A scenic spot locker management system and method based on face recognition

Abstract

The application discloses a scenic spot locker management system and method based on face recognition, and relates to the technical field of lockers. The management method comprises the following steps: setting face recognition rules to evaluate the similarity of each recognition record, and generating a recognition log of each user; dividing each recognition record in any recognition log into an abnormal part; extracting the difference characteristics of the recognition log, analyzing the occurrence frequency and similarity evaluation of the difference characteristics in each recognition log, and obtaining the influence degree of each difference characteristic; generating a recognition record and evaluating the similarity whenever a user performs face recognition, if the obtained similarity is lower than a similarity threshold, capturing the relevant information of the user about each difference characteristic, obtaining several difference characteristics satisfied by the user, adjusting the similarity, and if the adjusted similarity is higher than the set similarity threshold, providing adjustment suggestions to the user and performing re-recognition.

Description

Technical Field

[0001] This invention relates to the field of locker technology, specifically a scenic area locker management system and method based on facial recognition. Background Technology

[0002] Facial recognition lockers are an intelligent storage solution that uses advanced facial recognition technology to securely store and retrieve items. They not only improve the convenience of the storage process but also enhance the security of stored items, and are gradually being used in various settings.

[0003] Although it has advantages over traditional recognition methods, such as faster access and no dependence on the object being recognized, it also creates new problems. For example, when recognizing faces, it can be affected by various factors, including but not limited to lighting and facial decorations that obscure the face, causing face recognition to fail. Once recognition fails, it may continue to fail, affecting the user's access efficiency and mood, and significantly reducing the user's trust in the lockers. Summary of the Invention

[0004] The purpose of this invention is to provide a scenic area locker management system and method based on facial recognition, so as to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for managing lockers in scenic areas based on facial recognition, the method comprising the following steps:

[0006] Step S100: Establish a facial recognition management system in the lockers of the scenic area, record each facial recognition process of the lockers, and obtain several recognition records; based on the facial feature comparison during the facial recognition process, calculate the similarity of each recognition record, and determine the similarity threshold for judging whether the facial recognition is successful; the recognition records store information such as timestamps, the set of facial features entered by the user, the currently captured user facial image, recognition results, and matching degree;

[0007] Step S200: Generate recognition logs for each user based on the facial features stored in the face recognition management system; classify each recognition record in any recognition log as anomaly; extract the difference features of the recognition logs by comparing the differences between each recognition record; the recognition logs contain all recognition records with the same set of facial features entered by the user.

[0008] Step S300: Analyze the frequency of occurrence of differential features in each recognition log, classify each differential feature; compare the differential features and similarity assessment between each recognition record in each recognition log, and obtain the degree of influence of each differential feature on face recognition;

[0009] Step S400: Whenever a user performs face recognition, a recognition record is generated and a similarity assessment is performed. If the obtained similarity is lower than the similarity threshold, relevant information about each difference feature of the user is captured to obtain several difference features that the user meets. The similarity is adjusted according to the influence degree of the several difference features. If the adjusted similarity is higher than the similarity threshold, adjustment suggestions are provided to the user and recognition is performed again.

[0010] Furthermore, step S100 includes the following steps:

[0011] Step S101: Extract features from the user's facial image recorded when the user uses the locker for the first time to obtain the user's facial feature set and store it in the face recognition management system; randomly select a recognition record and compare the user's facial image captured in the recognition record with the pre-stored facial feature set to obtain the similarity S of the recognition record.

[0012] Step S102: Obtain the recognition results of each recognition record, mark the recognition records that failed to be recognized as having abnormal features; divide all recognition records according to whether there are abnormal feature marks to obtain a normal record set and an abnormal record set;

[0013] Step S103: Obtain the similarity of each identified record in the normal record set and the abnormal record set respectively, and extract the normal similarity threshold with the smallest value in the normal record set (S1). min The anomaly similarity threshold (S2) with the largest value in the set of anomaly records. max If (S1) min ≤(S2) max Then the normal similarity threshold (S1) will be applied. min As a similarity threshold for judging whether face recognition is successful, if (S1) min >(S2) max Then the abnormal similarity threshold (S2) will be set. max The similarity threshold is used to determine whether face recognition is successful. The similarity of the normal record set is the similarity range for successful recognition, while the similarity of the abnormal record set is the similarity range for failed recognition. Taking an appropriate value between the two similarity ranges will yield a reasonable similarity threshold to determine whether face recognition is successful.

[0014] Furthermore, step S200 includes the following steps:

[0015] Step S201: Obtain the facial feature set of each recognition record that is compared with the user's facial image. If two recognition records have the same facial feature set, then the two recognition records are summarized to generate a recognition log. Compare the facial feature sets of all recognition records to generate the recognition log corresponding to each user.

[0016] Step S202: Randomly select an identification log, divide each identification record in the identification log into a normal record set and an abnormal record set; extract one identification record from each set, extract key features from the user facial images captured in the two identification records, and obtain the key feature sets of the two identification records respectively.

[0017] Step S203: Preset several appearance types for the facial feature set stored in each recognition record, compare the key features in the two key feature sets with each appearance type to determine the appearance type corresponding to each key feature; summarize the several appearance types corresponding to the two key feature sets to obtain a appearance type set; appearance features include the shape of facial features, whether or not accessories are worn, etc. For example, if a user wears glasses when recording a facial image, wearing glasses can be considered as a appearance feature of the user;

[0018] Step S204: Randomly select one appearance type from the appearance type set, and select the key features corresponding to the appearance type from the two key feature sets respectively. Perform a difference comparison on the two selected key features to obtain the degree of difference between the two key features. Set a difference degree threshold. If the difference degree exceeds the difference degree threshold, set the appearance type corresponding to the two key features as a difference feature of the recognition log. Perform a difference comparison on each recognition record in the recognition log to generate the difference feature set of the recognition log.

[0019] Furthermore, step S300 includes the following steps:

[0020] Step S301: Randomly select an identification log and extract the set of differential features from the identification log. Randomly select a differential feature from the set of differential features and count the number of times the differential feature occurs in the i-th identification log, which is N. i Let M be the number of identification records in the i-th identification log. i According to the formula:

[0021]

[0022] Where i and j are positive integers, i∈(1,a), j∈(1,a), a is the number of recognition logs stored in the face recognition management system, and M is the number of recognition logs stored in the system. jLet f be the number of identification records in the j-th identification log; calculate the overall occurrence frequency f of the difference features. ’ The overall occurrence frequency reflects the frequency of a difference feature in each identification record. It is used to compare with the occurrence frequency in a single identification record to determine whether the occurrence frequency in an identification log is normal, which helps to determine the cause of the difference feature.

[0023] Step S302: Statistically calculate the difference features in each identification log that satisfy N i The number of logs ≠ 0 is b. The occurrence percentage of the differential feature in all identification logs is calculated as η = a / b. A threshold for occurrence percentage η is preset. max If η≥η max If the difference features are labeled with the first feature, then if η < η max Then, the abnormal frequency judgment value of the difference feature in each identification log is obtained as f. p =f ’ / η; obtain the frequency of occurrence of the differential feature in the i-th identification log as f. i =N i / M i When f is satisfied i ≥f p When the difference features are expressed, a second feature label is applied, and when f is satisfied... i <f p If the difference is found, the difference features will be marked with a third feature. The first feature indicates that the recognition failure is caused by objective factors existing in the scenic area lockers, such as light exposure or system fluctuations. The second feature indicates that the recognition failure is caused by subjective factors of the user corresponding to the recognition log, such as changes in the user's facial decorations or appearance. The third feature indicates that the recognition failure is caused by special circumstances.

[0024] Step S303: Randomly select the k-th difference feature from the set of difference features in the identification log, and set the influence degree of the k-th difference feature as x. k ; Select any identification record from the identification log, obtain the feature tags corresponding to each difference feature in the identification record, and apply the formula:

[0025]

[0026] Where k1, k2, and k3 are positive integers, and k1∈(1,c), k2∈(1,c), k3∈(1,c); Flag k1 The feature label is the feature label corresponding to the k1th differential feature. IF() is the judgment function. IF(Flag) k1=1) To determine whether the k1-th differential feature has a first feature label, if the first feature label exists, then IF(Flag) k1 =1)=1, otherwise, IF(Flag) k1 =1)=0; Flag k2 The feature label corresponding to the k2th differential feature, IF(Flag) k2 =2) To determine whether the k2th differential feature has a second feature label, if the second feature label exists, then IF(Flag) k2 =2)=1, otherwise, IF(Flag) k2 =2)=0; Flag k3 The feature label corresponding to the k3rd differential feature, IF(Flag) k3 =3) To determine whether the k3rd differential feature has a third feature marker, if the third feature marker exists, then IF(Flag) k3 =3)=1, otherwise, IF(Flag) k3 =3)=0; x k1 x represents the degree of influence of the k1th differential feature. k2 x represents the degree of influence of the k2th differential feature. k3 f represents the degree of influence of the k3rd differential feature; k2 η is the frequency of occurrence of the k2th difference feature in the identification log. k3 Let k3 be the proportion of occurrence of the differential feature in all identification logs; calculate the comprehensive influence Y1 of all differential features with the first feature label, the comprehensive influence Y2 of all differential features with the second feature label, and the comprehensive influence Y3 of all differential features with the third feature label in the identification logs.

[0027] Since the differential features are divided into three categories, the methods for analyzing the impact of each category of differential features also differ. Objective factors have problems in all identification logs, so the impact can be directly obtained. User subjective factors are differential features that exist independently in each identification log, so it is necessary to combine them with the frequency of occurrence in the identification logs to analyze the impact. Special differential features have a low probability of occurrence, so they can be analyzed by comprehensively analyzing the frequency of occurrence.

[0028] Step S304: Randomly select one identification record from the normal record set and one abnormal record set of the identification log, and obtain the similarity between the two selected identification records according to the formula:

[0029]

[0030] Wherein, S1 is the similarity of the selected identification records in the normal record set, and S2 is the similarity of the selected identification records in the abnormal record set;

[0031] Step S305: Obtain the similarity difference between any two identification records in the two record sets of the identification log, and determine the degree of influence x. k1 x k2 and x k3 The value; wherein, the average influence degree of each differential feature with a first feature marker and a second feature marker is calculated respectively to obtain the average influence degree of each differential feature.

[0032] Furthermore, step S400 includes the following steps:

[0033] Step S401: Perform facial recognition on the current user, match it with the user information stored in the facial recognition management system, generate a real-time recognition record, determine the recognition log where the real-time recognition record is located, and obtain the similarity S of the real-time recognition record. now ;

[0034] Step S402: Set the similarity threshold to S ’ If S now <S ’ Based on the preset attributes of different categories, the key information generated during the user's face recognition process is extracted to obtain the key information corresponding to each attribute; the key information corresponding to each attribute in each recognition record of the recognition log is obtained, and the similarity of the two sets of key information is compared according to the attributes to obtain the set of difference features satisfied by the current user.

[0035] Step S403: Obtain the influence degree of each differential feature in the differential feature set in the identification log, and set the influence degree of the d-th differential feature as x. d According to the formula:

[0036]

[0037] Where w is the number of differential features in the differential feature set satisfied by the current user; the adjusted similarity S of the real-time identification record is calculated. new If S new >S ’If the first recognition fails, the system will provide the user with adjustment suggestions and perform a second recognition. If the first recognition fails, the system can analyze the user's facial features to determine if any features are affecting facial recognition. If such features are found, the system can investigate the impact of these features and adjust the user's facial recognition results. If the adjusted similarity meets the requirements, it indicates that the user's recognition failure was caused by some factors. This helps the user proactively investigate the relevant reasons and perform a second recognition, ensuring recognition efficiency.

[0038] To better implement the above methods, a scenic area locker management system based on face recognition is also proposed. The management system includes a face recognition evaluation module, a differential feature analysis module, a feature impact analysis module, and a face recognition real-time analysis module.

[0039] The facial recognition evaluation module is used to establish a facial recognition management system in the lockers of the scenic area. It records each facial recognition process in the lockers and obtains several recognition records. Based on the facial feature comparison during the facial recognition process, it calculates the similarity of each recognition record and determines the similarity threshold for judging whether the facial recognition is successful.

[0040] The differential feature analysis module is used to generate recognition logs for each user based on the facial features of each user stored in the face recognition management system; to classify each recognition record in any recognition log as anomaly; and to extract the differential features of the recognition logs by comparing the differences between each recognition record.

[0041] The feature impact analysis module is used to analyze the frequency of occurrence of differential features in each recognition log, classify each differential feature, compare the differential features and similarity assessment between each recognition record in each recognition log, and obtain the degree of impact of each differential feature on face recognition.

[0042] The real-time facial recognition analysis module generates a recognition record and performs a similarity assessment whenever a user performs facial recognition. If the obtained similarity is lower than the similarity threshold, it captures relevant information about each difference feature of the user to obtain several difference features that the user meets. The similarity is adjusted according to the degree of influence of the several difference features. If the adjusted similarity is higher than the similarity threshold, it provides adjustment suggestions to the user and performs recognition again.

[0043] Furthermore, the face recognition evaluation module includes a recognition record construction unit and a similarity evaluation unit;

[0044] The recognition record construction unit is used to establish a face recognition management system in the lockers of the scenic area, and record each face recognition process of the lockers to obtain several recognition records; the similarity evaluation unit is used to calculate the similarity of each recognition record based on the facial feature comparison during the face recognition process, and determine the similarity threshold for judging whether the face recognition is successful.

[0045] Furthermore, the differential feature analysis module includes an abnormal record segmentation unit and a differential feature extraction unit;

[0046] The abnormal record segmentation unit is used to generate recognition logs for each user based on the facial features of each user stored in the face recognition management system; and to segment each recognition record in any recognition log as an anomaly; the difference feature extraction unit is used to extract the difference features of the recognition log by comparing the differences between each recognition record.

[0047] Furthermore, the feature impact analysis module includes a difference feature division unit and an impact degree analysis unit;

[0048] The differential feature segmentation unit is used to analyze the frequency of occurrence of differential features in each recognition log and segment each differential feature; the influence degree analysis unit is used to compare the differential features and similarity assessment between each recognition record in each recognition log to obtain the influence degree of each differential feature on face recognition.

[0049] Furthermore, the real-time facial recognition analysis module includes a feature capture unit and an adjustment suggestion provision unit;

[0050] The feature capture unit is used to generate a recognition record and perform similarity evaluation whenever a user performs face recognition. If the obtained similarity is lower than the similarity threshold, the relevant information of the user about each difference feature is captured to obtain several difference features that the user meets. The adjustment suggestion providing unit is used to adjust the similarity according to the influence degree of the several difference features. If the adjusted similarity is higher than the similarity threshold, the adjustment suggestion is provided to the user and recognition is performed again.

[0051] Compared with the prior art, the beneficial effects of the present invention are:

[0052] 1. This invention analyzes the differences between historical recognition records to help identify various factors that cause facial recognition failure, enabling users to make timely adjustments, avoid frequent recognition failures, improve user access efficiency and experience, and increase user trust in the lockers.

[0053] 2. This invention analyzes the degree of influence of each difference feature, takes into account the difference between objective and subjective difference features, and performs separate analysis to obtain a more accurate degree of influence for each difference feature. This helps to effectively analyze the similarity of user face recognition, obtain a more accurate similarity, and improve the user's recognition efficiency.

[0054] 3. When a user fails to identify a device for the first time, the present invention can proactively help the user analyze the reasons for the failure and predict the similarity after resolving the reasons for the failure. If the conditions for successful identification are met, the user will be reminded to make adjustments, which will help the user improve the accuracy of successful identification in the next attempt and improve the user's access efficiency. Attached Figure Description

[0055] Figure 1 A schematic diagram illustrating the steps of a scenic area locker management method based on facial recognition;

[0056] Figure 2 This is a schematic diagram of a scenic area locker management system based on facial recognition. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] Example: Figures 1 to 2 As shown, this invention provides a method for managing lockers in scenic areas based on facial recognition. The management method includes the following steps:

[0059] Step S100: Establish a face recognition management system in the lockers of the scenic area, record each face recognition process of the lockers, and obtain several recognition records; based on the facial feature comparison during the face recognition process, calculate the similarity of each recognition record, and determine the similarity threshold for judging whether the face recognition is successful.

[0060] Step S100 includes the following steps:

[0061] Step S101: Extract features from the user's facial image recorded when the user uses the locker for the first time to obtain the user's facial feature set and store it in the face recognition management system; randomly select a recognition record and compare the user's facial image captured in the recognition record with the pre-stored facial feature set to obtain the similarity S of the recognition record.

[0062] Step S102: Obtain the recognition results of each recognition record, mark the recognition records that failed to be recognized as having abnormal features; divide all recognition records according to whether there are abnormal feature marks to obtain a normal record set and an abnormal record set;

[0063] Step S103: Obtain the similarity of each identified record in the normal record set and the abnormal record set respectively, and extract the normal similarity threshold with the smallest value in the normal record set (S1). min The anomaly similarity threshold (S2) with the largest value in the set of anomaly records. max If (S1) min ≤(S2) max Then the normal similarity threshold (S1) will be applied. min As a similarity threshold for judging whether face recognition is successful, if (S1) min >(S2) max Then the abnormal similarity threshold (S2) will be set. max As a similarity threshold for determining whether facial recognition is successful;

[0064] Example 1: Set the similarity range obtained from the normal record set to (90%, 100%). If the similarity range of the abnormal record set is (0%, 80%), then the similarity threshold for judging whether face recognition is successful is 80%. If the similarity range of the abnormal record set is (0%, 92%), then the similarity threshold for judging whether face recognition is successful is 90%.

[0065] Step S200: Generate recognition logs for each user based on the facial features of each user stored in the face recognition management system; classify each recognition record in any recognition log as anomaly; extract the difference features of the recognition logs by comparing the differences between each recognition record;

[0066] Step S200 includes the following steps:

[0067] Step S201: Obtain the facial feature set of each recognition record that is compared with the user's facial image. If two recognition records have the same facial feature set, then the two recognition records are summarized to generate a recognition log. Compare the facial feature sets of all recognition records to generate the recognition log corresponding to each user.

[0068] Step S202: Randomly select an identification log, divide each identification record in the identification log into a normal record set and an abnormal record set; extract one identification record from each set, extract key features from the user facial images captured in the two identification records, and obtain the key feature sets of the two identification records respectively.

[0069] Step S203: Preset several appearance types for the facial feature set stored in each recognition record, compare the key features in the two key feature sets with each appearance type to determine the appearance type corresponding to each key feature; summarize the several appearance types corresponding to the two key feature sets to obtain a appearance type set.

[0070] Step S204: Randomly select one appearance type from the appearance type set, and select the key features corresponding to the appearance type from the two key feature sets respectively. Perform a difference comparison on the two selected key features to obtain the degree of difference between the two key features. Set a difference degree threshold. If the difference degree exceeds the difference degree threshold, set the appearance type corresponding to the two key features as a difference feature of the recognition log. Perform a difference comparison on each recognition record in the recognition log to generate the difference feature set of the recognition log.

[0071] Step S300: Analyze the frequency of occurrence of differential features in each recognition log, classify each differential feature; compare the differential features and similarity assessment between each recognition record in each recognition log, and obtain the degree of influence of each differential feature on face recognition;

[0072] Step S300 includes the following steps:

[0073] Step S301: Randomly select an identification log and extract the set of differential features from the identification log. Randomly select a differential feature from the set of differential features and count the number of times the differential feature occurs in the i-th identification log, which is N. i Let M be the number of identification records in the i-th identification log. i According to the formula:

[0074]

[0075] Where i and j are positive integers, i∈(1,a), j∈(1,a), a is the number of recognition logs stored in the face recognition management system, and M is the number of recognition logs stored in the system. j Let f be the number of identification records in the j-th identification log; calculate the overall occurrence frequency f of the difference features. ’ ;

[0076] Example 2: Given a discrepancy feature occurring 5, 6, and 3 times in 3 identification logs, with the number of identification records appearing in the 3 logs being 10, 15, and 10 times respectively, the overall occurrence frequency of this discrepancy feature is f. ’ =14 / 35=40%;

[0077] Step S302: Statistically calculate the difference features in each identification log that satisfy N i The number of logs ≠ 0 is b. The occurrence percentage of the differential feature in all identification logs is calculated as η = a / b. A threshold for occurrence percentage η is preset. max If η≥η max If the difference features are labeled with the first feature, then if η < η max Then, the abnormal frequency judgment value of the difference feature in each identification log is obtained as f. p =f ’ / η; obtain the frequency of occurrence of the differential feature in the i-th identification log as f. i =N i / M i When f is satisfied i ≥f p When the difference features are expressed, a second feature label is applied, and when f is satisfied... i <f p When this happens, the difference features are labeled with a third feature;

[0078] Example 3: Set the number of times a difference feature exists in each identification log to 9 times, and the number of identification logs to 10 times, so that the occurrence rate of the difference feature is 90%. Set the occurrence rate threshold to 70%, so the difference feature is marked as the first feature, which means that the identification failure is caused by objective factors.

[0079] Step S303: Randomly select the k-th difference feature from the set of difference features in the identification log, and set the influence degree of the k-th difference feature as x. k ; Select any identification record from the identification log, obtain the feature tags corresponding to each difference feature in the identification record, and apply the formula:

[0080]

[0081] Where k1, k2, and k3 are positive integers, and k1∈(1,c), k2∈(1,c), k3∈(1,c); Flag k1 The feature label is the feature label corresponding to the k1th differential feature. IF() is the judgment function. IF(Flag) k1 =1) To determine whether the k1-th differential feature has a first feature label, if the first feature label exists, then IF(Flag) k1 =1)=1, otherwise, IF(Flag) k1 =1)=0; Flag k2 The feature label corresponding to the k2th differential feature, IF(Flag) k2=2) To determine whether the k2th differential feature has a second feature label, if the second feature label exists, then IF(Flag) k2 =2)=1, otherwise, IF(Flag) k2 =2)=0; Flag k3 The feature label corresponding to the k3rd differential feature, IF(Flag) k3 =3) To determine whether the k3rd differential feature has a third feature marker, if the third feature marker exists, then IF(Flag) k3 =3)=1, otherwise, IF(Flag) k3 =3)=0; x k1 x represents the degree of influence of the k1th differential feature. k2 x represents the degree of influence of the k2th differential feature. k3 f represents the degree of influence of the k3rd differential feature; k2 η is the frequency of occurrence of the k2th difference feature in the identification log. k3 Let k3 be the proportion of occurrence of the differential feature in all identification logs; calculate the comprehensive influence Y1 of all differential features with the first feature label, the comprehensive influence Y2 of all differential features with the second feature label, and the comprehensive influence Y3 of all differential features with the third feature label in the identification logs.

[0082] Step S304: Randomly select one identification record from the normal record set and one abnormal record set of the identification log, and obtain the similarity between the two selected identification records according to the formula:

[0083]

[0084] Wherein, S1 is the similarity of the selected identification records in the normal record set, and S2 is the similarity of the selected identification records in the abnormal record set;

[0085] Step S305: Obtain the similarity difference between any two identification records in the two record sets of the identification log, and determine the degree of influence x. k1 x k2 and x k3 The value; wherein, the average influence degree of each differential feature with a first feature marker and a second feature marker is calculated respectively to obtain the average influence degree of each differential feature.

[0086] Step S400: Whenever a user performs face recognition, a recognition record is generated and a similarity assessment is performed. If the obtained similarity is lower than the similarity threshold, relevant information about each difference feature of the user is captured to obtain several difference features that the user meets. The similarity is adjusted according to the influence of the several difference features. If the adjusted similarity is higher than the similarity threshold, adjustment suggestions are provided to the user and recognition is performed again.

[0087] Step S400 includes the following steps:

[0088] Step S401: Perform facial recognition on the current user, match it with the user information stored in the facial recognition management system, generate a real-time recognition record, determine the recognition log where the real-time recognition record is located, and obtain the similarity S of the real-time recognition record. now ;

[0089] Step S402: Set the similarity threshold to S ’ If S now <S ’ Based on the preset attributes of different categories, the key information generated during the user's face recognition process is extracted to obtain the key information corresponding to each attribute; the key information corresponding to each attribute in each recognition record of the recognition log is obtained, and the similarity of the two sets of key information is compared according to the attributes to obtain the set of difference features satisfied by the current user.

[0090] Step S403: Obtain the influence degree of each differential feature in the differential feature set in the identification log, and set the influence degree of the d-th differential feature as x. d According to the formula:

[0091]

[0092] Where w is the number of differential features in the differential feature set satisfied by the current user; the adjusted similarity S of the real-time identification record is calculated. new If S new >S ’ If so, the user will be given adjustment suggestions and identified again;

[0093] Example 4: Set the similarity of the user's first face recognition to 70%, and the similarity threshold to 90%. Therefore, it is necessary to extract the user's differential features, such as the user wearing a mask and wearing glasses. The influence of wearing a mask is 20%, and the influence of wearing glasses is 10%. The adjusted similarity is calculated to be 80% × (1 + 20%) × (1 + 10%) = 92.4%. Since 92.4% > 90%, the system reminds the user to remove the mask and glasses and perform the detection again.

[0094] A scenic area locker management system based on facial recognition, the management system includes a facial recognition evaluation module, a differential feature analysis module, a feature impact analysis module, and a facial recognition real-time analysis module;

[0095] The facial recognition evaluation module is used to establish a facial recognition management system in the lockers of the scenic area. It records each facial recognition process in the lockers and obtains several recognition records. Based on the facial feature comparison during the facial recognition process, it calculates the similarity of each recognition record and determines the similarity threshold for judging whether the facial recognition is successful.

[0096] The differential feature analysis module is used to generate recognition logs for each user based on the facial features of each user stored in the face recognition management system; to classify each recognition record in any recognition log as anomaly; and to extract the differential features of the recognition logs by comparing the differences between each recognition record.

[0097] The feature impact analysis module is used to analyze the frequency of occurrence of differential features in each recognition log, classify each differential feature, compare the differential features and similarity assessment between each recognition record in each recognition log, and obtain the degree of impact of each differential feature on face recognition.

[0098] The real-time facial recognition analysis module generates a recognition record and performs a similarity assessment whenever a user performs facial recognition. If the obtained similarity is lower than the similarity threshold, it captures relevant information about each difference feature of the user to obtain several difference features that the user meets. The similarity is adjusted according to the degree of influence of the several difference features. If the adjusted similarity is higher than the similarity threshold, it provides adjustment suggestions to the user and performs recognition again.

[0099] The face recognition evaluation module includes a recognition record construction unit and a similarity evaluation unit.

[0100] The recognition record construction unit is used to establish a face recognition management system in the lockers of the scenic area, and record each face recognition process of the lockers to obtain several recognition records; the similarity evaluation unit is used to calculate the similarity of each recognition record based on the facial feature comparison during the face recognition process, and determine the similarity threshold for judging whether the face recognition is successful.

[0101] The differential feature analysis module includes an abnormal record segmentation unit and a differential feature extraction unit.

[0102] The abnormal record segmentation unit is used to generate recognition logs for each user based on the facial features of each user stored in the face recognition management system; and to segment each recognition record in any recognition log as an anomaly; the difference feature extraction unit is used to extract the difference features of the recognition log by comparing the differences between each recognition record.

[0103] The feature impact analysis module includes a difference feature division unit and an impact degree analysis unit;

[0104] The differential feature segmentation unit is used to analyze the frequency of occurrence of differential features in each recognition log and segment each differential feature; the influence degree analysis unit is used to compare the differential features and similarity assessment between each recognition record in each recognition log to obtain the influence degree of each differential feature on face recognition.

[0105] The real-time facial recognition analysis module includes a feature capture unit and an adjustment suggestion provision unit.

[0106] The feature capture unit is used to generate a recognition record and perform similarity evaluation whenever a user performs face recognition. If the obtained similarity is lower than the similarity threshold, the relevant information of the user about each difference feature is captured to obtain several difference features that the user meets. The adjustment suggestion providing unit is used to adjust the similarity according to the influence degree of the several difference features. If the adjusted similarity is higher than the similarity threshold, the adjustment suggestion is provided to the user and recognition is performed again.

[0107] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for managing lockers in scenic areas based on facial recognition, characterized in that: The management method includes the following steps: Step S100: Establish a face recognition management system in the lockers of the scenic area, record each face recognition process of the lockers, and obtain several recognition records; based on the facial feature comparison during the face recognition process, calculate the similarity of each recognition record, and determine the similarity threshold for judging whether the face recognition is successful. Step S200: Generate recognition logs for each user based on the facial features of each user stored in the face recognition management system; classify each recognition record in any recognition log as anomaly; extract the difference features of the recognition logs by comparing the differences between each recognition record; Step S300: Analyze the frequency of occurrence of differential features in each recognition log, classify each differential feature; compare the differential features and similarity assessment between each recognition record in each recognition log, and obtain the degree of influence of each differential feature on face recognition; Step S400: Whenever a user performs face recognition, a recognition record is generated and a similarity assessment is performed. If the obtained similarity is lower than the similarity threshold, relevant information about each difference feature of the user is captured to obtain several difference features that the user meets. The similarity is adjusted according to the influence degree of the several difference features. If the adjusted similarity is higher than the similarity threshold, adjustment suggestions are provided to the user and recognition is performed again.

2. The method for managing scenic area lockers based on facial recognition according to claim 1, characterized in that: Step S100 includes the following steps: Step S101: Extract features from the user's facial image recorded when the user uses the locker for the first time to obtain the user's facial feature set and store it in the face recognition management system; randomly select a recognition record and compare the user's facial image captured in the recognition record with the pre-stored facial feature set to obtain the similarity S of the recognition record. Step S102: Obtain the recognition results of each recognition record, mark the recognition records that failed to be recognized as having abnormal features; divide all recognition records according to whether there are abnormal feature marks to obtain a normal record set and an abnormal record set; Step S103: Obtain the similarity of each identified record in the normal record set and the abnormal record set respectively, and extract the normal similarity threshold with the smallest value in the normal record set (S1). min The anomaly similarity threshold (S2) with the largest value in the set of anomaly records. max If (S1) min ≤(S2) max Then the normal similarity threshold (S1) will be applied. min As a similarity threshold for judging whether face recognition is successful, if (S1) min >(S2) max Then the abnormal similarity threshold (S2) will be set. max This serves as a similarity threshold for determining whether facial recognition is successful.

3. The method for managing scenic area lockers based on facial recognition according to claim 2, characterized in that: Step S200 includes the following steps: Step S201: Obtain the facial feature set of each recognition record that is compared with the user's facial image. If two recognition records have the same facial feature set, then the two recognition records are summarized to generate a recognition log. Compare the facial feature sets of all recognition records to generate the recognition log corresponding to each user. Step S202: Randomly select an identification log, divide each identification record in the identification log into a normal record set and an abnormal record set; extract one identification record from each set, extract key features from the user facial images captured in the two identification records, and obtain the key feature sets of the two identification records respectively. Step S203: Preset several appearance types for the facial feature set stored in each recognition record, compare the key features in the two key feature sets with each appearance type to determine the appearance type corresponding to each key feature; summarize the several appearance types corresponding to the two key feature sets to obtain a appearance type set. Step S204: Randomly select one appearance type from the appearance type set, and select the key features corresponding to the appearance type from the two key feature sets respectively. Perform a difference comparison on the two selected key features to obtain the degree of difference between the two key features. Set a difference degree threshold. If the difference degree exceeds the difference degree threshold, set the appearance type corresponding to the two key features as a difference feature of the recognition log. Perform a difference comparison on each recognition record in the recognition log to generate the difference feature set of the recognition log.

4. The method for managing scenic area lockers based on facial recognition according to claim 3, characterized in that: Step S300 includes the following steps: Step S301: Randomly select an identification log and extract the set of differential features from the identification log. Randomly select a differential feature from the set of differential features and count the number of times the differential feature occurs in the i-th identification log, which is N. i Let M be the number of identification records in the i-th identification log. i According to the formula: Where i and j are positive integers, i∈(1,a), j∈(1,a), a is the number of recognition logs stored in the face recognition management system, and M is the number of recognition logs stored in the system. j Let f be the number of identification records in the j-th identification log; calculate the overall occurrence frequency f of the difference features. ’ ; Step S302: Statistically calculate the difference features in each identification log that satisfy N i The number of logs ≠ 0 is b. The occurrence percentage of the differential feature in all identification logs is calculated as η = a / b. A threshold for occurrence percentage η is preset. max If η≥η max If the difference features are labeled with the first feature, then if η < η max Then, the abnormal frequency judgment value of the difference feature in each identification log is obtained as f. p =f ’ / η; obtain the frequency of occurrence of the differential feature in the i-th identification log as f. i =N i / M i When f is satisfied i ≥f p When the difference features are expressed, a second feature label is applied, and when f is satisfied... i <f p When this happens, the difference features are labeled with a third feature; Step S303: Randomly select the k-th difference feature from the set of difference features in the identification log, and set the influence degree of the k-th difference feature as x. k ; Select any identification record from the identification log, obtain the feature tags corresponding to each difference feature in the identification record, and apply the formula: Where k1, k2, and k3 are positive integers, and k1∈(1,c), k2∈(1,c), k3∈(1,c); Flag k1 The feature label is the feature label corresponding to the k1th differential feature. IF() is the judgment function. IF(Flag) k1 =1) To determine whether the k1-th differential feature has a first feature label, if the first feature label exists, then IF(Flag) k1 =1)=1, otherwise, IF(Flag) k1 =1)=0; Flag k2 The feature label corresponding to the k2th differential feature, IF(Flag) k2 =2) To determine whether the k2th differential feature has a second feature label, if the second feature label exists, then IF(Flag) k2 =2)=1, otherwise, IF(Flag) k2 =2)=0; Flag k3 The feature label corresponding to the k3rd differential feature, IF(Flag) k3 =3) To determine whether the k3rd differential feature has a third feature marker, if the third feature marker exists, then IF(Flag) k3 =3)=1, otherwise, IF(Flag) k3 =3)=0; x k1 x represents the degree of influence of the k1th differential feature. k2 x represents the degree of influence of the k2th differential feature. k3 f represents the degree of influence of the k3rd differential feature; k2 η is the frequency of occurrence of the k2th difference feature in the identification log. k3 Let k3 be the proportion of occurrence of the third difference feature in all identification logs; calculate the comprehensive influence Y1 of all difference features with the first feature label, the comprehensive influence Y2 of all difference features with the second feature label, and the comprehensive influence Y3 of all difference features with the third feature label in the identification logs. Step S304: Randomly select one identification record from the normal record set and one abnormal record set of the identification log, and obtain the similarity between the two selected identification records according to the formula: Wherein, S1 is the similarity of the selected identification records in the normal record set, and S2 is the similarity of the selected identification records in the abnormal record set; Step S305: Obtain the similarity difference between any two identification records in the two record sets of the identification log, and determine the degree of influence x. k1 x k2 and x k3 The value; wherein, the average influence degree of each differential feature with a first feature marker and a second feature marker is calculated respectively to obtain the average influence degree of each differential feature.

5. A method for managing scenic area lockers based on facial recognition according to claim 4, characterized in that: Step S400 includes the following steps: Step S401: Perform facial recognition on the current user, match it with the user information stored in the facial recognition management system, generate a real-time recognition record, determine the recognition log where the real-time recognition record is located, and obtain the similarity S of the real-time recognition record. now ; Step S402: Set the similarity threshold to S ’ If S now <S ’ Based on the preset attributes of different categories, the key information generated during the user's face recognition process is extracted to obtain the key information corresponding to each attribute; the key information corresponding to each attribute in each recognition record of the recognition log is obtained, and the similarity of the two sets of key information is compared according to the attributes to obtain the set of difference features satisfied by the current user. Step S403: Obtain the influence degree of each differential feature in the differential feature set in the identification log, and set the influence degree of the d-th differential feature as x. d According to the formula: Where w is the number of differential features in the differential feature set satisfied by the current user; the adjusted similarity S of the real-time identification record is calculated. new If S new >S ’ If so, adjustment suggestions will be provided to the user and the user will be identified again.

6. A scenic area locker management system, used to execute the scenic area locker management method based on facial recognition as described in any one of claims 1-5, characterized in that: The management system includes a face recognition evaluation module, a differential feature analysis module, a feature impact analysis module, and a face recognition real-time analysis module; The facial recognition evaluation module is used to establish a facial recognition management system in the lockers of the scenic area, record each facial recognition process of the lockers, and obtain several recognition records; based on the facial feature comparison during the facial recognition process, the similarity of each recognition record is calculated, and the similarity threshold for judging whether the facial recognition is successful is determined. The differential feature analysis module is used to generate recognition logs for each user based on the facial features of each user stored in the face recognition management system; to classify each recognition record in any recognition log as anomaly; and to extract the differential features of the recognition logs by comparing the differences between each recognition record. The feature impact analysis module is used to analyze the frequency of occurrence of differential features in each identification log and to classify each differential feature. By comparing the differences and similarity assessments among the recognition records in each recognition log, the degree of influence of each difference feature on face recognition can be obtained. The real-time facial recognition analysis module is used to generate a recognition record and perform similarity evaluation whenever a user performs facial recognition. If the obtained similarity is lower than the similarity threshold, the module captures relevant information about each difference feature of the user to obtain several difference features that the user meets. The similarity is adjusted according to the influence of the several difference features. If the adjusted similarity is higher than the similarity threshold, the module provides adjustment suggestions to the user and performs recognition again.

7. A scenic area locker management system according to claim 6, characterized in that: The face recognition evaluation module includes a recognition record construction unit and a similarity evaluation unit; The identification record construction unit is used to establish a face recognition management system in the lockers of the scenic area, record each face recognition process of the lockers, and obtain several identification records; the similarity evaluation unit is used to calculate the similarity of each identification record based on the facial feature comparison during the face recognition process, and determine the similarity threshold for judging whether the face recognition is successful.

8. A scenic area locker management system according to claim 6, characterized in that: The differential feature analysis module includes an abnormal record segmentation unit and a differential feature extraction unit; The abnormal record segmentation unit is used to generate recognition logs for each user based on the facial features of each user stored in the face recognition management system; and to segment each recognition record in any recognition log as an anomaly; the difference feature extraction unit is used to extract the difference features of the recognition log by comparing the differences between each recognition record.

9. A scenic area locker management system according to claim 6, characterized in that: The feature impact analysis module includes a difference feature segmentation unit and an impact degree analysis unit; The difference feature segmentation unit is used to analyze the frequency of occurrence of difference features in each recognition log and segment each difference feature; the influence degree analysis unit is used to compare the difference features and similarity assessment between each recognition record in each recognition log to obtain the influence degree of each difference feature on face recognition.

10. A scenic area locker management system according to claim 6, characterized in that: The real-time face recognition analysis module includes a feature capture unit and an adjustment suggestion providing unit; The feature capture unit is used to generate a recognition record and perform similarity evaluation whenever a user performs face recognition. If the obtained similarity is lower than the similarity threshold, the relevant information of the user about each difference feature is captured to obtain several difference features that the user meets. The adjustment suggestion providing unit is used to adjust the similarity according to the influence degree of the several difference features. If the adjusted similarity is higher than the similarity threshold, the adjustment suggestion is provided to the user and recognition is performed again.

Images