CN111694825B

CN111694825B - Mechanism withdrawal data verification method, device, computer equipment and storage medium

Info

Publication number: CN111694825B
Application number: CN202010461771.6A
Authority: CN
Inventors: 李泽民
Original assignee: Ping An Bank Co Ltd
Current assignee: Ping An Bank Co Ltd
Priority date: 2020-05-27
Filing date: 2020-05-27
Publication date: 2024-04-09
Anticipated expiration: 2040-05-27
Also published as: CN111694825A

Abstract

The invention discloses a mechanism withdrawal data verification method, a device, computer equipment and a storage medium, which relate to the technical field of image recognition, wherein each inspection item in a withdrawal information table is firstly inspected to obtain a first type equipment inspection item field, a second type equipment inspection item field and a first type field inspection item field, a first trigger instruction is generated according to the correspondence of the first type equipment inspection item field and is sent to first type equipment, a second trigger instruction is generated according to the correspondence of the second type equipment inspection item field and is sent to second type equipment, a third trigger instruction is generated according to the correspondence of the first type field inspection item field and is sent to first type field equipment, then whether the first type equipment and the second type equipment are emptied is recognized by adopting a mode of combining image recognition and pressure parameter values, and whether the first type field is emptied is recognized by adopting image recognition. The method realizes the automatic inspection of each inspection item in the merge information table, and improves the efficiency of checking the data items.

Description

Mechanism withdrawal data verification method, device, computer equipment and storage medium

Technical Field

The present invention relates to the field of image recognition technologies, and in particular, to a method and apparatus for verifying organization withdrawal data, a computer device, and a storage medium.

Background

At present, when a server receives a data check list uploaded by a user and whether actual values of all fields in the data check list reach expected values or not is checked one by one, a common process is to manually check the actual values of each field on site and then manually compare the actual values with the corresponding expected values. The mode of checking whether the actual value of each field in the manual checking data check list reaches the expected value one by one is adopted, so that the efficiency is low, and errors are easy to occur in the checking process.

Disclosure of Invention

The embodiment of the invention provides a mechanism withdrawal data verification method, a mechanism withdrawal data verification device, computer equipment and a storage medium, which aim to solve the problems that in the prior art, when actual values of all fields in a data check table of an uploading server are verified one by one to reach expected values, the actual values of all the fields are manually checked on site and then manually compared with the corresponding expected values, so that the efficiency is low and errors are easy to occur in the verification process.

In a first aspect, an embodiment of the present invention provides a method for verifying organization withdrawal data, including:

receiving a withdrawal information table uploaded by a mechanism uploading terminal corresponding to a mechanism to be withdrawn; the disarming information table comprises a first type equipment check item field, a second type equipment check item field and a first type site check item field;

acquiring a first type equipment check item field, a second type equipment check item field and a first type field check item field which are included in the disarming information table, generating a first trigger instruction according to the first type equipment check item field to send to first type equipment, generating a second trigger instruction according to the second type equipment check item field to send to second type equipment, and generating a third trigger instruction according to the first type field check item field to send to first type field equipment;

receiving a first type equipment live-action picture and a first pressure parameter value of first type equipment, which are sent by the first type equipment according to a first trigger instruction, and acquiring a first field value corresponding to a first type equipment check item field through a first identification result corresponding to the first type equipment live-action picture and the first pressure parameter value so as to store the first field value into a cell corresponding to the first type equipment check item field in the withdrawal information table;

Receiving a second type equipment live-action picture and a second pressure parameter value of second type equipment, which are sent by the second type equipment according to a second trigger instruction, and acquiring a second field value corresponding to a second type equipment check item field through a second identification result corresponding to the second type equipment live-action picture and the second pressure parameter value so as to store the second field value into a cell corresponding to the second type equipment check item field in the withdrawal information table;

receiving a first type field live-action picture sent by first type field equipment according to a third trigger instruction, and acquiring a third field value corresponding to a first type field inspection item field through a third identification result corresponding to the first type field live-action picture so as to store the third field value into a cell corresponding to the first type field inspection item field in the disarming information table;

judging whether the value of the stored first field value, second field value and third field value in the withdrawal information table is unequal to the value of the corresponding field value in a preset withdrawal information expected value table; and

and if the value of the first field value, the value of the second field value and the value of the third field value stored in the withdrawal information table are not equal to the value of the corresponding field value in the withdrawal information expected value table, sending prompt information for prompting that the field value does not pass to the mechanism uploading terminal.

In a second aspect, embodiments of the present invention provide a facility evacuation data verification apparatus, comprising:

the withdrawal information table acquisition unit is used for receiving the withdrawal information table uploaded by the mechanism uploading terminal corresponding to the mechanism to be withdrawn; the disarming information table comprises a first type equipment check item field, a second type equipment check item field and a first type site check item field;

the trigger instruction sending unit is used for obtaining a first type equipment check item field, a second type equipment check item field and a first type field check item field which are included in the withdrawal information table, generating a first trigger instruction according to the first type equipment check item field to send to the first type equipment, generating a second trigger instruction according to the second type equipment check item field to send to the second type equipment, and generating a third trigger instruction according to the first type field check item field to send to the first type field equipment;

the first field value obtaining unit is used for receiving a first type equipment live-action picture and a first pressure parameter value of first type equipment, which are sent by the first type equipment according to a first trigger instruction, and obtaining a first field value corresponding to a first type equipment inspection item field through a first identification result corresponding to the first type equipment live-action picture and the first pressure parameter value so as to store the first field value corresponding to the first type equipment inspection item field in the parallel removing information table;

A second field value obtaining unit, configured to receive a second type device live-action picture and a second pressure parameter value of a second type device, where the second type device live-action picture and the second pressure parameter value are sent by a second trigger instruction, and obtain, through a second identification result corresponding to the second type device live-action picture and the second pressure parameter value, a second field value corresponding to a second type device inspection item field, so that the second field value is stored in a cell corresponding to the second type device inspection item field in the parallel removing information table;

a third field value obtaining unit, configured to receive a first type field live-action picture sent by the first type field device according to a third trigger instruction, obtain, according to a third identification result corresponding to the first type field live-action picture, a third field value corresponding to a first type field inspection item field, and store the third field value in a cell corresponding to the first type field inspection item field in the disarming information table;

the field value comparison unit is used for judging whether the value of the stored first field value, second field value and third field value in the withdrawal information table is unequal to the value of the corresponding field value in the preset withdrawal information expected value table; and

And the prompt information sending unit is used for sending prompt information for prompting that the value of the field is not passed to the mechanism uploading terminal if the value of the first field, the value of the second field and the value of the third field stored in the withdrawal information table are not equal to the value of the corresponding field in the withdrawal information expected value table.

In a third aspect, an embodiment of the present invention further provides a computer apparatus, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the mechanism withdrawal data verification method described in the first aspect when the processor executes the computer program.

In a fourth aspect, embodiments of the present invention also provide a computer readable storage medium, where the computer readable storage medium stores a computer program, which when executed by a processor, causes the processor to perform the mechanism disarming data verification method described in the first aspect.

The embodiment of the invention provides a mechanism withdrawal data verification method, a device, a computer device and a storage medium, wherein a server is informed to automatically check each check item in a withdrawal information table in the day before withdrawal of a mechanism to be withdrawn, a first type equipment check item field, a second type equipment check item field and a first type field check item field which are included in the withdrawal information table are obtained, a first trigger instruction is generated according to the first type equipment check item field to send the first trigger instruction to first type equipment, a second trigger instruction is generated according to the second type equipment check item field to send the second trigger instruction to second type equipment, a third trigger instruction is generated according to the first type field check item field to send the third trigger instruction to first type field equipment, then whether the first type equipment and the second type equipment are emptied is identified by adopting a mode of combining image identification and pressure parameter values, and whether the first type field is emptied is identified by adopting image identification. The method realizes the automatic inspection of each inspection item in the merge information table, and improves the efficiency of checking the data items.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic diagram of an application scenario of a mechanism withdrawal data verification method provided by an embodiment of the present invention;

FIG. 2 is a schematic flow chart of a method for verifying organization withdrawal data according to an embodiment of the present invention;

FIG. 3 is a schematic view of a sub-flow of a method for verifying organization withdrawal data according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of another sub-flow of the method for verifying the organization retirement data according to the embodiment of the present invention;

FIG. 5 is a schematic block diagram of a facility evacuation data verification apparatus provided by an embodiment of the present invention;

FIG. 6 is a schematic block diagram of a subunit of the facility evacuation data verification apparatus provided by an embodiment of the present invention;

FIG. 7 is a schematic block diagram of another subunit of the facility evacuation data verification apparatus provided by an embodiment of the present invention;

fig. 8 is a schematic block diagram of a computer device according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

It should be understood that the terms "comprises" and "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

It is also to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.

Referring to fig. 1 and fig. 2, fig. 1 is a schematic view of an application scenario of a mechanism merge data verification method according to an embodiment of the present invention; fig. 2 is a schematic flow chart of a method for verifying organization withdrawal data, which is provided by an embodiment of the present invention, wherein the method for verifying organization withdrawal data is applied to a server, and the method is executed by application software installed in the server.

As shown in fig. 2, the method includes steps S110 to S170.

S110, receiving a withdrawal information table uploaded by a mechanism uploading terminal corresponding to a mechanism to be withdrawn; the disarming information table comprises a first type equipment check item field, a second type equipment check item field and a first type site check item field.

In this embodiment, the institution evacuation is to plan and adjust the behavior for the business hall for external service, for example, taking a banking system as an example, where a banking institution refers to an atomic institution without a lower level, such as a branch business hall of a bank; withdrawal refers to the conversion of withdrawal and incorporation, i.e. the delivery of data between two banking institutions, the withdrawal process transferring the data of the withdrawn institution to the receiving institution. In order to ensure that clients, accounts and related services of the revocation authority transferred to the receiving authority can still be processed normally, it is necessary to perform finishing and batch conversion processing on service data of the revocation authority.

At present, each item of data inspection of the revocation organization is realized based on manual inspection, namely each item of data to be inspected is inspected in a manual inspection mode according to each item of data to be inspected withdrawn by the organization. In order to improve the checking efficiency of the item data to be checked in the mechanism withdrawal process, the method is carried out by combining with image recognition.

In order to more clearly understand the technical scheme of the application, the following describes the terminal related to the application in detail.

The first is an organization uploading terminal, which can be understood as a local server deployed in a place where an organization is revoked, and is configured to receive and cache data uploaded by devices such as teller terminals, ATM machines (i.e., self-service teller machines), teller boxes, and the like deployed in the place where the organization is revoked, and then send the cached data to a total server.

The second type of equipment is first type equipment, and when the implementation is carried out, the first type equipment can be a teller trunk provided with a camera and a pressure sensor in the trunk, articles such as certificates, cash and the like are placed in the teller trunk, and when a withdrawal mechanism for placing the teller trunk needs to be withdrawn, whether the articles in the teller trunk are empty or not needs to be checked.

And the third is a second type device, and in specific implementation, the second type device can be an ATM (automatic teller machine) with a camera and a pressure sensor arranged in a cabinet body, cash is placed in the ATM, and when a withdrawal mechanism for placing the ATM needs to withdraw, whether the interior of each ATM is empty or not needs to be checked.

Fourth, the first type of field device, when embodied, may be a 360 degree panoramic camera disposed within the warehouse for checking real-time scenes in the warehouse, such as checking whether the warehouse is empty.

And fifth, the server, when embodied, can be understood as a total server, for receiving data uploaded by various terminals.

In this embodiment, if a certain revocation authority needs to perform an institutional revocation, at this time, a plurality of items of data of the revocation authority need to be checked, at this time, the authority needs to upload a terminal to upload a revocation list, and the revocation list includes at least fields of a first type equipment check item field, a second type equipment check item field, and a first type site check item field, so as to notify the total server that the fields are items to be checked in the checking process. In specific implementation, for example, the first type equipment inspection item field is a teller trunk inspection item field (i.e., whether the teller trunk is empty is judged, if empty, the value of the field is 0, and if not empty, the value of the field is 1); the second type equipment check item field is an ATM check item field (i.e., whether the ATM is empty is determined, if empty, the value of the field is 0, and if not empty, the value of the field is 1); the first type field is a warehouse field check item field (i.e., whether the warehouse is empty is determined, if empty, the value of the field is 0, and if not empty, the value of the field is 1).

S120, a first type equipment inspection item field, a second type equipment inspection item field and a first type field inspection item field which are included in the disarming information table are obtained, a first trigger instruction is generated according to the first type equipment inspection item field and is sent to the first type equipment, a second trigger instruction is generated according to the second type equipment inspection item field and is sent to the second type equipment, and a third trigger instruction is generated according to the first type field inspection item field and is sent to the first type field equipment.

In this embodiment, after the server receives the merge information table, it may analyze and learn which fields need to be checked to determine whether the specific values are the same as the preset expected values of the fields, for example, the preset expected value of the first type equipment inspection item field is 0, and the value of the first type equipment inspection item field after the specific check is actually performed is 1, which indicates that the check content corresponding to the first type equipment inspection item field is not passed, and a prompt mechanism is required to timely prompt a user corresponding to the uploading terminal to review.

In order to facilitate understanding of the technical solution of the present application, the present application focuses on the 3 inspections of teller's tail box, ATM machine and warehouse. In order to drive a teller trunk, an ATM and a storehouse to correspondingly upload a current field picture, a first trigger instruction is generated according to the corresponding field of the checking item of the first type of equipment to be sent to the first type of equipment, a second trigger instruction is generated according to the corresponding field of the checking item of the second type of equipment to be sent to the second type of equipment, and a third trigger instruction is generated according to the corresponding field of the checking item of the first type of equipment to be sent to the first type of field equipment.

S130, receiving a first type equipment live-action picture and a first pressure parameter value of the first type equipment, which are sent by the first type equipment according to a first trigger instruction, and acquiring a first field value corresponding to a first type equipment check item field through a first identification result corresponding to the first type equipment live-action picture and the first pressure parameter value so as to store the first field value into a cell corresponding to the first type equipment check item field in the withdrawal information table.

In this embodiment, after the server receives the first type equipment live-action picture and the first pressure parameter value of the first type equipment, the first identification result corresponding to the first type equipment live-action picture may be obtained through image identification, and the first field value corresponding to the first type equipment inspection item field may be obtained by combining the first pressure parameter value. In this way, automatic remote inventory of the teller's tail box is achieved.

In one embodiment, the first type of device is a bank teller boot; as shown in fig. 3, the step S130 includes:

s131, receiving a first type equipment live-action picture and a first pressure parameter value of the first type equipment, which are sent by the first type equipment according to a first trigger instruction, and carrying out image recognition on the first type equipment live-action picture through a trained first depth convolution neural network model to obtain a first recognition result corresponding to the first type equipment live-action picture;

S132, judging whether the first identification result corresponds to an empty box or not, and judging whether the first pressure parameter value is zero or not;

s133, if the first identification result corresponds to an empty box and the first pressure parameter value is zero, setting the value of a first field corresponding to a first type equipment inspection item field to be 0, and storing the value of the first field into a cell corresponding to the first type equipment inspection item field in the parallel removing information table;

and S134, if the first identification result corresponds to a non-empty box and the first pressure parameter value is greater than zero, setting the value of a first field corresponding to a first type equipment inspection item field to be 1, and storing the first field into a cell corresponding to the first type equipment inspection item field in the parallel removing information table.

In this embodiment, the first deep convolutional neural network model is an improved deep convolutional neural network model, and compared with the deep convolutional neural network model, the difference is that the activation function adopts a leakage rectification function (leakage-Rectified Linear Unit, leakage-ReLU), and the pooling layer adopts a multi-level spatial pyramid pooling layer. Through the improved deep convolutional neural network model, whether the tail box of the teller is cleaned can be accurately identified.

In one embodiment, as shown in fig. 4, step S131 includes:

s1311, obtaining a first brightness-adjusted picture by carrying out brightness correction on the first type equipment live-action picture;

s1312, performing geometric correction on the picture with the first brightness adjusted to obtain a first preprocessed picture;

s1313, performing convolution with a step length of 1*1 on input arrays corresponding to RGB channels corresponding to the first preprocessed picture through convolution kernels of 3*3 in the first depth convolution neural network model to obtain a feature map corresponding to the first preprocessed picture;

s1314, pooling the feature images corresponding to the first preprocessed picture through a multi-level space pyramid pooling layer consisting of 1*1 blocks, 1*3 blocks, 3*1 blocks and 4*4 blocks to obtain picture features corresponding to the first preprocessed picture;

s1315, inputting the picture features to a full-connection layer in a first depth convolutional neural network model to obtain one-dimensional vectors, and classifying the one-dimensional vectors by a softmax layer to obtain a first recognition result corresponding to the first type equipment live-action picture.

In this embodiment, in order to identify whether there is a banknote in the first type equipment live-action picture, brightness correction and geometric correction are required to be performed on the first type equipment live-action picture. And after the first preprocessing picture is obtained by carrying out brightness correction and geometric correction on the first type equipment live-action picture, the input arrays respectively corresponding to the RGB channels corresponding to the first preprocessing picture are convolved through the first depth convolution neural network model.

The convolution process is to effectively extract the picture characteristics of the first preprocessed picture, and the activation function used in the convolution process is a leak ReLU function (i.e. a rectifier function with leakage), and the specific expression of the leak ReLU function is as follows:

although the ReLU function (i.e. the linear rectification function) can reduce the parameter dependence and alleviate the over-fitting problem, the output of the ReLU function is easy to fall into a hard saturation region so as to generate a neuron necrosis phenomenon, so that partial information loss of paper money images occurs in the training process, and the accuracy of a neural network is reduced. The leak ReLU function used in the present application not only has all the advantages of the ReLU function, but also solves the problem of partial neuron "necrosis" caused by the ReLU function.

And then, adopting a multi-level spatial pyramid pooling algorithm can extract banknote image features from different angles, and simultaneously, an image processor (GPU) can be fully utilized, so that the instantaneity of the algorithm is ensured.

Finally, the picture features are input to the fully connected and softmax layers (the softmax function used in the softmax layers can "compress" one K-dimensional vector z containing any real number into another K-dimensional real vector σ (z) such that each element ranges between (0, 1) and the sum of all elements is 1) is prior art and will not be discussed further herein. The result corresponding to the first recognition result includes a null value (i.e., representing a null box), a rmb (a specific denomination can be recognized when the embodiment is performed), dollars (a specific denomination can be recognized when the embodiment is performed), euros (a specific denomination can be recognized when the embodiment is performed), and so on.

After the identification of the first type equipment live-action picture is completed, whether articles such as paper money, passbooks or certificates and the like are still left in the tail box of the teller can be judged.

In one embodiment, step S1311 includes:

performing image cutting on the first type equipment live-action picture through an image cutting model based on the image to obtain a first cut picture;

carrying out image normalization processing on the first cut picture to obtain a first normalized picture;

calling a preset brightness layering number to correspondingly perform image layering on the first normalized picture to obtain a first layered picture;

carrying out brightness layer combination on the first cut picture according to the first layered picture to obtain a first combined picture;

acquiring the pixel point proportion of each brightness layer in the first combined picture, and acquiring the brightness layer with the maximum pixel point proportion as a target brightness layer;

and taking each brightness layer with the brightness value smaller than that of the target brightness layer as a brightness layer to be adjusted, and adjusting the brightness value corresponding to the brightness layer to be adjusted to be equal to that of the target brightness layer so as to obtain a picture after the first brightness adjustment.

In the embodiment, the Graph-Based Segmentation model is an image cutting model based on a Graph, and the algorithm is a greedy clustering algorithm based on the Graph, so that the implementation is simple and the speed is high. This algorithm is prior art and will not be discussed further herein.

Image normalization refers to the process of transforming an image into a fixed standard form by performing a series of standard process transformations on the image, which standard image is referred to as a normalized image. The original image can obtain various duplicate images after being subjected to some processing or attack, and the images can obtain standard images in the same form after being subjected to image normalization processing of the same parameters.

When a pixel point is subjected to pixel value normalization processing, referring to a formula such as X '= (X-x_min)/(x_max-x_min), wherein X' is a pixel value after normalization processing of a R channel value (also a G channel value or a B channel value) of the pixel point; x is the initial value of the R channel value (also can be G channel value or B channel value) of the pixel point, X_max is the maximum value of the R channel value (also can be G channel value or B channel value), and X_min is the minimum value of the R channel value (also can be G channel value or B channel value).

And calling a preset brightness layering number to correspondingly perform image layering on the first normalized picture, and dividing the first normalized picture into 11 layers according to brightness 0.0, 0.1, 0.2 and 1.0.

And carrying out luminance layer combination on the first cut picture according to the first layered picture, and when the first combined picture is obtained, specifically, combining the picture layers with the same luminance value in the first cut picture (for example, combining all the partitions with the luminance value of 0.1 in the first cut picture into the same luminance layer), so as to obtain the first combined picture.

If 11 brightness layers exist in the first combined picture, the number of the pixel points of each brightness layer is obtained, and the proportion of the pixel points of each brightness layer is calculated. For example, only the pixels of 7 luminance layers, i.e., 0.2, 0.4, 0.6, 0.7, 0.8, 0.9, and 1.0, remain in the first combined picture, wherein the proportion of the pixels of the luminance layer with the luminance value of 0.6 is the highest.

Finally, for example, in the first combined picture, only the pixels of 7 brightness layers, namely 0.2, 0.4, 0.6, 0.7, 0.8, 0.9 and 1.0, are left, the highest pixel proportion of the brightness layer with the brightness value of 0.6 is recorded as the target brightness layer, and at this time, the proportion of each pixel of the two brightness layers, namely 0.2 and 0.4, is adjusted to be equal to the target brightness layer, namely, all the pixels are adjusted to be 0.6. By means of the brightness adjustment mode, the picture features can be conveniently and accurately extracted later.

In one embodiment, step S1312 includes:

acquiring a control point in the picture after the first brightness adjustment, and calling a preset distortion function corresponding to the control point;

and correcting the pixel points of the first brightness-adjusted picture according to the distortion function to obtain a first preprocessed picture corresponding to the first brightness-adjusted picture.

In this embodiment, the control point in the first luminance-adjusted picture is a base reference point whose coordinates remain unchanged after the first luminance-adjusted picture passes through the coordinate adjustment; for example, in the coordinate system corresponding to the first luminance-adjusted picture, coordinates (x, y) corresponding to a certain control point are adjusted by a distortion function to obtain a corrected picture, spatial coordinates of a corrected image corresponding to the control point (x, y) are changed to be (ζ, η), and a distortion relationship between the corrected picture and the first luminance-adjusted picture may be described as:

According to the distortion relation, the correction coordinate value of each pixel of the corrected picture can be calculated in sequence, and the pixel value is kept unchanged, so that a corrected image is generated. By geometrically correcting the picture, the picture features can be conveniently and accurately extracted later.

And S140, receiving a second type equipment live-action picture and a second pressure parameter value of the second type equipment, which are sent by the second type equipment according to a second trigger instruction, and acquiring a second field value corresponding to a second type equipment check item field through a second identification result corresponding to the second type equipment live-action picture and the second pressure parameter value so as to store the second field value into a cell corresponding to the second type equipment check item field in the withdrawal information table.

In this embodiment, after the server receives the second type equipment live-action picture and the second pressure parameter value of the second type equipment, the second identification result corresponding to the second type equipment live-action picture may be obtained through image identification, and the second field value corresponding to the second type equipment inspection item field may be obtained by combining the second pressure parameter value. In this way, automatic remote inventory of the bank ATM is achieved.

In an embodiment, the second type of device is a bank ATM; the step S140 includes:

receiving a second type equipment live-action picture and a second pressure parameter value of the second type equipment, which are sent by the second type equipment according to a second trigger instruction, and carrying out image recognition on the second type equipment live-action picture through a trained second depth convolution neural network model to obtain a second recognition result corresponding to the second type equipment live-action picture;

judging whether the second identification result corresponds to an empty box or not, and judging whether the second pressure parameter value is zero or not;

if the second identification result corresponds to an empty box and the second pressure parameter value is zero, setting the value of a second field corresponding to a second type equipment inspection item field to be 0, and storing the value of the second field into a cell corresponding to the second type equipment inspection item field in the parallel removing information table;

and if the second identification result corresponds to a non-empty box and the second pressure parameter value is greater than zero, setting the value of a second field corresponding to a second type equipment inspection item field to be 1, and storing the second field into a cell corresponding to the second type equipment inspection item field in the parallel removing information table.

In this embodiment, reference is made to detecting whether the teller's tail box is empty in the step of detecting whether the ATM is empty, and the discussion will not be repeated here.

And S150, receiving a first type field live-action picture sent by the first type field equipment according to a third trigger instruction, and acquiring a third field value corresponding to a first type field inspection item field through a third identification result corresponding to the first type field live-action picture so as to store the third field value into a cell corresponding to the first type field inspection item field in the parallel removing information table.

In this embodiment, after the server receives the first type field live-action picture, a third recognition result corresponding to the first type field live-action picture may be obtained through image recognition, so as to obtain a third field value corresponding to the first type field live-action picture. In this way, automatic remote checking of the bank warehouse is achieved.

In an embodiment, the first type of venue is a bank vault monitoring camera; the step S150 includes:

receiving a first type field live-action picture sent by first type field equipment according to a third trigger instruction, and carrying out image recognition on the first type field live-action picture through a trained third depth convolution neural network model to obtain a third recognition result corresponding to the first type field live-action picture;

Judging whether the third recognition result corresponds to an empty field;

if the third identification result corresponds to an empty field, setting the value of a third field corresponding to a first type field inspection item field to 0, and storing the value of the third field into a cell corresponding to the first type field inspection item field in the parallel removing information table;

and if the third identification result corresponds to a non-empty field, setting the value of a third field corresponding to the first type field inspection item field to be 1, and storing the value of the third field into a cell corresponding to the first type field inspection item field in the parallel removing information table.

In this embodiment, the step of detecting whether the warehouse is empty refers to the step of detecting whether the tail box of the teller is empty, and the difference here is that the value of the pressure sensor is not required to be comprehensively considered, and only whether the photo of the empty place (i.e. no articles such as a safe) exists in the place is required to be judged.

And S160, judging whether the value of the stored first field value, the second field value and the third field value in the withdrawal information table is not equal to the value of the corresponding field value in the preset withdrawal information expected value table.

In this embodiment, when each organization is evacuated, the server also preselects and stores an expected value table of the evacuation information which is identical to the field structure of the evacuation information, and the value of each field in the expected value table of the evacuation information is a preset expected value, which indicates that the checking item of the field finally reaches the value and passes the evacuation checking.

At this time, the call script automatically compares the values of the first field, the second field and the third field stored in the withdrawal information table with the corresponding fields in the preset withdrawal information expected value table, so as to break whether the values of the first field, the second field and the third field stored in the withdrawal information table are unequal to the values of the corresponding fields in the withdrawal information expected value table.

And S170, if the value of the first field value, the value of the second field value and the value of the third field value stored in the withdrawal information table are not equal to the value of the corresponding field value in the withdrawal information expected value table, sending prompt information for prompting that the field value is not passed to the mechanism uploading terminal.

In this embodiment, once the values of the first field, the second field, and the third field are different from the values of the corresponding fields in the expected value table of the withdrawal information, it indicates that the items of the inspection items corresponding to the withdrawal information table do not pass, and at this time, prompt information of failing inspection needs to be sent to the mechanism uploading terminal in time.

And if all the values of the first field value, the second field value and the third field value stored in the withdrawal information table are equal to the values of the corresponding fields in the withdrawal information expected value table, indicating that the failed item does not exist in the inspection item corresponding to the withdrawal information table, and sending prompt information of inspection pass to the mechanism uploading terminal. Through the mode, automatic checking and comparison of all inspection items are realized, and the checking efficiency is improved.

The method realizes that whether the first type equipment and the first type equipment are empty or not is identified by adopting a mode of combining image identification and pressure parameter values, and whether the first type site is empty or not is detected by adopting image identification, so that various items of examination is not needed by people, and the efficiency of checking the data items is improved.

The embodiment of the invention also provides a mechanism withdrawal data verification device which is used for executing any embodiment of the mechanism withdrawal data verification method. In particular, referring to fig. 5, fig. 5 is a schematic block diagram of a mechanism evacuation data verification apparatus according to an embodiment of the present invention. The institution disarmed data verification device 100 may be configured in a server.

As shown in fig. 5, the organization withdrawal data verification apparatus 100 includes: the device comprises a disarming information table acquisition unit 110, a trigger instruction sending unit 120, a first field value acquisition unit 130, a second field value acquisition unit 140, a third field value acquisition unit 150, a field value comparison unit 160 and a prompt information sending unit 170.

A withdrawal information table obtaining unit 110, configured to receive a withdrawal information table uploaded by a mechanism uploading terminal corresponding to a mechanism to be withdrawn; the disarming information table comprises a first type equipment check item field, a second type equipment check item field and a first type site check item field.

The trigger instruction sending unit 120 is configured to obtain a first type equipment check item field, a second type equipment check item field, and a first type field check item field included in the merge information table, generate a first trigger instruction according to the first type equipment check item field and send the first trigger instruction to the first type equipment, generate a second trigger instruction according to the second type equipment check item field and send the second trigger instruction to the second type equipment, and generate a third trigger instruction according to the first type field check item field and send the third trigger instruction to the first type field equipment.

The first field value obtaining unit 130 is configured to receive the first type equipment live-action picture and the first pressure parameter value of the first type equipment sent by the first type equipment according to the first trigger instruction, obtain, through a first identification result corresponding to the first type equipment live-action picture and the first pressure parameter value, a first field value corresponding to the first type equipment inspection item field, and store the first field value in a cell corresponding to the first type equipment inspection item field in the parallel removal information table.

In one embodiment, the first type of device is a bank teller boot; as shown in fig. 6, the first field value obtaining unit 130 includes:

A first recognition result obtaining unit 131, configured to receive a first type device live-action picture and a first pressure parameter value of a first type device sent by a first trigger instruction, and perform image recognition on the first type device live-action picture through a trained first depth convolutional neural network model, so as to obtain a first recognition result corresponding to the first type device live-action picture;

a first parameter determining unit 132, configured to determine whether the first identification result corresponds to an empty box, and determine whether the first pressure parameter value is zero;

a first parameter first filling unit 133, configured to set a value of a first field corresponding to a first type equipment inspection item field to 0 if the first identification result corresponds to an empty box and the first pressure parameter value is zero, and store the value of the first field to a cell corresponding to the first type equipment inspection item field in the parallel removal information table;

and a first parameter second filling unit 134, configured to set a value of a first field corresponding to a first type equipment inspection item field to 1 and store the first field in a cell corresponding to the first type equipment inspection item field in the parallel removal information table if the first identification result corresponds to a non-empty box and the first pressure parameter value is greater than zero.

In an embodiment, as shown in fig. 7, the first recognition result acquisition unit 131 includes:

a brightness correction unit 1311, configured to obtain a first brightness-adjusted picture by performing brightness correction on the first type equipment live-action picture;

a geometry correcting unit 1312, configured to obtain a first preprocessed picture by performing geometry correction on the first luminance-adjusted picture;

the feature map obtaining unit 1313 is configured to perform convolution with a step length of 1*1 on input arrays corresponding to RGB channels corresponding to the first preprocessed picture through a convolution kernel of 3*3 in the first deep convolutional neural network model, so as to obtain a feature map corresponding to the first preprocessed picture;

the image feature obtaining unit 1314 is configured to pool, by a multi-level spatial pyramid pooling layer composed of 1*1 blocks, 1*3 blocks, 3*1 blocks, and 4*4 blocks, feature images corresponding to the first preprocessed image to obtain image features corresponding to the first preprocessed image;

the classifying unit 1315 is configured to input the image feature to a full-connection layer in the first deep convolutional neural network model to obtain a one-dimensional vector, and classify the one-dimensional vector by a softmax layer to obtain a first recognition result corresponding to the first type equipment live-action image.

In one embodiment, the brightness correction unit 1311 includes:

the image cutting unit is used for cutting the image of the first type equipment live-action picture through an image cutting model based on the image to obtain a first cut picture;

the image normalization unit is used for carrying out image normalization processing on the first cut picture to obtain a first normalized picture;

the image layering unit is used for calling a preset brightness layering number to correspondingly layer the image of the first normalized picture to obtain a first layered picture;

the brightness layer merging unit is used for carrying out brightness layer merging on the first cut picture according to the first layered picture to obtain a first merged picture;

the target brightness layer obtaining unit is used for obtaining the pixel point proportion of each brightness layer in the first combined picture and obtaining the brightness layer with the maximum pixel point proportion as the target brightness layer;

and the brightness layer adjusting unit is used for taking each brightness layer with the brightness value smaller than that of the target brightness layer as the brightness layer to be adjusted, and adjusting the brightness value corresponding to the brightness layer to be adjusted to be equal to that of the target brightness layer so as to obtain the picture after the first brightness adjustment.

In one embodiment, the geometry correction unit 1312 includes:

The control point acquisition unit is used for acquiring a control point in the picture after the first brightness adjustment and calling a preset distortion function corresponding to the control point;

and the pixel point correction unit is used for correcting the pixel points of the first brightness-adjusted picture according to the distortion function to obtain a first preprocessed picture corresponding to the first brightness-adjusted picture.

The second field value obtaining unit 140 is configured to receive the second type equipment live-action picture and the second pressure parameter value of the second type equipment sent by the second type equipment according to the second trigger instruction, obtain, through a second identification result corresponding to the second type equipment live-action picture and the second pressure parameter value, a second field value corresponding to the second type equipment inspection item field, and store the second field value in a cell corresponding to the second type equipment inspection item field in the parallel removing information table.

In an embodiment, the second type of device is a bank ATM; the second field value obtaining unit 140 includes:

the second recognition result acquisition unit is used for receiving a second type equipment live-action picture and a second pressure parameter value of the second type equipment, which are sent by the second type equipment according to a second trigger instruction, and carrying out image recognition on the second type equipment live-action picture through a trained second depth convolution neural network model to obtain a second recognition result corresponding to the second type equipment live-action picture;

A second parameter judging unit, configured to judge whether the second identification result corresponds to an empty box, and judge whether the second pressure parameter value is zero;

a second parameter first filling unit, configured to, if the second identification result corresponds to an empty box and the second pressure parameter value is zero, set a value of a second field corresponding to a second type equipment inspection item field to 0, and store the value of the second field to a cell corresponding to the second type equipment inspection item field in the parallel removing information table;

and the second parameter first filling unit is used for setting the value of the second field corresponding to the second type equipment inspection item field to be 1 and storing the second field into the cell corresponding to the second type equipment inspection item field in the withdrawal information table if the second identification result corresponds to a non-empty box and the second pressure parameter value is greater than zero.

The third field value obtaining unit 150 is configured to receive the first type field live-action picture sent by the first type field device according to the third trigger instruction, obtain, according to a third identification result corresponding to the first type field live-action picture, a third field value corresponding to the first type field inspection item field, and store the third field value in a cell corresponding to the first type field inspection item field in the parallel information table.

In an embodiment, the first type of venue is a bank vault monitoring camera; the third field value obtaining unit 150 includes:

the third recognition result acquisition unit is used for receiving a first type field live-action picture sent by the first type field equipment according to a third trigger instruction, and carrying out image recognition on the first type field live-action picture through a trained third depth convolution neural network model to obtain a third recognition result corresponding to the first type field live-action picture;

a third parameter judging unit, configured to judge whether the third identification result corresponds to an empty site;

a third parameter first filling unit, configured to set a value of a third field corresponding to a first type field inspection item field to 0 if the third identification result corresponds to an empty field, and store the value of the third field to a cell corresponding to the first type field inspection item field in the parallel removal information table;

and the third parameter first filling unit is used for setting the value of the third field corresponding to the first type field inspection item field to be 1 if the third identification result corresponds to the non-empty field, and storing the value of the third field into the cell corresponding to the first type field inspection item field in the parallel removing information table.

The field value comparison unit 160 is configured to determine whether any of the stored first field value, second field value, and third field value in the merge information table is unequal to the corresponding field value in the preset merge information expected value table.

And the prompt information sending unit 170 is configured to send, to the mechanism uploading terminal, a prompt message for prompting that the value of the field is not passed if the value of the first field, the value of the second field, and the value of the third field stored in the withdrawal information table are not equal to the value of the corresponding field in the withdrawal information expected value table.

The device realizes that whether the first type equipment and the first type equipment are empty or not is identified by adopting a mode of combining image identification and pressure parameter values, and whether the first type site is empty or not is checked by adopting image identification, so that various items of checking are not needed by people, and the efficiency of checking data items is improved.

The above-described mechanism disarming data verification apparatus may be implemented in the form of a computer program that is executable on a computer device as shown in fig. 8.

Referring to fig. 8, fig. 8 is a schematic block diagram of a computer device according to an embodiment of the present invention. The computer device 500 is a server, and the server may be a stand-alone server or a server cluster formed by a plurality of servers.

With reference to FIG. 8, the computer device 500 includes a processor 502, memory, and a network interface 505 connected by a system bus 501, where the memory may include a non-volatile storage medium 503 and an internal memory 504.

The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer program 5032, when executed, may cause the processor 502 to perform a mechanism retirement data verification method.

The processor 502 is used to provide computing and control capabilities to support the operation of the overall computer device 500.

The internal memory 504 provides an environment for the execution of a computer program 5032 in the non-volatile storage medium 503, which computer program 5032, when executed by the processor 502, causes the processor 502 to perform a mechanism-withdrawal and data verification method.

The network interface 505 is used for network communication, such as providing for transmission of data information, etc. It will be appreciated by those skilled in the art that the architecture shown in fig. 8 is merely a block diagram of some of the architecture relevant to the present inventive arrangements and is not limiting of the computer device 500 to which the present inventive arrangements may be implemented, as a particular computer device 500 may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

The processor 502 is configured to execute a computer program 5032 stored in a memory to implement the mechanism disarming and data verification method disclosed in the embodiment of the present invention.

Those skilled in the art will appreciate that the embodiment of the computer device shown in fig. 8 is not limiting of the specific construction of the computer device, and in other embodiments, the computer device may include more or less components than those shown, or certain components may be combined, or a different arrangement of components. For example, in some embodiments, the computer device may include only a memory and a processor, and in such embodiments, the structure and function of the memory and the processor are consistent with the embodiment shown in fig. 8, and will not be described again.

It should be appreciated that in embodiments of the present invention, the processor 502 may be a central processing unit (Central Processing Unit, CPU), the processor 502 may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf Programmable gate arrays (FPGAs) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. Wherein the general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

In another embodiment of the invention, a computer-readable storage medium is provided. The computer readable storage medium may be a non-volatile computer readable storage medium. The computer readable storage medium stores a computer program, wherein the computer program when executed by a processor implements the mechanism disarming data verification method disclosed by the embodiment of the present invention.

It will be clearly understood by those skilled in the art that, for convenience and brevity of description, specific working procedures of the apparatus, device and unit described above may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein. Those of ordinary skill in the art will appreciate that the elements and algorithm steps described in connection with the embodiments disclosed herein may be embodied in electronic hardware, in computer software, or in a combination of the two, and that the elements and steps of the examples have been generally described in terms of function in the foregoing description to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

In the several embodiments provided by the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, and for example, the division of the units is merely a logical function division, there may be another division manner in actual implementation, or units having the same function may be integrated into one unit, for example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. In addition, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices, or elements, or may be an electrical, mechanical, or other form of connection.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiment of the present invention.

In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units may be stored in a storage medium if implemented in the form of software functional units and sold or used as stand-alone products. Based on such understanding, the technical solution of the present invention is essentially or a part contributing to the prior art, or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a magnetic disk, an optical disk, or other various media capable of storing program codes.

While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.

Claims

1. A method of verifying institutional disarming data, comprising:

2. The institutional disarming data verification method as claimed in claim 1, wherein said first type of device is a bank teller trunk;

the receiving the first type equipment obtains a first field value corresponding to a first type equipment check item field according to a first identification result corresponding to the first type equipment live-action picture and the first pressure parameter value of the first type equipment sent by the first trigger instruction, so as to store the first field value corresponding to the first type equipment check item field in the withdrawal information table, and the receiving the first field value comprises the following steps:

receiving a first type equipment live-action picture and a first pressure parameter value of first type equipment, which are sent by the first type equipment according to a first trigger instruction, and carrying out image recognition on the first type equipment live-action picture through a trained first depth convolution neural network model to obtain a first recognition result corresponding to the first type equipment live-action picture;

Judging whether the first identification result corresponds to an empty box or not, and judging whether the first pressure parameter value is zero or not;

if the first identification result corresponds to an empty box and the first pressure parameter value is zero, setting the value of a first field corresponding to a first type equipment inspection item field to be 0, and storing the value of the first field into a cell corresponding to the first type equipment inspection item field in the parallel removing information table;

and if the first identification result corresponds to a non-empty box and the first pressure parameter value is greater than zero, setting the value of a first field corresponding to a first type equipment inspection item field to be 1, and storing the first field into a cell corresponding to the first type equipment inspection item field in the withdrawal information table.

3. The facility evacuation data verification method of claim 1, wherein the second type of device is a bank self-service cash dispenser;

the receiving the second type equipment obtains a second field value corresponding to a second type equipment checking item field according to a second identification result corresponding to the second type equipment live-action picture and the second pressure parameter value of the second type equipment sent by the second trigger instruction, and stores the second field value corresponding to the second type equipment checking item field in a unit cell corresponding to the second type equipment checking item field in the withdrawal information table, and the receiving the second type equipment live-action picture and the second pressure parameter value of the second type equipment, wherein the receiving the second field value comprises:

4. The facility evacuation data verification method of claim 1, wherein the first type of venue equipment is a bank vault monitoring camera;

The receiving the first type field live-action picture sent by the first type field equipment according to the third trigger instruction, obtaining a third field value corresponding to the first type field inspection item field through a third identification result corresponding to the first type field live-action picture, and storing the third field value into a cell corresponding to the first type field inspection item field in the decomplex information table, wherein the receiving the first type field live-action picture comprises the following steps:

judging whether the third recognition result corresponds to an empty field;

5. The method for verifying the organization disarming data according to claim 2, wherein the performing image recognition on the first type of equipment live-action picture through the trained first depth convolutional neural network model to obtain a first recognition result corresponding to the first type of equipment live-action picture comprises:

obtaining a first brightness-adjusted picture by carrying out brightness correction on the first type equipment live-action picture;

obtaining a first preprocessed picture by geometrically correcting the picture with the first brightness adjusted;

carrying out convolution with the step length of 1*1 on input arrays corresponding to RGB channels corresponding to the first preprocessing picture through a convolution kernel of 3*3 in the first depth convolution neural network model to obtain a feature map corresponding to the first preprocessing picture;

pooling the feature images corresponding to the first preprocessed image through a multi-level space pyramid pooling layer consisting of 1*1 blocks, 1*3 blocks, 3*1 blocks and 4*4 blocks to obtain image features corresponding to the first preprocessed image;

and inputting the picture features to a full-connection layer in a first depth convolutional neural network model to obtain one-dimensional vectors, and classifying the one-dimensional vectors by a softmax layer to obtain a first recognition result corresponding to the first type equipment live-action picture.

6. The method for verifying the organization disarming data as claimed in claim 5, wherein the obtaining the first brightness-adjusted picture by performing brightness correction on the first type of equipment live-action picture comprises:

7. The method for verifying the organization disarming data as set forth in claim 5, wherein the obtaining the first preprocessed picture by geometrically correcting the first brightness-adjusted picture comprises:

8. A facility evacuation data verification apparatus, comprising:

9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of institutional disarming data verification as claimed in any one of claims 1 to 7 when the computer program is executed.

10. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, causes the processor to perform the institution-withdrawal data verification method of any one of claims 1 to 7.