CN111339964A

CN111339964A - Image processing method and device, electronic equipment and storage medium

Info

Publication number: CN111339964A
Application number: CN202010130108.8A
Authority: CN
Inventors: 黄青虬; 杨磊; 黄怀毅; 吴桐; 林达华
Original assignee: Beijing Sensetime Technology Development Co Ltd
Current assignee: Beijing Sensetime Technology Development Co Ltd
Priority date: 2020-02-28
Filing date: 2020-02-28
Publication date: 2020-06-26

Abstract

The present disclosure relates to an image processing method and apparatus, an electronic device, and a storage medium, wherein the method includes: acquiring a target image to be identified; carrying out face recognition on the target image by using the trained face recognition model to obtain a face recognition result of the target image; the face recognition model is obtained by training based on training samples in an extended sample pool, and the training samples in the extended sample pool are obtained by labeling at least one face image in a first image according to a face recognition result of the first image and image description information. The embodiment of the disclosure can identify the face by using the face identification model with higher accuracy.

Description

Image processing method and device, electronic equipment and storage medium

Technical Field

The present disclosure relates to the field of computer vision technologies, and in particular, to an image processing method and apparatus, an electronic device, and a storage medium.

Background

With the development of electronic technology, more and more work can be completed by using electronic equipment, for example, a human face can be automatically recognized by using the electronic equipment, so that convenience is provided for people. At present, the accuracy of the face recognition technology exceeds that of human eyes, and a reliable condition is further provided for popularization of the face recognition technology.

In general, face recognition techniques rely on a large amount of labeled training data. As the amount of training data is larger, the cost of labeling the training data is higher. This brings a very large limitation to the practical application of the face recognition technology.

Disclosure of Invention

The present disclosure proposes an image processing technical solution.

According to an aspect of the present disclosure, there is provided an image processing method including:

acquiring a target image to be identified;

carrying out face recognition on the target image by using the trained face recognition model to obtain a face recognition result of the target image; the face recognition model is obtained by training based on training samples in an extended sample pool, and the training samples in the extended sample pool are obtained by labeling at least one face image in a first image according to a face recognition result of the first image and image description information.

In one or more possible implementations, the method further includes:

and under the condition that the accuracy of the face recognition result of the target image is greater than a preset accuracy threshold, searching related information of the face in the target image according to the face recognition result of the target image, wherein the related information comprises one or more of identity information, whereabouts information and character images.

In one or more possible implementations, the extended sample pool includes:

acquiring an unmarked first image in at least one detection image;

labeling at least one face image in the first image according to the face recognition result of the first image and the image description information;

and adding the labeled human face image serving as a training sample into a sample pool, and expanding the sample pool to obtain the expanded sample pool.

In one or more possible implementations, before the acquiring the first image that is not labeled in the at least one detection image, the method further includes:

acquiring a second image in at least one detection image, wherein the second image comprises a face image;

labeling the face image in the second image according to the image description information of the second image;

and constructing the sample pool by taking the face image labeled in the second image as a training sample.

In one or more possible implementations, the acquiring a second image of the at least one detected image includes:

for one detection image in the at least one detection image, determining a first number of face images included in the detection image and a second number of identity marks included in image description information of the detection image;

determining the detection image as a second image in a case where the first number is equal to 1 and the second number is equal to 1.

In one or more possible implementation manners, the labeling, according to the image description information of the second image, a face image in the second image, includes:

extracting the identity included in the image description information of the second image;

and taking the extracted identity as a label of the face image in the second image to label the face image in the second image.

In one or more possible implementations, the method further includes:

training the face recognition model by using the face image labeled in the second image to obtain a preliminarily trained face recognition model;

and carrying out face recognition on the first image by using the preliminarily trained face recognition model to obtain a face recognition result of the first image.

In one or more possible implementation manners, the labeling at least one facial image in the first image according to the face recognition result of the first image and the image description information includes:

and for one face image in the first image, labeling the face image under the condition that the face recognition result of the face image meets a preset labeling condition according to the face recognition result of the face image and the image description information of the first image.

In one or more possible implementations, the preset labeling condition includes:

the accuracy of the face recognition result is greater than a preset accuracy threshold; and the face recognition result is matched with the identity included in the image description information of the first image.

In one or more possible implementations, after labeling the face image satisfying the labeling condition, the method further includes:

determining a third number of the face images which are not marked in the first image, and determining a fourth number of the identity labels which are not matched with the face recognition result of the face images in the image description information;

and under the condition that the third number is equal to 1 and the fourth number is equal to 1, labeling the unlabeled face image by using the unmatched identity.

In one or more possible implementations, the method further includes:

training the face recognition model by using the training samples in the extended sample pool, and returning to the step of obtaining the first image in the at least one detection image until the training samples in the sample pool are not increased any more, so as to obtain the trained face recognition model.

According to an aspect of the present disclosure, there is also provided an image processing apparatus including:

the first acquisition module is used for acquiring a target image to be identified;

the recognition module is used for carrying out face recognition on the target image by utilizing the trained face recognition model to obtain a face recognition result of the target image; the face recognition model is obtained by training based on training samples in an extended sample pool, and the training samples in the extended sample pool are obtained by labeling at least one face image in a first image according to a face recognition result of the first image and image description information.

In one or more possible implementations, the apparatus further includes:

and the searching module is used for searching related information of the face in the target image according to the face recognition result of the target image under the condition that the accuracy of the face recognition result of the target image is greater than a preset accuracy threshold, wherein the related information comprises one or more of identity information, whereabouts information and figure images.

In one or more possible implementations, the apparatus further includes:

the second acquisition module is also used for acquiring the unmarked first image in the at least one detection image;

the marking module is used for marking at least one face image in the first image according to the face recognition result of the first image and the image description information;

and the adding module is used for adding the labeled human face image serving as a training sample into a sample pool, and expanding the sample pool to obtain the expanded sample pool.

In one or more possible implementation manners, the second obtaining module is further configured to obtain a second image of the at least one detected image, where the second image includes a face image; the labeling module is further used for labeling the face image in the second image according to the image description information of the second image; the adding module is further configured to construct the sample pool by using the face image labeled in the second image as a training sample.

In one or more possible implementation manners, the second obtaining module is specifically configured to determine, for one of the at least one detection image, a first number of face images included in the detection image and a second number of identifiers included in image description information of the detection image; determining the detection image as a second image in a case where the first number is equal to 1 and the second number is equal to 1.

In one or more possible implementation manners, the labeling module is specifically configured to extract an identity included in the image description information of the second image; and taking the extracted identity as a label of the face image in the second image to label the face image in the second image.

In one or more possible implementations, the apparatus further includes:

the first training module is used for training the face recognition model by using the face image labeled in the second image to obtain a preliminarily trained face recognition model; and carrying out face recognition on the first image by using the preliminarily trained face recognition model to obtain a face recognition result of the first image.

In one or more possible implementation manners, the labeling module is specifically configured to label, for a face image in the first image, the face image according to a face recognition result of the face image and image description information of the first image, under a condition that it is determined that the face recognition result of the face image meets a preset labeling condition.

In one or more possible implementations, the preset labeling condition includes: the accuracy of the face recognition result is greater than a preset accuracy threshold; and the face recognition result is matched with the identity included in the image description information of the first image.

In one or more possible implementation manners, the labeling module is further configured to determine, after labeling the face images meeting the labeling condition, a third number of face images that are not labeled in the first image, and a fourth number of identifiers that are unmatched with the face recognition result of the face image in the image description information; and under the condition that the third number is equal to 1 and the fourth number is equal to 1, labeling the unlabeled face image by using the unmatched identity.

In one or more possible implementations, the apparatus further includes:

and the second training module is used for training the face recognition model by using the training samples in the extended sample pool and returning to the step of obtaining the first image in the at least one detection image until the training samples in the sample pool are not increased any more, so that the trained face recognition model is obtained.

According to an aspect of the present disclosure, there is provided an electronic device including:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to: the above-described image processing method is performed.

According to an aspect of the present disclosure, there is provided a computer-readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the above-described image processing method.

In the embodiment of the disclosure, a target image to be recognized may be obtained, and then the trained face recognition model is used to perform face recognition on the target image, so as to obtain a face recognition result of the target image. The face recognition model is obtained by training based on training samples in the extended sample pool, and the training samples in the extended sample pool are obtained by labeling at least one face image in the first image according to the face recognition result of the first image and the image description information. Therefore, the face image can be automatically marked through the image description information to continuously expand the sample pool, and then the face recognition model obtained through training of a large number of training samples in the sample pool has high accuracy, so that an accurate face recognition result can be obtained.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure.

Fig. 1 shows a flowchart of an image processing method according to an embodiment of the present disclosure.

FIG. 2 shows a flow diagram of an expanded cuvette process according to an embodiment of the disclosure.

Fig. 3 illustrates an exemplary schematic diagram in accordance with an embodiment of the present disclosure.

Fig. 4 illustrates a block diagram of an image processing apparatus according to an embodiment of the present disclosure.

Fig. 5 shows a block diagram of an example of an electronic device according to an embodiment of the present disclosure.

Fig. 6 shows a block diagram of an example of an electronic device according to an embodiment of the present disclosure.

Detailed Description

Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. In the drawings, like reference numbers can indicate functionally identical or similar elements. While the various aspects of the embodiments are presented in drawings, the drawings are not necessarily drawn to scale unless specifically indicated.

The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

The term "and/or" herein is merely an association describing an associated object, meaning that three relationships may exist, e.g., a and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B, C, and may mean including any one or more elements selected from the group consisting of A, B and C.

Furthermore, in the following detailed description, numerous specific details are set forth in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, methods, means, elements and circuits that are well known to those skilled in the art have not been described in detail so as not to obscure the present disclosure.

According to the image processing scheme provided by the embodiment of the disclosure, the target image to be recognized can be obtained, and then the trained face recognition model is used for carrying out face recognition on the target image to obtain the face recognition result of the target image. The face recognition model is obtained by training based on training samples in the extended sample pool, and the training samples in the extended sample pool are obtained by labeling at least one face image in the first image according to the face recognition result of the first image and the image description information. By the method, the face recognition model obtained by training the automatically labeled training sample can be used for carrying out face recognition on the target image, and the face recognition result with higher accuracy is obtained.

In the related art, the face recognition model is obtained by training a large number of labeled training samples. Some implementations obtain training samples by way of manual labeling, but training a face recognition model requires a large number of training samples, so that the manual labeling consumes a large amount of human resources. Some implementations are a way to train face recognition models by combining partially labeled training samples with training samples for labeling, i.e., semi-supervised training. However, the semi-supervised training mode still needs to label part of training samples manually, and training the face recognition model by using unlabelled training samples cannot ensure the accuracy of the face recognition model, and is very limited in practical application. The face recognition model can be obtained by training the training sample which is automatically marked, so that the training sample can be added under the condition of no need of manual marking, the performance of the face recognition model is improved, the face recognition model which is trained is used for carrying out face recognition on the target image, and the face recognition result with high accuracy can be obtained.

The technical scheme provided by the embodiment of the disclosure can be applied to the expansion of application scenes such as face recognition, neural network training, sample pool expansion, security monitoring and the like, and the embodiment of the disclosure does not limit the application scenes. For example, character recognition in multimedia material may be performed, such as recognition of actors in movies, television shows, and the like. For another example, people in the surveillance image may be identified, such as searching for a video segment containing the appearance of a suspect in the surveillance video.

Fig. 1 shows a flowchart of an image processing method according to an embodiment of the present disclosure. The image processing method may be performed by a terminal device, a server, or other types of electronic devices, where the terminal device may be a User Equipment (UE), a mobile device, a user terminal, a cellular phone, a cordless phone, a Personal Digital Assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, or the like. In some possible implementations, the image processing method may be implemented by a processor calling computer readable instructions stored in a memory. The following describes an image processing method according to an embodiment of the present disclosure, taking an electronic device as an execution subject.

In step S11, a target image to be recognized is acquired.

In the embodiment of the present disclosure, the electronic device may have an image capturing function, and may perform image capturing on a scene to obtain a target image to be recognized. Alternatively, the electronic device may acquire the target image to be recognized from another device, for example, the electronic device may acquire one or more target images to be recognized from a device such as an image capturing device, a monitoring device, or a network server. The target image may have a face image therein, so that the target image to be recognized may be understood as an image to be subjected to face recognition.

Step S12, carrying out face recognition on the target image by using the trained face recognition model to obtain a face recognition result of the target image; the face recognition model is obtained by training based on training samples in an extended sample pool, and the training samples in the extended sample pool are obtained by labeling at least one face image in a first image according to a face recognition result of the first image and image description information.

In the embodiment of the present disclosure, the acquired target image may be input into a trained face recognition model, and the face recognition model is used to perform face recognition on the target image, so as to obtain a face recognition result output by the face recognition model. The face recognition result may be an identification corresponding to the face in the target image, such as a name, a title, a certificate number, and the like. The face recognition model is obtained by training based on training samples in the extended sample pool, the training samples of the face recognition model are face images which are automatically labeled, a large number of face images can be automatically labeled according to the face recognition result of the first image and the image description information, so that human resources are saved, and a large number of training samples with labels are arranged in the sample pool of the face recognition model. The face recognition model obtained by training a large number of labeled training samples has high face recognition performance, and an accurate face recognition result can be obtained.

In one possible implementation manner, under the condition that the accuracy of the face recognition result of the target image is greater than a preset accuracy threshold, relevant information of the face in the target image is searched according to the face recognition result of the target image, wherein the relevant information includes one or more of identity information, whereabouts information and a person image.

In this implementation manner, the face recognition model may further output an accuracy of the face recognition result, so that after the face recognition result of the target image is obtained, the accuracy of the face recognition result of the target image may be compared with a preset accuracy threshold to determine whether the face recognition result of the target image is authentic. Under the condition that the accuracy of the face recognition result of the target image is greater than the preset accuracy threshold, the obtained face recognition result of the target image can be considered to be accurate, and further, related information of the face in the target image can be searched in the database according to the face recognition result of the target image, so that more information of the face in the target image can be obtained. Wherein the related information comprises one or more of identity information, whereabouts information and person images.

For example, when the accuracy of the face recognition result of the target image is greater than the preset accuracy threshold, the identity information of the face in the target image, for example, the information such as the identification number, work, age, address, and the like, may be obtained according to the face recognition result of the target image, so that more detailed information of the person to which the face belongs may be further obtained. For example, in a security scene, the trace information of the person to which the face belongs in the target image, such as the action route, the frequently-occurring location, the last-occurring location, and the like of the person to which the face belongs in the target image, can be acquired according to the face recognition result of the target image, so that the person to which the face belongs in the target image can be tracked, security personnel can be assisted in handling cases, and the case handling efficiency can be improved.

The face recognition model in the embodiment of the disclosure is obtained by training based on the training samples in the extended sample pool, and the extended sample pool provides a large number of training samples for the face recognition model. The process of obtaining the expanded cuvette is described below.

Fig. 2 shows a flowchart of an extended cuvette process according to an embodiment of the present disclosure, which may include the following steps:

step S21, acquiring a first image that is not labeled in at least one detection image.

In the embodiment of the present disclosure, the detection image may have a face image, and the detection image is divided according to whether the face image in the detection image has a label, where the detection image having one or more unmarked face images may be the first image. The test images may originate from advertisements, posters, movies or television shows, photo albums, etc., and typically have corresponding image description information. The image description information may be a caption of the detected image, for example, in the case where the detected image is a movie drama, the image description information may be a movie character, a drama introduction, or the like.

Step S22, labeling at least one face image in the first image according to the face recognition result of the first image and the image description information.

In the embodiment of the present disclosure, the face images in the first image may be subjected to face recognition to obtain face recognition results of one or more face images, and one face image may correspond to one face recognition result. The face recognition result may be an identification corresponding to a face in the face image, such as a name, a title, a certificate number, and the like. In some implementations, in a case that one or more face images in the first image cannot be recognized, the corresponding face recognition result may be null or an identifier indicating that the face image cannot be recognized, and the embodiment of the present disclosure does not limit a specific implementation. According to the face recognition result of the face image in the first image and the image description information of the first image, the labels of one or more face images in the first image can be determined, so that the one or more face images in the first image can be labeled. For example, the face recognition result of one face image may be matched with the image description information of the first image, and in the case that the face recognition result of the face image is matched with the image description information of the first image, the face recognition result of the face image may be used as a label of the face image to label the face image. The matching of the face recognition result and the image description information of the first image can be understood as that the image description information of the first image includes the face recognition result.

Here, the face image in the first image may be recognized using a face recognition model. The face recognition model can be obtained by training the training samples in the sample pool, and the training samples in the sample pool can be obtained by a manual labeling mode or an automatic labeling mode. The manner of automatic labeling will be described below, and will not be described herein.

And step S23, adding the labeled human face image as a training sample into a sample pool, and expanding the sample pool, wherein the training sample in the expanded sample pool is used for training a human face recognition model to obtain the trained human face recognition model.

In the embodiment of the present disclosure, the labeled face image may be added to the sample pool as a training sample, the training sample in the sample pool may be used to train the face recognition model, and the sample pool may be expanded by continuously adding the labeled face image to the sample pool as a training sample. And then, training the face recognition model by using the training samples in the extended sample pool to obtain the trained face recognition model. By continuously expanding the training samples in the sample pool, the face recognition model can have a large number of training samples, so that the accuracy of face recognition of the face recognition model after training is finished can be improved.

In the embodiment of the present disclosure, the accuracy of the face recognition model can be improved by expanding the training samples in the sample pool. Here, the initial sample pool can also be obtained by means of automatic labeling. The process of obtaining training samples in the initial pool of samples is described below in one implementation.

In a possible implementation manner, a second image in at least one detection image may be obtained, where the second image includes a face image, then the face image in the second image is labeled according to image description information of the second image, and then the labeled face image in the second image is used as a training sample to construct a sample pool.

In this implementation, a second image having one face image among the plurality of detection images may be acquired. For a second image having one face image, the label of the face image in the second image may be determined according to the image description information of the second image, for example, the name of a person in the image description information of the second image may be used as the label of the face image in the second image. And then, the determined label can be used for labeling the face image in the second image, and the labeled face image can be used as a training image and a sample pool is constructed. By the method, the face image in the second image can be labeled according to the image description information of the second image, so that a sample pool for training a face recognition model can be automatically constructed, and human resources are saved.

In one example of this implementation, for one detection image of at least one detection image, a first number of face images included in the detection image and a second number of identities included in image description information of the detection image are determined. In a case where the first number is equal to 1 and the second number is equal to 1, the detection image is determined as a second image.

In this example, the number of face images included in at least one detection image may be determined by performing face detection on the detection image, and the number of face images included in the detection image may be the first number. By performing identity detection on the image description information of the detected image, the number of identities included in the image description information of the detected image can be determined, and the number of identities included in the image description information can be a second number. In the case where the first number is equal to 1 and the second number is equal to 1, that is, in the case where the detected image includes only one face image and, at the same time, only one identification is included in the image description information of the detected image, the detected image may be regarded as the second image. In this way, the second one of the images can be detected with a quick determination.

In an example of this implementation, in a case that the face image in the second image is labeled according to the image description information of the second image, the identity included in the image description information of the second image is extracted, and then the extracted identity is used as a label of the face image in the second image to label the face image in the second image.

In this example, the second image may have a face image, and the image description information of the second image has an identity, so that the identity in the image description information may be considered as the identity of the face image in the second image, so that the identity included in the image description information of the second image may be extracted, and then the extracted identity may be determined as a tag of the face image in the second image, and the face image in the second image is labeled by using the identity. Therefore, the face image in the second image can be automatically labeled through the image description information of the second image, so that the sample pool can be constructed by using the face image labeled in the second image, and the automatic construction of the sample pool is realized.

In an example of this implementation manner, after the face image in the second image is labeled, the face image labeled in the second image may be used to train the face recognition model, so as to obtain a preliminarily trained face recognition model. And then, carrying out face recognition on the first image by using the preliminarily trained face recognition model to obtain a face recognition result of the first image.

In this example, the face image labeled in the second image may be used to train the face recognition model, that is, the training image in the preliminarily constructed sample pool may be used to train the face recognition model, for example, the training image in the preliminarily constructed sample pool may be output to the face recognition model, the face recognition model is used to perform face recognition on the training image, so as to obtain a face recognition result of the training image, then the face recognition result of the training image may be compared with the label of the training image, so as to obtain a comparison result, and the network parameter of the face recognition model is adjusted according to the comparison result, so that the face recognition result of the training image approaches to the label of the training image. The face recognition model is trained by using the face image labeled in the second image, so that the preliminarily trained face recognition model can be obtained. After the face recognition model after the preliminary training is obtained, the face recognition model after the preliminary training can be used for carrying out face recognition on the first image to obtain a face recognition result of the first image. By the method, the face recognition model can be preliminarily trained through the face image labeled in the second image, and other face images which do not enter the sample pool are deduced by utilizing the preliminarily trained face recognition model in the year, namely, the face image in the first image is recognized, so that the recognition result of the face image in the first image can be accurately obtained.

In the embodiment of the disclosure, at least one face image can be labeled according to the face recognition results of a plurality of face images and the image description information of the first image. The following describes a process of labeling at least one face image in the first image by a possible implementation manner.

In a possible implementation manner, for a face image in the plurality of face images in the first image, the face image may be labeled according to the face recognition result of the face image and the image description information of the first image, under the condition that the face recognition result of the face image meets a preset labeling condition.

In this implementation manner, the face detection may be performed on the plurality of face images in the first image by using the preliminarily trained face recognition model, so as to obtain face detection results of the plurality of face images. For a face image of a plurality of face images in the first image, whether the face recognition result of the face image meets a preset labeling condition can be judged according to the face recognition result of the face image and the image description information of the first image. And under the condition that the face recognition result of the face image meets the preset labeling condition, labeling the face image.

Here, the preset labeling conditions may include: the accuracy of the face recognition result is greater than a preset accuracy threshold, and the face recognition result is matched with the identity included in the image description information of the first image. Under the condition of judging whether the face identification result of the face image meets the preset labeling condition, the accuracy of the face identification result can be compared with a preset accuracy threshold value, and whether the accuracy of the face identification result is greater than the accuracy threshold value is judged. Meanwhile, the face recognition result can be compared with the identity included in the image description information of the first image, and whether the face recognition result is matched with the identity included in the image description information of the first image or not can be judged. When the accuracy of the face identification result of the face image is greater than a preset accuracy threshold and the face identification result is matched with the identity included in the image description information of the first image, it can be determined that the face identification result of the face image meets a preset labeling condition, the identity included in the image description information of the first image can be further determined as a label of the face image, and the face image can be labeled by using the determined label. The preset accuracy threshold may be set according to an actual application scenario, for example, the accuracy threshold may be set to 0.85, 0.9, and the like. The face recognition model can output the accuracy of the face recognition result while outputting the face recognition result, and the accuracy can represent the accuracy of the corresponding face recognition result. Generally, the higher the accuracy of the face recognition result is, the more accurate the face recognition result can be considered, and the lower the accuracy of the face recognition result is, the less accurate the face recognition result can be considered, so that whether the face recognition result is credible can be judged by setting an accuracy threshold value for the accuracy of the face recognition result.

The face image meeting the labeling condition in the first image can be labeled by determining whether the face recognition result of the face image in the first image meets the labeling condition, so that after the face image meeting the labeling condition in the first image is labeled, the labeled face image can be added into a sample pool as a training image, and the sample pool is expanded.

In an example of this implementation, after the face images satisfying the annotation condition are annotated, a third number of face images that are not annotated in the first image may be determined, and a fourth number of ids that do not match face recognition results of the plurality of face images in the image description information may be determined. And in the case that the third number is equal to 1 and the fourth number is equal to 1, labeling the unlabeled face images by using the unmatched identity identifications.

In this example, after the face images satisfying the annotation condition in the first image are annotated, the number of the face images that are not annotated in the first image may be determined, and the number may be a third number. Accordingly, the number of the identifiers in the image description information of the first image that do not match the face recognition results of the plurality of face images in the first image may be determined, and the number may be a fourth number. Then, in the case that the third number is equal to 1 and the fourth number is equal to 1, that is, in the case that one facial image is not labeled in the first image and one identification in the image description information of the first image is not matched with the facial recognition results of the plurality of facial images in the first image, it may be considered that the identification in the image description information not matched with the facial recognition results of the plurality of facial images in the first image may be used as a label of the facial image not labeled in the first image, and the labeled facial image may be labeled. In this way, the face image which is not labeled in the first image can be labeled.

In a possible implementation manner, the training samples in the extended sample pool are used for training the face recognition model, and the step of obtaining the first image including the plurality of face images in the at least one detection image is returned until the number of the training samples in the sample pool is not increased any more.

In this implementation, the training samples in the extended sample pool can be used to train the face recognition model, so as to improve the face recognition rate of the face recognition model, and meanwhile, the above step S11 can be returned, the first images including a plurality of face images in the detection images are obtained again, the training samples in the sample pool continue to be extended until the training samples in the sample pool do not increase any more, so that the face images in the detection images can be automatically labeled, the sample pool is continuously extended, and the face recognition model obtained by the training samples in the extended sample pool can have higher face recognition accuracy.

Fig. 3 shows a block diagram of an example of image processing according to an embodiment of the present disclosure. The image processing scheme provided by the present example may include the steps of:

in step S301, a second image of the detection images is acquired.

Here, it may be determined that there is only one face image in the detection image and there is only one identity (e.g., person name) in the image description information of the second image by performing face detection on the detection image and identity detection on the image description information of the detection image.

Step S302, labeling the face image in the second image according to the description information of the second image, and constructing an initial sample pool by using the labeled face image in the second image.

Here, for only one face image detected in the second image and only one identity identifier mentioned in the image description information, it may be considered that the face image in the second image matches the identity identifier in the image description information, that is, the tag corresponding to the face image in the second picture may be the identity identifier mentioned only in the image description information. Further, an initial sample pool can be constructed by using the labeled face image in the second image.

Step S303, training a face recognition model by using an initial sample pool, and performing face recognition on an unmarked first image in the detection image by using the preliminarily trained face recognition model to obtain a face recognition result.

Step S304, judging whether the accuracy of the face recognition result of the face image in the first image is greater than an accuracy threshold value, and whether the face recognition result of the face image is matched with the identity in the image description information of the first image.

Step S305, under the condition that the accuracy of the face recognition result of the face image in the first image is greater than the accuracy threshold and the face recognition result of the face image is matched with the identity in the image description information of the first image, taking the identity matched with the face recognition result as a label of the face image in the first image, and adding the face image labeled in the first image into a sample pool.

Step S306, obtaining only one unmarked face image and obtaining a first image of which only one identity label in the image description information is not matched with the face image, taking the identity label of the unmarked face image in the image description information as a label of the unmarked face image in the first image, and adding the marked face image into a sample pool.

Here, after performing step S306, it may return to step S303 to further expand the training samples in the sample pool, train the face recognition model on the basis of the expanded sample pool, until the sample pool is no longer expanded. Therefore, the face recognition model can be trained by using the identity in the image description and the face in the image, so that a face recognition model with higher accuracy (equivalent to the accuracy of a model obtained by using a full-annotation training sample) can be trained without additional manual annotation.

It is understood that the above-mentioned method embodiments of the present disclosure can be combined with each other to form a combined embodiment without departing from the logic of the principle, which is limited by the space, and the detailed description of the present disclosure is omitted.

In addition, the present disclosure also provides an apparatus, an electronic device, a computer-readable storage medium, and a program, which can be used to implement any image processing method provided by the present disclosure, and the corresponding technical solutions and descriptions and corresponding descriptions in the method sections are not repeated.

It will be understood by those skilled in the art that in the method of the present invention, the order of writing the steps does not imply a strict order of execution and any limitations on the implementation, and the specific order of execution of the steps should be determined by their function and possible inherent logic.

Fig. 4 shows a block diagram of an image processing apparatus according to an embodiment of the present disclosure, which includes, as shown in fig. 4:

a first obtaining module 41, configured to obtain a target image to be identified;

the recognition module 42 is configured to perform face recognition on the target image by using the trained face recognition model to obtain a face recognition result of the target image; the face recognition model is obtained by training based on training samples in an extended sample pool, and the training samples in the extended sample pool are obtained by labeling at least one face image in a first image according to a face recognition result of the first image and image description information.

In one or more possible implementations, the apparatus further includes:

the accuracy of the face recognition result is greater than a preset accuracy threshold; and the number of the first and second electrodes,

the face recognition result is matched with the identity included in the image description information of the first image.

In one or more possible implementations, the apparatus further includes:

In some embodiments, functions of or modules included in the apparatus provided in the embodiments of the present disclosure may be used to execute the method described in the above method embodiments, and specific implementation thereof may refer to the description of the above method embodiments, and for brevity, will not be described again here.

Fig. 5 is a block diagram illustrating an apparatus 800 for observed behavior control in accordance with an example embodiment. For example, the apparatus 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like.

Referring to fig. 5, the apparatus 800 may include one or more of the following components: processing component 802, memory 804, power component 806, multimedia component 808, audio component 810, input/output (I/O) interface 812, sensor component 814, and communication component 816.

The processing component 802 generally controls overall operation of the device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing components 802 may include one or more processors 820 to execute instructions to perform all or a portion of the steps of the methods described above. Further, the processing component 802 can include one or more modules that facilitate interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.

The memory 804 is configured to store various types of data to support operations at the apparatus 800. Examples of such data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 804 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.

Power components 806 provide power to the various components of device 800. The power components 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the apparatus 800.

The multimedia component 808 includes a screen that provides an output interface between the device 800 and a user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front facing camera and/or a rear facing camera. The front camera and/or the rear camera may receive external multimedia data when the device 800 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the apparatus 800 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.

The sensor assembly 814 includes one or more sensors for providing various aspects of state assessment for the device 800. For example, the sensor assembly 814 may detect the open/closed status of the device 800, the relative positioning of components, such as a display and keypad of the device 800, the sensor assembly 814 may also detect a change in the position of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, the orientation or acceleration/deceleration of the device 800, and a change in the temperature of the device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The communication component 816 is configured to facilitate communications between the apparatus 800 and other devices in a wired or wireless manner. The device 800 may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, the apparatus 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.

In an exemplary embodiment, a non-transitory computer-readable storage medium, such as the memory 804, is also provided that includes computer program instructions executable by the processor 820 of the device 800 to perform the above-described methods.

An embodiment of the present disclosure further provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein the processor is configured as the above method.

The electronic device may be provided as a terminal, server, or other form of device.

Fig. 6 is a block diagram illustrating an electronic device 1900 according to an example embodiment. For example, the electronic device 1900 may be provided as a server. Referring to fig. 6, electronic device 1900 includes a processing component 1922 further including one or more processors and memory resources, represented by memory 1932, for storing instructions, e.g., applications, executable by processing component 1922. The application programs stored in memory 1932 may include one or more modules that each correspond to a set of instructions. Further, the processing component 1922 is configured to execute instructions to perform the above-described method.

The electronic device 1900 may also include a power component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input/output (I/O) interface 1958. The electronic device 1900 may operate based on an operating system stored in memory 1932, such as Windows Server, Mac OSXTM, UnixTM, LinuxTM, FreeBSDTM, or the like.

In an exemplary embodiment, a non-transitory computer readable storage medium, such as the memory 1932, is also provided that includes computer program instructions executable by the processing component 1922 of the electronic device 1900 to perform the above-described methods.

The present disclosure may be systems, methods, and/or computer program products. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied thereon for causing a processor to implement various aspects of the present disclosure.

The computer readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: 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), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as punch cards or in-groove projection structures having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or electrical signals transmitted through electrical wires.

The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device via a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.

The computer program instructions for carrying out operations of the present disclosure may be assembler instructions, Instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, the electronic circuitry that can execute the computer-readable program instructions implements aspects of the present disclosure by utilizing the state information of the computer-readable program instructions to personalize the electronic circuitry, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA).

Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.

These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, 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/acts specified in the flowchart and/or block diagram block or blocks. These computer-readable program instructions may also be stored in a computer-readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer-readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.

The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Having described embodiments of the present disclosure, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or technical improvements to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. An image processing method, comprising:

acquiring a target image to be identified;

2. The method of claim 1, further comprising:

3. The method of claim 1 or 2, wherein obtaining an expanded sample pool comprises:

acquiring an unmarked first image in at least one detection image;

4. The method of claim 3, wherein prior to obtaining the first image that is not labeled in the at least one detection image, further comprising:

5. The method of claim 4, wherein said acquiring a second image of the at least one inspection image comprises:

6. The method according to claim 4 or 5, wherein the labeling the face image in the second image according to the image description information of the second image comprises:

7. The method according to any one of claims 3 to 6, further comprising:

8. The method according to any one of claims 3 to 7, wherein the labeling at least one face image in the first image according to the face recognition result of the first image and image description information comprises:

9. The method according to claim 8, wherein the preset labeling condition comprises:

10. The method according to claim 8 or 9, wherein after labeling the face image satisfying the labeling condition, the method further comprises:

11. The method according to any one of claims 3 to 10, further comprising:

12. An image processing apparatus characterized by comprising:

13. The apparatus of claim 12, further comprising:

14. The apparatus of claim 12 or 13, further comprising:

15. The apparatus of claim 14,

the second acquisition module is further configured to acquire a second image of the at least one detection image, where the second image includes a face image;

the labeling module is further used for labeling the face image in the second image according to the image description information of the second image;

the adding module is further configured to construct the sample pool by using the face image labeled in the second image as a training sample.

16. The apparatus according to claim 15, wherein the second obtaining module is specifically configured to determine, for one of the at least one detection image, a first number of face images included in the detection image and a second number of identifiers included in image description information of the detection image; determining the detection image as a second image in a case where the first number is equal to 1 and the second number is equal to 1.

17. The apparatus according to claim 15 or 16, wherein the labeling module is specifically configured to extract an identity included in the image description information of the second image; and taking the extracted identity as a label of the face image in the second image to label the face image in the second image.

18. The apparatus of any one of claims 14 to 17, further comprising:

19. An electronic device, comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to invoke the memory-stored instructions to perform the method of any of claims 1 to 11.

20. A computer readable storage medium having computer program instructions stored thereon, which when executed by a processor implement the method of any one of claims 1 to 11.