the other external factors are at a conventional level, and record it as an unconventional scenario; obtain the recognition quality index based on the verification results , where R represents the current recognition accuracy, and T represents the average value of the current recognition time; based on the recognition quality index and the total number of data in the unconventional scene selected in real time, the recognition curve corresponding to the unconventional scene is constructed; Set the standard deviation threshold St, select several monitoring points on the recognition curve, and obtain the standard deviation S of the recognition curve based on each monitoring point; Select the external factors at the unconventional level in the unconventional scenario of S≥St in the identification curve and record them as influencing factors; Error-prone scenario judgment module: obtains the average value of the quality index (QI) when the influencing factors are at different levels in unconventional scenarios ave ; Set the index threshold QIt and select QI ave The level of the factors affecting QIt is recorded as the error-prone level. When each influencing factor is at the error-prone level, the barcode scanner is recorded as being in an error-prone scenario. Verification module: A shooting point is set at the scanning end of the barcode scanner, and the shooting point respectively captures the barcode image facing the scanning end in error-prone scenarios and normal scenarios; when the barcode scanner is verified, if the barcode scanner successfully recognizes the barcode image, the barcode scanner is recorded as qualified; The external factors include angle, distance, light intensity and barcode quality, the difference external factors include light intensity, and the common external factors include angle, distance and barcode quality.

Images