CN113688794A - Identity recognition method and device, electronic equipment and computer readable storage medium - Google Patents
Identity recognition method and device, electronic equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN113688794A CN113688794A CN202111123316.6A CN202111123316A CN113688794A CN 113688794 A CN113688794 A CN 113688794A CN 202111123316 A CN202111123316 A CN 202111123316A CN 113688794 A CN113688794 A CN 113688794A
- Authority
- CN
- China
- Prior art keywords
- person
- information
- human body
- face
- recognition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
Abstract
The invention discloses an identity recognition method, an identity recognition device, electronic equipment and a computer readable storage medium, and belongs to the technical field of artificial intelligence, wherein the identity recognition method comprises the following steps: acquiring a first acquisition image; carrying out personnel detection on the first collected image, and extracting the human body characteristics of each detected personnel; respectively tracking the position of each person based on the extracted human body features; according to the image quality of the face image of the first person, determining to perform face recognition or human body feature recognition on the tracked first person, and determining first identity information of the first person according to a recognition result. The invention can improve the success rate of identity recognition.
Description
Technical Field
The invention belongs to the technical field of artificial intelligence, and particularly relates to an identity recognition method, an identity recognition device, electronic equipment and a computer-readable storage medium.
Background
In the related technology, a camera is used for collecting personnel images in a visual field range, and faces in the personnel images are identified, so that the identified face information is matched with a face database stored in advance to determine the identity information of the personnel.
However, in practical applications, the quality of the face collected by the camera is not stable, and in many cases, only the side face and the target back shadow can be collected, so that face recognition cannot be performed, and the success rate of identity recognition is not high.
From the above, the identity recognition method in the related art has the problem of low success rate of identity recognition.
Disclosure of Invention
The invention aims to provide an identity recognition method, an identity recognition device, electronic equipment and a computer readable storage medium, which can improve the success rate of identity recognition.
In order to solve the technical problem, the invention is realized as follows:
in a first aspect, the present invention provides an identity recognition method, including:
acquiring a first acquisition image;
carrying out personnel detection on the first collected image, and extracting the human body characteristics of each detected personnel;
respectively tracking the position of each person based on the extracted human body features;
according to the image quality of the face image of the first person, determining to perform face recognition or human body feature recognition on the tracked first person, and determining first identity information of the first person according to a recognition result.
In a second aspect, the present invention further provides an identity recognition apparatus, including:
the first acquisition module is used for acquiring a first acquisition image;
the detection module is used for carrying out personnel detection on the first collected image and extracting the human body characteristics of each detected personnel;
the characteristic extraction module is used for respectively tracking the position of each person based on the extracted human body characteristics;
and the identity recognition module is used for determining face recognition or human body feature recognition of the tracked first person according to the image quality of the face image of the first person and determining first identity information of the first person according to a recognition result.
In a third aspect, the present invention also provides an electronic device comprising a processor, a memory, and a program or instructions stored on the memory and executable on the processor, which when executed by the processor implements the steps of the method according to the first aspect.
In a fourth aspect, the invention also provides a computer readable storage medium on which a program or instructions are stored, which program or instructions, when executed by a processor, implement the steps of the method according to the first aspect.
In the embodiment of the invention, a first collected image is obtained; carrying out personnel detection on the first collected image, and extracting the human body characteristics of each detected personnel; respectively tracking the position of each person based on the extracted human body features; according to the image quality of the face image of the first person, determining to perform face recognition or human body feature recognition on the tracked first person, and determining first identity information of the first person according to a recognition result. Like this, can carry out feature extraction to the pedestrian in the first collection image to the pedestrian that each human body feature corresponds is tracked, like this, under the condition that detects the people face in the first collection image, can confirm the human body feature that this pedestrian corresponds according to the position of this people face, and discern this pedestrian's identity information based on face identification or human body feature identification, realized making up through human body feature identification: when the image quality of the face image is poor, the face image is low in integrity, shielded and poor in definition, face recognition fails, the problem that identity information recognition cannot be completed is caused, and therefore the success rate of identity recognition can be improved.
Drawings
Fig. 1 is a flowchart of an identity recognition method according to an embodiment of the present invention;
FIG. 2 is a flow chart of another method for identifying an identity provided by an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an identification apparatus according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of another electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms first, second and the like in the description and in the claims of the present application are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that embodiments of the application may be practiced in sequences other than those illustrated or described herein, and that the terms "first," "second," and the like are generally used herein in a generic sense and do not limit the number of terms, e.g., the first term can be one or more than one. In addition, "and/or" in the specification and claims means at least one of connected objects, a character "/" generally means that a preceding and succeeding related objects are in an "or" relationship.
The identity recognition method provided by the embodiment of the invention can be applied to various scenes, such as: for convenience of description, in the following embodiments, an application scenario in which the identity recognition method provided by the embodiment of the present invention is applied to a robot to perform identity recognition on a pedestrian is taken as an example to illustrate, that is, in the following embodiments, an execution main body for executing the identity recognition method provided by the embodiment of the present invention may be a robot. Of course, the following embodiments do not limit the identity recognition method provided by the embodiments of the present invention to be only applicable to an application scenario in which a robot performs identity recognition on a pedestrian.
The visitor reception method, the robot, and the computer-readable storage medium provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings by specific embodiments and application scenarios thereof.
Referring to fig. 1, an identity recognition method provided in an embodiment of the present invention is applied to a greeting robot, and as shown in fig. 1, the identity recognition method may include the following steps:
In a specific implementation, the first captured image may be a video image, or at least two photographs captured within a preset time period (e.g., 1 second, 2 seconds, etc.), and the like. In addition, the first collected image may be shot by a monitoring camera or other devices and transmitted to a robot executing the identity recognition method provided by the embodiment of the present invention; or an image captured by the image capturing device on the greeting robot, which is not limited in detail herein.
And 102, carrying out personnel detection on the first collected image, and extracting the human body characteristics of each detected personnel.
In a specific implementation, the above-mentioned person detection can be understood as: and detecting the first collected image to determine a pixel area where each person in the first collected image is located, so that people and objects, people and animals, and people in the first collected image are distinguished. For example: inputting the first collected image into a human body target detection algorithm, so as to detect human body actions, human body forms and the like in the first collected image through the human body target detection algorithm, thereby distinguishing people and objects, people and animals, and people in the first collected image, and further enabling the human body target detection algorithm to output the positions of the detected people in the first collected image. The human target detection algorithm may be an algorithm that can be thought of by those skilled in the art, for example: the human body detection algorithm and the human body detection and tracking algorithm are not limited herein.
In addition, the above-mentioned extracting the detected human body feature of each person may be understood as performing human body feature recognition on a pixel region of each person in the first captured image, for example: and (3) taking the human body image as input, and calculating through a neural network to obtain a high-dimensional feature vector. For example: the human body features are extracted by using a pedestrian re-IDentification (ReID) algorithm, but in practical applications, the human body features may also be extracted by using other feature extraction algorithms, which is not limited herein.
And 103, respectively tracking the position of each person based on the extracted human body features.
In a specific implementation, after a certain human body feature is extracted, the human body feature of each pedestrian in the visual field range can be periodically extracted, and the extracted human body feature is compared with the previously extracted human body feature to determine whether the pedestrian is a target person identified previously, so that even if the position of the person changes in the video image or the multiple continuously collected photos, the actual position of the person corresponding to the human body feature in the video image or the multiple continuously collected photos can be determined during the walking process of the person.
And step 104, determining to perform face recognition or human body feature recognition on the tracked first person according to the image quality of the face image of the first person, and determining first identity information of the first person according to a recognition result.
Alternatively, the first person may be a person located at the target location, for example: and the personnel are positioned in the range of 0.5 to 2 meters in front of the welcome robot.
In this embodiment, when it is determined that identity recognition needs to be performed on a first person located at a target position in a first captured image, a first human body feature of the first person located at the target position may be obtained based on position tracking, and a first face image of the first person located at the target position in the first captured image may be captured; and then according to the image quality of the face image, judging whether the first identity information of the first person is determined by matching the first human body characteristics with the human body characteristic information in the human body characteristic database or determining the first identity information of the first person by matching the face identification information of the first face image with the face information in the face information database.
Of course, in practical applications, the first person may also include each person in the first captured image, and the following embodiment takes the first person as an example of the person located at the target position in the first captured image, and the first person is not limited herein.
In implementation, the face detection may be performed on the first captured image to determine a face area of each person in the first captured image, and the screenshot may be performed to capture a face image (i.e., a small face image) of each person in the first captured image, and the image quality of each small face image is determined, and in a subsequent face recognition process, the subsequent face recognition process may be a face recognition process on the small face image.
In practical application, in view of factors such as a shooting angle, a shooting distance, a middle blocking object, and the like, a situation that a face in a first acquired image is in an improper orientation, such as a side face and a back shadow, or a situation that an image quality of a face image is poor, such as an unclear face, a blocked face, a dark brightness, an improper size of a face, and the like, may be caused, at this time, a human feature corresponding to a person to be recognized may be determined according to a position of the face image without performing face recognition on the face image, and identity information of the person to be recognized may be determined based on human feature recognition (e.g., ReID recognition technology).
Correspondingly, under the condition that the image quality of the face image is good, the face image can be subjected to face recognition, and the identity information of the person to be recognized is determined according to the face recognition result.
In other words, the identity information of the person to be identified can be identified preferentially by adopting a face identification mode, but the premise is that the image quality of the face image must reach the standard; if the image quality of the face image does not meet the standard, the identity information of the person to be identified can be identified by adopting a human body feature identification mode.
In specific implementation, in order to achieve the purpose of judging whether the image quality of the face image reaches the standard, the definition and the integrity of the face image can be detected, so that the image quality of the face image reaches the standard under the condition that the definition of the face image is greater than or equal to the preset definition and the integrity of the face image is greater than or equal to the preset proportion; otherwise, determining that the image quality of the face image does not reach the standard.
As an optional implementation manner, the determining, according to the image quality of the face image of the first person, to perform face recognition or human body feature recognition on the tracked first person, and determining, according to a recognition result, first identity information of the first person includes:
determining to perform human body feature recognition on a first person under the condition that the image quality of a face image of the first person is smaller than a preset quality;
matching the human body characteristic result with human body characteristic information in a preset human body characteristic information base;
and determining first identity information of the first person according to the human body characteristic information matching result.
In a specific implementation, after determining that the image quality of the face image does not meet the standard, the following human body feature recognition process may be adopted to determine the first identity information of the first person:
and matching the extracted human body features with a preset human body feature information base, and if target feature information matched with the human body features exists in the preset human body feature information base, determining first identity information of the first person according to identity information corresponding to the target feature information.
Of course, if the preset human body feature information base does not have the target feature information matched with the human body features, the first person is determined to be a stranger. For example: assuming that the database of the greeting robot stores the feature information of the person a in advance, when the robot recognizes that the image quality of a certain face image meets the standard, the human feature information corresponding to the position of the face image (the human feature information of the first person extracted in step 102) may be matched with the feature information stored in the database in advance, and if the human feature information of the first person is successfully matched with the feature information of the person a, it may be determined that the first person is the person a.
Of course, in implementation, the first identity information of the first person may also include other information, such as: the first identity information may further include: whether the first person is a person who has been served by the welcome robot within a preset time period (past 1 hour and the like) is judged, and the type of the first identity information is not particularly limited.
It should be noted that, when it is determined that the image quality of the face image does not meet the criterion, in view of the fact that the position of the face image in the first captured image is fixed, at this time, according to the target position of the face image in the first captured image and the previous position tracking process based on the human body features, the human body features corresponding to the target position may be determined as the human body features of the first person, so that the human body features are matched with the pre-stored feature information, and confusion of the human body features of different persons when the human body features of a plurality of persons are extracted from the first captured image may be avoided.
As an optional implementation manner, the determining, according to image quality of a face image of a first person, face recognition or human body feature recognition on the tracked first person, and determining, according to a recognition result, first identity information of the first person includes:
determining to perform face recognition on a first person under the condition that the image quality of a face image of the first person is greater than or equal to a preset quality;
matching the face recognition result with face information in a preset face information base;
and determining first identity information of the first person according to a face information matching result, and associating the human body characteristics of the first person with the face image of the first person.
In a specific implementation, after determining that the image quality of the face image meets the standard, the following face recognition process may be adopted to determine the first identity information of the first person:
performing face recognition on the face image to obtain face recognition information, matching the face recognition information with a preset face information base, and if target face information matched with the face recognition information exists in the preset face information base, determining first identity information of the first person according to identity information corresponding to the target face information; and if target face information matched with the face identification information does not exist in a preset face information base, determining that the first person is a stranger. For example: if the database of the robot stores the face information of the person a in advance, when the robot recognizes that the image quality of a certain face image meets the standard, the face image can be subjected to face recognition so as to match the recognized face recognition information with the face information stored in the database in advance, and if the face recognition information is successfully matched with the face information of the person a, the first person can be determined to be the person a.
It should be noted that the above process of determining identity information through face recognition and human body feature recognition is similar to the process of face recognition and human body feature recognition in the related art, and will not be described herein too much.
In addition, the above-mentioned association of the human body feature of the first person with the face image of the first person can be understood as: based on the first identity information of the first person corresponding to the face image of the first person, the human body characteristics of the first person can be associated with the first identity information of the first person by associating the human body characteristics of the first person with the face image of the first person. In this way, the characteristics of the first person in the database may be supplemented or updated based on the currently identified human characteristics of the first person.
In an optional implementation manner, in a case that the face identification information of the first person does not exist in the preset face information base, when the first identity information of the first person is determined based on the face matching result, the face image of the first person and the body feature of the first person may be stored in association, so that in a subsequent identity identification process, the identity information of the first person may be identified based on the face image and the body feature of the first person that are stored in association.
In another optional implementation manner, when the face identification information of the first person is determined based on the face matching result in the preset face information base, the body feature of the first person is associated with the pre-stored face identification information of the first person, so that the body feature of the first person is updated or supplemented, and in a subsequent identification process, the first person can be identified based on the updated or supplemented body feature.
Optionally, the determining, according to a face information matching result, first identity information of the first person, and associating the human body feature of the first person with the face image of the first person includes:
under the condition that the target face information of the first person is successfully matched with first face information in a person information base, determining first identity information of the first person based on the first face information, wherein the person information base comprises face information and human body characteristic information;
and associating the first face information with first human body characteristic information of the first person, or updating human body characteristic information associated with the first face information in advance into the first human body characteristic information.
In practical applications, the preset body feature information base may not store the feature information of the first person currently, or the feature information of the first person pre-stored in the preset body feature information base may not match with the actual body feature information of the first person, such as the current hairstyle and clothes of the first person. In the embodiment, the currently recognized human body feature information of the first person is used for supplementing or updating the feature information of the first person in the preset human body feature information base, so that the accuracy and the real-time performance of the feature information stored in the preset human body feature information base can be improved.
In this embodiment, under the condition that the first identity information of the first person can be determined through face recognition, the extracted human body characteristics of the first person are used to supplement or update the human body characteristics in the preset human body characteristic information base, so that the accuracy and the real-time performance of the characteristic information stored in the preset human body characteristic information base can be improved, and the accuracy of the subsequent human body characteristic recognition process is further improved.
Further, after the associating the first face information with the first human body feature information of the target person or updating the human body feature information associated with the first face information in advance to the first human body feature information, the method further includes:
when the first human body feature information is detected from a second collected image, identity information of a person corresponding to the first human body feature information in the second collected image is determined based on the first human face information associated with the first human body feature information.
In a specific implementation, the second captured image is a video image or a plurality of photographs captured after a preset time (e.g., 5 minutes, 30 minutes, etc.) of the first captured image, and the detection of the first human body feature information from the second captured image may be understood as: the robot serves the first person again after preset time (for example, 5 minutes, 30 minutes, and the like), and at this time, the identity information of the first person is determined based on the first person information stored in association with the first person feature information after the previous identity recognition without performing face recognition on the person, so that the calculation power and the calculation time in the face recognition process can be reduced, and the efficiency of the identity recognition process can be improved.
Specifically, in the previous identity recognition process, the first identity information of the first person is determined through face recognition, at this time, the association relationship between the first identity information and the first face information is inevitably stored in advance, and after the identity recognition process is completed, the first human body feature information of the first person and the first face information of the first person are also stored in association. Thus, when the first person is identified again, the identity information of the first person can be identified according to the first person feature extracted in the person feature extraction process shown in step 102, that is, the first person feature is associated with the first person information, and the first person information is associated with the first identity information, so that the identity information corresponding to the first person feature is determined to be the first identity information, that is, the identity information of the first person is determined to be the first identity information.
In the embodiment, when the identity of the same person is recognized for multiple times, the identity information of the person to be recognized can be determined based on the incidence relation between the human body features and the human face information which are pre-stored in the first identity recognition process and the currently extracted human body features of the person to be recognized, and the processes of a series of human face detection, human face image quality judgment, human face recognition and the like do not need to be performed on the person, so that the identity recognition process can be simplified, and the identity recognition efficiency is improved.
Certainly, in practical application, when the identity of the same person is recognized for multiple times, the person detection and the position tracking can be performed based on the human body characteristics updated in the previous identity recognition process, then, the face detection and the face image quality judgment can be performed again, and under the condition that the face image quality reaches the standard, the identity information of the person is determined by matching the recognized face characteristic information with the face information stored in advance; and under the condition that the quality of the face image does not meet the standard, determining the identity information of the person through matching between the updated human body features and the extracted human body features in the last identity recognition process.
As an optional implementation, after the determining the first identity information of the first person, the method further comprises:
acquiring the destination information of the first person;
associating the destination information with the first identity information.
In a specific implementation, the above-mentioned obtaining of the heading information of the first person may be to obtain the reverse information of the departure of the first person in a similar manner to the location tracking in step 103.
Of course, in a specific application, the robot may provide guidance, leading and other services to the first person after determining the first identity information of the first person to guide or lead the first person to smoothly reach the destination to which the first person intends to reach, and in this case, the going information of the first person may include the first person to smoothly reach the destination to which the first person intends to reach.
In this embodiment, after the destination information of the first person is obtained, the destination information is associated with the first identity information, so that a more complete service can be provided based on the destination information of the first person in a subsequent service process for the first person or other persons. For example: inquiring whether the first person leads the first person to move to the destination of the previous leading service when the first person is served next time; or when the colleague of the first person is served, the colleague is informed of the going direction of the first person, so that the colleague can find the colleague of the first person in time.
As an optional implementation, after the associating the destination information with the first identity information, the method further comprises:
and under the condition that second identity information of a second person is identified, outputting prompt information, wherein the prompt information comprises: the first identity information and the destination information of the first person, wherein the second identity information and the first identity information have a preset relationship.
In a specific implementation, the outputting the first identity information and the destination information of the first person may be understood as: and informing the second person of the destination information of the first person by means of interface display, voice output and the like.
In addition, the second identity information and the first identity information have a preset relationship, and it can be understood that: and judging that the second identity information and the first identity information are related identity information through a preset rule, for example: indicating that the first person and the second person are employees of the same company, employees of the same department, friends, people who need to go to the same destination and know each other, and so on.
For example: the identity information may include information of a department where the corresponding person is located, and when the robot detects that a plurality of persons in the same department move in the same direction within a preset time interval (for example, 5 minutes), prompt information is sent to the currently identified person in the department, such as: your buddy XXX goes to the XXX direction.
For another example: after the identity information of the person is identified by the identity identification method provided by the embodiment of the application, the department corresponding to the identity information can be inquired in a pre-stored person registration information table, so that when a plurality of persons in the same department walk in the same direction within a preset time interval (for example: 5 minutes), prompt information is sent to the currently identified person in the department, such as: your buddy XXX goes to the XXX direction.
In the embodiment, the destination information of other related personnel can be prompted to the current personnel, so that the intelligent effect of the robot service can be improved.
As an optional implementation manner, the first identity information includes first indication information, where the first indication information is used to indicate that the first person is: strangers, preset personnel for the first service or preset personnel who have been served;
after determining the first identity information of the first person, the method further comprises:
and sending out the calling information corresponding to the first indication information to the first person.
In a specific implementation, the strangers can understand that: and the preset human body characteristic information base and the preset human face information base do not have personnel corresponding to the human body characteristic information base and the human face information base.
In this case, if the first person is a stranger, the call information may be call information for the stranger, for example: "you are good, welcome".
The aforementioned pre-set personnel for the primary service can be understood as: the human body feature information and/or the human face information corresponding to the preset human body feature information base and/or the preset human face information base exist in the preset human body feature information base and/or the preset human face information base, and the person is a person who serves for the first time within a specified time period (for example, 1 day).
In this case, if the first person is a preset person for the first service, the call information may be call information for the preset person for the first service, for example: "hello, XXX (e.g., the name of the first person, the name of the company or employee code, etc.)".
The above-mentioned pre-set persons that have been served can be understood as: the human body feature information and/or the human face information corresponding to the preset human body feature information base and/or the preset human face information base exist in the preset human body feature information base and/or the preset human face information base, and at least 1 service is provided for the personnel within a specified time period (for example, 1 day).
In this case, if the first person is a preset person who has been served, the call information may be call information for the preset person who has been served, for example: "what you are good, can serve you again".
In the embodiment, the robot can send different call information to people with different identities or different historical service backgrounds, and the artificial intelligence of the robot service can be improved.
In the embodiment of the invention, a first collected image is obtained; carrying out personnel detection on the first collected image, and extracting the human body characteristics of each detected personnel; respectively tracking the position of each person based on the extracted human body features; according to the image quality of the face image of the first person, determining to perform face recognition or human body feature recognition on the tracked first person, and determining first identity information of the first person according to a recognition result. Like this, can carry out feature extraction to the pedestrian in the first collection image to the position of the pedestrian that each human body feature corresponds is tracked, like this, under the condition that detects the people face in the first collection image, can confirm the human body feature that this pedestrian corresponds according to the position of this people face, and discern the pedestrian's of this position identity information based on face identification or human body feature recognition, realized making up through human body feature recognition: when the image quality of the face image is poor, the face image is low in integrity, shielded and poor in definition, face recognition fails, the problem that identity information recognition cannot be completed is caused, and therefore the success rate of identity recognition can be improved.
For convenience of description, in the following embodiments, taking the example that the robot executes the flow of the identity recognition method shown in fig. 2, the identity recognition method provided by the embodiment of the present invention is described as an example, and as shown in fig. 2, the identity recognition method may include the following steps:
This step differs from step 101 in the embodiment of the method shown in fig. 1 in that the captured image in this step is an image captured by an image capturing device on the robot.
This step is similar to that in the embodiment of the method shown in fig. 1: and carrying out personnel detection on the first collected image, wherein the personnel detection has the same meaning and effect, and the details are not repeated.
And step 203, extracting human body features.
This step is similar to that in the embodiment of the method shown in fig. 1: the detected human body features of each person are extracted, have the same meaning and effect, and are not described in detail herein.
And step 204, ReID tracking.
This step is similar to that in the embodiment of the method shown in fig. 1: and respectively tracking the position of each person based on the extracted human body features, wherein the person has the same meaning and effect, and the person is not described any more herein.
And step 205, detecting the human face.
This step represents acquiring a face region in the acquired image, thereby capturing a face thumbnail.
And step 206, judging the quality of the face image.
In this step, if the judgment result is that the image quality of the face image is poor, that is, the image quality does not reach the standard, step 207 is executed; otherwise, step 208 is performed.
In addition, this step is similar to that in the embodiment of the method shown in fig. 1: according to the image quality of the face image of the first person, the face recognition or the human body feature recognition is carried out on the first person, the meaning and the effect are the same, and the description is omitted.
And step 207, comparing the human body characteristics.
This step can be understood as in the embodiment of the method shown in fig. 1: and performing human body feature identification on the first person to determine the first identity information process of the first person, which is not described herein again.
And step 208, comparing the face features.
This step can be understood as in the embodiment of the method shown in fig. 1: and performing face recognition on the first person to determine the first identity information of the first person, which is not described herein again.
In this step, if the determination result is yes (i.e. there is face information matching with the face recognition information in the preset face information base), step 210 is executed; otherwise, step 211 is executed.
And step 210, updating the human body characteristics.
And step 211, updating the stranger information to a database.
The identity recognition method provided by the embodiment of the invention can extract the features of the pedestrians in the collected image and track the position of the pedestrian corresponding to each human body feature, so that under the condition that the face is detected from the collected image, the human body feature corresponding to the pedestrian can be determined according to the position of the face, and the identity information of the pedestrian at the position is recognized based on the face recognition or the human body feature recognition, thereby realizing the compensation through the human body feature recognition: when the image quality of the face image is poor, the face image is low in integrity, shielded and poor in definition, face recognition fails, the problem that identity information recognition cannot be completed is caused, and therefore the success rate of identity recognition can be improved.
Referring to fig. 3, an identification apparatus 300 according to an embodiment of the present invention may include:
a first obtaining module 301, configured to obtain a first captured image;
a detection module 302, configured to perform person detection on the first captured image, and extract a human body feature of each detected person;
a feature extraction module 303, configured to perform location tracking on each person based on the extracted human body features;
and the identity recognition module 304 is configured to determine, according to image quality of a face image of a first person, to perform face recognition or human body feature recognition on the tracked first person, and determine first identity information of the first person according to a recognition result.
Optionally, the identity module 304 includes:
the characteristic identification unit is used for carrying out human body characteristic identification on a first person under the condition that the image quality of a face image of the first person is smaller than the preset quality;
the characteristic matching unit is used for matching the human body characteristic result with human body characteristic information in a preset human body characteristic information base;
and the first determining unit is used for determining the first identity information of the first person according to the human body characteristic information matching result.
Optionally, the identity module 304 includes:
the face recognition unit is used for carrying out face recognition on a first person under the condition that the image quality of a face image of the first person is greater than or equal to the preset quality;
the face matching unit is used for matching the face recognition result with face information in a preset face information base;
the second determining unit is used for determining the first identity information of the first person according to the face information matching result;
and the association unit is used for associating the human body characteristics of the first person with the face image of the first person.
Optionally, the associating unit includes:
a determining subunit, configured to determine first identity information of the first person based on first face information when target face information of the first person is successfully matched with the first face information in a person information base, where the person information base includes face information and body feature information;
and the data processing unit is used for associating the first face information with first human body characteristic information of the first person, or updating human body characteristic information associated with the first face information in advance into the first human body characteristic information.
Optionally, the identification apparatus 300 further includes:
the second determination module is used for determining the identity information of the person corresponding to the first human body feature information in the second collected image based on the first human face information associated with the first human body feature information under the condition that the first human body feature information is detected from the second collected image.
Optionally, the identification apparatus 300 further includes:
the second acquisition module is used for acquiring the destination information of the first person;
and the association module associates the destination information with the first identity information.
Optionally, the identification apparatus 300 further includes:
the first output module is used for outputting prompt information under the condition that second identity information of a second person is identified, and the prompt information comprises: the first identity information and the destination information of the first person, wherein the second identity information and the first identity information have a preset relationship.
Optionally, the first identity information includes first indication information, where the first indication information is used to indicate that the first person is: strangers, preset personnel for the first service or preset personnel who have been served;
the identification device 300 further includes:
and the second output module is used for sending out the call information corresponding to the first indication information to the first person.
The identity recognition apparatus 300 provided in the embodiment of the present application can implement each process implemented by the method embodiment shown in fig. 1 or fig. 2, and can obtain the same beneficial effects, and for avoiding repetition, the details are not repeated here.
Optionally, as shown in fig. 4, an electronic device 400 is further provided in this embodiment of the present application, and includes a processor 401, a memory 402, and a program or an instruction stored in the memory 402 and executable on the processor 401, where the program or the instruction is executed by the processor 401 to implement each process in the method embodiment shown in fig. 1 or fig. 2, and can achieve the same technical effect, and in order to avoid repetition, it is not described here again.
In a specific implementation, the electronic device may be a robot, or it may also be other electronic devices with an identity recognition function, and is not limited in particular herein.
An embodiment of the present application further provides a computer-readable storage medium, where a program or an instruction is stored on the computer-readable storage medium, and when the program or the instruction is executed by a processor, the process of the method embodiment shown in fig. 1 or fig. 2 is implemented, and the same technical effect can be achieved, and in order to avoid repetition, details are not repeated here.
The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and so on.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. Further, it should be noted that the scope of the methods and apparatus of the embodiments of the present application is not limited to performing the functions in the order illustrated or discussed, but may include performing the functions in a substantially simultaneous manner or in a reverse order based on the functions involved, e.g., the methods described may be performed in an order different than that described, and various steps may be added, omitted, or combined. In addition, features described with reference to certain examples may be combined in other examples.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present application.
While the present embodiments have been described with reference to the accompanying drawings, it is to be understood that the invention is not limited to the precise embodiments described above, which are meant to be illustrative and not restrictive, and that various changes may be made therein by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (11)
1. An identity recognition method, the method comprising:
acquiring a first acquisition image;
carrying out personnel detection on the first collected image, and extracting the human body characteristics of each detected personnel;
respectively tracking the position of each person based on the extracted human body features;
according to the image quality of the face image of the first person, determining to perform face recognition or human body feature recognition on the tracked first person, and determining first identity information of the first person according to a recognition result.
2. The method of claim 1, wherein the determining of face recognition or human body feature recognition of the tracked first person according to the image quality of the face image of the first person and determining the first identity information of the first person according to the recognition result comprises:
determining to perform human body feature recognition on a first person under the condition that the image quality of a face image of the first person is smaller than a preset quality;
matching the human body characteristic result with human body characteristic information in a preset human body characteristic information base;
and determining first identity information of the first person according to the human body characteristic information matching result.
3. The method of claim 1, wherein the determining of face recognition or human body feature recognition of the tracked first person according to the image quality of the face image of the first person and determining the first identity information of the first person according to the recognition result comprises:
determining to perform face recognition on a first person under the condition that the image quality of a face image of the first person is greater than or equal to a preset quality;
matching the face recognition result with face information in a preset face information base;
and determining first identity information of the first person according to a face information matching result, and associating the human body characteristics of the first person with the face image of the first person.
4. The method of claim 3, wherein the determining first identity information of the first person according to the face information matching result and associating the human body feature of the first person with the face image of the first person comprises:
under the condition that the target face information of the first person is successfully matched with first face information in a person information base, determining first identity information of the first person based on the first face information, wherein the person information base comprises face information and human body characteristic information;
and associating the first face information with first human body characteristic information of the first person, or updating human body characteristic information associated with the first face information in advance into the first human body characteristic information.
5. The method according to claim 4, wherein after the human characteristic information associated with the first human characteristic information of the target person or previously associated with the first human characteristic information is updated as the first human characteristic information, the method further comprises:
when the first human body feature information is detected from a second collected image, identity information of a person corresponding to the first human body feature information in the second collected image is determined based on the first human face information associated with the first human body feature information.
6. The method of claim 1, wherein after the determining the first identity information of the first person, the method further comprises:
acquiring the destination information of the first person;
associating the destination information with the first identity information.
7. The method of claim 6, wherein after said associating said destination information with said first identity information, said method further comprises:
and under the condition that second identity information of a second person is identified, outputting prompt information, wherein the prompt information comprises: the first identity information and the destination information of the first person, wherein the second identity information and the first identity information have a preset relationship.
8. The method of claim 1, wherein the first identity information comprises first indication information indicating that the first person is: strangers, preset personnel for the first service or preset personnel who have been served;
after determining the first identity information of the first person, the method further comprises:
and sending out the calling information corresponding to the first indication information to the first person.
9. An identification device, comprising:
the first acquisition module is used for acquiring a first acquisition image;
the detection module is used for carrying out personnel detection on the first collected image and extracting the human body characteristics of each detected personnel;
the characteristic extraction module is used for respectively tracking the position of each person based on the extracted human body characteristics;
and the identity recognition module is used for determining face recognition or human body feature recognition of the tracked first person according to the image quality of the face image of the first person and determining first identity information of the first person according to a recognition result.
10. An electronic device comprising a processor, a memory and a computer program stored on the memory and executable on the processor, the computer program, when executed by the processor, implementing the steps in the identification method according to any one of claims 1 to 8.
11. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method of identification according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111123316.6A CN113688794A (en) | 2021-09-24 | 2021-09-24 | Identity recognition method and device, electronic equipment and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111123316.6A CN113688794A (en) | 2021-09-24 | 2021-09-24 | Identity recognition method and device, electronic equipment and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113688794A true CN113688794A (en) | 2021-11-23 |
Family
ID=78587170
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111123316.6A Pending CN113688794A (en) | 2021-09-24 | 2021-09-24 | Identity recognition method and device, electronic equipment and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113688794A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114743155A (en) * | 2022-03-10 | 2022-07-12 | 慧之安信息技术股份有限公司 | Mall pedestrian recognition method based on combination of face recognition and pedestrian re-recognition |
| CN115083004A (en) * | 2022-08-23 | 2022-09-20 | 浙江大华技术股份有限公司 | Identity recognition method and device and computer readable storage medium |
| CN116205952A (en) * | 2023-04-19 | 2023-06-02 | 齐鲁空天信息研究院 | Face recognition and tracking method and device, electronic equipment and storage medium |
-
2021
- 2021-09-24 CN CN202111123316.6A patent/CN113688794A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114743155A (en) * | 2022-03-10 | 2022-07-12 | 慧之安信息技术股份有限公司 | Mall pedestrian recognition method based on combination of face recognition and pedestrian re-recognition |
| CN114743155B (en) * | 2022-03-10 | 2022-11-15 | 慧之安信息技术股份有限公司 | Mall pedestrian recognition method based on combination of face recognition and pedestrian re-recognition |
| CN115083004A (en) * | 2022-08-23 | 2022-09-20 | 浙江大华技术股份有限公司 | Identity recognition method and device and computer readable storage medium |
| CN115083004B (en) * | 2022-08-23 | 2022-11-22 | 浙江大华技术股份有限公司 | Identity recognition method and device and computer readable storage medium |
| CN116205952A (en) * | 2023-04-19 | 2023-06-02 | 齐鲁空天信息研究院 | Face recognition and tracking method and device, electronic equipment and storage medium |
| CN116205952B (en) * | 2023-04-19 | 2023-08-04 | 齐鲁空天信息研究院 | Face recognition and tracking method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113688794A (en) | Identity recognition method and device, electronic equipment and computer readable storage medium | |
| CN106991395B (en) | Information processing method and device and electronic equipment | |
| US10949657B2 (en) | Person's behavior monitoring device and person's behavior monitoring system | |
| CN108885698B (en) | Face recognition method and device and server | |
| CN113038018B (en) | Method and device for assisting user in shooting vehicle video | |
| WO2019071664A1 (en) | Human face recognition method and apparatus combined with depth information, and storage medium | |
| CN110516623B (en) | Face recognition method and device and electronic equipment | |
| JP2021108149A (en) | Person detection system | |
| CN112785769B (en) | Queuing and calling management method, device, equipment and storage medium based on face recognition | |
| CN112949396B (en) | Self-adaptive method for searching for people on the same trip in scenic spot | |
| US20240127626A1 (en) | Information processing apparatus, information processing method, and storage medium | |
| CN110599129A (en) | Campus attendance checking method, device, identification terminal and system based on image tracking | |
| CN110674680A (en) | Living body identification method, living body identification device and storage medium | |
| CN113677409A (en) | Treasure hunting game guiding technology | |
| CN110941992B (en) | Smile expression detection method and device, computer equipment and storage medium | |
| CN110580708B (en) | Rapid movement detection method and device and electronic equipment | |
| CN112528261A (en) | Method and device for identifying user identity of SIM card | |
| CN111813995A (en) | Pedestrian article extraction behavior detection method and system based on space-time relationship | |
| CN112070185A (en) | Re-ID-based non-contact fever person tracking system and tracking method thereof | |
| CN115457449B (en) | Early warning system based on AI video analysis and monitoring security protection | |
| CN109665387B (en) | Intelligent elevator boarding method and device, computer equipment and storage medium | |
| US10824868B2 (en) | Image processing for determining relationships between tracked objects | |
| CN111667664B (en) | Lost child prevention support system | |
| CN110991316B (en) | Method for automatically acquiring shape and identity information applied to open environment | |
| US11494942B2 (en) | Information processing device, information processing method, program, recording medium, and camera system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |