CN109829706A - Transfer account method, device, computer equipment and storage medium based on recognition of face - Google Patents

Abstract

The invention discloses a kind of transfer account methods based on recognition of face, device, equipment and medium, this method comprises: receiving paying party initiates transfer request, according to transfer request, transfer information, which is sent, to client fills in instruction, receive the target shroff account number and target transfer amounts uploaded, each current face's picture of acquisition paying party in real time, when the corresponding target person of target payment account number in transfer request and paying party are the same person, each current face's picture is identified using preset face identification method, obtain each current face face movement, the threshold interval that the target transfer amounts are fallen into is selected in preset each threshold interval, when whether the movement of each set current face face is the corresponding preset each set target face face movement of threshold interval that target transfer amounts are fallen into, from target payment account Target transfer amounts are transferred in number to target shroff account number, improve the safety transferred accounts.

Description

Transfer account method, device, computer equipment and storage medium based on recognition of face

Technical field

The present invention relates to artificial intelligence field more particularly to a kind of transfer account methods based on recognition of face, device, computer Equipment and storage medium.

Background technique

Currently, with the development of mobile internet, move and transfer accounts are increasingly by the welcome of user, and many users are using shifting Dynamic equipment carries out move and transfer accounts, such as carries out move and transfer accounts using smart phone.

In general, in conventional methods where, when carrying out move and transfer accounts, bank may require that user inputs secret number, in this, as The means of security verification.But when the secret number is revealed, the people that leads to take the secret number is from target account It is middle that wealth is transferred to another target account, or the people for taking the secret number is caused to take wealth away from target account, And then under causing the safety transferred accounts to reduce.

Therefore, the problem of safer transfer account method of one kind becomes those skilled in the art's urgent need to resolve is found.

Summary of the invention

The embodiment of the present invention provides a kind of transfer account method based on recognition of face, device, computer equipment and storage medium, To solve the problems, such as that it is lower that the safety transferred accounts reduces.

A kind of transfer account method based on recognition of face, comprising:

The transfer request that real-time reception paying party is initiated in client, the transfer request include target payment account number；

According to the transfer request, Xiang Suoshu client sends transfer information and fills in instruction；

Receive that the paying party fills in target shroff account number that instruction uploads according to the transfer information and target is transferred accounts gold Volume；

Each current face's picture of the paying party is acquired in real time；

When the corresponding target person of target payment account number in the transfer request and the paying party are the same person, adopt Each current face's picture is identified with preset face identification method, obtains each current face face movement, The movement of current face face refers to that face are according to movement caused by reflection on the face of the paying party；

The threshold interval that the target transfer amounts are fallen into is selected in preset each threshold interval, it is described each A threshold interval is there are at least two and mutually disjoints；

Judge whether each set current face face movement is that the threshold interval that the target transfer amounts are fallen into is corresponding Preset each set target face face movement；

When each set current face face movement is that the threshold interval that the target transfer amounts are fallen into is corresponding default Each set target face face movement when, the target transfer amounts to the target is transferred from the target payment account number and is received Money account.

A kind of transfer device based on recognition of face, comprising:

First receiving module, for the transfer request that real-time reception paying party is initiated in client, the transfer request packet Include target payment account number；

First sending module, for according to the transfer request, Xiang Suoshu client to send transfer information and fills in instruction；

Second receiving module fills in the target gathering that instruction uploads according to the transfer information for receiving the paying party Account and target transfer amounts；

Acquisition module, for acquiring each current face's picture of the paying party in real time；

Identification module, in the transfer request the corresponding target person of target payment account number with the paying party be When the same person, each current face's picture is identified using preset face identification method, is obtained each current Facial face movement, the movement of current face face refer to that face are according to movement caused by reflection on the face of the paying party；

Choose module, the threshold fallen into for selecting the target transfer amounts in preset each threshold interval It is worth section, each threshold interval is there are at least two and mutually disjoints；

Second judgment module, for judging whether each set current face face movement is that the target transfer amounts are fallen The corresponding preset each set target face face movement of the threshold interval entered；

Module is transferred, for being the threshold zone that the target transfer amounts are fallen into when each set current face face movement Between corresponding preset each set target face face act when, the target transfer amounts are transferred from the target payment account number To the target shroff account number.

A kind of computer equipment, including memory, processor and storage are in the memory and can be in the processing The computer program run on device, the processor realize above-mentioned transferring accounts based on recognition of face when executing the computer program The step of method.

A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter The step of calculation machine program realizes the above-mentioned transfer account method based on recognition of face when being executed by processor.

The above-mentioned transfer account method based on recognition of face, device, computer equipment and storage medium, are paid the bill by real-time reception The transfer request for the mobile banking that side is initiated in client, the transfer request includes target payment account number, is transferred accounts according to described Request, Xiang Suoshu client send transfer information and fill in instruction, receive the paying party according to the transfer information and fill in instruction The target shroff account number and target transfer amounts of upload acquire each current face's picture of the paying party, when described in real time When the corresponding target person of target payment account number and the paying party are the same person in transfer request, known using preset face Other method identifies each current face's picture, obtains each current face face movement, and current face face are dynamic As face on the face of the paying party according to movement caused by reflection, institute is selected in preset each threshold interval The threshold interval that target transfer amounts are fallen into is stated, there are at least two, each threshold interval mutually disjoints threshold interval, sentences Whether each set current face face movement of breaking is that the threshold interval that the target transfer amounts are fallen into is corresponding preset each The movement of target face face is covered, when whether each set current face face movement is threshold value that the target transfer amounts are fallen into When the corresponding preset each set target face face movement of section, the target is transferred from the target payment account number and is transferred accounts gold Volume is to the target shroff account number.Because when the corresponding target person of target payment account number and the payment in the transfer request When side is the same person, just each current face's picture is identified using preset face identification method, is obtained each A current face face movement, while when whether the target transfer amounts are fallen into for each set current face face movement When the corresponding preset each set target face face of threshold interval act, the target just is transferred from the target payment account number Transfer amounts are to the target shroff account number.And target transfer amounts are bigger, then belong to the target face face movement The movement of current face face is more, improves the safety that the amount of money is transferred accounts.

Detailed description of the invention

In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings Obtain other attached drawings.

Fig. 1 is an application environment schematic diagram of the transfer account method in one embodiment of the invention based on recognition of face；

Fig. 2 is a flow chart of the transfer account method in one embodiment of the invention based on recognition of face；

Fig. 3 is to determine whether to transfer according to destination number in transfer account method in one embodiment of the invention based on recognition of face One flow chart of the amount of money；

Fig. 4 is in transfer account method in one embodiment of the invention based on recognition of face to determine current mood value and according to this Current mood value determines whether to transfer a flow chart of the amount of money；

Fig. 5 is to judge the corresponding mesh of target payment account number in transfer account method in one embodiment of the invention based on recognition of face It marks personage and paying party is the flow chart of the same person；

Fig. 6 is a flow chart of step S60 in transfer account method in one embodiment of the invention based on recognition of face；

Fig. 7 is a schematic diagram of the transfer device in one embodiment of the invention based on recognition of face；

Fig. 8 is another schematic diagram of the transfer device in one embodiment of the invention based on recognition of face；

Fig. 9 is yet another schematic diagram of the transfer device in one embodiment of the invention based on recognition of face；

Figure 10 is a schematic diagram of computer equipment in one embodiment of the invention.

Specific embodiment

Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts Example, shall fall within the protection scope of the present invention.

Transfer account method provided by the present application based on recognition of face, can be applicable in the application environment such as Fig. 1, wherein visitor Family end is communicated by network with server.Wherein, client can be, but not limited to various personal computers, notebook electricity Brain, smart phone, tablet computer and portable wearable device.Server can use the either multiple services of independent server The server cluster of device composition is realized.

In one embodiment, as shown in Fig. 2, providing a kind of transfer account method based on recognition of face, method application should be based on In financial industry, applies be illustrated for the server in Fig. 1 in this way, include the following steps:

The transfer request that S10, real-time reception paying party are initiated in client；

In the present embodiment, paying party is the natural person for the transfer request initiated in client, such as " Li Si ".

Specifically, usual paying party clicks transfer request button on the transfer request page of client, then client It generates bank transfer request and is sent to server-side, the transfer request that server-side real-time reception paying party is initiated in client, If paying party clicks transfer request button in the homepage of Bank of China's application program of smart phone, then in smart phone State's bank application just generates only one China bank transfer request and is sent to server-side.

It is understood that transfer request includes target payment account number, target payment account number is to draw target transfer amounts Divide to the payment account number of target shroff account number, such as " 621661280000447287 ".Target shroff account number is to collect target payment Account divides the shroff account number for carrying out target transfer amounts.

It should be noted that client can be smart phone or tablet computer etc., target transfer amounts can be 1000 Yuan etc., paying party, client, target payment account number, target transfer amounts and target shroff account number particular content, can To be set according to practical application, herein with no restrictions.

S20, according to transfer request, send transfer information to client and fill in instruction；

In the present embodiment, it is to fill in the instruction of transfer information User Interface, transfer information packet that transfer information, which fills in instruction, Target shroff account number and target transfer amounts are included, transfer information can also include other information.

Specifically, the target shroff account number and target transfer amounts, server-side in order to get paying party need root According to transfer request, transfer information is sent to client and fills in instruction.

S30, target shroff account number and target transfer amounts that paying party fills in instruction upload according to transfer information are received；

Specifically, paying party fills in the target shroff account number and mesh for instructing and uploading according to transfer information in receiving step S20 Transfer amounts are marked, if target shroff account number is " 621661280000447287 ", target transfer amounts are 2500 yuans.

Each current face's picture of S40, in real time acquisition paying party；

Specifically, the current face's picture for acquiring paying party in real time, if the camera using client is on scanning circle Acquire current face's picture of paying party.

S50, judge that the corresponding target person of target payment account number and paying party is the same persons in transfer request, when transferring accounts When the corresponding target person of target payment account number and paying party are the same person in request, step S60 is executed, when in transfer request When the corresponding target person of target payment account number and paying party are not the same persons, step S100 is executed；

Specifically, the target face picture of the corresponding target person of target payment account number in transfer request is obtained first, so The target face key point extracted in the target face picture afterwards is set, and working as current face's picture of applicant is next extracted Preceding face key point is set, when current face's key point location information and consistent target face key point confidence breath, Determine that the corresponding target person of target payment account number and paying party are the same person in transfer request, are performed simultaneously step S60, when When the corresponding target person of target payment account number and paying party are not the same persons in transfer request, step S100 is executed.

S60, each current face's picture is identified using preset face identification method, obtains each set current face Face movement；

In the present embodiment, the movement of current face face refers to that face are according to dynamic caused by reflection on the face of paying party Make, a set of current face face movement includes that more than one current face face acts, and preset face identification method can be Neural network algorithm etc., it should be noted that the particular content of preset face identification method can be carried out according to practical application Setting, herein with no restrictions.

Specifically, first when the corresponding target person of target payment account number in transfer request and paying party are the same person It is then that each face is special first using each face feature point extracted in preset face identification method on current face's picture Position of the sign point on face is input in preparatory trained classifier, is obtained each current face face movement, is such as blinked Eye is opened one's mouth.

S70, the threshold interval that target transfer amounts are fallen into is selected in preset each threshold interval；

Specifically, each threshold interval is pre-set first, then selects step in preset each threshold interval The threshold interval that the target transfer amounts uploaded in rapid S30 are fallen into, such as a threshold interval are [0,1000], another threshold Being worth section is [1001,2000], and target transfer amounts are 1500, then selects 1500 from [0,1000] and [1001,2000] The threshold interval fallen into.

It should be noted that each threshold interval is there are at least two and mutually disjoints, preset threshold interval it is specific Content can be set, herein with no restrictions according to practical application.

S80, judge whether each set current face face movement is that the threshold interval that target transfer amounts are fallen into is corresponding pre- If the movement of each set target face face, when the movement of each set current face face is the threshold interval pair that falls into of target transfer amounts When the preset each set target face face movement answered, step S90 is executed, when the movement of each set current face face is not that target turns When the corresponding preset each set target face face of the threshold interval that the account amount of money is fallen into act, step S121 is executed to step S129；

In the present embodiment, preset target face face movement is face on the face of natural person according to produced by reflection Movement, such as blink.A set of target face face movement includes that more than one target face face acts.Wherein, preset The particular content of target face face movement, can be set, herein with no restrictions according to practical application.

Specifically, respectively cover whether the movement of current face face is that target transfer amounts are fallen into obtained in judgment step S60 The corresponding preset each set target face face movement of threshold interval, when the movement of each set current face face is that target is transferred accounts gold When the corresponding preset each set target face face of the threshold interval that volume is fallen into act, step 90 is executed, when each set current face When face movement is not that the corresponding preset each set target face face of threshold interval that target transfer amounts are fallen into act, execute Step S121 to step S129.

S90, target transfer amounts are transferred from target payment account number to target shroff account number；

It specifically, is the threshold interval pair that target transfer amounts are fallen into when respectively covering the movement of current face face in step S60 When the preset each set target face face movement answered, target transfer amounts are transferred from target payment account number to target gathering account Number, if the movement of current face face is blink, target transfer amounts are that 1800,1800 threshold intervals fallen into are [0,2000], [0,2000] corresponding target face face movement is blink and opens one's mouth, then target transfer amounts are transferred from target payment account number To target shroff account number.

Further, target transfer amounts are bigger, then the threshold interval that target transfer amounts are fallen into is corresponding preset each It is more to cover the movement of target face face.

S100, when the corresponding target person of target payment account number and paying party are not the same persons in transfer request, to Client sends transfer error prompting message, is transferred accounts with present the bill for payment side to bank counter.

Specifically, when the corresponding target person of target payment account number in transfer request and paying party are not the same persons, Transfer error prompting message is sent to client, is transferred accounts with present the bill for payment side to bank counter.Such as to client send " when Before the face that detects be not inconsistent, bank counter please be arrive and transferred accounts.".

It should be noted that the particular content of transfer error prompting message, can be set, herein according to practical application With no restrictions.

In the corresponding embodiment of Fig. 2, asked by real-time reception paying party in the transferring accounts for mobile banking that client is initiated It asks, transfer request includes target payment account number, according to transfer request, sends transfer information to client and fills in instruction, receive and pay Money root fills in the target shroff account number and target transfer amounts that instruction uploads according to transfer information, acquires each of paying party in real time Current face's picture is used when the corresponding target person of target payment account number and paying party are the same person in transfer request Preset face identification method identifies each current face's picture, obtains each current face face movement, works as front Face movement in portion's is that face are chosen in preset each threshold interval according to movement caused by reflection on the face of paying party The threshold interval that target transfer amounts are fallen into out, there are at least two, each threshold interval mutually disjoints threshold interval, sentences Whether each set current face face movement of breaking is the corresponding preset each set target face of threshold interval that target transfer amounts are fallen into Portion's face movement, when the movement of each set current face face is the corresponding preset each set of threshold interval that target transfer amounts are fallen into When target face face act, target transfer amounts are transferred from target payment account number to target shroff account number.Because when transferring accounts When the corresponding target person of target payment account number and paying party are the same person in request, preset face identification method is just used Each current face's picture is identified, obtains each current face face movement, while when each set current face face are dynamic When work is that the corresponding preset each set target face face of threshold interval that target transfer amounts are fallen into act, just pay the bill from target Target transfer amounts are transferred in account to target shroff account number.And target transfer amounts are bigger, then belong to target face face The current face face movement of movement is more, improves the safety that the amount of money is transferred accounts.

In one embodiment, it is somebody's turn to do the transfer account method based on recognition of face to apply in financial industry, Fig. 2 pairs as shown in Figure 3 It answers in embodiment and determines whether that transfer the amount of money applies at one according to destination number in a kind of transfer account method based on recognition of face Flow chart under scene, specifically comprises the following steps:

S121, statistics are not belonging to the number of targets of the current face face movement of preset each set target face face movement Amount；

Specifically, statistics is not belonging to the number of targets of the current face face movement of preset each set target face face movement Amount, so that can determine whether that client shows password User Interface according to the destination number and preset first threshold.

It should be noted that preset first threshold can be 3 etc., and the particular content of preset first threshold, Ke Yigen It is set according to practical application, herein with no restrictions.

S122, judge whether destination number is less than or equal to preset first threshold, when destination number is less than or equal in advance If first threshold when, execute step S123, when destination number be greater than preset first threshold when, execute step S128；

Specifically, whether the destination number counted in judgment step S121 is less than or equal to preset first threshold, When the destination number counted in step S121 is less than or equal to preset first threshold, step S123 is executed, step is worked as When the destination number counted in S121 is greater than preset first threshold, step S128 is executed.

S123, the secret number that paying party uploads is received；

Specifically, the secret number that paying party uploads is received, is such as received " 88012345 " that " Li Si " uploads.

Target secret number in S124, the first appointing system of acquisition；

In the present embodiment, the first appointing system is the database of special storage target secret number, such as safety bank number According to library herein, it should be noted that the particular content of the first appointing system can be set according to practical application, not limited System.

Specifically, store path of the target secret number in the first appointing system is obtained first, then according to the storage The path extraction target secret number.

S125, judge whether the secret number uploaded and target secret number are consistent, when the secret number and target of upload When secret number is consistent, step S126 is executed, when the secret number of upload and target secret number are inconsistent, executes step S127；

Specifically, the secret number uploaded in judgment step S123 and the target secret number got in step S124 are It is no consistent, when the secret number uploaded in step S123 is consistent with target secret number in step S124, step S126 is executed, When target secret number is inconsistent in the secret number and step S124 uploaded in step S123, step S127 is executed.

S126, target transfer amounts are transferred from target payment account number to target shroff account number.

Specifically, when the secret number uploaded in step S124 is consistent with target secret number in step S125, from mesh Transferring target transfer amounts to target shroff account number, such as target payment account number in mark payment account number is " 621661280000447287 ", target transfer amounts are 2000 yuans, and target shroff account number is " 621661280000440055 " then transfer 2000 yuans extremely from " 621661280000447287 " “621661280000440055”。

S127, password miscue message is sent to client, secret number is uploaded again with present the bill for payment side；

Specifically, when target secret number is inconsistent in the secret number and step S125 uploaded in step S124, to Client sends password miscue message, uploads secret number again with present the bill for payment side, if secret number is " 12456 ", Target secret number is " 23568 ", it is clear that secret number and target secret number are inconsistent, then " you input for client transmission Password is wrong, please re-enter ".

It should be noted that the particular content of password miscue message, can be set, herein according to practical application With no restrictions.

S128, transfer error prompting message is sent to client, is transferred accounts with present the bill for payment side to bank counter.

Specifically, it when the destination number counted in step S121 is greater than preset first threshold, is sent out to client Transfer error prompting message is sent, is transferred accounts with present the bill for payment side to bank counter, if destination number is 8, preset first threshold Value is 5, then " quantity of the incongruent current face detected arrived upper limit value, please arrive bank counter for client transmission It transfers accounts.".

In the corresponding embodiment of Fig. 3, by count be not belonging to preset each set target face face movement work as front The destination number of portion's face movement, judges whether destination number is less than or equal to preset first threshold, when destination number is less than Or when being equal to preset first threshold, the secret number that paying party uploads is received, the target obtained in the first appointing system is transferred accounts Password judges whether the secret number uploaded and target secret number are consistent, when the secret number and target secret number of upload When consistent, target transfer amounts are transferred from target payment account number to target shroff account number, when the secret number and target of upload When secret number is inconsistent, password miscue message is sent to client, secret number is uploaded again with present the bill for payment side, when Transfer error prompting message is sent to client when destination number is greater than preset first threshold, bank's cabinet is arrived with present the bill for payment side Platform is transferred accounts.Because when the destination number that count on is less than or equal to preset first threshold, and the secret number uploaded with When the target secret number got in the first appointing system, target transfer amounts are just transferred from target payment account number to target Shroff account number sends transfer error prompting message to client when the destination number counted on is greater than preset first threshold, It is transferred accounts with present the bill for payment side to bank counter, to improve the safety transferred accounts.

In one embodiment, it is somebody's turn to do the transfer account method based on recognition of face to apply in financial industry, Fig. 2 pairs as shown in Figure 4 It answers and determines current mood value in a kind of transfer account method based on recognition of face in embodiment and determined according to the current mood value Flow chart of the amount of money under an application scenarios whether is transferred, is specifically comprised the following steps:

S131, the micro- expression for extracting paying party from current face's picture using preset micro- expression extraction algorithm；

In the present embodiment, preset micro- expression extraction algorithm can be neural network algorithm or deep learning algorithm etc., It should be noted that the particular content of preset micro- expression extraction algorithm, can be set according to practical application, not done herein Limitation.

Specifically, schemed using the current face of preset micro- expression extraction algorithm collected paying party from step S50 In piece extract paying party micro- expression, mainly acquire paying party Facial action unit, it is subsequent can be from Facial action unit Middle extraction obtains each motor unit of micro- expression.As using neural network algorithm from paying party collected in step S50 Micro- expression of paying party, the movements such as micro- expression can raise up for outer eyebrow or cheek lifts are extracted in current face's picture.

Each motor unit in S132, the micro- expression of extraction；

Specifically, the type of motor unit mainly includes 19 kinds of motor units general in the world in following table 1:

1 19 kinds of AU of table

It is understood that can therefrom be extracted in above-mentioned 19 kinds after collecting micro- expression of paying party One, two or more motor units (AU).It is found that it is the current facial characteristics of paying party and movement that motor unit is corresponding, These motor units one have reacted paying party current mood or mood.For example, when paying party happiness, face action meeting It include: the motor units such as the corners of the mouth raises up, cheek lifts, eyelid is tightened, i.e., above-mentioned AU12, AU6, AU7.

S133, the mood value that each motor unit is obtained according to preset micro- expression code of points, preset micro- expression are commented Divider then has recorded the corresponding relationship of motor unit Yu mood value；

Specifically, system can preset the corresponding mood value of 19 kinds of AU, be recorded as micro- expression standards of grading, for example, In a concrete application scene, micro- expression standards of grading are as shown in table 2 below:

2 19 kinds of AU of table

As it can be seen that by micro- expression standards of grading, in extracting micro- expression after each motor unit AU, Ke Yicha It askes micro- expression standards of grading and obtains the corresponding mood value of each motor unit.In above-mentioned concrete application scene, mood value is got over Greatly, indicate that the mood of motor unit reaction paying party is better.

S134, the current mood value that paying party is calculated is carried out according to mood value；

Specifically, after getting the corresponding mood value of each motor unit, the sum of these mood values can be calculated The mood value current as the paying party.For example, accept the example above, if the motor unit of the paying party be respectively AU12, AU6, AU7, then the corresponding mood value got is respectively 3,1,4, so that the mood value that the paying party is calculated is 8.

S135, judge whether the current mood value of paying party is greater than or equal to preset second threshold, when working as paying party When preceding mood value is greater than or equal to preset second threshold, step S136 is executed, is preset when the current mood value of paying party is less than Second threshold when, execute step S137；

In the present embodiment, preset second threshold is pre-set second threshold, can be 9 etc., need to illustrate It is that the particular content of preset second threshold can be set, herein with no restrictions according to practical application.

Specifically, judge whether the current mood value of paying party is greater than or equal to preset second threshold, when paying party When current mood value is greater than or equal to preset second threshold, step S136 is executed, when the current mood value of paying party is less than in advance If second threshold when, execute step S137.

S136, step S100 is executed；

It should be noted that executing step when the current mood value of paying party is greater than or equal to preset second threshold S100, if the current mood value of paying party is 9, preset second threshold is 6,9 greater than 6, thens follow the steps S100.

S137, when the current mood value of paying party be less than preset second threshold when, receive paying party upload transfer accounts it is close Code obtains the target secret number in the first appointing system, judges whether the secret number uploaded and target secret number are consistent, When the secret number of upload is consistent with target secret number, target transfer amounts to target is transferred from target payment account number and is received Money account.

Specifically, when the current mood value of paying party is less than preset second threshold, transferring accounts for paying party upload is received Password, obtain the first appointing system in target secret number, judge upload secret number and target secret number whether one It causes, when the secret number of upload is consistent with target secret number, target transfer amounts is transferred from target payment account number to mesh Shroff account number is marked, when the secret number of upload and target secret number are inconsistent, password miscue are sent to client and disappears Breath, uploads secret number with present the bill for payment side again, it should be noted that turn for receiving paying party and uploading in this step S137 The content of account password is similar with the content of secret number for receiving paying party upload in step S123, in this step S137 Obtain the target in the content of the target secret number in the first appointing system and the first appointing system of acquisition in step S124 The content of secret number is similar, and whether the secret number for judging to upload and target secret number in this step S137 are consistent Whether consistent content and the secret number for judging to upload in step S125 be similar with target secret number content, this step Transferred in slave target payment account number in S137 target transfer amounts to target shroff account number content and step S126 in from The content for transferring target transfer amounts to target shroff account number in target payment account number is similar, no longer illustrates herein.

In the corresponding embodiment of Fig. 4, extracted from current face's picture by using preset micro- expression extraction algorithm Micro- expression of paying party extracts each motor unit in micro- expression, obtains according to preset micro- expression code of points each dynamic Make the mood value of unit, preset micro- expression code of points has recorded the corresponding relationship of motor unit Yu mood value, according to mood Value carries out the current mood value that paying party is calculated, and judges whether the current mood value of paying party is greater than or equal to preset Two threshold values transfer mesh from target payment account number when the current mood value of paying party is greater than or equal to preset second threshold It marks transfer amounts and receives paying party when the current mood value of paying party is less than preset second threshold to target shroff account number The secret number of upload obtains the target secret number in the first appointing system, judges that the secret number uploaded is transferred accounts with target Whether password is consistent, when the secret number of upload is consistent with target secret number, target is transferred from target payment account number and is turned The account amount of money is to target shroff account number.Because micro- expression is that paying party heart really reflects, working as paying party is calculated Preceding mood value is really, when the current mood value of paying party is greater than or equal to preset second threshold, namely to work as paying party Current mood when be positive, target transfer amounts are just transferred from target payment account number to target shroff account number, when paying When the current mood value of money side is less than preset second threshold, the secret number that paying party uploads is received, obtains the first specified system Target secret number in system is transferred from target payment account number when the secret number of upload is consistent with target secret number Target transfer amounts improve the safety transferred accounts anyway transferring accounts when the current mood of paying party is bad to target shroff account number Property.

In one embodiment, it is somebody's turn to do the transfer account method based on recognition of face to apply in financial industry, Fig. 2 pairs as shown in Figure 5 It answers in embodiment and judges the corresponding target person of target payment account number and paying party in a kind of transfer account method based on recognition of face For flow chart of the same person under an application scenarios, specifically comprise the following steps:

S601, face key point is carried out based on current face picture of the preset face critical point detection tool to paying party Detection, obtains current face's key point position on current face's picture；

In the present embodiment, preset face critical point detection tool can be dlib tool, and dlib refers to that one includes The C++ Open-Source Tools packet of machine learning algorithm is the cross-platform library C++, contains many machine learning algorithms.Face is crucial Point includes spectacle rims, nose frame, mouth frame and the face frame etc. of personage.Face key point location information is face pass The coordinate value of key point.It should be noted that preset face critical point detection tool can also be other face critical point detections Tool, the particular content of preset face critical point detection tool, can be set according to actual needs, not limited herein System.

Specifically, face key is carried out based on current face picture of the preset face critical point detection tool to paying party Point detection, obtains current face's key point position on current face's picture.

S602, from the second appointing system obtain transfer request in the corresponding target person of target payment account number target person Face picture；

In the present embodiment, the second appointing system is specially to store the corresponding target of target payment account number in transfer request The database of the target face picture of personage, such as public security system database.It should be noted that the second appointing system can be SQL database etc., the particular content of third appointing system can be set, herein with no restrictions according to practical application.

Specifically, first from obtaining target payment account number corresponding target person in transfer request in the second appointing system Then it is corresponding to extract target payment account number in the acquisition transfer request according to the store path for the store path of target face picture Target person target face picture.

S603, face critical point detection is carried out to target face picture based on preset face critical point detection tool, obtained Target face key point on to target face picture is set；

In the present embodiment, pre- in the content of preset face critical point detection tool and step S601 in this step S603 If face critical point detection tool content it is consistent, no longer illustrate herein.

Specifically, face critical point detection is carried out to target face picture based on preset face critical point detection tool, The target face key point obtained in target face picture is set.

S604, judge whether current face's key point location information and target face key point confidence breath are consistent, when this When current face's key point location information and consistent target face key point confidence breath, step S605 is executed, when deserving forefathers When face key point location information and inconsistent target face key point confidence breath, step S606 is executed；

Specifically, it the current face's key point location information extracted in judgment step S601 and is extracted in step S603 Target face key point confidence breath it is whether consistent, when current face's key point location information for being extracted in step S601 with When the target face key point confidence breath extracted in step S603 is consistent, step S605 is executed, is extracted when in step S601 To current face's key point location information and step S603 in extract target face key point confidence breath it is inconsistent when, Execute step S606.

S605, determine that the corresponding target person of target payment account number and paying party are the same person in transfer request；

Specifically, when extracting in current face's key point location information for being extracted in step S601 and step S603 When target face key point confidence breath is consistent, the corresponding target person of target payment account number and paying party in transfer request are determined For the same person.

S606, determine that the corresponding target person of target payment account number and paying party are not the same persons in transfer request.

Specifically, when extracting in current face's key point location information for being extracted in step S601 and step S603 When target face key point confidence breath is inconsistent, the corresponding target person of target payment account number and payment in transfer request are determined Side is not the same person.

In the corresponding embodiment of Fig. 5, by working as forefathers to paying party based on preset face critical point detection tool Face picture carries out face critical point detection, obtains current face's key point position on current face's picture, from the second specified system The target face picture of the corresponding target person of target payment account number in transfer request is obtained in system, it is crucial based on preset face Point detection instrument carries out face critical point detection to target face picture, obtains the target face key point in target face picture Position judges whether current face's key point location information is ceased with target face key point confidence consistent, as the current face When key point location information and consistent target face key point confidence breath, determine that target payment account number is corresponding in transfer request Target person and paying party are the same person, when current face's key point location information and target face key point confidence cease When inconsistent, determine that the corresponding target person of target payment account number and paying party are not the same persons in transfer request.Because pre- If face critical point detection tool include machine learning algorithm, with self-teaching ability, can be accurately from current face Picture extracts current face's key point position, and accurately can extract target face key point from target face picture It sets, when current face's key point location information and consistent target face key point confidence breath, determines mesh in transfer request It marks the corresponding target person of payment account number and paying party is the same person, improve target payment account number pair in identification transfer request The accuracy of the target person and paying party answered.

In one embodiment, it is somebody's turn to do the transfer account method based on recognition of face to apply in financial industry, Fig. 2 as shown in Figure 6 is extremely In any of Fig. 5 figure corresponding embodiment in a kind of transfer account method based on recognition of face step S60 under an application scenarios Flow chart, specifically comprise the following steps:

S701, the face feature point on current face's picture is extracted for the preset facial characteristics point extracting method of use, Face feature point is behaved the profile essential characteristic points of face on the face；

In the present embodiment, face feature point is behaved the profile essential characteristic point of face on the face, and face feature point includes should Face feature point is in the position on current face's picture namely the face feature point is in the coordinate on current face's picture.

Specifically, using in preset facial characteristics point extracting method extraction step S50 collected applicant it is current Each face feature point in face picture, wherein preset facial characteristics point extracting method can be deep learning method etc., The particular content of preset facial characteristics point extracting method, can be set, herein with no restrictions according to actual needs.

S702, the position for the face feature point extracted to be on current face's picture are input to preparatory training In good classifier, each current face face movement is obtained.

In inventive embodiments, the movement of current face face is face movement on the face of applicant, such as blink.

Specifically, each current face face movement of applicant in order to obtain, needs each face that will have been extracted The position that portion's characteristic point is on face is input in preparatory trained classifier.

It should be noted that the movement of current face face is that face are moved according to caused by reflection on the face of applicant Make, such as blinks or open one's mouth.The classifier can be recurrence classifier, Naive Bayes Classifier or neural network classifier Deng the particular content of the classifier can be set according to practical application, herein with no restrictions.

In the corresponding embodiment of Fig. 6, current face's picture is extracted by using preset facial characteristics point extracting method On face feature point, face feature point includes the position that the face feature point is on current face's picture, face feature point For the profile essential characteristic point of people's face on the face, the position that the face feature point extracted is on current face's picture is defeated Enter into preparatory trained classifier, obtains each current face face movement.Because preparatory trained classifier has The position that face feature point is in face picture is accurately sorted out to the function of facial face movement, so can be from having mentioned The position that the face feature point got is on current face's picture accurately sorts out each current face face movement, improves Obtain current face face accuracy of action.

It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit It is fixed.

In one embodiment, a kind of transfer device based on recognition of face is provided, it should the transfer device based on recognition of face It is corresponded with the transfer account method in above-described embodiment based on recognition of face.As shown in fig. 7, should the dress of transferring accounts based on recognition of face It sets and judges mould including the first receiving module 701, the first sending module 702, the second receiving module 703, acquisition module 704, first Block 705, identification module 706, selection module 707, the second judgment module 708, first transfer module 709 and the second sending module 710.Detailed description are as follows for each functional module:

First receiving module 701, for the transfer request that real-time reception paying party is initiated in client, transfer request includes Target payment account number；

First sending module 702, for sending transfer information to client and filling in instruction according to transfer request；

Second receiving module 703 fills in the target shroff account number that instruction uploads according to transfer information for receiving paying party With target transfer amounts；

Acquisition module 704, for acquiring each current face's picture of paying party in real time；

First judgment module 705, for judging the corresponding target person of target payment account number and paying party in transfer request For the same person；

Identification module 706 is same for the corresponding target person of target payment account number in the transfer request and paying party When personal, each current face's picture is identified using preset face identification method, obtains each set current face face Movement, the movement of current face face refer to that face are according to movement caused by reflection on the face of paying party；

Choose module 707, the threshold fallen into for selecting target transfer amounts in preset each threshold interval It is worth section, each threshold interval is there are at least two and mutually disjoints；

Second judgment module 708, for judging whether each set current face face movement is that target transfer amounts are fallen into The corresponding preset each set target face face movement of threshold interval；

First transfers module 709, for being the threshold zone that target transfer amounts are fallen into when the movement of each set current face face Between corresponding preset each set target face face act when, transferred from target payment account number target transfer amounts to target receive Money account；

Second sending module 710, not for the corresponding target person of target payment account number in the transfer request and paying party When being the same person, transfer error prompting message is sent to client, is transferred accounts with present the bill for payment side to bank counter.

Further, the transfer device based on recognition of face can also include:

Statistical module 711, for count be not belonging to preset each set target face face movement current face face it is dynamic The destination number of work；

Third judgment module 712, for judging whether destination number is less than or equal to preset first threshold；

Third receiving module 713, for receiving the secret number of paying party upload；

First obtains module 714, for obtaining the target secret number in the first appointing system；

Whether the 4th judgment module 715, the secret number and target secret number for judging to upload are consistent；

Second transfers module 716, for when the secret number of upload is consistent with target secret number, pays the bill account from target Target transfer amounts are transferred in number to target shroff account number.

Third sending module 717 is uploaded for sending password miscue message to client with present the bill for payment side again Secret number；

4th sending module 718 arrives bank's cabinet for sending transfer error prompting message to client with present the bill for payment side Platform is transferred accounts.

Further, the transfer device based on recognition of face can also include:

First extraction module 719, for extracting payment from current face's picture using preset micro- expression extraction algorithm Micro- expression of side；

Second extraction module 720, for extracting each motor unit in micro- expression；

Second obtains module 721, for obtaining the mood value of each motor unit according to preset micro- expression code of points, Preset micro- expression code of points has recorded the corresponding relationship of motor unit Yu mood value；

Computing module 722, the current mood value for according to mood value carrying out that paying party is calculated；

5th judgment module 723, for judging whether the current mood value of paying party is greater than or equal to preset second threshold Value, when the current mood value of paying party is greater than or equal to preset second threshold；

Third transfers module 724, for holding when the current mood value of paying party is greater than or equal to preset second threshold The step of row transfers target transfer amounts to target shroff account number from target payment account number；

4th receiving module 725, for receiving payment when the current mood value of paying party is less than preset second threshold The secret number of Fang Shangchuan obtains the target secret number in the first appointing system, judges that the secret number uploaded and target turn Whether account password is consistent, when the secret number of upload is consistent with target secret number, transfers target from target payment account number Transfer amounts are to target shroff account number.

Further, the transfer device based on recognition of face can also include:

First detection module, for based on preset face critical point detection tool to current face's picture of paying party into Pedestrian's face critical point detection obtains current face's key point position on current face's picture；

Third obtain module, for from the second appointing system obtain transfer request in the corresponding target of target payment account number The target face picture of personage；

Second detection module, for carrying out face pass to target face picture based on preset face critical point detection tool The detection of key point, the target face key point obtained in target face picture are set；

6th judgment module, for judging that current face's key point location information and target face key point confidence breath is It is no consistent；

First determining module, for when current face's key point location information and target face key point confidence breath one When cause, determine that the corresponding target person of target payment account number and paying party are the same person in transfer request；

Second determining module, for being ceased not when current face's key point location information and target face key point confidence When consistent, determine that the corresponding target person of target payment account number and paying party are not the same persons in transfer request.

Further, identification module 707 may include:

Submodule is refined, it is special for extracting the face on current face's picture using preset facial characteristics point extracting method Levy point, face feature point is behaved the profile essential characteristic points of face on the face；

Input submodule, the position for the face feature point extracted to be on current face's picture are input to pre- First in trained classifier, each current face face movement is obtained.

Specific restriction about the transfer device based on recognition of face may refer to above for based on recognition of face The restriction of transfer account method, details are not described herein.Modules in the above-mentioned transfer device based on recognition of face can whole or portion Divide and is realized by software, hardware and combinations thereof.Above-mentioned each module can be embedded in the form of hardware or independently of computer equipment In processor in, can also be stored in a software form in the memory in computer equipment, in order to processor calling hold The corresponding operation of the above modules of row.

In one embodiment, a kind of computer equipment is provided, which can be server, internal junction Composition can be as shown in Figure 10.The computer equipment include by system bus connect processor, memory, network interface and Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating The database of machine equipment is for storing data involved by the transfer account method based on recognition of face.The network of the computer equipment Interface is used to communicate with external terminal by network connection.To realize that one kind is based on when the computer program is executed by processor The transfer account method of recognition of face.

In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory And the computer program that can be run on a processor, processor realize that above-described embodiment is known based on face when executing computer program The step of other transfer account method, such as step S10 shown in Fig. 2 to step S100.Alternatively, when processor executes computer program Realize the function of each module/unit of the transfer device based on recognition of face in above-described embodiment, such as the first reception shown in Fig. 7 The function of 701 to the second sending module 710 of module.To avoid repeating, which is not described herein again.

In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated The transfer account method based on recognition of face in above method embodiment is realized when machine program is executed by processor, alternatively, the computer The function of each module/unit in the transfer device based on recognition of face in above-mentioned apparatus embodiment is realized when program is executed by processor Energy.To avoid repeating, which is not described herein again.Those of ordinary skill in the art will appreciate that realizing complete in above-described embodiment method Portion or part process are relevant hardware can be instructed to complete by computer program, and the computer program can be deposited It is stored in a non-volatile computer read/write memory medium, the computer program is when being executed, it may include such as above-mentioned each method Embodiment process.Wherein, used in each embodiment provided herein to memory, storage, database or its Any reference of its medium, may each comprise non-volatile and/or volatile memory.Nonvolatile memory may include read-only deposits Reservoir (ROM), programming ROM (PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms, such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate (DDR) SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory are total Line (Rambus) directly RAM (RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..

It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing The all or part of function of description.

Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all It is included within protection scope of the present invention.

Claims (10)

1. a kind of transfer account method based on recognition of face, which is characterized in that the transfer account method based on recognition of face includes:

The transfer request that real-time reception paying party is initiated in client, the transfer request include target payment account number；

According to the transfer request, Xiang Suoshu client sends transfer information and fills in instruction；

It receives the paying party and the target shroff account number and target transfer amounts that instruction uploads is filled according to the transfer information；

Each current face's picture of the paying party is acquired in real time；

When the corresponding target person of target payment account number in the transfer request and the paying party are the same person, using pre- If face identification method each current face's picture is identified, obtain the movement of each set current face face, currently Facial face movement refers to that face are according to movement caused by reflection on the face of the paying party；

The threshold interval that the target transfer amounts are fallen into, each threshold are selected in preset each threshold interval Value section is there are at least two and mutually disjoints；

Judge whether each set current face face movement is that the threshold interval that the target transfer amounts are fallen into is corresponding pre- If each set target face face movement；

When each set current face face movement is that the threshold interval that the target transfer amounts are fallen into is corresponding preset each When covering the movement of target face face, the target transfer amounts are transferred from the target payment account number to target gathering account Number.

2. as described in claim 1 based on the transfer account method of recognition of face, which is characterized in that when each set current face five Official's movement is not that threshold interval corresponding preset each set target face face that the target transfer amounts are fallen into are held when acting Row following steps:

Statistics is not belonging to the destination number of the current face face movement of preset each set target face face movement；

Judge whether the destination number is less than or equal to preset first threshold；

When the destination number is less than or equal to preset first threshold, the secret number that the paying party uploads is received；

Obtain the target secret number in the first appointing system；

Judge whether secret number and the target secret number of the upload are consistent；

When the secret number of the upload is consistent with the target secret number, transferred from the target payment account number described Target transfer amounts are to the target shroff account number.

3. as described in claim 1 based on the transfer account method of recognition of face, which is characterized in that pay the bill described from the target Before transferring the target transfer amounts to the target shroff account number in account, the transfer account method based on recognition of face is also Include:

Micro- expression of the paying party is extracted from current face's picture using preset micro- expression extraction algorithm；

Extract each motor unit in micro- expression；

The mood value of each motor unit, preset micro- expression scoring are obtained according to preset micro- expression code of points The regular record corresponding relationship of motor unit and mood value；

The current mood value that the paying party is calculated is carried out according to the mood value；

When the current mood value of the paying party is greater than or equal to preset second threshold, execute described from target payment The step of target transfer amounts to the target shroff account number is transferred in account；

When the current mood value of the paying party is less than preset second threshold, it is close to receive transferring accounts for the paying party upload Code obtains the target secret number in the first appointing system, judges the secret number and the target secret number of the upload It is whether consistent, when the secret number of the upload is consistent with the target secret number, drawn from the target payment account number The target transfer amounts are dialled to the target shroff account number.

4. as described in claim 1 based on the transfer account method of recognition of face, which is characterized in that judge mesh in the transfer request The corresponding target person of payment account number is marked with the paying party includes: for the same person

Face critical point detection is carried out based on current face picture of the preset face critical point detection tool to the paying party, Obtain current face's key point position on current face's picture；

From the target face figure for obtaining the corresponding target person of target payment account number in the transfer request in the second appointing system Piece；

Face critical point detection is carried out to the target face picture based on preset face critical point detection tool, is obtained described Target face key point in target face picture is set；

Judge whether current face's key point location information is ceased with the target face key point confidence consistent；

When current face's key point location information and consistent target face key point confidence breath, described turn is determined The corresponding target person of target payment account number and the paying party are the same person in account request.

5. according to any one of claims 1 to 4 based on the transfer account method of recognition of face, which is characterized in that the use Preset face identification method identifies each current face's picture, obtains each current face face movement packet It includes:

Face feature point on current face's picture is extracted using preset facial characteristics point extracting method, the face is special Sign point is the profile essential characteristic point of people's face on the face；

The position that the face feature point extracted is on current face's picture is input in preparatory trained classifier, Obtain each current face face movement.

6. a kind of transfer device based on recognition of face, which is characterized in that the transfer device based on recognition of face includes:

First receiving module, for the transfer request that real-time reception paying party is initiated in client, the transfer request includes mesh Mark payment account number；

First sending module, for according to the transfer request, Xiang Suoshu client to send transfer information and fills in instruction；

Second receiving module fills in the target shroff account number that instruction uploads according to the transfer information for receiving the paying party With target transfer amounts；

Acquisition module, for acquiring each current face's picture of the paying party in real time；

Identification module is same for the corresponding target person of target payment account number in the transfer request and the paying party When personal, each current face's picture is identified using preset face identification method, obtains each current face Face movement, the movement of current face face refer to that face are according to movement caused by reflection on the face of the paying party；

Choose module, the threshold zone fallen into for selecting the target transfer amounts in preset each threshold interval Between, each threshold interval is there are at least two and mutually disjoints；

Second judgment module, for judging whether the target transfer amounts are fallen into for each set current face face movement The corresponding preset each set target face face movement of threshold interval；

First transfers module, for being the threshold zone that the target transfer amounts are fallen into when each set current face face movement Between corresponding preset each set target face face act when, the target transfer amounts are transferred from the target payment account number To the target shroff account number.

7. as claimed in claim 6 based on the transfer device of recognition of face, which is characterized in that described to be turned based on recognition of face Account device further include:

Statistical module, for counting the target for being not belonging to the current face face movement of preset each set target face face movement Quantity；

Third judgment module, for judging whether the destination number is less than or equal to preset first threshold；

Third receiving module, for when whether the destination number is less than or equal to preset first threshold, receiving described pay The secret number that money side uploads；

First obtains module, for obtaining the target secret number in the first appointing system；

Whether the 4th judgment module, the secret number and the target secret number for judging the upload are consistent；

Second transfers module, for when the secret number of the upload is consistent with the target secret number, from the target The target transfer amounts are transferred in payment account number to the target shroff account number.

8. the transfer device based on recognition of face as described in any one of claim 6 to 7, which is characterized in that described to be based on The transfer device of recognition of face further include:

First extraction module, for extracting the payment from current face's picture using preset micro- expression extraction algorithm Micro- expression of side；

Second extraction module, for extracting each motor unit in micro- expression；

Second obtains module, for obtaining the mood value of each motor unit, institute according to preset micro- expression code of points State the corresponding relationship that preset micro- expression code of points has recorded motor unit Yu mood value；

Computing module, the current mood value for according to the mood value carrying out that the paying party is calculated；

Third transfers module, for executing when the current mood value of the paying party is greater than or equal to preset second threshold The described the step of target transfer amounts to the target shroff account number is transferred from the target payment account number；

4th receiving module, for receiving described pay when the current mood value of the paying party is less than preset second threshold The secret number that money side uploads obtains the target secret number in the first appointing system, judge the secret number of the upload with Whether the target secret number is consistent, when the secret number of the upload is consistent with the target secret number, from described The target transfer amounts are transferred in target payment account number to the target shroff account number.

9. a kind of computer equipment, including memory, processor and storage are in the memory and can be in the processor The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to The step of transfer account method described in any one of 5 based on recognition of face.

10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists In realization is turned as described in any one of claims 1 to 5 based on recognition of face when the computer program is executed by processor The step of account method.