CN116824339A

CN116824339A - Image processing method and device

Info

Publication number: CN116824339A
Application number: CN202310861449.6A
Authority: CN
Inventors: 孔计胤杰
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2023-07-13
Filing date: 2023-07-13
Publication date: 2023-09-29

Abstract

The embodiment of the specification provides an image processing method and device, wherein the image processing method comprises the following steps: and constructing an image set according to the acquired image to be detected and the associated image, calculating image classification parameters based on image features of each image in the target image set obtained by construction, updating an evaluation index threshold and a confidence coefficient threshold if the image classification parameters are in a preset parameter interval, obtaining a target evaluation index threshold and a target confidence coefficient threshold, calculating image confidence coefficient according to the image evaluation result under the condition that the image evaluation result obtained by carrying out image evaluation on the image to be detected does not accord with the target evaluation index threshold, and finally determining the image detection result of the image to be detected by means of the image confidence coefficient and the target confidence coefficient threshold.

Description

Image processing method and device

Technical Field

The present disclosure relates to the field of image processing technologies, and in particular, to an image processing method and apparatus.

Background

With the continuous development of internet technology, more and more online services are generated, users do not need to go to offline service institutions to perform service processing, but perform corresponding processing directly through online services, in the process of using online services, in order to improve the safety of online services, more and more online services need users to provide corresponding user images, identity verification is performed on the users through user images uploaded by the users, such as identity credential images and driving credential images uploaded by the users, and in this process, how to better manage the user images uploaded by the users gradually becomes the focus of attention of each party.

Disclosure of Invention

One or more embodiments of the present specification provide an image processing method including: and acquiring an image to be detected and an associated image, constructing an image set, and calculating image classification parameters based on image characteristics of each image in a target image set obtained by construction. And if the image classification parameters are in the preset parameter interval, updating the evaluation index threshold and the confidence coefficient threshold to obtain a target evaluation index threshold and a target confidence coefficient threshold. And carrying out image evaluation on the image to be detected to obtain an image evaluation result, and calculating the image confidence coefficient according to the image evaluation result under the condition that the image evaluation result does not accord with the target evaluation index threshold value. And determining an image detection result of the image to be detected based on the image confidence and the target confidence threshold.

One or more embodiments of the present specification provide an image processing apparatus including: and the parameter calculation module is configured to acquire the image to be detected and the associated image, construct an image set and calculate image classification parameters based on the image characteristics of each image in the target image set obtained by construction. And the threshold updating module is configured to update the evaluation index threshold and the confidence coefficient threshold if the image classification parameter is in a preset parameter interval, so as to obtain a target evaluation index threshold and a target confidence coefficient threshold. The confidence coefficient calculating module is configured to perform image evaluation on the image to be detected to obtain an image evaluation result, and calculate the image confidence coefficient according to the image evaluation result under the condition that the image evaluation result does not accord with the target evaluation index threshold value. And a result determining module configured to determine an image detection result of the image to be detected based on the image confidence and the target confidence threshold.

One or more embodiments of the present specification provide an image processing apparatus including: a processor; and a memory configured to store computer-executable instructions that, when executed, cause the processor to: and acquiring an image to be detected and an associated image, constructing an image set, and calculating image classification parameters based on image characteristics of each image in a target image set obtained by construction. And if the image classification parameters are in the preset parameter interval, updating the evaluation index threshold and the confidence coefficient threshold to obtain a target evaluation index threshold and a target confidence coefficient threshold. And carrying out image evaluation on the image to be detected to obtain an image evaluation result, and calculating the image confidence coefficient according to the image evaluation result under the condition that the image evaluation result does not accord with the target evaluation index threshold value. And determining an image detection result of the image to be detected based on the image confidence and the target confidence threshold.

One or more embodiments of the present specification provide a storage medium storing computer-executable instructions that, when executed by a processor, implement the following: and acquiring an image to be detected and an associated image, constructing an image set, and calculating image classification parameters based on image characteristics of each image in a target image set obtained by construction. And if the image classification parameters are in the preset parameter interval, updating the evaluation index threshold and the confidence coefficient threshold to obtain a target evaluation index threshold and a target confidence coefficient threshold. And carrying out image evaluation on the image to be detected to obtain an image evaluation result, and calculating the image confidence coefficient according to the image evaluation result under the condition that the image evaluation result does not accord with the target evaluation index threshold value. And determining an image detection result of the image to be detected based on the image confidence and the target confidence threshold.

Drawings

For a clearer description of one or more embodiments of the present description or of the solutions of the prior art, the drawings that are needed in the description of the embodiments or of the prior art will be briefly described below, it being obvious that the drawings in the description that follow are only some of the embodiments described in the present description, from which other drawings can be obtained, without inventive faculty, for a person skilled in the art;

FIG. 1 is a schematic diagram of an environment in which an image processing method according to one or more embodiments of the present disclosure is implemented;

FIG. 2 is a process flow diagram of an image processing method according to one or more embodiments of the present disclosure;

FIG. 3 is a process flow diagram of an image processing method for an identity credential scenario provided in one or more embodiments of the present disclosure;

FIG. 4 is a schematic diagram of an embodiment of an image processing apparatus according to one or more embodiments of the present disclosure;

fig. 5 is a schematic structural diagram of an image processing apparatus according to one or more embodiments of the present disclosure.

Detailed Description

In order to enable a person skilled in the art to better understand the technical solutions in one or more embodiments of the present specification, the technical solutions in one or more embodiments of the present specification will be clearly and completely described below with reference to the drawings in one or more embodiments of the present specification, and it is obvious that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments, which can be made by one or more embodiments of the present disclosure without inventive effort, are intended to be within the scope of the present disclosure.

Referring to fig. 1, one or more embodiments of the present disclosure provide a schematic diagram of an implementation environment of an image processing method.

The image processing method provided in one or more embodiments of the present disclosure may be applied to an implementation environment for performing image detection on an image to be detected and an associated image and determining an image detection result, where the implementation environment includes at least the image detection server 101.

In addition, the implementation environment may further include a user terminal 102, where the user terminal 102 may configure a client that interacts with the image detection server 101, and the specific form of the client may be an application program, a sub-program in the application program, a service module in the application program, or a web page program.

The image detection server 101 may correspond to one server, or corresponds to a server cluster formed by a plurality of servers, or corresponds to one or more cloud servers in the cloud computing platform, and is configured to perform image detection on an image to be detected and an associated image and determine an image detection result.

The user terminal 102 may be a mobile phone, a personal computer, a tablet computer, an electronic book reader, a VR (Virtual Reality technology) -based device for information interaction, an in-vehicle terminal, an IoT device, a wearable smart device, a laptop portable computer, a desktop computer, etc., and the user terminal 102 is configured to upload an image to be detected to the image detection server 101.

In this implementation environment, after obtaining the image to be detected uploaded by the user terminal 102, the image detection server 101 may obtain an associated image of the image to be detected, perform image set construction according to the image to be detected and the associated image of the image to be detected, calculate an image classification parameter according to image features of each image in the constructed target image set, update a subsequent evaluation index threshold and a confidence threshold under the condition that the image classification parameter is within a preset parameter interval, obtain a target evaluation index threshold and a target confidence threshold, if an image evaluation result obtained by performing image evaluation on the image to be detected does not conform to the target evaluation index threshold, determine an image detection result of the image to be detected according to the target confidence threshold and the image confidence obtained based on the image evaluation result, and update the evaluation index threshold and the confidence threshold according to the image classification parameter, thereby improving accuracy of image detection, and improving comprehensiveness and flexibility of image detection by combining the image to be detected and the associated image.

One or more embodiments of an image processing method provided in the present specification are as follows:

Referring to fig. 2, the image processing method provided in the present embodiment specifically includes steps S202 to S208.

Step S202, obtaining an image to be detected and an associated image, constructing an image set, and calculating image classification parameters based on image features of each image in a target image set obtained by construction.

The image to be detected in this embodiment refers to an image to be detected; the image to be detected comprises an image to be detected uploaded by a user; optionally, the image to be detected includes an identity image to be detected, such as an identity document image, a bank card image, a driving document image, a medical insurance document image, and the like. The image to be detected can be RGB (Red Green Blue) image, and the image to be detected can also comprise card image to be detected. The card image to be detected may in particular represent a card image and/or a credential image of the user.

The image to be detected can also be any image applied to identity verification, for example, the image to be detected can be an image containing biological characteristics of a user, and the specific image to be detected can be an image containing iris characteristics of the user, an image containing fingerprint characteristics of the user, an image containing facial characteristics of the user and the like.

The associated image refers to an image associated with the image to be detected; the associated image comprises a history image of undetected passing of a user under the target service; for example, the image to be detected is uploaded through the target service, and the associated image represents a history image of undetected passing before the image to be detected under the target service; the target service here includes any service that needs to perform image detection of an image to be detected, for example, in an account registration service, the user face image needs to be collected for identity verification. The related images acquired in this embodiment may be one or more.

Optionally, the associated image may be obtained from a database; the database comprises a database for storing associated images of users; the database may be determined according to the service type of the target service, each target service may correspond to a different database, and in addition, the database may not be determined according to the service type of the target service, and each target service may correspond to the same database. In an optional implementation manner provided in this embodiment, the associated image is obtained by:

determining a user identifier and/or a device identifier for uploading the image to be detected;

And based on the user identification and/or the equipment identification, reading the associated image of the image to be detected from a database.

The user identification of the image to be detected comprises uploading a user account of the image to be detected; the equipment identifier comprises a terminal identifier of a user terminal uploading the image to be detected.

Optionally, the database stores the associated image corresponding to the user identifier and/or the equipment identifier; and eliminating target images in the associated images corresponding to the user identifications and/or the equipment identifications according to time sequence under the condition that the number of the images of the associated images corresponding to the user identifications and/or the equipment identifications exceeds an image number threshold value.

For example, in order to reduce the load amount of the database, the threshold of the number of images of the database is set to m, and in the case that the number of images of the associated image corresponding to the user identifier a exceeds m, m images of the associated images that are later in time can be determined in time sequence, and the target image other than the m images can be removed from the associated images.

In addition, in the case that each target service may correspond to the same database, the associated image may be obtained by: determining a user account for uploading the image to be detected; and based on the user account, reading the associated image of the image to be detected from a database. The user account comprises an application account of the user.

The image classification parameters in this embodiment include parameters that characterize the image classification of the target image set, such as a score that characterizes the image classification or a confidence that characterizes the image classification; in particular, the image classification parameters may include parameters that characterize whether the target image set is a multiple attempt by a normal user (multiple upload of images) or a multiple attack by a malicious user (multiple upload of images). The image features, including feature maps of the images, may specifically include feature maps of the images in at least one dimension, where the dimension may include a quality dimension, a text dimension, and/or a counterfeit dimension, and where the dimension may include other types of dimensions; wherein the counterfeit dimension includes a dimension representing whether each image is a counterfeit image.

In an actual identity verification scene, the situation that a user uploads an image to be detected for identity verification for many times and the detection results of the image detection for many times are all failed is often existed, in the process, two situations may exist, one situation that a common user uploads the image to be detected for many times for identity verification and the other situation that a malicious user uploads the image to be detected for many times for malicious attack; if the image detection is performed only through the image to be detected uploaded in real time, multiple interruptions to the common user can be caused, the enthusiasm of the common user is reduced, and the recognition accuracy of multiple attacks to the malicious user is lower.

In order to improve the recognition accuracy of multiple image attacks on malicious users and the recognition accuracy of multiple image attempts on common users, the method reduces the disturbance of multiple attempts on the common users, can acquire the associated image of the image to be detected on the basis of acquiring the image to be detected, represents the image which is uploaded by the users before the image to be detected and is not passed through the image detection, and determines the image detection result of the image to be detected by combining the image to be detected and the associated image.

In practical application, the process of performing image detection on the image to be detected and the associated image can set the preset image number, for example, the preset image number is set to n, and under the condition that the total image number of the image to be detected and the associated image does not reach the preset image number n, the image filling is performed on the image to be detected and the associated image, so that the total image number reaches the preset image number n, and the detection convenience of performing image detection on the image to be detected and the associated image is improved; in an optional implementation manner provided in this embodiment, the image set construction includes:

constructing an image set based on the image to be detected and the associated image, and detecting whether the number of images of the image set is smaller than a preset number of images;

And if not, taking the image set as the target image set.

The preset number of images may be the threshold number of images used in the process of removing the target image from the associated image.

In an optional implementation manner provided in this embodiment, if the result of executing the detection that whether the number of images in the image set is smaller than the preset number of images is yes, the following operations are executed:

and performing image filling on the image set based on the image number and the preset image number, and taking the image set subjected to image filling as the target image set.

Specifically, the images to be detected and the associated images can be combined according to a time sequence, and whether the number of images of the image set obtained by combination is smaller than the preset number of images or not is detected; if not, taking the image set as the target image set, if so, filling specific images into the image set based on the image number and the preset image number, and taking the filled image set as the target image set; wherein the specific image includes a blank image; in the process of filling the specific image into the image set obtained by combination, blank images can be randomly filled into the image set obtained by combination, in addition, in the process of filling the specific image into the image set based on the image number and the preset image number, the filling number of the blank images can be determined based on the image number and the preset image number, and the blank images with the filling number are filled into the image set obtained by combination at intervals according to the arrangement sequence of the images in the image set obtained by combination, for example, the filling number is a, and a blank images are filled into the image set obtained by combination at intervals from front to back in sequence; the number of intervals used herein is not particularly limited, and is determined according to the actual application scenario.

In practical application, in the process that an ordinary user uploads an image for many times but the image detection fails, the difference of the image uploaded for many times is not large, at most, only the angle, illumination and the like of the image shooting have slight differences, but in the process that a malicious user uploads the image for many times but the image detection fails, the difference of the image uploaded each time is possibly large, for example, the shape, angle or area of a shielding object of the image are different; for this, the image classification parameters can be calculated through the image feature differences of the images in the target image set, so that the accuracy of the image classification parameters is improved; in an optional implementation manner provided in this embodiment, in a process of calculating an image classification parameter based on image features of each image in a target image set obtained by construction, the following operations are performed:

calculating a first characteristic residual error in a time dimension and a second characteristic residual error in a space dimension based on the image characteristic diagrams of the images;

and calculating the image classification parameters of the target image set according to the first characteristic residual error and the second characteristic residual error.

In the process of calculating the first feature residual in the time dimension and the second feature residual in the space dimension based on the image feature map of each image, the following operations are performed in an optional implementation manner provided in this embodiment:

Extracting target feature blocks from the image feature graphs of the images according to time sequence, and calculating the first feature residual errors based on the target feature blocks; the method comprises the steps of,

and calculating the second characteristic residual error based on the image characteristic diagram of each image.

The target feature block includes target feature blocks in each image feature map, for example, each image feature map includes 9 feature blocks, the target feature blocks are feature blocks in the same position in each image feature map, and the specific target feature blocks may be a first feature block, a second feature block and/or a third feature block in each image feature map. The first feature residual comprises differences of pixel values of pixel points in a target feature block of the adjacent image feature map. The second feature residual comprises differences of pixel values of pixel points in each feature block of the image feature map.

For example, each image feature map includes 9 feature blocks, a difference value of pixel values of pixel points in a target feature block at the same position of an adjacent image feature map in each image feature map is calculated as a first feature residual, and a difference value of pixel values of pixel points in each feature block of each image feature map is calculated as a second feature residual; that is, in the process of calculating the first feature residual in the time dimension and the second feature residual in the space dimension based on the image feature map of each image, in order to improve the accuracy of residual calculation, the image classification parameters can be calculated more accurately, and the following operations can be performed: extracting target feature blocks from the image feature images of the images according to time sequence, and calculating the difference value of pixel values of pixel points in the target feature blocks at the same position of the adjacent image feature images in the image feature images as a first feature residual; and calculating the difference value of the pixel values of the pixel points in each feature block of each image feature map as a second feature residual.

In addition, in calculating the first feature residual in the time dimension and the second feature residual in the space dimension based on the image feature map of each image, the following operations may also be performed: extracting target feature blocks from the image feature graphs of the images according to a time sequence, inputting the extracted target feature blocks into a time network for feature residual calculation in a time dimension, and obtaining a first feature residual in the time dimension; and inputting each feature block in the image feature map of each image into a spatial network to perform feature residual calculation in the spatial dimension, and obtaining a second feature residual in the spatial dimension.

Wherein, the time network can adopt a transformer neural network structure; the spatial network may also employ a transformer neural network structure.

In the process of calculating the image classification parameters of the target image set according to the first feature residual and the second feature residual, the following operations may be performed:

and carrying out feature residual fusion on the first feature residual and the second feature residual, and calculating image classification parameters of the target image set based on the fused feature residual.

In addition, in calculating the image classification parameters of the target image set according to the first feature residual and the second feature residual, the following operations may also be performed: inputting the first characteristic residual error and the second characteristic residual error into the full-connection layer for parameter calculation, and obtaining the image classification parameters of the target image set.

In addition, in the process of calculating the image classification parameters based on the image features of each image in the target image set obtained by construction, in order to improve the flexibility of the image classification parameters, the following operations may be performed: calculating a first characteristic residual error in a time dimension and/or a second characteristic residual error in a space dimension based on the image characteristic diagrams of the images; and calculating the image classification parameters of the target image set according to the first characteristic residual error and/or the second characteristic residual error.

And step S204, if the image classification parameters are in the preset parameter interval, updating the evaluation index threshold and the confidence coefficient threshold to obtain a target evaluation index threshold and a target confidence coefficient threshold.

Acquiring an image to be detected and an associated image, constructing an image set, and calculating image classification parameters based on image characteristics of each image in a target image set obtained by construction; if the image classification parameter is not in the preset parameter interval, the current evaluation index threshold and the current confidence coefficient threshold are respectively used as the target evaluation index threshold and the target confidence coefficient threshold without processing, and the following step S206 is executed; in the step, if the image classification parameters are in a preset parameter interval, updating the evaluation index threshold and the confidence coefficient threshold to obtain a target evaluation index threshold and a target confidence coefficient threshold.

The preset parameter interval in this embodiment refers to an interval of preset image classification parameters, for example, parameter intervals are set to be a-b, b-c, and c-d, and preset parameter intervals are set to be a-b; for example, the image classification parameters are classified confidence, the parameter intervals are a high confidence interval, a middle confidence interval and a low confidence interval, and the preset parameter interval is a high confidence interval.

The evaluation index threshold comprises an index threshold required for image evaluation of the image to be detected, and optionally, the evaluation index threshold can comprise an image quality score threshold, a text confidence threshold and/or an image falsification index threshold, such as an image quality score threshold, wherein the image quality score threshold comprises a definition threshold, an integrity threshold and the like; in addition, the evaluation index threshold may also include other types of index thresholds. The confidence threshold comprises the probability or degree of passing of image detection of the image to be detected or the probability of whether the image to be detected is an image uploaded by a common user or an image uploaded by a malicious user.

In the specific implementation, in order to improve the image detection passing rate and success rate of the image to be detected under the condition that the image classification parameters are in a preset parameter interval; in an optional implementation manner provided in this embodiment, in a process of updating the evaluation index threshold and the confidence threshold, the following operations are performed:

And carrying out down-regulating processing on the confidence coefficient threshold value, the quality score threshold value in the evaluation index threshold value, the text confidence coefficient threshold value and/or the image falsification index threshold value.

Specifically, through the index threshold and the confidence threshold which are adjusted downwards, the image detection passing rate of the image to be detected is improved, the situation that the common user still fails to try for many times and the malicious user attacks for many times is avoided, the image detection safety is improved, the index threshold and the confidence threshold are adjusted downwards under the condition that the image classification parameters are in the preset parameter interval, the probability that the target image set is the common user tries for many times or uploads the image for many times is relatively high, the probability that the malicious user attacks through uploading the image for many times is relatively low, so that the detection requirement on the image detection of the common user is relaxed through adjusting the index threshold and the confidence threshold, the flexibility and the effectiveness of the image detection are improved, the user experience is improved, and the long-time detection of the common user is avoided without disturbing and poor experience caused to the common user.

In addition, in the process of updating the evaluation index threshold and the confidence coefficient threshold to obtain the target evaluation index threshold and the target confidence coefficient threshold, the following operations may also be performed: and updating the evaluation index threshold or the confidence coefficient threshold to obtain the target evaluation index threshold and/or the target confidence coefficient threshold.

Step S206, performing image evaluation on the image to be detected to obtain an image evaluation result, and calculating an image confidence coefficient according to the image evaluation result under the condition that the image evaluation result does not accord with the target evaluation index threshold.

In the step, firstly, performing image evaluation on the image to be detected to obtain an image evaluation result, and calculating the image confidence coefficient by means of the image evaluation result of the image to be detected under the condition that the image evaluation result does not accord with the target evaluation index threshold value.

The image confidence level in this embodiment includes a probability or a degree of passing of image detection that characterizes an image to be detected. The image evaluation result does not accord with the target evaluation index threshold, and comprises that the image quality score is smaller than a preset score threshold, the text confidence is smaller than a preset confidence threshold and/or the image falsification index is smaller than a preset index threshold.

In the implementation, in the process of performing image evaluation on an image to be detected to obtain an image evaluation result, the following operations may be performed: performing image quality evaluation on the image to be detected to obtain an image quality score; and/or carrying out text analysis on the image to be detected, and calculating text confidence coefficient based on each text block obtained by recognition; and/or calculating a forgery score of the image to be detected; wherein the image quality score includes a sharpness score, an integrity score, and the like; the forgery index includes a probability or score that characterizes the image to be detected as a forgery image. The text confidence coefficient comprises the confidence coefficient of the image to be detected from the text angle, such as text information (certificate number of xx identity certificate, address information and the like) is obtained from the image to be detected by recognition, the confidence coefficient of the image to be detected is calculated based on each text information obtained by recognition, and specifically, the text confidence coefficient of the image to be detected can be reduced under the condition that each text information is not matched, and the text confidence coefficient of the image to be detected can be increased under the condition that each text information is matched.

In an optional implementation manner provided in this embodiment, after performing image evaluation on the image to be detected to obtain an image evaluation result, the following operations are further performed: and under the condition that the image evaluation result accords with the target evaluation index threshold value, determining that the image detection result of the image to be detected is passing detection.

Wherein the image evaluation result meets the target evaluation index threshold, comprising: the image quality score is greater than or equal to a preset score threshold, the text confidence is greater than or equal to a preset confidence threshold, and/or the image forgery index is greater than or equal to a preset index threshold. That is, if any one or more of the image quality score being greater than or equal to a preset score threshold, the text confidence being greater than or equal to a preset confidence threshold, and the image falsification index being greater than or equal to a preset index threshold is established, it is determined that the image evaluation result meets the target evaluation index threshold.

In a specific implementation process, in order to calculate the image confidence level with finer granularity, in an alternative implementation manner provided in this embodiment, in a process of calculating the image confidence level according to the image evaluation result, the following operations are performed:

Determining the evaluation index distribution of each image and/or the evaluation index difference value of every two images in each image;

and calculating the image confidence according to the evaluation index distribution and/or the evaluation index difference value.

Optionally, the evaluating the index distribution includes at least one of: image quality score distribution, text confidence distribution, and image forgery score distribution.

Optionally, the evaluation index difference comprises at least one of the following: image quality score difference, text confidence difference, image forgery score difference.

Specifically, the evaluation index distribution of each image can be determined, the evaluation index difference value of each two images is calculated according to the evaluation index distribution of each image, and if the evaluation index difference value exceeds a preset difference threshold value, the image confidence is determined to be a first confidence; if the difference value of the evaluation index does not exceed the preset difference value threshold value, determining the image confidence coefficient as a second confidence coefficient; optionally, the second confidence level is less than the first confidence level; in addition, in the process of calculating the image confidence according to the evaluation index score and/or the evaluation index difference, the evaluation index difference of each two images can be calculated according to the evaluation index distribution of each image, and the corresponding confidence is determined based on the evaluation index difference obtained by calculation.

In a specific execution process, in order to improve the effectiveness and comprehensiveness of the image confidence coefficient obtained by calculation, the image confidence coefficient can be calculated based on multi-mode features; in another optional implementation manner provided in this embodiment, in the process of calculating the image confidence coefficient according to the image evaluation result, the following operations are performed: and calculating the image confidence according to the images and the character recognition results of the images.

Specifically, the image confidence coefficient can be calculated according to the image characteristics of each image and the character recognition result of each image; the image confidence coefficient is calculated from two modes of the image and the text, and the effectiveness of the image confidence coefficient is improved.

In addition, in the process of calculating the image confidence according to the image evaluation result, the image confidence can be calculated according to the image characteristics of each image and/or the character recognition result of each image; the process of calculating the image confidence coefficient according to the image features of each image is similar to the above, a first feature residual coefficient in a time dimension and a second feature residual coefficient in a space dimension can be calculated based on the image feature map of each image, and the image confidence coefficient is calculated according to the first feature residual coefficient and the second feature residual coefficient; the above-mentioned image classification parameters, here, the image confidence level, and other processing procedures are similar, and reference may be made to reading, and will not be repeated here.

In the process of calculating the image confidence coefficient according to the image characteristics of each image and/or the character recognition result of each image, calculating a first confidence coefficient according to the image characteristics of each image and/or calculating a second confidence coefficient according to the character recognition result of each image, and calculating the image confidence coefficient based on the first confidence coefficient and/or the second confidence coefficient; the process of calculating the first confidence coefficient according to the image features of each image is similar to the process of calculating the image classification parameters according to the image features of each image, and will not be repeated here; in the process of calculating the second confidence coefficient according to the character recognition result of each image, the matching coefficient can be used as the second confidence coefficient according to the matching coefficient between the character recognition information of each image; the content of calculating the confidence coefficient according to the text recognition result of each image in this embodiment can refer to the implementation process herein. Confidence calculation is carried out through data in two modes, namely multiple modes, of the image and the text, and comprehensiveness and effectiveness of the confidence calculation are improved.

Step S208, determining an image detection result of the image to be detected based on the image confidence and the target confidence threshold.

After the image confidence coefficient is obtained, in the step, an image detection result of the image to be detected is determined by means of the image confidence coefficient and a target confidence coefficient threshold value.

In an optional implementation manner provided in this embodiment, in a process of determining an image detection result of an image to be detected based on an image confidence level and a target confidence level threshold, the following operations are performed:

if the image confidence coefficient is greater than or equal to the target confidence coefficient threshold value, determining that the image detection result is passing detection;

and if the image confidence coefficient is smaller than the target confidence coefficient threshold value, determining that the image detection result is that the detection fails.

It should be noted that, in this embodiment, the steps S202 to S206 may be replaced by: acquiring an image to be detected, and performing image evaluation on the image to be detected to acquire an image evaluation result; reading the associated image of the image to be detected from the database based on the user identification and/or the equipment identification of the uploaded image to be detected under the condition that the image evaluation result does not accord with the evaluation index threshold (the current evaluation index threshold); constructing an image set based on the image to be detected and the associated image to obtain a target image set; accordingly, step S208 may be replaced by calculating an image confidence based on the image features of each image in the target image set, or calculating an image confidence based on each image and the text recognition result of each image, or calculating an image confidence based on the evaluation index distribution of each image and/or the evaluation index difference value of each two images in each image; after that, if the calculated image confidence coefficient is greater than or equal to the confidence coefficient threshold value, determining that the image detection result of the image to be detected is detection passing, and if the calculated image confidence coefficient is less than the confidence coefficient threshold value, determining that the image detection result of the image to be detected is detection failing; and forms a new implementation manner with other processing steps provided in the present embodiment;

Alternatively, the step S206 may be replaced by performing image evaluation on the image to be detected to obtain an image evaluation result, performing update processing on the target confidence level threshold based on the image evaluation result to obtain an updated confidence level threshold (updated confidence level threshold), and calculating the image confidence level according to the image evaluation result if the image evaluation result does not conform to the target evaluation index threshold; accordingly, step S208 may be replaced by determining an image detection result of the image to be detected based on the image confidence and the update confidence threshold; and forms a new implementation manner with other processing steps provided in the present embodiment; the process of updating the target confidence coefficient threshold based on the image evaluation result can be realized by performing down-regulation processing on the target confidence coefficient threshold if the image evaluation result accords with the target evaluation index threshold, and performing no processing if the image evaluation result does not accord with the target evaluation index threshold.

Or, step S204 may be replaced by updating the evaluation index threshold to obtain the target evaluation index threshold if the image classification parameter is within the preset parameter interval; step S206 may be replaced by performing image evaluation on the image to be detected to obtain an image evaluation result, updating the confidence coefficient threshold based on the image evaluation result to obtain a target confidence coefficient threshold, and calculating an image confidence coefficient according to the image evaluation result if the image evaluation result does not conform to the target evaluation index threshold; and forms a new implementation manner with other processing steps provided in the present embodiment; the process of updating the confidence coefficient threshold based on the image evaluation result can be realized by performing down-regulation processing on the confidence coefficient threshold if the image evaluation result accords with the target evaluation index threshold, and performing no processing if the image evaluation result does not accord with the target evaluation index threshold.

Or, the step S204 may be replaced by updating the confidence threshold to obtain the target confidence threshold if the image classification parameter is within the preset parameter interval; step S206 may be replaced by performing image evaluation on the image to be detected to obtain an image evaluation result, and calculating an image confidence according to the image evaluation result if the image evaluation result does not meet an evaluation index threshold; and forms a new implementation manner with other processing steps provided in the present embodiment;

or, the step S204 may be replaced by updating the confidence threshold to obtain the target confidence threshold if the image classification parameter is within the preset parameter interval; step S206 may be replaced by performing image evaluation on the image to be detected to obtain an image evaluation result, performing update processing on the target confidence coefficient threshold based on the image evaluation result to obtain an update confidence coefficient threshold, and calculating an image confidence coefficient according to the image evaluation result if the image evaluation result does not conform to the evaluation index threshold; step S208 may be replaced by determining an image detection result of the image to be detected based on the image confidence and the updated confidence threshold; and forms a new implementation with the other processing steps provided in this embodiment. The process of updating the target confidence coefficient threshold based on the image evaluation result can be realized by performing down-regulation processing on the confidence coefficient threshold if the image evaluation result accords with the evaluation index threshold, and performing no processing if the image evaluation result does not accord with the evaluation index threshold.

The above-mentioned process of updating the target confidence coefficient threshold based on the image evaluation result may be implemented by performing a down-adjustment process on the target confidence coefficient threshold if the image evaluation result meets the target evaluation index threshold, or performing no-process if the image evaluation result does not meet the target evaluation index threshold; it should be noted that, the process of updating the confidence threshold based on the image evaluation result is similar to the process of updating the target confidence threshold based on the image evaluation result, and the process of updating the target confidence threshold based on the image evaluation result is not repeated here.

In summary, in the image processing method provided in this embodiment, an image to be detected is first obtained, based on a user identifier and/or a device identifier of the uploaded image to be detected, an associated image of the image to be detected is read from a database, an image set is constructed based on the image to be detected and the associated image, and an image classification parameter is calculated based on image features of each image in a target image set obtained by the construction;

secondly, if the image classification parameters are in a preset parameter interval, updating the evaluation index threshold and the confidence coefficient threshold to obtain a target evaluation index threshold and a target confidence coefficient threshold; performing image evaluation on the image to be detected to obtain an image evaluation result, and calculating the image confidence coefficient according to the image evaluation result under the condition that the image evaluation result does not accord with the target evaluation index threshold; if the image confidence coefficient is greater than or equal to the target confidence coefficient threshold value, determining that the image detection result is passing detection; if the image confidence coefficient is smaller than the target confidence coefficient threshold value, determining that the image detection result is that the detection fails, updating the evaluation index threshold value and the confidence coefficient threshold value through the image classification parameters to improve the accuracy of image detection, and combining the image to be detected and the associated image to perform image detection to improve the comprehensiveness and flexibility of image detection.

The following further describes the image processing method provided in this embodiment by taking the application of the image processing method provided in this embodiment to an identity credential scene as an example, and referring to fig. 3, the image processing method applied to an identity credential scene specifically includes the following steps.

Step S302, acquiring an identity credential image and reading an associated identity credential image of the identity credential image from a database based on the device identifier of the uploaded identity credential image.

Step S304, constructing an image set according to the identity credential image and the associated identity credential image, and calculating image classification parameters based on the image characteristics of each image in the target image set obtained by construction.

And step S306, if the image classification parameters are in the preset parameter interval, updating the evaluation index threshold and the confidence coefficient threshold to obtain a target evaluation index threshold and a target confidence coefficient threshold.

Step S308, performing image evaluation on the identity document image to obtain an image evaluation result.

Step S310, determining an evaluation index distribution of each image and an evaluation index difference value of each two images in each image, in the case that the image evaluation result does not meet the target evaluation index threshold.

Step S312, calculating the image confidence according to the evaluation index distribution and the evaluation index difference value.

Step S314, if the image confidence coefficient is greater than or equal to the target confidence coefficient threshold value, determining that the image detection result of the identity document image is the passing of detection.

The step S314 may be replaced by determining that the image detection result is that the detection is failed if the image confidence is smaller than the target confidence threshold, and forming a new implementation manner with other processing steps provided in the present embodiment.

An embodiment of an image processing apparatus provided in the present specification is as follows:

in the above-described embodiments, an image processing method and an image processing apparatus corresponding thereto are provided, and the following description is made with reference to the accompanying drawings.

Referring to fig. 4, a schematic diagram of an embodiment of an image processing apparatus provided in this embodiment is shown.

Since the apparatus embodiments correspond to the method embodiments, the description is relatively simple, and the relevant portions should be referred to the corresponding descriptions of the method embodiments provided above. The device embodiments described below are merely illustrative.

The present embodiment provides an image processing apparatus including:

a parameter calculation module 402, configured to acquire an image to be detected and an associated image and perform image set construction, and calculate an image classification parameter based on image features of each image in a target image set obtained by the construction;

The threshold updating module 404 is configured to update the evaluation index threshold and the confidence coefficient threshold if the image classification parameter is within a preset parameter interval, so as to obtain a target evaluation index threshold and a target confidence coefficient threshold;

a confidence coefficient calculating module 406, configured to perform image evaluation on the image to be detected to obtain an image evaluation result, and calculate an image confidence coefficient according to the image evaluation result if the image evaluation result does not meet the target evaluation index threshold;

a result determination module 408 configured to determine an image detection result of the image to be detected based on the image confidence and the target confidence threshold.

in correspondence to the above-described image processing method, one or more embodiments of the present disclosure further provide an image processing apparatus for performing the above-provided image processing method, based on the same technical concept, and fig. 5 is a schematic structural diagram of the image processing apparatus provided by the one or more embodiments of the present disclosure.

An image processing apparatus provided in this embodiment includes:

As shown in fig. 5, the image processing apparatus may have a relatively large difference due to different configurations or performances, and may include one or more processors 501 and a memory 502, where one or more storage applications or data may be stored in the memory 502. Wherein the memory 502 may be transient storage or persistent storage. The application programs stored in memory 502 may include one or more modules (not shown), each of which may include a series of computer executable instructions in the image processing apparatus. Still further, the processor 501 may be configured to communicate with the memory 502 and execute a series of computer executable instructions in the memory 502 on the image processing device. The image processing device may also include one or more power supplies 503, one or more wired or wireless network interfaces 504, one or more input/output interfaces 505, one or more keyboards 506, and the like.

In a particular embodiment, an image processing apparatus includes a memory, and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the image processing apparatus, and configured to be executed by the one or more processors, the one or more programs comprising computer-executable instructions for:

Acquiring an image to be detected and an associated image, constructing an image set, and calculating image classification parameters based on image characteristics of each image in a target image set obtained by construction;

if the image classification parameters are in the preset parameter interval, updating the evaluation index threshold and the confidence coefficient threshold to obtain a target evaluation index threshold and a target confidence coefficient threshold;

performing image evaluation on the image to be detected to obtain an image evaluation result, and calculating an image confidence coefficient according to the image evaluation result under the condition that the image evaluation result does not accord with the target evaluation index threshold;

and determining an image detection result of the image to be detected based on the image confidence and the target confidence threshold.

An embodiment of a storage medium provided in the present specification is as follows:

in correspondence with the above-described image processing method, one or more embodiments of the present specification further provide a storage medium based on the same technical idea.

The storage medium provided in this embodiment is configured to store computer executable instructions that, when executed by a processor, implement the following flow:

It should be noted that, in the present specification, an embodiment of a storage medium and an embodiment of an image processing method in the present specification are based on the same inventive concept, so that a specific implementation of the embodiment may refer to an implementation of the foregoing corresponding method, and a repetition is omitted.

In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment focuses on the differences from other embodiments, for example, an apparatus embodiment, and a storage medium embodiment, which are all similar to a method embodiment, so that description is relatively simple, and relevant content in reading apparatus embodiments, and storage medium embodiments is referred to the part description of the method embodiment.

The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

In the 30 s of the 20 th century, improvements to one technology could clearly be distinguished as improvements in hardware (e.g., improvements to circuit structures such as diodes, transistors, switches, etc.) or software (improvements to the process flow). However, with the development of technology, many improvements of the current method flows can be regarded as direct improvements of hardware circuit structures. Designers almost always obtain corresponding hardware circuit structures by programming improved method flows into hardware circuits. Therefore, an improvement of a method flow cannot be said to be realized by a hardware entity module. For example, a programmable logic device (Programmable Logic Device, PLD) (e.g., field programmable gate array (Field Programmable Gate Array, FPGA)) is an integrated circuit whose logic function is determined by the programming of the device by a user. A designer programs to "integrate" a digital system onto a PLD without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Moreover, nowadays, instead of manually manufacturing integrated circuit chips, such programming is mostly implemented by using "logic compiler" software, which is similar to the software compiler used in program development and writing, and the original code before the compiling is also written in a specific programming language, which is called hardware description language (Hardware Description Language, HDL), but not just one of the hdds, but a plurality of kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), lava, lola, myHDL, PALASM, RHDL (Ruby Hardware Description Language), etc., VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog are currently most commonly used. It will also be apparent to those skilled in the art that a hardware circuit implementing the logic method flow can be readily obtained by merely slightly programming the method flow into an integrated circuit using several of the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable logic controllers, and embedded microcontrollers, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, atmel AT91SAM, microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic of the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller in a pure computer readable program code, it is well possible to implement the same functionality by logically programming the method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Such a controller may thus be regarded as a kind of hardware component, and means for performing various functions included therein may also be regarded as structures within the hardware component. Or even means for achieving the various functions may be regarded as either software modules implementing the methods or structures within hardware components.

The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. One typical implementation is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being functionally divided into various units, respectively. Of course, the functions of each unit may be implemented in the same piece or pieces of software and/or hardware when implementing the embodiments of the present specification.

One skilled in the relevant art will recognize that one or more embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, one or more embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present description can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present description is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the specification. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

One or more embodiments of the present specification may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. One or more embodiments of the specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.

The foregoing description is by way of example only and is not intended to limit the present disclosure. Various modifications and changes may occur to those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. that fall within the spirit and principles of the present document are intended to be included within the scope of the claims of the present document.

Claims

1. An image processing method, comprising:

2. The method of claim 1, the calculating an image confidence from the image evaluation result, comprising:

3. The method of claim 2, the evaluating an index profile comprising at least one of:

image quality score distribution, text confidence distribution, and image forgery index distribution.

4. The method of claim 1, the calculating image classification parameters based on image features of each image in the set of constructed obtained target images, comprising:

5. The method of claim 4, wherein calculating a first feature residual in a temporal dimension and a second feature residual in a spatial dimension based on the image feature map of each image comprises:

6. The method of claim 1, the associated image obtained by:

7. The method according to claim 6, wherein the database stores associated images corresponding to the user identifications and/or the device identifications;

and eliminating target images in the associated images according to the time sequence under the condition that the number of images of the associated images corresponding to the user identification and/or the equipment identification exceeds the threshold value of the number of images.

8. The method of claim 1, the image set construction, comprising:

and if not, taking the image set as the target image set.

9. The method of claim 8, wherein if the execution result after the operation of detecting whether the number of images in the image set is less than the preset number of images is yes, the following operations are executed:

10. The method of claim 1, the calculating an image confidence from the image evaluation result, comprising:

and calculating the image confidence according to the images and the character recognition results of the images.

11. The method of claim 1, the determining an image detection result of the image to be detected based on the image confidence and the target confidence threshold, comprising:

12. The method of claim 1, further comprising, after performing the image evaluation operation for the image to be detected to obtain an image evaluation result:

and under the condition that the image evaluation result accords with the target evaluation index threshold value, determining that the image detection result of the image to be detected is passing detection.

13. The method of claim 1, the updating the evaluation index threshold and the confidence threshold, comprising:

14. An image processing apparatus comprising:

the parameter calculation module is configured to acquire an image to be detected and an associated image, construct an image set and calculate image classification parameters based on image characteristics of each image in a target image set obtained by construction;

the threshold updating module is configured to update the evaluation index threshold and the confidence coefficient threshold if the image classification parameter is in a preset parameter interval, so as to obtain a target evaluation index threshold and a target confidence coefficient threshold;

The confidence coefficient calculating module is configured to carry out image evaluation on the image to be detected to obtain an image evaluation result, and calculate the image confidence coefficient according to the image evaluation result under the condition that the image evaluation result does not accord with the target evaluation index threshold value;

and a result determining module configured to determine an image detection result of the image to be detected based on the image confidence and the target confidence threshold.

15. An image processing apparatus comprising:

a processor; and a memory configured to store computer-executable instructions that, when executed, cause the processor to: