CN114882571A

CN114882571A - Method for acquiring image information and method for training image detection model

Info

Publication number: CN114882571A
Application number: CN202210616055.XA
Authority: CN
Inventors: 王珂尧
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2022-05-31
Filing date: 2022-05-31
Publication date: 2022-08-09

Abstract

The disclosure provides a method and a device for acquiring image information, relates to the technical field of artificial intelligence, in particular to the technical field of deep learning, image processing and computer vision, and can be applied to scenes such as human faces. The implementation scheme is as follows: acquiring a target image; extracting a first feature belonging to at least one of a plurality of classes from the target image; and acquiring the category of the target image according to the first characteristic.

Description

Method for acquiring image information and method for training image detection model

Technical Field

The present disclosure relates to the field of artificial intelligence technologies, in particular to the field of deep learning, image processing, and computer vision technologies, and may be applied to scenes such as human faces, and in particular to a method for acquiring image detection information, and a method, an apparatus, an electronic device, a computer-readable storage medium, and a computer program product for training an image detection model.

Background

Artificial intelligence is the subject of research that makes computers simulate some human mental processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.), both at the hardware level and at the software level. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing, and the like: the artificial intelligence software technology mainly comprises a computer vision technology, a voice recognition technology, a natural language processing technology, machine learning/deep learning, a big data processing technology, a knowledge map technology and the like.

Image detection techniques based on artificial intelligence have penetrated into various fields. The human face living body detection technology based on artificial intelligence judges whether the image data is from a human face living body or not according to the image data input by a user.

The approaches described in this section are not necessarily approaches that have been previously conceived or pursued. Unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section. Similarly, unless otherwise indicated, the problems mentioned in this section should not be considered as having been acknowledged in any prior art.

Disclosure of Invention

The present disclosure provides a method of acquiring image detection information, a method for training an image detection model, an apparatus, an electronic device, a computer readable storage medium and a computer program product.

According to an aspect of the present disclosure, there is provided a method of acquiring image detection information, including: acquiring a target image; extracting a first feature belonging to at least one of a plurality of classes from the target image; and acquiring the category of the target image according to the first characteristic.

According to another aspect of the present disclosure, there is provided a method for training an image detection model, comprising: obtaining a plurality of second sample images corresponding to each of a plurality of categories; for each of the multiple categories, acquiring multiple first sample images according to multiple second sample images corresponding to the category, wherein each first sample image is obtained by splicing partial areas of at least two second sample images; and training an image detection model based on a plurality of first sample images corresponding to each of the plurality of classes.

According to another aspect of the present disclosure, there is provided an apparatus for acquiring image information, including: a target image acquisition unit configured to acquire a target image; a feature extraction unit configured to extract a first feature belonging to at least one of a plurality of classes from the target image; and a detection result acquisition unit configured to acquire a category of the target image according to the first feature.

According to another aspect of the present disclosure, there is provided an apparatus for training an image detection model, including: a training image acquisition unit configured to obtain a plurality of second sample images corresponding to each of a plurality of classes; an updated image obtaining unit, configured to obtain, for each of the multiple categories, multiple first sample images according to multiple second sample images corresponding to the category, where each first sample image is obtained by splicing partial regions of at least two second sample images; and a first model training unit configured to train an image detection model based on a plurality of first sample images corresponding to each of the plurality of classes.

According to another aspect of the present disclosure, there is provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to implement a method according to the above.

According to another aspect of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to implement the method according to the above.

According to another aspect of the present disclosure, a computer program product is provided comprising a computer program, wherein the computer program realizes the method according to the above when executed by a processor.

According to one or more embodiments of the present disclosure, by extracting a first feature from a target image, since the first feature belongs to at least one of a plurality of categories, the extracted first feature is a separability feature that can distinguish between the plurality of categories, and thus, a detection result obtained according to the extracted first feature is accurate.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the embodiments and, together with the description, serve to explain the exemplary implementations of the embodiments. The illustrated embodiments are for purposes of illustration only and do not limit the scope of the claims. Throughout the drawings, identical reference numbers designate similar, but not necessarily identical, elements.

Fig. 1 illustrates a schematic diagram of an exemplary system in which various methods described herein may be implemented, in accordance with embodiments of the present disclosure;

FIG. 2 shows a flow diagram of a method of obtaining image information according to an embodiment of the present disclosure;

FIG. 3 shows a flow diagram of a method for training an image detection model according to an embodiment of the present disclosure;

FIG. 4 shows a flow chart of a process of acquiring a plurality of first sample images from a plurality of second sample images corresponding to classes in a method for training an image detection model according to an embodiment of the present disclosure;

FIG. 5 shows a flow diagram of a process for training an image detection model based on a plurality of first sample images corresponding to each of the plurality of classes in a method for training an image detection model according to an embodiment of the disclosure;

fig. 6 shows a block diagram of a structure of an apparatus for acquiring image information according to an embodiment of the present disclosure;

FIG. 7 shows a block diagram of an apparatus for training an image detection model according to an embodiment of the present disclosure; and

FIG. 8 illustrates a block diagram of an exemplary electronic device that can be used to implement embodiments of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

In the present disclosure, unless otherwise specified, the use of the terms "first", "second", etc. to describe various elements is not intended to limit the positional relationship, the timing relationship, or the importance relationship of the elements, and such terms are used only to distinguish one element from another. In some examples, a first element and a second element may refer to the same instance of the element, and in some cases, based on the context, they may also refer to different instances.

The terminology used in the description of the various described examples in this disclosure is for the purpose of describing particular examples only and is not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the elements may be one or more. Furthermore, the term "and/or" as used in this disclosure is intended to encompass any and all possible combinations of the listed items.

Embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

Fig. 1 illustrates a schematic diagram of an exemplary system 100 in which various methods and apparatus described herein may be implemented in accordance with embodiments of the present disclosure. Referring to fig. 1, the system 100 includes one or

more client devices

101, 102, 103, 104, 105, and 106, a server 120, and one or more communication networks 110 coupling the one or more client devices to the server 120.

Client devices

101, 102, 103, 104, 105, and 106 may be configured to execute one or more applications.

In embodiments of the present disclosure, the server 120 may run one or more services or software applications that enable the image detection method to be performed.

In some embodiments, the server 120 may also provide other services or software applications that may include non-virtual environments and virtual environments. In certain embodiments, these services may be provided as web-based services or cloud services, for example, provided to users of

client devices

101, 102, 103, 104, 105, and/or 106 under a software as a service (SaaS) model.

In the configuration shown in fig. 1, server 120 may include one or more components that implement the functions performed by server 120. These components may include software components, hardware components, or a combination thereof, which may be executed by one or more processors. A user operating a

client device

101, 102, 103, 104, 105, and/or 106 may, in turn, utilize one or more client applications to interact with the server 120 to take advantage of the services provided by these components. It should be understood that a variety of different system configurations are possible, which may differ from system 100. Accordingly, fig. 1 is one example of a system for implementing the various methods described herein and is not intended to be limiting.

The user may receive the detection results using

client devices

101, 102, 103, 104, 105, and/or 106. The client device may provide an interface that enables a user of the client device to interact with the client device. The client device may also output information to the user via the interface. Although fig. 1 depicts only six client devices, those skilled in the art will appreciate that any number of client devices may be supported by the present disclosure.

Client devices

101, 102, 103, 104, 105, and/or 106 may include various types of computer devices, such as portable handheld devices, general purpose computers (such as personal computers and laptops), workstation computers, wearable devices, smart screen devices, self-service terminal devices, service robots, gaming systems, thin clients, various messaging devices, sensors or other sensing devices, and so forth. These computer devices may run various types and versions of software applications and operating systems, such as MICROSOFT Windows, APPLE iOS, UNIX-like operating systems, Linux, or Linux-like operating systems (e.g., GOOGLE Chrome OS); or include various Mobile operating systems such as MICROSOFT Windows Mobile OS, iOS, Windows Phone, Android. Portable handheld devices may include cellular telephones, smart phones, tablets, Personal Digital Assistants (PDAs), and the like. Wearable devices may include head-mounted displays (such as smart glasses) and other devices. The gaming system may include a variety of handheld gaming devices, internet-enabled gaming devices, and the like. The client device is capable of executing a variety of different applications, such as various Internet-related applications, communication applications (e.g., email applications), Short Message Service (SMS) applications, and may use a variety of communication protocols.

Network 110 may be any type of network known to those skilled in the art that may support data communications using any of a variety of available protocols, including but not limited to TCP/IP, SNA, IPX, etc. By way of example only, one or more networks 110 may be a Local Area Network (LAN), an ethernet-based network, a token ring, a Wide Area Network (WAN), the internet, a virtual network, a Virtual Private Network (VPN), an intranet, an extranet, a Public Switched Telephone Network (PSTN), an infrared network, a wireless network (e.g., bluetooth, WIFI), and/or any combination of these and/or other networks.

The server 120 may include one or more general purpose computers, special purpose server computers (e.g., PC (personal computer) servers, UNIX servers, mid-end servers), blade servers, mainframe computers, server clusters, or any other suitable arrangement and/or combination. The server 120 may include one or more virtual machines running a virtual operating system, or other computing architecture involving virtualization (e.g., one or more flexible pools of logical storage that may be virtualized to maintain virtual storage for the server). In various embodiments, the server 120 may run one or more services or software applications that provide the functionality described below.

The computing units in server 120 may run one or more operating systems including any of the operating systems described above, as well as any commercially available server operating systems. The server 120 may also run any of a variety of additional server applications and/or middle tier applications, including HTTP servers, FTP servers, CGI servers, JAVA servers, database servers, and the like.

In some implementations, the server 120 may include one or more applications to analyze and consolidate data feeds and/or event updates received from users of the

client devices

101, 102, 103, 104, 105, and 106. Server 120 may also include one or more applications to display data feeds and/or real-time events via one or more display devices of

client devices

101, 102, 103, 104, 105, and 106.

In some embodiments, the server 120 may be a server of a distributed system, or a server incorporating a blockchain. The server 120 may also be a cloud server, or a smart cloud computing server or a smart cloud host with artificial intelligence technology. The cloud Server is a host product in a cloud computing service system, and is used for solving the defects of high management difficulty and weak service expansibility in the traditional physical host and Virtual Private Server (VPS) service.

The system 100 may also include one or more databases 130. In some embodiments, these databases may be used to store data and other information. For example, one or more of the databases 130 may be used to store information such as audio files and object files. The data store 130 may reside in various locations. For example, the data store used by the server 120 may be local to the server 120, or may be remote from the server 120 and may communicate with the server 120 via a network-based or dedicated connection. The data store 130 may be of different types. In certain embodiments, the data store used by the server 120 may be a database, such as a relational database. One or more of these databases may store, update, and retrieve data to and from the database in response to the command.

In some embodiments, one or more of the databases 130 may also be used by applications to store application data. The databases used by the application may be different types of databases, such as key-value stores, object stores, or regular stores supported by a file system.

The system 100 of fig. 1 may be configured and operated in various ways to enable application of the various methods and apparatus described in accordance with the present disclosure.

Referring to fig. 2, an image detection method 200 according to some embodiments of the present disclosure includes:

step S210: acquiring a target image;

step S220: extracting a first feature belonging to at least one of a plurality of classes from the target image;

step S230: and acquiring the category of the target image according to the first characteristic.

By extracting the first feature from the target image, the extracted first feature is a separability feature which can distinguish among a plurality of categories because the first feature belongs to at least one of the plurality of categories, and therefore the detection result obtained according to the extracted first feature is accurate.

In the related art, by extracting various features in an image, a comprehensive judgment is performed based on the extracted various features to obtain a detection result of the image. The extracted features include features that cannot be distinguished among various categories, so that the accuracy of the obtained detection result is not high.

For example, in the living human face detection process, feature extraction is performed on an image containing a human face to determine whether the image is from a living human face. However, in the image including the face, in addition to the feature of judging whether it comes from a living body of the face, the feature of judging whether it is the face (i.e., the face structural feature), the feature of which face it is (i.e., the face identification feature), and the like are included; these face structure features and face identity features can also be present in images from non-living human faces (e.g., screen shot images), and thus do not belong to living human face or non-living human face features. Based on the face structure features and the face identity features, whether the face structure features and the face identity features come from a living face or not is judged, so that the judgment result is often inaccurate.

In one example, the image obtained after the person is photographed is subjected to feature extraction, and whether the person passes the verification is judged based on the extracted features and the face identity information which is recorded in advance. In the judgment process, the human face identity characteristics are strongly depended. If the photo of the person is used as the image obtained after the person is photographed for verification, the photo of the person also contains the face identity information, so that the verification can be passed only by photographing the photo of the person, and the verification result is inaccurate.

According to the embodiment of the present disclosure, by extracting the first feature from the target image, since the first feature belongs to at least one of the plurality of categories, that is, the first feature does not exist in the image of each of the plurality of categories at the same time, the first feature is enabled to distinguish between the plurality of categories, so that the detection result obtained based on the first feature is accurate.

For example, in the process of face live body detection, features belonging to at least one of a face live body class and a screen attack class are extracted. For example, the feature belonging to the living human face class is a reflection feature of a human face included in an image against light, and the feature belonging to the screen attack class is a screen moire feature. In the process of extracting the features of the target image, one of the reflection features of the face to the light rays and the moire features of the screen in the target image is extracted. So that the extracted features are the separability features which can be clearly distinguished between the live human face class and the screen class. Therefore, whether the target image comes from the human face living body or not is judged more accurately based on the extracted features.

In the technical scheme of the disclosure, the collection, storage, use, processing, transmission, provision, disclosure and other processing of the personal information of the related user are all in accordance with the regulations of related laws and regulations and do not violate the good customs of the public order.

In some embodiments, the target image comprises an image comprising a human face, and the plurality of categories comprise a living human face score, a screen attack category, a head die attack category, or a paper attack category.

It should be noted that the target image is an image containing a human face, and the plurality of categories including a living human face category, a screen attack category, a head pattern attack category or a paper attack category are merely exemplary, and those skilled in the art will understand that the image detection method according to the present disclosure can be applied to a multi-classification process of any image.

In some embodiments, the step S210 of acquiring the target image includes: performing face detection on the first image in response to receiving the first image; in response to detecting a face in the first image, the first image is determined to be a target image.

In some embodiments, the first image may be an image captured by a camera or an image read from a memory or transmitted by another data transmission device.

In some embodiments, the step S220 of extracting a first feature belonging to at least one of a plurality of classes from the target image comprises:

inputting the target image into a feature extraction network to obtain the first feature; wherein the content of the first and second substances,

the feature extraction network is trained by using a first sample image, the first sample image is obtained by splicing partial regions in at least two second sample images, the first sample image corresponds to a first category in the plurality of categories, and the second sample image corresponds to the first category.

Since the first sample image is formed by splicing partial regions of at least two second sample images, the first sample includes features belonging to a first class corresponding to the at least two second sample images, and since the first sample image is formed by splicing partial regions of the at least two second sample images, attention (for example, face structure information, face identity features, and the like) to inseparable features contained in the at least two second sample images, which are difficult to distinguish between the respective classes, is destroyed or removed. The feature extraction network is trained by adopting the first sample image, so that the trained feature extraction network can extract features which belong to a first class corresponding to the image and are contained in the image (namely, separability features which can be distinguished among the classes), and the attention to the inseparability features which are contained in the image and are difficult to distinguish among the classes (such as face structure information, face identity features and the like) is reduced, so that a classification result obtained based on the features extracted by the feature extraction network is more accurate; and further the accuracy of the detection result obtained based on the classification result is improved.

In some embodiments, the feature extraction network may include, without limitation, convolution networks in Resnet-18, MobileNet V2, VGG11, VGG15, and the like.

In some embodiments, the first sample image is a die attack image corresponding to a die attack class, a paper attack image corresponding to a paper attack class, or a screen attack image corresponding to a screen attack class.

It is to be appreciated that the training is performed using a plurality of sample images in the training of the feature extraction network, wherein the plurality of sample images includes the first sample image. In some embodiments, the first sample image is a set of images.

In some embodiments, the first sample image corresponding to the second sample image is obtained by dividing each of the at least two second sample images into a plurality of image blocks, and replacing, for each second sample image, one or more image blocks in the second sample image with an image block from another second sample image of the at least two second sample images.

For example, in the case where the first sample image is a screen attack image, by obtaining two screen attack images corresponding to the screen attack class, the two screen attack images are respectively divided into a plurality of image blocks. And aiming at least one image block in a plurality of image blocks included in a first screen attack image in the two screen attack images, replacing at least one image block in a plurality of image blocks included in a second screen attack image in the two screen attack images, so as to obtain two first sample images corresponding to the two screen attack images.

Because the first screen attack image and the second screen attack image both include features belonging to the screen attack class, in the process of replacing image blocks of the first screen attack image and the second screen attack image, the features belonging to the screen attack class (for example, the screen moire features) are still retained, and the face identity features, the face structure features and the like existing in the first screen attack image and the second screen attack image are at least partially removed due to the replacement of the image blocks. Therefore, the feature extraction network trained according to the obtained first sample image can accurately extract features belonging to the screen attack class without paying attention to face identity features, face structure features and the like.

In some embodiments, a first partial area of the at least two partial areas from the at least two second sample images that are stitched into the first sample image accounts for no less than 50% of the total area of the first sample image.

In the process of forming a first sample image, because too many image areas for forming the image by splicing are formed, the characteristics of the second sample images belonging to the first class are lost in each image area, and the characteristic extraction process is influenced.

In some embodiments, each image is divided into 8 image blocks, wherein the number of image blocks replaced does not exceed 4.

In some embodiments, the step S230, obtaining the detection result of the target image based on the first feature, includes: inputting the first features into a classification network to obtain a confidence level that the target image corresponds to each of a plurality of classes; and taking the category with the highest corresponding confidence coefficient in the plurality of categories in the target image as a classification result of the target image, and obtaining a detection result of the target image based on the classification result.

In some embodiments, the classification network may be a fully connected layer or a pooled layer, and the like, without limitation.

In some embodiments, the classification result is taken as the detection result.

In some embodiments, the detection result indicates whether the target image corresponds to the target category.

In some embodiments, the target class is a first class of a plurality of classes. For example, the target image includes an image including a human face, the plurality of categories include a human face live class, a screen attack class, a head model attack class, or a paper attack class, and the target is classified as the human face live class.

In some embodiments, in response to the classification result of the target image being any of the plurality of classes other than the target class, a detection result is obtained indicating that the target image does not correspond to the target class.

For example, when the classification result of the target image is obtained as any one of a screen attack class, a head pattern attack class, or a paper attack class, a detection result indicating that the target image does not correspond to a living human face class, i.e., does not come from a living human face is determined.

According to another aspect of the present disclosure, there is also provided a method for training an image detection model, as shown in fig. 3, the method 300 includes:

step S310: obtaining a plurality of second sample images corresponding to each of a plurality of categories;

step S320: for each of the multiple categories, acquiring multiple first sample images according to multiple second sample images corresponding to the category, wherein each first sample image is obtained by splicing partial areas of at least two second sample images; and

step S330: training an image detection model based on a plurality of first sample images corresponding to each of the plurality of classes.

Since each first sample image used for training the image detection model is formed by splicing partial regions of the at least two second sample images belonging to the same class, each first sample image includes features belonging to the same class, and since the first sample image is formed by splicing partial regions of the at least two second sample images, attention (e.g., face structure information, face identity features, and the like) to inseparable features contained in the at least two second sample images, which are difficult to distinguish between the classes, is destroyed or removed. The image detection model is trained by adopting a plurality of first sample images corresponding to each of a plurality of classes, so that the trained image detection model can extract features (namely, separability features capable of distinguishing among the classes) belonging to the class corresponding to the image contained in the image, the attention (such as face structure information, face identity features and the like) to the inseparability features which are contained in the image and are difficult to distinguish among the classes is reduced, and the accuracy of the detection result obtained based on the image detection model is improved. Meanwhile, the generalization capability of the image detection model is improved.

In the related art, an image detection model is directly trained by using a plurality of obtained second sample images corresponding to each of a plurality of classes, and each second sample image includes not only a separability feature for distinguishing between the classes but also an inseparability feature which cannot distinguish between the classes, and in the training process, the adopted supervision information is limited and cannot be subdivided into the features (for example, two-classification supervision), so that the trained model obtains a detection result based on the separability feature and the inseparability feature at the same time, and the accuracy of the detection result is not high. Meanwhile, the generalization capability of the image detection model is poor.

For example, in the process of training a face verification model, a training image set from a living human face and a training image set from each attack type are used for training the model, and whether the living human face passes verification or not is judged by extracting the image features of each image in the training image set and comparing the image features with the pre-stored face identity features. In the model training process, two-classification supervision training is adopted, so that the features extracted by the model comprise separability features for distinguishing human face living bodies and various attack types and also comprise human face identity features. And the trained model is made to strongly depend on the identity characteristics of the human face for judgment. If the photo of the person is used as an input image and the model is input for verification, the photo of the person can pass the verification because the photo of the person also contains the identity characteristics of the face, so that the verification result is inaccurate.

In some embodiments, the image detection model is used for face liveness detection, and the plurality of categories include a face liveness category, a screen attack category, a head pattern attack category, or a paper attack category.

In some embodiments, the image detection model includes a feature extraction network and a classification network, and the feature extraction network may be Resnet-18, MobileNet V2, VGG11, VGG15, and the like, but is not limited thereto; the classification network may be, for example, a fully connected layer, a pooled layer, or the like.

In some embodiments, as shown in fig. 4, the obtaining of the updated image set corresponding to the training image set in step S320 includes:

step S410: dividing the third sample image into a first number of first image blocks;

step S420: dividing a fourth sample image into a first number of second image blocks, the second sample image comprising: the third sample image and the fourth sample image; and

step S430: and replacing a second number of the first image blocks with the second number of the second image blocks to obtain the first sample image.

The first sample image is obtained by dividing a third sample image and a fourth sample image in the plurality of second sample images into a preset number of image blocks and replacing the image blocks between the third sample image and the fourth sample image, so that the process method of obtaining the first sample image is simple and the data processing amount is small.

For example, in the case where the first sample image is obtained for the screen attack class, each of the two screen attack images is divided into a plurality of image blocks by obtaining two screen attack images corresponding to the screen attack class. And aiming at least one image block in a plurality of image blocks included in a first screen attack image of the two screen attack images, replacing the image block by at least one image block in a plurality of image blocks included in a second screen attack image of the two screen attack images, so as to generate a first sample image corresponding to the screen attack class.

Since the first screen attack image and the second screen attack image both include features belonging to the screen attack class, in the process of replacing image blocks of the first screen attack image and the second screen attack image, the features belonging to the screen attack class (for example, the screen moire features) are still retained, and the face identity features, the face structure features and the like existing in the first screen attack image and the second screen attack image are at least partially removed due to the replacement of the image blocks. Therefore, the feature extraction network trained according to the first sample image obtained after replacement can accurately extract features belonging to the screen attack class without paying attention to face identity features, face structure features and the like.

In some embodiments, in step S410, by performing face detection on each of the plurality of second sample images, after a face region is detected, the face region image is intercepted, and after the face region image is adjusted to a preset size, the image with the preset size is divided into a preset number of image blocks.

In some embodiments, before dividing the image with the preset size into the preset number of image blocks, the image with the preset size is further subjected to normalization processing, and the normalized image is subjected to random data enhancement processing.

In some embodiments, the ratio between the second number and the first number is less than or equal to 0.5.

In the process of forming the first sample image aiming at the third sample image, the phenomenon that the feature extraction process in the model training process is influenced because the first sample image area loses the features belonging to the category in the second sample image due to the fact that the image areas of the first sample image formed by splicing are too many is avoided.

In some embodiments, the third sample image and the fourth sample image are divided into 8 tiles, of which up to 4 tiles are replaced for each image.

In some embodiments, step S330, training an image detection model based on a plurality of first sample images corresponding to each of the plurality of classes, as shown in fig. 5, performs:

step S510: inputting a fifth sample image to the image detection model for each of the plurality of classes to obtain a prediction class corresponding to the fifth sample image, wherein the plurality of first sample images of the class include the fifth sample image; and

step S520: and adjusting the parameters of the image detection model based on the prediction category and the category corresponding to the fifth sample image.

By adjusting the model parameters based on the corresponding prediction categories and the categories according to the fifth sample images included in the plurality of first sample images in each category, the monitoring is performed on each category in the model training process, the monitoring is performed on each category in the plurality of categories in a category-by-category monitoring process, and the separability characteristic extracted by the trained model is more accurate.

In some embodiments, in training the image detection model based on the plurality of updated image sets, a binary supervision is performed. For example, when the image detection model is used for human face living body detection, for an image input to the image detection model in a training process, a binary prediction result indicating whether the image is from a human face living body is obtained, and a parameter of the image detection model is adjusted based on the binary prediction result and a category corresponding to the image.

In some embodiments, the method for training an image detection model according to the present disclosure further comprises: training the image detection model based on a plurality of second sample images corresponding to each of the plurality of classes.

In the process of training the image detection model, training is performed based on a plurality of second sample images corresponding to each of a plurality of classes, so that the accuracy of the separability characteristics extracted by the model is further improved, and meanwhile, the data volume of the attack class can be expanded.

According to the embodiment of the present disclosure, the plurality of second sample images and the plurality of first sample images corresponding to each of the plurality of categories correspond one-to-one, that is, each second sample image and the corresponding first sample image constitute an image pair. In the process of training the image detection model based on the plurality of second sample images corresponding to each of the plurality of categories, simultaneously training the image detection model based on the plurality of first sample images corresponding to each of the plurality of categories, and realizing the effect of supervising the training process of training the second sample images by adopting the corresponding first sample images; the image features extracted by the trained image detection model are features belonging to at least one of a plurality of categories, and the separability features can be distinguished among the categories, so that the accuracy of the detection result obtained by the image detection model is improved. Meanwhile, the features extracted by the image monitoring model are separability features which can be distinguished among a plurality of categories, so that the generalization capability of the model is improved.

In some embodiments, training the image detection model based on the plurality of second sample images corresponding to each of the plurality of classes comprises inputting each of the plurality of second sample images corresponding to each of the plurality of classes to the image detection model to obtain a prediction class corresponding to the second sample image; and adjusting parameters of the image detection model based on the prediction type corresponding to the second sample image and the type corresponding to the second sample image.

According to another aspect of the present disclosure, there is also provided an apparatus for acquiring image information, as shown in fig. 6, the apparatus 600 includes: a target image acquisition unit 610 configured to acquire a target image; a feature extraction unit 620 configured to extract a first feature belonging to at least one of a plurality of classes from the target image; and a detection result acquiring unit 630 configured to acquire a category of the target image according to the first feature.

In some embodiments, the feature extraction unit 620 includes: an image input unit configured to input the target image to a feature extraction network to obtain the first feature; wherein the feature extraction network is trained by using a first sample image, the first sample image is obtained by splicing partial regions in at least two second sample images, the first sample image corresponds to a first category in the plurality of categories, and the second sample image corresponds to the first category.

In some embodiments, the target image comprises an image containing a human face, and the plurality of categories comprise a live human face category, a screen attack category, a head model attack category, or a paper attack category.

According to another aspect of the present disclosure, there is also provided an apparatus for training an image detection model, as shown in fig. 7, the apparatus 700 includes: a training image acquisition unit 710 configured to obtain a plurality of second sample images corresponding to each of a plurality of classes; an updated image obtaining unit 720, configured to obtain, for each of the multiple categories, multiple first sample images according to multiple second sample images corresponding to the category, where each first sample image is obtained by splicing partial regions of at least two second sample images; and a first model training unit 730 configured to train an image detection model based on a plurality of first sample images corresponding to each of the plurality of classes.

In some embodiments, the update image obtaining unit 720 includes: a first image dividing unit configured to divide the third sample image into a first number of first image blocks; a second image dividing unit configured to divide a fourth sample image into a first number of second image blocks, the second sample image comprising: the third sample image and the fourth sample image; and the image block replacing unit is configured to replace a second number of the first image blocks with the second number of the second image blocks to obtain the first sample image.

In some embodiments, the first model training unit 730 comprises: a result prediction unit configured to input, for each of the plurality of classes, a fifth sample image to the image detection model to obtain a prediction class corresponding to the fifth sample image, wherein the plurality of first sample images of the class include the fifth sample image; and a parameter adjusting unit configured to adjust a parameter of the image detection model based on the prediction category and the category corresponding to the fifth sample image.

In some embodiments, the apparatus 700 further comprises: a second model training unit configured to train the image detection model based on a plurality of second sample images corresponding to each of the plurality of classes.

In some embodiments, the plurality of categories include a live face category, a screen attack category, a head die attack category, or a paper attack category.

According to another aspect of the present disclosure, there is also provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a method according to the present disclosure.

According to another aspect of the present disclosure, there is also provided a non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method according to the present disclosure.

According to another aspect of the present disclosure, there is also provided a computer program product comprising a computer program, wherein the computer program, when executed by a processor, implements the method according to the present disclosure.

Referring to fig. 8, a block diagram of a structure of an electronic device 800, which may be a server or a client of the present disclosure, which is an example of a hardware device that may be applied to aspects of the present disclosure, will now be described. Electronic device is intended to represent various forms of digital electronic computer devices, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 8, the electronic device 800 includes a computing unit 801 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM)802 or a computer program loaded from a storage unit 808 into a Random Access Memory (RAM) 803. In the RAM803, various programs and data required for the operation of the electronic apparatus 800 can also be stored. The calculation unit 801, the ROM 802, and the RAM803 are connected to each other by a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.

A number of components in the electronic device 800 are connected to the I/O interface 805, including: an input unit 806, an output unit 807, a storage unit 808, and a communication unit 809. The input unit 806 may be any type of device capable of inputting information to the electronic device 800, and the input unit 806 may receive input numeric or character information and generate key signal inputs related to user settings and/or function controls of the electronic device, and may include, but is not limited to, a mouse, a keyboard, a touch screen, a track pad, a track ball, a joystick, a microphone, and/or a remote controller. Output unit 807 can be any type of device capable of presenting information and can include, but is not limited to, a display, speakers, an object/audio output terminal, a vibrator, and/or a printer. The storage unit 808 may include, but is not limited to, a magnetic disk, an optical disk. The communication unit 809 allows the electronic device 800 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers and/or chipsets, such as bluetooth (TM) devices, 802.11 devices, WiFi devices, WiMax devices, cellular communication devices, and/or the like.

Computing unit 801 may be a variety of general and/or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and the like. The computing unit 801 performs the various methods and processes described above, such as the method 200. For example, in some embodiments, the method 200 may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and/or installed onto the electronic device 800 via the ROM 802 and/or the communication unit 809. When loaded into RAM803 and executed by computing unit 801, may perform one or more of the steps of method 200 described above. Alternatively, in other embodiments, the computing unit 801 may be configured to perform the method 200 in any other suitable manner (e.g., by way of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server with a combined blockchain.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be performed in parallel, sequentially or in different orders, and are not limited herein as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved.

Although embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it is to be understood that the above-described methods, systems and apparatus are merely exemplary embodiments or examples and that the scope of the present invention is not limited by these embodiments or examples, but only by the claims as issued and their equivalents. Various elements in the embodiments or examples may be omitted or may be replaced with equivalents thereof. Further, the steps may be performed in an order different from that described in the present disclosure. Further, various elements in the embodiments or examples may be combined in various ways. It is important that as technology evolves, many of the elements described herein may be replaced with equivalent elements that appear after the present disclosure.

Claims

1. A method of acquiring image information, comprising:

acquiring a target image;

extracting a first feature belonging to at least one of a plurality of classes from the target image; and

and acquiring the category of the target image according to the first characteristic.

2. The method of claim 1, wherein said extracting from the target image a first feature belonging to at least one of a plurality of classes comprises:

3. The method according to claim 2, wherein a first partial area of the at least two partial areas from the at least two second sample images that are stitched into the first sample image accounts for no less than 50% of the total area of the first sample image.

4. The method of any of claims 1-3, wherein the target image comprises an image containing a human face, and the plurality of categories comprise a live human face category, a screen attack category, a head-mounted attack category, or a paper attack category.

5. A method for training an image detection model, comprising:

obtaining a plurality of second sample images corresponding to each of a plurality of categories;

for each of the multiple categories, acquiring multiple first sample images according to multiple second sample images corresponding to the category, wherein each first sample image is obtained by splicing partial areas of at least two second sample images; and

training an image detection model based on a plurality of first sample images corresponding to each of the plurality of classes.

6. The method of claim 5, wherein the obtaining a plurality of second sample images corresponding to the category and obtaining a plurality of first sample images comprises:

dividing the third sample image into a first number of first image blocks;

dividing a fourth sample image into a first number of second image blocks, the second sample image comprising: the third sample image and the fourth sample image; and

and replacing a second number of the first image blocks with the second number of the second image blocks to obtain the first sample image.

7. The method of claim 6, wherein a ratio between the second quantity and the first quantity is less than or equal to 0.5.

8. The method of any of claims 5-7, wherein training an image detection model based on a plurality of first sample images corresponding to each of the plurality of classes comprises:

for each of the plurality of categories,

inputting a fifth sample image into the image detection model to obtain a prediction category corresponding to the fifth sample image, wherein the plurality of first sample images of the category comprise the fifth sample image; and

and adjusting the parameters of the image detection model based on the prediction category and the category corresponding to the fifth sample image.

9. The method according to any one of claims 5-8, further comprising:

training the image detection model based on a plurality of second sample images corresponding to each of the plurality of classes.

10. The method of any of claims 5-9, wherein the plurality of categories include a live face category, a screen attack category, a head-mounted attack category, or a paper attack category.

11. An apparatus for acquiring image information, comprising:

a target image acquisition unit configured to acquire a target image;

a feature extraction unit configured to extract a first feature belonging to at least one of a plurality of classes from the target image; and

a detection result acquisition unit configured to acquire a category of the target image according to the first feature.

12. The apparatus of claim 11, wherein the feature extraction unit comprises:

an image input unit configured to input the target image to a feature extraction network to obtain the first feature; wherein the content of the first and second substances,

13. The apparatus according to claim 12, wherein a first partial area of the at least two partial areas from the at least two second sample images that are stitched into the first sample image accounts for no less than 50% of the total area of the first sample image.

14. The apparatus of claim 11, wherein the target image comprises an image containing a human face, and the plurality of categories comprise a live human face category, a screen attack category, a head model attack category, or a paper attack category.

15. An apparatus for training an image detection model, comprising:

a training image acquisition unit configured to obtain a plurality of second sample images corresponding to each of a plurality of classes;

an updated image obtaining unit, configured to obtain, for each of the multiple categories, multiple first sample images according to multiple second sample images corresponding to the category, where each first sample image is obtained by splicing partial regions of at least two second sample images; and

a first model training unit configured to train an image detection model based on a plurality of first sample images corresponding to each of the plurality of classes.

16. The apparatus of claim 15, wherein the update image acquisition unit comprises:

a first image dividing unit configured to divide the third sample image into a first number of first image blocks;

a second image dividing unit configured to divide a fourth sample image into a first number of second image blocks, the second sample image comprising: the third sample image and the fourth sample image; and

an image block replacing unit configured to replace a second number of the first image blocks with the second number of the second image blocks, resulting in the first sample image.

17. The apparatus of claim 16, wherein a ratio between the second number and the first number is less than or equal to 0.5.

18. The apparatus according to any one of claims 15-17, wherein the first model training unit comprises:

a result prediction unit configured to input, for each of the plurality of classes, a fifth sample image to the image detection model to obtain a prediction class corresponding to the fifth sample image, wherein the plurality of first sample images of the class include the fifth sample image; and

a parameter adjusting unit configured to adjust a parameter of the image detection model based on the prediction category and the category corresponding to the fifth sample image.

19. The apparatus of any of claims 15-18, further comprising:

a second model training unit configured to train the image detection model based on a plurality of second sample images corresponding to each of the plurality of classes.

20. The apparatus of any of claims 15-19, wherein the plurality of categories include a live human face category, a screen attack category, a head-mounted attack category, or a paper attack category.

21. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein

The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-10.

22. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-10.

23. A computer program product comprising a computer program, wherein the computer program realizes the method of any one of claims 1-10 when executed by a processor.