Fraud early warning method and device in bank outlets
阅读说明：本技术 一种银行网点内的诈骗预警方法及装置 (Fraud early warning method and device in bank outlets ) 是由 黄文强 于 2021-09-14 设计创作，主要内容包括：本发明公开了一种银行网点内的诈骗预警方法及装置,可应用于人工智能领域或金融领域,在检测到客户进入银行网点后,通过判断包括客户在柜台停留时间、客户在自助设备停留时间以及客户与银行网点内其他人接触时间的行为特征信息是否符合预设规则,过滤掉被诈骗风险较小的客户行为特征信息。对于行为特征信息符合预设规则的客户,通过识别客户的接触人员是否符合银行工作人员的着装特点以及客户在银行网点的时间内账户余额是否减少,进一步判断客户是否存在被伪装为银行工作人员的不法分子诱骗而办理非银行业务的风险,在存在风险的情况下及时向客户发出诈骗预警信息,进而在客户确认被诈骗时及时拦截,避免造成客户的财产损失。(The invention discloses a fraud early warning method and a fraud early warning device in a bank outlet, which can be applied to the field of artificial intelligence or the field of finance. For the customers with behavior characteristic information in accordance with the preset rules, whether contact personnel of the customers conform to dressing characteristics of bank workers or not and whether account balances of the customers are reduced in time of bank outlets or not are identified, whether the customers have risks of being disguised as lawless persons of the bank workers to handle non-banking services or not is further judged, fraud early warning information is timely sent to the customers under the condition of the risks, and then the customers are timely intercepted when confirming that the customers are swindled, so that property loss of the customers is avoided.)
1. A fraud early warning method in a bank outlet is characterized by comprising the following steps:
after detecting that a client enters a banking outlet, acquiring identity information of the client and behavior characteristic information in the banking outlet, wherein the behavior characteristic information comprises: the time of the customer staying at the counter, the time of the customer staying at the self-service equipment and the time of the customer contacting with other people in the bank outlet;
judging whether the behavior characteristic information accords with a preset rule or not;
if the behavior characteristic information accords with a preset rule, acquiring image information of contact personnel when a customer contacts with other people in the bank outlets;
inputting the image information of the contact person into a pre-constructed dressing identification model of the bank worker, and determining whether the contact person accords with the dressing characteristics of the bank worker according to the identification result of the dressing identification model of the bank worker;
under the condition that the contact person accords with the dressing characteristics of bank workers, whether the account balance of the customer is reduced in the time of a bank outlet is judged;
and under the condition that the account balance of the customer is reduced in the time of the bank outlet, sending fraud early warning prompts to the customer according to the identity information of the customer.
2. The method of claim 1, wherein the obtaining of the behavior feature information of the customer at the banking site comprises:
acquiring client video information acquired by image acquisition equipment corresponding to a counter in a bank outlet, client video information acquired by image acquisition equipment corresponding to self-service equipment and client video information acquired by other image acquisition equipment in the bank outlet;
extracting the time of a client staying on a counter from client video information acquired by image acquisition equipment corresponding to the counter;
extracting the staying time of a customer in the self-service equipment from the customer video information acquired by the image acquisition equipment corresponding to the self-service equipment;
and extracting the contact time of the customer and other people in the bank outlets from the customer video information acquired by other image acquisition equipment in the bank outlets.
3. The method of claim 1, wherein determining whether the behavior feature information complies with a predetermined rule comprises:
judging whether the behavior characteristic information meets the following conditions:
the staying time of the client on the counter is less than the fastest transaction time of the counter business;
the staying time of the client in the self-service equipment is less than the fastest service handling time of the self-service equipment;
the contact time of the customer and other people in the banking outlet is within the non-banking risk business handling time interval;
if the behavior characteristic information simultaneously meets all the conditions, judging that the behavior characteristic information accords with a preset rule;
and if the behavior characteristic information does not meet all the conditions at the same time, judging that the behavior characteristic information does not meet a preset rule.
4. The method of claim 1, wherein constructing the bank worker dress identification model comprises:
acquiring sample data, wherein the sample data is image data marked with dresses of female bank workers, dresses of male bank workers or dresses of non-bank workers;
dividing the sample data into a training set and a verification set;
training a neural network model by using the training set;
verifying the trained neural network model by using the verification set;
and under the condition that the verification result meets the preset requirement, the construction of the dress identification model of the bank staff is completed.
5. The method as recited in claim 1, wherein said sending a fraud alert to the customer based on the customer's identity information comprises:
inquiring the contact information of the client in a client information database according to the identity information of the client;
and sending fraud early warning prompts to the customer according to the contact information of the customer.
6. The method as recited in claim 5, wherein said sending a fraud alert prompt to a customer comprises:
and sending fraud early warning prompts to the customers in a mode of short messages, mobile banking or intelligent voice calls.
7. A fraud early warning device in a bank outlet, comprising:
the behavior characteristic acquiring unit is used for acquiring the identity information of the client and the behavior characteristic information in the bank website after detecting that the client enters the bank website, and the behavior characteristic information comprises: the time of the customer staying at the counter, the time of the customer staying at the self-service equipment and the time of the customer contacting with other people in the bank outlet;
the first judging unit is used for judging whether the behavior characteristic information accords with a preset rule or not;
the image information acquisition unit is used for acquiring the image information of contact personnel when the customer contacts with other people in the bank outlets under the condition that the behavior characteristic information accords with a preset rule;
the image recognition unit is used for inputting the image information of the contact person into a pre-constructed dressing recognition model of the bank worker and determining whether the contact person accords with the dressing characteristics of the bank worker according to the recognition result of the dressing recognition model of the bank worker;
the second judgment unit is used for judging whether the account balance of the customer is reduced in the time of the bank branch or not under the condition that the contact person accords with the dressing characteristics of bank workers;
and the fraud early warning unit is used for sending fraud early warning prompts to the customer according to the identity information of the customer under the condition that the account balance of the customer is reduced in the time of the bank outlet.
8. The apparatus according to claim 7, wherein the behavior feature obtaining unit is specifically configured to:
acquiring client video information acquired by image acquisition equipment corresponding to a counter in a bank outlet, client video information acquired by image acquisition equipment corresponding to self-service equipment and client video information acquired by other image acquisition equipment in the bank outlet;
extracting the time of a client staying on a counter from client video information acquired by image acquisition equipment corresponding to the counter;
extracting the staying time of a customer in the self-service equipment from the customer video information acquired by the image acquisition equipment corresponding to the self-service equipment;
and extracting the contact time of the customer and other people in the bank outlets from the customer video information acquired by other image acquisition equipment in the bank outlets.
9. The apparatus according to claim 7, wherein the first determining unit is specifically configured to:
judging whether the behavior characteristic information meets the following conditions:
the staying time of the client on the counter is less than the fastest transaction time of the counter business;
the staying time of the client in the self-service equipment is less than the fastest service handling time of the self-service equipment;
the contact time of the customer and other people in the banking outlet is within the non-banking risk business handling time interval;
if the behavior characteristic information simultaneously meets all the conditions, judging that the behavior characteristic information accords with a preset rule;
and if the behavior characteristic information does not meet all the conditions at the same time, judging that the behavior characteristic information does not meet a preset rule.
10. The apparatus according to claim 7, further comprising an identification model construction unit, in particular for:
acquiring sample data, wherein the sample data is image data marked with dresses of female bank workers, dresses of male bank workers or dresses of non-bank workers;
dividing the sample data into a training set and a verification set;
training a neural network model by using the training set;
verifying the trained neural network model by using the verification set;
and under the condition that the verification result meets the preset requirement, the construction of the dress identification model of the bank staff is completed.
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a fraud early warning method and device in a bank outlet.
Background
With the popularization of the setting equipment of the bank outlets, in the bank outlets, besides counter staff, staff can be generally set in a hall to help clients to handle business by using intelligent equipment.
Some lawbreakers can imitate wearing of bank workers to carry out fraud on customers at a bank network site, if the customers meet the lawbreakers pretending to be bank workers at the bank network site, the customers are difficult to distinguish, especially old customers are likely to handle non-banking services under the deception of the lawbreakers, and once the situation occurs, if the lawbreakers cannot find and intercept the lawbreakers in time, property loss of the customers is likely to be caused.
Disclosure of Invention
In view of the above, the invention provides a fraud early warning method and device in a bank outlet, which can accurately detect fraud in the bank, send fraud early warning information to a customer in time, and reduce the risk of property loss caused by fraud of the customer in the bank outlet.
In order to achieve the above purpose, the invention provides the following specific technical scheme:
a fraud early warning method in a banking outlet, comprising:
after detecting that a client enters a banking outlet, acquiring identity information of the client and behavior characteristic information in the banking outlet, wherein the behavior characteristic information comprises: the time of the customer staying at the counter, the time of the customer staying at the self-service equipment and the time of the customer contacting with other people in the bank outlet;
judging whether the behavior characteristic information accords with a preset rule or not;
if the behavior characteristic information accords with a preset rule, acquiring image information of contact personnel when a customer contacts with other people in the bank outlets;
inputting the image information of the contact person into a pre-constructed dressing identification model of the bank worker, and determining whether the contact person accords with the dressing characteristics of the bank worker according to the identification result of the dressing identification model of the bank worker;
under the condition that the contact person accords with the dressing characteristics of bank workers, whether the account balance of the customer is reduced in the time of a bank outlet is judged;
and under the condition that the account balance of the customer is reduced in the time of the bank outlet, sending fraud early warning prompts to the customer according to the identity information of the customer.
Optionally, the acquiring the behavior feature information of the customer in the banking outlet includes:
acquiring client video information acquired by image acquisition equipment corresponding to a counter in a bank outlet, client video information acquired by image acquisition equipment corresponding to self-service equipment and client video information acquired by other image acquisition equipment in the bank outlet;
extracting the time of a client staying on a counter from client video information acquired by image acquisition equipment corresponding to the counter;
extracting the staying time of a customer in the self-service equipment from the customer video information acquired by the image acquisition equipment corresponding to the self-service equipment;
and extracting the contact time of the customer and other people in the bank outlets from the customer video information acquired by other image acquisition equipment in the bank outlets.
Optionally, the determining whether the behavior feature information conforms to a preset rule includes:
judging whether the behavior characteristic information meets the following conditions:
the staying time of the client on the counter is less than the fastest transaction time of the counter business;
the staying time of the client in the self-service equipment is less than the fastest service handling time of the self-service equipment;
the contact time of the customer and other people in the banking outlet is within the non-banking risk business handling time interval;
if the behavior characteristic information simultaneously meets all the conditions, judging that the behavior characteristic information accords with a preset rule;
and if the behavior characteristic information does not meet all the conditions at the same time, judging that the behavior characteristic information does not meet a preset rule.
Optionally, constructing the dress identification model of the bank worker includes:
acquiring sample data, wherein the sample data is image data marked with dresses of female bank workers, dresses of male bank workers or dresses of non-bank workers;
dividing the sample data into a training set and a verification set;
training a neural network model by using the training set;
verifying the trained neural network model by using the verification set;
and under the condition that the verification result meets the preset requirement, the construction of the dress identification model of the bank staff is completed.
Optionally, the sending a fraud early warning prompt to the customer according to the identity information of the customer includes:
inquiring the contact information of the client in a client information database according to the identity information of the client;
and sending fraud early warning prompts to the customer according to the contact information of the customer.
Optionally, the sending a fraud early warning prompt to the customer includes:
and sending fraud early warning prompts to the customers in a mode of short messages, mobile banking or intelligent voice calls.
A fraud early warning apparatus in a banking outlet, comprising:
the behavior characteristic acquiring unit is used for acquiring the identity information of the client and the behavior characteristic information in the bank website after detecting that the client enters the bank website, and the behavior characteristic information comprises: the time of the customer staying at the counter, the time of the customer staying at the self-service equipment and the time of the customer contacting with other people in the bank outlet;
the first judging unit is used for judging whether the behavior characteristic information accords with a preset rule or not;
the image information acquisition unit is used for acquiring the image information of contact personnel when the customer contacts with other people in the bank outlets under the condition that the behavior characteristic information accords with a preset rule;
the image recognition unit is used for inputting the image information of the contact person into a pre-constructed dressing recognition model of the bank worker and determining whether the contact person accords with the dressing characteristics of the bank worker according to the recognition result of the dressing recognition model of the bank worker;
the second judgment unit is used for judging whether the account balance of the customer is reduced in the time of the bank branch or not under the condition that the contact person accords with the dressing characteristics of bank workers;
and the fraud early warning unit is used for sending fraud early warning prompts to the customer according to the identity information of the customer under the condition that the account balance of the customer is reduced in the time of the bank outlet.
Optionally, the behavior feature obtaining unit is specifically configured to:
acquiring client video information acquired by image acquisition equipment corresponding to a counter in a bank outlet, client video information acquired by image acquisition equipment corresponding to self-service equipment and client video information acquired by other image acquisition equipment in the bank outlet;
extracting the time of a client staying on a counter from client video information acquired by image acquisition equipment corresponding to the counter;
extracting the staying time of a customer in the self-service equipment from the customer video information acquired by the image acquisition equipment corresponding to the self-service equipment;
and extracting the contact time of the customer and other people in the bank outlets from the customer video information acquired by other image acquisition equipment in the bank outlets.
Optionally, the first determining unit is specifically configured to:
judging whether the behavior characteristic information meets the following conditions:
the staying time of the client on the counter is less than the fastest transaction time of the counter business;
the staying time of the client in the self-service equipment is less than the fastest service handling time of the self-service equipment;
the contact time of the customer and other people in the banking outlet is within the non-banking risk business handling time interval;
if the behavior characteristic information simultaneously meets all the conditions, judging that the behavior characteristic information accords with a preset rule;
and if the behavior characteristic information does not meet all the conditions at the same time, judging that the behavior characteristic information does not meet a preset rule.
Optionally, the apparatus further includes an identification model building unit, specifically configured to:
acquiring sample data, wherein the sample data is image data marked with dresses of female bank workers, dresses of male bank workers or dresses of non-bank workers;
dividing the sample data into a training set and a verification set;
training a neural network model by using the training set;
verifying the trained neural network model by using the verification set;
and under the condition that the verification result meets the preset requirement, the construction of the dress identification model of the bank staff is completed.
Optionally, the fraud early warning unit is specifically configured to:
inquiring the contact information of the client in a client information database according to the identity information of the client;
and sending fraud early warning prompts to the customer according to the contact information of the customer.
Optionally, the fraud early warning unit is specifically configured to:
inquiring the contact information of the client in a client information database according to the identity information of the client;
and sending fraud early warning prompts to the customer in a short message mode, a mobile banking mode or an intelligent voice telephone mode according to the contact mode of the customer.
Compared with the prior art, the invention has the following beneficial effects:
the invention discloses a fraud early warning method in a bank outlet, which filters out the behavior characteristic information of a customer with less fraud risk by judging whether the behavior characteristic information comprising the staying time of the customer on a counter, the staying time of the customer on self-service equipment and the contact time of the customer and other people in the bank outlet accords with a preset rule or not after detecting that the customer enters the bank outlet. For the customers with behavior characteristic information in accordance with the preset rules, whether contact personnel of the customers conform to dressing characteristics of bank workers or not and whether account balances of the customers are reduced in time of bank outlets or not are identified, whether the customers have risks of being disguised as lawless persons of the bank workers to handle non-banking services or not is further judged, and therefore fraud early warning information is timely sent to the customers under the condition of the risks, and the customers are timely intercepted when confirming that the customers are swindled, and property loss of the customers is avoided.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a schematic flow chart of a fraud early warning method in a bank outlet according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a fraud warning device in a bank outlet according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention discloses a fraud early warning method in a bank outlet, which realizes accurate detection of fraud in a bank, sends fraud early warning information to a customer in time and reduces the risk of property loss caused by fraud of the customer in the bank outlet.
Specifically, please refer to fig. 1, the fraud early warning method in a bank outlet disclosed in this embodiment includes the following steps:
s101: after detecting that the client enters the banking outlet, acquiring identity information of the client and behavior characteristic information in the banking outlet, wherein the behavior characteristic information comprises: the time of the customer staying at the counter, the time of the customer staying at the self-service equipment and the time of the customer contacting with other people in the bank outlet;
when the customer takes a number by swiping a bank card or an identity card, the system detects that the customer enters a bank outlet. Under the condition that a customer takes a number by swiping a bank card, obtaining the identity information of the customer according to the corresponding relation between the bank card number stored in the database and the identity information of the customer; and under the condition that the client takes the number by swiping the identity card, obtaining the identity information of the client according to the corresponding relation between the identity card number stored in the database and the identity information of the client.
It should be noted that a plurality of image capturing devices, such as cameras, are generally deployed at a banking outlet, and the image capturing devices are generally deployed in a counter area and a self-service device (such as a self-service teller machine and a self-service transaction device).
The fraud early warning method in the bank outlets disclosed by the embodiment reuses the image acquisition function of image acquisition equipment deployed in the bank outlets to acquire the video information of each customer entering the bank outlets. After detecting that the customer enters a bank outlet, acquiring behavior characteristic information of the customer in the bank outlet through each image acquisition device in the bank outlet, specifically comprising the following steps:
the method comprises the steps of obtaining client video information collected by image collection equipment corresponding to a counter in a bank outlet of a client, client video information collected by image collection equipment corresponding to self-service equipment and client video information collected by other image collection equipment in the bank outlet.
And extracting the time of the client staying at the counter from the client video information collected by the image collecting equipment corresponding to the counter.
And extracting the staying time of the customer in the self-service equipment from the customer video information acquired by the image acquisition equipment corresponding to the self-service equipment.
And extracting the contact time of the customer and other people in the bank outlets from the customer video information acquired by other image acquisition equipment in the bank outlets.
It should be noted that when the distance between the customer and the other people in the banking outlet is within the preset range, it is determined that the customer is in contact with the person, and the contact time between the customer and the person needs to be extracted. According to the embodiment, the track information of the customer can be obtained through a video processing technology, and the residence time of the customer in different areas, such as the residence time of a counter area, the residence time of self-service equipment and the contact time of the customer with other people, can be extracted according to the track information.
S102: judging whether the behavior characteristic information accords with a preset rule or not;
specifically, whether the behavior characteristic information of the customer in the bank outlets meets the following conditions is respectively judged:
the staying time of the client on the counter is shorter than the fastest transaction time of the counter service, namely, the client is determined not to transact the service on the counter;
the staying time of the client in the self-service equipment is shorter than the fastest service handling time of the self-service equipment, namely the client does not handle the service in the self-service equipment;
the contact time of the client and other people in the bank outlets is in the non-banking risk business handling time interval, namely the client is possible to handle the non-banking risk business in the contact time with other people in the bank outlets;
if the behavior characteristic information of the customer in the bank outlets meets all the conditions, judging that the behavior characteristic information of the customer in the bank outlets meets preset rules;
and if the behavior characteristic information of the customer in the bank branch does not meet all the conditions at the same time, judging that the behavior characteristic information of the customer in the bank branch does not meet the preset rule.
The fastest transaction time of counter business, the fastest transaction time of self-service equipment business and the transaction time of non-bank risk business are obtained by carrying out statistical analysis on historical data.
It can be understood that the fact that the residence time of the customer at the counter is less than the fastest transaction time of the counter business, the residence time of the customer at the self-service equipment is less than the fastest transaction time of the self-service equipment business, and the contact time of the customer with other people in the bank outlets is within the non-banking risk business transaction time interval indicates that the customer does not transact business at the counter or transact business at the self-service equipment, but the customer may transact non-banking risk business within the contact time with other people in the bank outlets, and further screening is needed to determine whether the customer is at risk of being swindled in the bank outlets.
If the behavior feature information does not conform to the preset rule, executing S103: subsequent fraud identification processing is not performed;
if the behavior feature information meets the preset rule, executing S104: acquiring image information of contact personnel when a customer contacts with other people in a bank outlet;
specifically, the image information of the contact person in the non-banking risk business handling time interval with the contact time of the client is obtained by capturing the video information of the contact of the client and other people in the banking outlet.
S105: inputting image information of the contact person into a pre-constructed dressing identification model of the bank worker, and determining whether the contact person accords with the dressing characteristics of the bank worker according to the identification result of the dressing identification model of the bank worker;
the construction method of the dress identification model for bank workers comprises the following steps:
acquiring sample data, wherein the sample data is image data marked with dresses of female bank workers, dresses of male bank workers or dresses of non-bank workers;
dividing sample data into a training set and a verification set;
training the neural network model by using a training set;
verifying the trained neural network model by using a verification set;
and under the condition that the verification result meets the preset requirement, namely under the condition that the verification result meets the preset model precision requirement, finishing constructing the dressing identification model by bank staff.
It should be noted that, different banks have different dress characteristics, such as different dress colors, different dress styles and different dress arrangements, and in the process of constructing a bank worker dress identification model, sample data needs to be prepared according to the dress characteristics of a target bank worker, so as to improve the identification precision of the bank worker dress identification model.
S106: under the condition that the contact personnel accord with the dressing characteristics of bank workers, whether the account balance of the customer is reduced in the time of a bank outlet is judged;
in the case that the account balance of the customer is decreased in the time of the banking outlet, executing S107: sending fraud early warning prompts to the customer according to the identity information of the customer;
in the case that the account balance of the customer is not decreased in the time of the banking outlet, S103 is executed: subsequent fraud identification processing is not performed.
It can be understood that the substantial fraud possibility can only occur when the account balance of the customer is reduced, therefore, in the case that the contact person of the customer is identified to be in accordance with the dressing characteristics of the bank staff, the embodiment further inquires whether the account balance of the customer is reduced within the time of the bank branch, and sends a fraud early warning prompt to the customer according to the identity information of the customer when the account balance of the customer is reduced within the time of the bank branch.
Specifically, according to the identity information of the customer, the contact information of the customer is inquired in the customer information database, and according to the contact information of the customer, a fraud early warning prompt is sent to the customer, for example, the fraud early warning prompt is sent to the customer through a short message, a mobile banking or an intelligent voice telephone.
When the fraud information fed back by the customer is received, the transfer account of the customer in the time of the bank outlet can be frozen, and property loss of the customer is avoided. When the information fed back by the customer is not defrauded or any information fed back by the customer is not received within the preset time, the transfer account of the customer within the time of the bank outlet is not frozen.
In the fraud early warning method in the bank outlets disclosed by the embodiment, after the fact that the customer enters the bank outlets is detected, the behavior characteristic information of the customer with smaller fraud risk is filtered by judging whether the behavior characteristic information comprising the time of the customer staying at the counter, the time of the customer staying at the self-service equipment and the contact time of the customer and other people in the bank outlets meets the preset rules. For the customers with behavior characteristic information in accordance with the preset rules, whether contact personnel of the customers conform to dressing characteristics of bank workers or not and whether account balances of the customers are reduced in time of bank outlets or not are identified, whether the customers have risks of being disguised as lawless persons of the bank workers to handle non-banking services or not is further judged, and therefore fraud early warning information is timely sent to the customers under the condition of the risks, and the customers are timely intercepted when confirming that the customers are swindled, and property loss of the customers is avoided.
Based on the above embodiment, the embodiment discloses a fraud early warning method in a bank outlet, and correspondingly discloses a fraud early warning device in a bank outlet, please refer to fig. 2, where the device includes:
a behavior feature obtaining unit 201, configured to obtain identity information of a client and behavior feature information in a banking website after detecting that the client enters the banking website, where the behavior feature information includes: the time of the customer staying at the counter, the time of the customer staying at the self-service equipment and the time of the customer contacting with other people in the bank outlet;
a first determining unit 202, configured to determine whether the behavior feature information meets a preset rule;
the image information acquisition unit 203 is used for acquiring image information of contact personnel when a customer contacts with other people in a bank outlet under the condition that the behavior characteristic information accords with a preset rule;
the image recognition unit 204 is used for inputting the image information of the contact person into a pre-constructed dressing recognition model of the bank worker, and determining whether the contact person accords with the dressing characteristics of the bank worker according to the recognition result of the dressing recognition model of the bank worker;
a second determining unit 205, configured to determine whether the account balance of the customer decreases within the time of the banking outlet when the contact person meets the dressing characteristics of the banking staff;
and the fraud early warning unit 206 is configured to send a fraud early warning prompt to the customer according to the identity information of the customer when the account balance of the customer decreases within the time of the banking outlet.
Optionally, the behavior feature obtaining unit 201 is specifically configured to:
acquiring client video information acquired by image acquisition equipment corresponding to a counter in a bank outlet, client video information acquired by image acquisition equipment corresponding to self-service equipment and client video information acquired by other image acquisition equipment in the bank outlet;
extracting the time of a client staying on a counter from client video information acquired by image acquisition equipment corresponding to the counter;
extracting the staying time of a customer in the self-service equipment from the customer video information acquired by the image acquisition equipment corresponding to the self-service equipment;
and extracting the contact time of the customer and other people in the bank outlets from the customer video information acquired by other image acquisition equipment in the bank outlets.
Optionally, the first determining unit 202 is specifically configured to:
judging whether the behavior characteristic information meets the following conditions:
the staying time of the client on the counter is less than the fastest transaction time of the counter business;
the staying time of the client in the self-service equipment is less than the fastest service handling time of the self-service equipment;
the contact time of the customer and other people in the banking outlet is within the non-banking risk business handling time interval;
if the behavior characteristic information simultaneously meets all the conditions, judging that the behavior characteristic information accords with a preset rule;
and if the behavior characteristic information does not meet all the conditions at the same time, judging that the behavior characteristic information does not meet a preset rule.
Optionally, the apparatus further includes an identification model building unit, specifically configured to:
acquiring sample data, wherein the sample data is image data marked with dresses of female bank workers, dresses of male bank workers or dresses of non-bank workers;
dividing the sample data into a training set and a verification set;
training a neural network model by using the training set;
verifying the trained neural network model by using the verification set;
and under the condition that the verification result meets the preset requirement, the construction of the dress identification model of the bank staff is completed.
Optionally, the fraud early warning unit 206 is specifically configured to:
inquiring the contact information of the client in a client information database according to the identity information of the client;
and sending fraud early warning prompts to the customer according to the contact information of the customer.
Optionally, the fraud early warning unit 206 is specifically configured to:
inquiring the contact information of the client in a client information database according to the identity information of the client;
and sending fraud early warning prompts to the customer in a short message mode, a mobile banking mode or an intelligent voice telephone mode according to the contact mode of the customer.
According to the fraud early warning device in the bank outlets disclosed by the embodiment, after a customer is detected to enter the bank outlets, the behavior characteristic information of the customer with smaller fraud risk is filtered by judging whether the behavior characteristic information comprising the time of the customer staying at the counter, the time of the customer staying at the self-service equipment and the contact time of the customer and other people in the bank outlets conforms to the preset rule or not. For the customers with behavior characteristic information in accordance with the preset rules, whether contact personnel of the customers conform to dressing characteristics of bank workers or not and whether account balances of the customers are reduced in time of bank outlets or not are identified, whether the customers have risks of being disguised as lawless persons of the bank workers to handle non-banking services or not is further judged, and therefore fraud early warning information is timely sent to the customers under the condition of the risks, and the customers are timely intercepted when confirming that the customers are swindled, and property loss of the customers is avoided.
It should be noted that the fraud early warning method and device in the bank outlets provided by the invention can be applied to the field of artificial intelligence or the field of finance. The above description is only an example, and does not limit the application field of the fraud early warning method and device in a bank outlet provided by the present invention.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The above embodiments can be combined arbitrarily, and the features described in the embodiments in the present specification can be replaced or combined with each other in the above description of the disclosed embodiments, so that those skilled in the art can implement or use the present application.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.