CN111898515A - Method and device for identifying identity of pedestrian - Google Patents
Method and device for identifying identity of pedestrian Download PDFInfo
- Publication number
- CN111898515A CN111898515A CN202010725980.7A CN202010725980A CN111898515A CN 111898515 A CN111898515 A CN 111898515A CN 202010725980 A CN202010725980 A CN 202010725980A CN 111898515 A CN111898515 A CN 111898515A
- Authority
- CN
- China
- Prior art keywords
- whole
- real
- body image
- shot
- feature vector
- 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
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000000875 corresponding effect Effects 0.000 claims abstract description 40
- 230000002596 correlated effect Effects 0.000 claims abstract description 4
- 239000013598 vector Substances 0.000 claims description 243
- 238000012795 verification Methods 0.000 claims description 21
- 230000001815 facial effect Effects 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 13
- 230000004927 fusion Effects 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012790 confirmation Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005034 decoration Methods 0.000 description 2
- 238000007499 fusion processing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
- G06F16/532—Query formulation, e.g. graphical querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Library & Information Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Collating Specific Patterns (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method for identifying the identity of a pedestrian, which comprises the following steps: acquiring a first real shooting whole body image and a corresponding first real shooting face image; determining identity information of a person to which the first real shot face image belongs based on base database data, wherein the base database data comprises a plurality of base face images and corresponding identity information; the first real shot whole-body image is correlated with the identity information of the person to which the first real shot face image belongs, and the first real shot whole-body image is stored in a bottom library as a bottom library whole-body image; acquiring a second real-shot whole-body image of the pedestrian to be identified; and determining the identity information of the pedestrian to be identified based on the first real-shot whole-body image and the second real-shot whole-body image. The invention also discloses a device for identifying the identity of the pedestrian. The invention can realize the correlation between the photographed whole-body image and the identity information of the base database and can be used as the basis for the next authentication, and the identity information is authenticated by photographing the whole-body image.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method and a device for identifying the identity of a pedestrian.
Background
At present, when the identity of a pedestrian is identified, the identity identification result is generally obtained through the face. In a non-matching scene, the included angle between the face of the pedestrian and the camera is large, or the acquisition definition of the camera is insufficient, so that a high-definition image of the face is difficult to distinguish, and the face comparison success rate is low. Meanwhile, the bottom library only establishes a connection between the face image and the identity information, so that the identity confirmation rate is low.
Disclosure of Invention
In order to solve the above problems, an object of the present invention is to provide a method and an apparatus for identifying a pedestrian, which can implement the association between a photographed whole-body image and identity information of an underlying database and serve as a basis for the next authentication, and authenticate the identity information by photographing the whole-body image.
The invention provides a method for identifying the identity of a pedestrian, which comprises the following steps:
acquiring a first real shooting whole body image and a corresponding first real shooting face image;
determining identity information of a person to which the first real shot face image belongs based on base database data, wherein the base database data comprises a plurality of base face images and corresponding identity information;
the first real shot whole-body image is correlated with the identity information of the person to which the first real shot face image belongs, and the first real shot whole-body image is stored in a bottom library as a bottom library whole-body image;
acquiring a second real-shot whole-body image of the pedestrian to be identified;
and determining the identity information of the pedestrian to be identified based on the first real-shot whole-body image and the second real-shot whole-body image.
As a further improvement of the present invention, determining the identity information of the person to which the first live-shot face image belongs based on the base database data includes:
extracting the characteristic vector of the first real shot facial image and the characteristic vector of the bottom base facial image as a first characteristic vector and a second characteristic vector;
calculating a first face similarity of the first feature vector and the second feature vector;
and determining the identity information of the person to which the first real shot face image belongs based on the first face similarity.
As a further improvement of the present invention, determining the identity information of the person to which the first real shot face image belongs based on the first face similarity includes:
comparing the first face similarity with a face similarity setting threshold;
and if the first face similarity is larger than or equal to the face similarity set threshold, determining the identity information of the person to which the first real shot face image belongs.
As a further improvement of the present invention, determining the identity information of the pedestrian to be identified based on the first real-shot whole-body image and the second real-shot whole-body image includes:
extracting the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image as a third feature vector and a fourth feature vector;
calculating the whole-body similarity of the third feature vector and the fourth feature vector;
and determining the identity information of the pedestrian to be identified based on the whole-body similarity.
As a further improvement of the present invention, the method further comprises:
acquiring a second real shooting face image corresponding to a second real shooting whole body image of the pedestrian to be identified;
extracting a feature vector of the second real shot face image and a feature vector of the first real shot face image as a fifth feature vector and a sixth feature vector;
calculating a second face similarity of the fifth feature vector and the sixth feature vector;
and displaying the identity information of the person to be recognized, the whole-body similarity and the second face similarity.
As a further improvement of the present invention, determining the identity information of the pedestrian to be identified based on the whole-body similarity includes:
comparing the whole-body similarity with a whole-body similarity setting threshold;
and if the whole-body similarity is larger than or equal to the whole-body similarity set threshold, determining the identity information of the pedestrian to be identified according to the identity information of the person to which the first real shot whole-body image belongs.
As a further improvement of the present invention, the method further comprises:
determining whether to update the bottom library whole-body images in the bottom library based on the whole-body similarity.
As a further improvement of the present invention, determining whether to update the bottom library whole-body images in the bottom library based on the whole-body similarity includes:
when the whole-body similarity is greater than or equal to the whole-body similarity set threshold, judging whether the whole-body similarity is greater than or equal to a first updating threshold;
if the whole-body similarity is larger than or equal to the first updating threshold, determining to update the whole-body image in the bottom library;
and taking the second real shot whole-body image as a new bottom base whole-body image to be stored in the bottom base, or fusing the characteristic vector of the first real shot whole-body image and the characteristic vector of the second real shot whole-body image to generate a new bottom base whole-body image, and storing the fused characteristic vector and the new bottom base whole-body image in the bottom base.
As a further improvement of the present invention, storing the second real-shot whole-body image as a new bottom-library whole-body image in the bottom library, or fusing the feature vector of the first real-shot whole-body image and the feature vector of the second real-shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image in the bottom library, includes:
if the whole-body similarity is greater than or equal to a first updating threshold and less than or equal to a second updating threshold, determining to fuse the feature vector of the first real shot whole-body image and the feature vector of the second real shot whole-body image to generate a new whole-body image of the bottom base, and storing the fused feature vector and the new whole-body image of the bottom base into the bottom base;
and if the whole-body similarity is larger than a second updating threshold, determining that the second real-shot whole-body image is stored in the bottom base as a new bottom base whole-body image.
As a further improvement of the present invention, fusing the feature vector of the first real-shot whole-body image and the feature vector of the second real-shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image in the bottom library, includes:
respectively determining weights corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image according to the whole-body similarity;
and according to the weight corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image, performing weighted fusion on the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image into the bottom library.
The invention also relates to a device for identifying the identity of a pedestrian, which comprises:
the first real shooting module is used for acquiring a first real shooting whole-body image and a corresponding first real shooting face image;
the first real shooting verification module is used for determining the identity information of a person to which the first real shooting face image belongs based on bottom database data, wherein the bottom database data comprises a plurality of bottom database face images and corresponding identity information;
the first real shooting association module is used for associating the first real shooting whole-body image with the identity information of the person to which the first real shooting face image belongs and storing the first real shooting whole-body image as a bottom library whole-body image into a bottom library;
the second real shooting module is used for acquiring a second real shooting whole-body image of the pedestrian to be identified;
and the second real shooting verification module is used for determining the identity information of the pedestrian to be identified based on the first real shooting whole-body image and the second real shooting whole-body image.
As a further improvement of the present invention, the first live beat verification module is configured to:
extracting the characteristic vector of the first real shot facial image and the characteristic vector of the bottom base facial image as a first characteristic vector and a second characteristic vector;
calculating a first face similarity of the first feature vector and the second feature vector;
and determining the identity information of the person to which the first real shot face image belongs based on the first face similarity.
As a further improvement of the present invention, determining the identity information of the person to which the first real shot face image belongs based on the first face similarity includes:
comparing the first face similarity with a face similarity setting threshold;
and if the first face similarity is larger than or equal to the face similarity set threshold, determining the identity information of the person to which the first real shot face image belongs.
As a further improvement of the present invention, the second real-time shooting verification module is configured to:
extracting the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image as a third feature vector and a fourth feature vector;
calculating the whole-body similarity of the third feature vector and the fourth feature vector;
and determining the identity information of the pedestrian to be identified based on the whole-body similarity.
As a further improvement of the present invention, the apparatus further comprises:
the second real shooting face image acquisition module is used for acquiring a second real shooting face image corresponding to a second real shooting whole-body image of the pedestrian to be identified;
a feature vector extraction module, configured to extract a feature vector of the second real shot face image and a feature vector of the first real shot face image as a fifth feature vector and a sixth feature vector;
the similarity calculation module is used for calculating second face similarity of the fifth feature vector and the sixth feature vector;
and the display module is used for displaying the identity information of the person to be identified, the whole-body similarity and the second face similarity.
As a further improvement of the present invention, determining the identity information of the pedestrian to be identified based on the whole-body similarity includes:
comparing the whole-body similarity with a whole-body similarity setting threshold;
and if the whole-body similarity is larger than or equal to the whole-body similarity set threshold, determining the identity information of the pedestrian to be identified according to the identity information of the person to which the first real shot whole-body image belongs.
As a further improvement of the present invention, the apparatus further comprises:
and the updating module is used for determining whether to update the whole-body image of the bottom library in the bottom library according to the whole-body similarity.
As a further improvement of the present invention, the update module is configured to:
when the whole-body similarity is greater than or equal to the whole-body similarity set threshold, judging whether the whole-body similarity is greater than or equal to a first updating threshold;
if the whole-body similarity is larger than or equal to the first updating threshold, determining to update the whole-body image in the bottom library;
and taking the second real shot whole-body image as a new bottom base whole-body image to be stored in the bottom base, or fusing the characteristic vector of the first real shot whole-body image and the characteristic vector of the second real shot whole-body image to generate a new bottom base whole-body image, and storing the fused characteristic vector and the new bottom base whole-body image in the bottom base.
As a further improvement of the present invention, storing the second real-shot whole-body image as a new bottom-library whole-body image in the bottom library, or fusing the feature vector of the first real-shot whole-body image and the feature vector of the second real-shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image in the bottom library, includes:
if the whole-body similarity is greater than or equal to a first updating threshold and less than or equal to a second updating threshold, determining to fuse the feature vector of the first real shot whole-body image and the feature vector of the second real shot whole-body image to generate a new whole-body image of the bottom base, and storing the fused feature vector and the new whole-body image of the bottom base into the bottom base;
and if the whole-body similarity is larger than a second updating threshold, determining that the second real-shot whole-body image is stored in the bottom base as a new bottom base whole-body image.
As a further improvement of the present invention, fusing the feature vector of the first real-shot whole-body image and the feature vector of the second real-shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image in the bottom library, includes:
respectively determining weights corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image according to the whole-body similarity;
and according to the weight corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image, performing weighted fusion on the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image into the bottom library.
The invention also provides an electronic device comprising a memory and a processor, wherein the memory is configured to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method.
The invention also provides a computer-readable storage medium, on which a computer program is stored, characterized in that the computer program is executed by a processor to implement the method.
The invention has the beneficial effects that: through the real shooting of the face image and the real shooting of the whole body image, the correlation between the real shooting whole body image and the identity information of the base can be realized and used as the basis of the next identity verification, and the identity information is verified through the real shooting of the whole body image.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It is to be understood that the drawings in the following description are merely exemplary of the disclosure, and that other drawings may be derived from those drawings by one of ordinary skill in the art without undue inventive faculty.
Fig. 1 is a schematic flow chart of a method for identifying a pedestrian according to an exemplary embodiment of the present disclosure;
fig. 2 is a schematic diagram illustrating the association between a base library image, a first live-shot whole-body image, and a second live-shot whole-body image according to an exemplary embodiment of the present disclosure;
fig. 3 is a schematic flow chart illustrating an implementation of a method for identifying a pedestrian according to an exemplary embodiment of the present disclosure;
fig. 4 is a schematic flow chart of an implementation of a method for identifying a pedestrian according to another exemplary embodiment of the present disclosure;
fig. 5 is a schematic flowchart of updating an underlying whole-body image according to an exemplary embodiment of the disclosure.
Detailed Description
The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
It should be noted that, if directional indications (such as up, down, left, right, front, and back … …) are involved in the disclosed embodiment, the directional indications are only used to explain the relative position relationship between the components, the motion situation, and the like in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indications are changed accordingly.
In addition, in the description of the present disclosure, the terms used are for illustrative purposes only and are not intended to limit the scope of the present disclosure. The terms "comprises" and/or "comprising" are used to specify the presence of stated elements, steps, operations, and/or components, but do not preclude the presence or addition of one or more other elements, steps, operations, and/or components. The terms "first," "second," and the like may be used to describe various elements, not necessarily order, and not necessarily limit the elements. In addition, in the description of the present disclosure, "a plurality" means two or more unless otherwise specified. These terms are only used to distinguish one element from another. These and/or other aspects will become apparent to those of ordinary skill in the art in view of the following drawings, and the description of the embodiments of the disclosure will be more readily understood by those of ordinary skill in the art. The drawings are only for purposes of illustrating the described embodiments of the disclosure. One skilled in the art will readily recognize from the following description that alternative embodiments of the structures and methods illustrated in the present disclosure may be employed without departing from the principles described in the present disclosure.
As shown in fig. 1, a method for identifying a pedestrian identity according to an embodiment of the present disclosure includes:
acquiring a first real shooting whole body image and a corresponding first real shooting face image;
determining identity information of a person to which the first real shot face image belongs based on the base database data, wherein the base database data comprises a plurality of base face images and corresponding identity information;
the first real shot whole body image is correlated with the identity information of the person to which the first real shot face image belongs, and the first real shot whole body image is stored in a bottom library as a bottom library whole body image;
acquiring a second real-shot whole-body image of the pedestrian to be identified;
and determining the identity information of the pedestrian to be identified based on the first shot whole-body image and the second shot whole-body image.
The method adds the real-shot whole-body image to perform identity verification, as shown in fig. 2, obtains a first real-shot face image and a corresponding first real-shot whole-body image of a clear shot (for example, a first real shot), associates the first real-shot face image with the bottom library face image, and can obtain the identity information of a person to which the first real-shot face image belongs, and the first real-shot face image and the first real-shot whole-body image are associated correspondingly. When the second real-shot whole-body image is shot next time, the second real-shot whole-body image can be compared with the whole-body image in the bottom library to acquire the identity information of the person to which the second real-shot whole-body image belongs, and the association between the second real-shot whole-body image and the first real-shot whole-body image is realized. According to the method, the association between the shot whole-body image and the identity information of the base can be realized and used as the basis for the next authentication through shooting the face image and the shot whole-body image, so that the identity information can be acquired through shooting the whole-body image.
In an alternative embodiment, determining the identity information of the person to which the first live-shot facial image belongs based on the underlying database data includes:
extracting a feature vector of the first real shot face image and a feature vector of the bottom base face image as a first feature vector and a second feature vector;
calculating a first face similarity of the first feature vector and the second feature vector;
and determining the identity information of the person to which the first real shot face image belongs based on the first face similarity.
In an alternative embodiment, determining the identity information of the person to which the first live-shot facial image belongs based on the first face similarity includes:
comparing the first face similarity with a face similarity setting threshold;
and if the first face similarity is larger than or equal to a face similarity set threshold, determining the identity information of the person to which the first real shot face image belongs.
The base library has pre-stored a plurality of base library face images and corresponding identity information as base library data. When the identity of the first real shot image is verified, the first face similarity is calculated one by one with the bottom library face images in the bottom library for comparison, and when the first face similarity reaches or exceeds a face similarity set threshold, the first real shot face image and the bottom library face images in the bottom library have high similarity, namely the identity information of the person to which the first real shot face image belongs can be determined. The face similarity threshold is preset, and the face similarity threshold is not specifically limited by the disclosure. In the face similarity calculation, for example, the face similarity may be obtained by calculating cosine distances of feature vectors of the first live-shot face image and the bottom library face image, and the calculation of the first face similarity is not particularly limited in the present disclosure.
In an alternative embodiment, determining the identity information of the pedestrian to be identified based on the first photographed whole-body image and the second photographed whole-body image includes:
extracting a feature vector of the second real-shot whole-body image and a feature vector of the first real-shot whole-body image as a third feature vector and a fourth feature vector;
calculating the whole-body similarity of the third feature vector and the fourth feature vector;
and determining the identity information of the pedestrian to be identified based on the similarity of the whole body.
In another optional embodiment, the determining the identity information of the pedestrian to be identified based on the whole-body similarity includes:
comparing the whole-body similarity with a whole-body similarity setting threshold;
and if the whole-body similarity is greater than or equal to the whole-body similarity set threshold, determining the identity information of the pedestrian to be identified according to the identity information of the person to which the first real shot whole-body image belongs.
As shown in fig. 3, when the photographed image of the pedestrian to be identified is obtained, for example, a second photographed whole-body image may be obtained, and the second photographed whole-body image and the first photographed whole-body image are compared by calculating the whole-body similarity, so as to verify the identity information. When the whole-body similarity reaches or exceeds the whole-body similarity setting threshold, it can be understood that the second real-shot whole-body image has high similarity with the first real-shot whole-body image, that is, the identity information of the person to which the second real-shot whole-body image belongs, that is, the identity information of the pedestrian to be identified, can be determined according to the identity information of the person to which the first real-shot whole-body image belongs. The whole-body similarity threshold is preset, and is not specifically limited by the disclosure. When the whole-body similarity is calculated, for example, the calculation may be performed by calculating a cosine distance of a feature vector of the second real-shot whole-body image and the first real-shot whole-body image. After the identity information and the whole-body similarity of the person to which the second real-shot whole-body image belongs are obtained, the identity information and the whole-body similarity can be displayed to the pedestrian to be identified.
In an optional embodiment, the method further comprises:
acquiring a second real shooting face image corresponding to a second real shooting whole body image of the pedestrian to be identified;
extracting a feature vector of the second real shot face image and a feature vector of the first real shot face image as a fifth feature vector and a sixth feature vector;
calculating a second face similarity of the fifth feature vector and the sixth feature vector;
and displaying the identity information, the whole-body similarity and the second face similarity of the person to be identified.
As shown in fig. 4, when the photographed image of the pedestrian to be identified is obtained, for example, a second photographed whole-body image may be obtained, and the second photographed whole-body image and the first photographed whole-body image are compared by calculating the whole-body similarity, so as to verify the identity information. When the whole-body similarity reaches or exceeds the whole-body similarity setting threshold, it can be understood that the second real-shot whole-body image has high similarity with the first real-shot whole-body image, that is, the identity information of the person to which the second real-shot whole-body image belongs, that is, the identity information of the pedestrian to be identified, can be determined according to the identity information of the person to which the first real-shot whole-body image belongs. The whole-body similarity threshold is preset, and is not specifically limited by the disclosure. When the whole-body similarity is calculated, for example, the calculation may be performed by calculating a cosine distance of a feature vector of the second real-shot whole-body image and the first real-shot whole-body image. After the second real shot whole-body image is obtained, a second real shot face image corresponding to the second real shot whole-body image can be obtained, the second face similarity of the second real shot face image and the first real shot face image is calculated, and the second face similarity is used for being displayed to the pedestrian to be identified. The second face similarity may be obtained, for example, by calculating a cosine distance of a feature vector of the second real-shot face image and the feature vector of the first real-shot face image, and the calculation of the second face similarity is not specifically limited in this disclosure. After the identity information, the whole-body similarity and the second face similarity of the person to which the second real-shot whole-body image belongs are obtained, the identity information, the whole-body similarity and the second face similarity can be displayed to the pedestrian to be identified.
In an optional embodiment, the method further comprises: and determining whether to update the whole-body images in the bottom library based on the whole-body similarity.
In an alternative embodiment, determining whether to update the whole-body image in the fundus library based on the whole-body similarity includes:
when the whole-body similarity is greater than or equal to the whole-body similarity set threshold, judging whether the whole-body similarity is greater than or equal to a first updating threshold;
if the whole-body similarity is larger than or equal to the first updating threshold, determining to update the whole-body image in the bottom library;
and taking the second real-shot whole-body image as a new bottom base whole-body image to be stored in a bottom base, or fusing the feature vector of the first real-shot whole-body image and the feature vector of the second real-shot whole-body image to generate a new bottom base whole-body image, and storing the fused feature vector and the new bottom base whole-body image in the bottom base.
In various different scenes, such as clothing change, head decoration change and the like of the pedestrian to be identified, when the whole-body image of the pedestrian to be identified is photographed, the situation that the whole-body information changes greatly may occur, so that the passing rate of the identity authentication through the whole-body photographed image is low. As shown in fig. 5, the whole-body image can be updated as a new bottom-library whole-body image according to the situation, and when the real-shot whole-body image is shot next time, the whole-body image can be compared with the new bottom-library whole-body image, so that the identity confirmation rate is improved.
In an optional implementation manner, the storing the second actually shot whole-body image as a new bottom-library whole-body image in the bottom library, or fusing the feature vector of the first actually shot whole-body image and the feature vector of the second actually shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image in the bottom library includes:
if the whole-body similarity is greater than or equal to the first updating threshold and less than or equal to the second updating threshold, determining to fuse the feature vector of the first real shot whole-body image and the feature vector of the second real shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image into a bottom library;
and if the whole-body similarity is larger than the second updating threshold, determining that the second real-shot whole-body image is stored in the bottom library as a new bottom library whole-body image.
The selection of the feature vector for replacing the first real-shot whole-body image or for weighted fusion of the first real-shot whole-body image can be made according to the size of the whole-body similarity value. For example, a first update threshold and a second update threshold may be set, and when the whole-body similarity value is within the two update threshold intervals, it may be understood that the whole-body similarity has not reached the degree of complete replacement, the feature vector of the first real-shot whole-body image may be selected to be subjected to fusion processing, so as to weaken (reduce) the association between the first real-shot whole-body image and the identity of the corpus. When the whole-body similarity value is greater than the second update threshold, it can be understood that the second real-shot whole-body image has very high similarity, and the first real-shot whole-body image can be replaced to remove the association between the first real-shot whole-body image and the identity of the bottom library. The first update threshold and the second update threshold are preset, and the two update thresholds are not specifically limited in the present disclosure.
In an alternative embodiment, fusing the feature vector of the first live-shot whole-body image and the feature vector of the second live-shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image into the bottom library, including:
determining the weight corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image according to the whole-body similarity;
and according to the weight corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image, performing weighted fusion on the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image to generate a new whole-body image of the bottom library, and storing the fused feature vector and the new whole-body image of the bottom library into the bottom library.
When the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image are subjected to weighted fusion, weights corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image can be respectively determined according to the whole-body similarity, the sum of the weights of the second real-shot whole-body image and the first real-shot whole-body image is 1, and then the first real-shot whole-body image and the bottom library identity are subjected to weighted fusion according to the weights, so that the association between the first real-shot whole-body image and the bottom.
The device of pedestrian's identification of this disclosed embodiment, the device includes:
the first real shooting module is used for acquiring a first real shooting whole-body image and a corresponding first real shooting face image;
the first real shooting verification module is used for determining the identity information of a person to which the first real shooting face image belongs based on the bottom database data, wherein the bottom database data comprises a plurality of bottom database face images and corresponding identity information;
the first real shooting association module is used for associating the first real shooting whole-body image with the identity information of the person to which the first real shooting face image belongs and storing the first real shooting whole-body image as a bottom library whole-body image into a bottom library;
the second real shooting module is used for acquiring a second real shooting whole-body image of the pedestrian to be identified;
and the second real shooting verification module is used for determining the identity information of the pedestrian to be identified based on the first real shooting whole-body image and the second real shooting whole-body image.
In an optional implementation manner, the first live beat verification module is configured to:
extracting a feature vector of the first real shot face image and a feature vector of the bottom base face image as a first feature vector and a second feature vector;
calculating a first face similarity of the first feature vector and the second feature vector;
and determining the identity information of the person to which the first real shot face image belongs based on the first face similarity.
The device disclosed by the disclosure adds a real-shot whole-body image to perform identity verification, as shown in fig. 2, obtains a first real-shot face image and a corresponding first real-shot whole-body image which are clearly shot (for example, a first real shot), associates the first real-shot face image with a bottom library face image, and can obtain identity information of a person to which the first real-shot face image belongs, and the first real-shot face image and the first real-shot whole-body image are correspondingly associated. When the second real-shot whole-body image is shot next time, the second real-shot whole-body image can be compared with the whole-body image in the bottom library to acquire the identity information of the person to which the second real-shot whole-body image belongs, and the association between the second real-shot whole-body image and the first real-shot whole-body image is realized. According to the method, the association between the shot whole-body image and the identity information of the base can be realized and used as the basis for the next authentication through shooting the face image and the shot whole-body image, so that the identity information can be acquired through shooting the whole-body image.
In an optional implementation manner, the determining, by the first live-shot verification module, the identity information of the person to which the first live-shot facial image belongs based on the first face similarity includes:
comparing the first face similarity with a face similarity setting threshold;
and if the first face similarity is larger than or equal to a face similarity set threshold, determining the identity information of the person to which the first real shot face image belongs.
The base library has pre-stored a plurality of base library face images and corresponding identity information as base library data. When the first real shooting verification module performs identity verification on the first real shooting image, the first face similarity is calculated one by one with the bottom library face image in the bottom library for comparison, and when the first face similarity reaches or exceeds a face similarity set threshold, the first real shooting face image and the bottom library face image in the bottom library have high similarity, namely identity information of a person to which the first real shooting face image belongs can be determined. The face similarity threshold is preset, and the face similarity threshold is not specifically limited by the disclosure. In the face similarity calculation, for example, the face similarity may be obtained by calculating cosine distances of feature vectors of the first live-shot face image and the bottom library face image, and the calculation of the first face similarity is not particularly limited in the present disclosure.
In an optional implementation manner, the second real-time beat verification module is configured to:
extracting a feature vector of the second real-shot whole-body image and a feature vector of the first real-shot whole-body image as a third feature vector and a fourth feature vector;
calculating the whole-body similarity of the third feature vector and the fourth feature vector;
and determining the identity information of the pedestrian to be identified based on the similarity of the whole body.
In another optional embodiment, the determining the identity information of the pedestrian to be identified based on the whole-body similarity includes:
comparing the whole-body similarity with a whole-body similarity setting threshold;
and if the whole-body similarity is greater than or equal to the whole-body similarity set threshold, determining the identity information of the pedestrian to be identified according to the identity information of the person to which the first real shot whole-body image belongs.
As shown in fig. 3, when the second real-shot module obtains the real-shot image of the pedestrian to be identified, for example, the second real-shot whole-body image may be obtained, and the second real-shot verification module calculates the whole-body similarity through the second real-shot whole-body image and the first real-shot whole-body image for comparison, so as to verify the identity information. When the whole-body similarity reaches or exceeds the whole-body similarity setting threshold, it can be understood that the second real-shot whole-body image has high similarity with the first real-shot whole-body image, that is, the identity information of the person to which the second real-shot whole-body image belongs, that is, the identity information of the pedestrian to be identified, can be determined according to the identity information of the person to which the first real-shot whole-body image belongs. The whole-body similarity threshold is preset, and is not specifically limited by the disclosure. When the whole-body similarity is calculated, for example, the calculation may be performed by calculating a cosine distance of a feature vector of the second real-shot whole-body image and the first real-shot whole-body image. After the identity information and the whole-body similarity of the person to which the second real-shot whole-body image belongs are obtained, the identity information and the whole-body similarity can be displayed to the pedestrian to be identified.
In an alternative embodiment, the apparatus further comprises:
the second real shooting face image acquisition module is used for acquiring a second real shooting face image corresponding to a second real shooting whole-body image of the pedestrian to be identified;
a feature vector extraction module, configured to extract a feature vector of the second real shot face image and a feature vector of the first real shot face image as a fifth feature vector and a sixth feature vector;
the similarity calculation module is used for calculating second face similarity of the fifth feature vector and the sixth feature vector;
and the display module is used for displaying the identity information of the person to be identified, the whole-body similarity and the second face similarity.
As shown in fig. 4, when the second real-shot module obtains the real-shot image of the pedestrian to be identified, for example, the second real-shot whole-body image may be obtained, and the second real-shot verification module calculates the whole-body similarity through the second real-shot whole-body image and the first real-shot whole-body image for comparison, so as to verify the identity information. When the whole-body similarity reaches or exceeds the whole-body similarity setting threshold, it can be understood that the second real-shot whole-body image has high similarity with the first real-shot whole-body image, that is, the identity information of the person to which the second real-shot whole-body image belongs, that is, the identity information of the pedestrian to be identified, can be determined according to the identity information of the person to which the first real-shot whole-body image belongs. The whole-body similarity threshold is preset, and is not specifically limited by the disclosure. When the whole-body similarity is calculated, for example, the calculation may be performed by calculating a cosine distance of a feature vector of the second real-shot whole-body image and the first real-shot whole-body image. The second real shooting module can also obtain a second real shooting face image corresponding to the second real shooting whole body image after obtaining the second real shooting whole body image, the second real shooting verification module calculates the second face similarity of the second real shooting face image and the first real shooting face image, and the second face similarity is used for being displayed to the pedestrian to be identified. The second face similarity may be obtained, for example, by calculating a cosine distance of a feature vector of the second real-shot face image and the feature vector of the first real-shot face image, and the calculation of the second face similarity is not specifically limited in this disclosure. After the identity information, the whole-body similarity and the second face similarity of the person to which the second real-shot whole-body image belongs are obtained, the identity information, the whole-body similarity and the second face similarity can be displayed to the pedestrian to be identified.
In an alternative embodiment, the apparatus further comprises:
and the updating module is used for determining whether to update the whole-body image of the bottom library according to the whole-body similarity.
In an alternative embodiment, the update module is configured to:
when the whole-body similarity is greater than or equal to the whole-body similarity set threshold, judging whether the whole-body similarity is greater than or equal to a first updating threshold;
if the whole-body similarity is larger than or equal to the first updating threshold, determining to update the whole-body image in the bottom library;
and taking the second real-shot whole-body image as a new bottom base whole-body image to be stored in a bottom base, or fusing the feature vector of the first real-shot whole-body image and the feature vector of the second real-shot whole-body image to generate a new bottom base whole-body image, and storing the fused feature vector and the new bottom base whole-body image in the bottom base.
In various different scenes, such as clothing change, head decoration change and the like of the pedestrian to be identified, when the whole-body image of the pedestrian to be identified is photographed, the situation that the whole-body information changes greatly may occur, so that the passing rate of the identity authentication through the whole-body photographed image is low. As shown in fig. 5, the whole-body image can be updated as a new bottom-library whole-body image according to the situation, and when the real-shot whole-body image is shot next time, the whole-body image can be compared with the new bottom-library whole-body image, so that the identity confirmation rate is improved.
In an optional implementation manner, the storing the second actually shot whole-body image as a new bottom-library whole-body image in the bottom library, or fusing the feature vector of the first actually shot whole-body image and the feature vector of the second actually shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image in the bottom library includes:
if the whole-body similarity is greater than or equal to the first updating threshold and less than or equal to the second updating threshold, determining to fuse the feature vector of the first real shot whole-body image and the feature vector of the second real shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image into a bottom library;
and if the whole-body similarity is larger than the second updating threshold, determining that the second real-shot whole-body image is stored in the bottom library as a new bottom library whole-body image.
The updating module can select to replace the first real-shot whole-body image or to weight-fuse the feature vector of the first real-shot whole-body image according to the size of the whole-body similarity value. For example, a first update threshold and a second update threshold may be set, and when the whole-body similarity value is within the two update threshold intervals, it may be understood that the whole-body similarity has not reached the degree of complete replacement, the feature vector of the first real-shot whole-body image may be selected to be subjected to fusion processing, so as to weaken (reduce) the association between the first real-shot whole-body image and the identity of the corpus. When the whole-body similarity value is greater than the second update threshold, it can be understood that the second real-shot whole-body image has very high similarity, and the first real-shot whole-body image can be replaced to remove the association between the first real-shot whole-body image and the identity of the bottom library. The first update threshold and the second update threshold are preset, and the two update thresholds are not specifically limited in the present disclosure.
In an alternative embodiment, fusing the feature vector of the first live-shot whole-body image and the feature vector of the second live-shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image into the bottom library, including:
respectively determining the weight corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image according to the whole-body similarity;
and according to the weight corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image, performing weighted fusion on the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image to generate a new whole-body image of the bottom library, and storing the fused feature vector and the new whole-body image of the bottom library into the bottom library.
The disclosure also relates to an electronic device comprising a server, a terminal and the like. The electronic device includes: at least one processor; a memory communicatively coupled to the at least one processor; and a communication component communicatively coupled to the storage medium, the communication component receiving and transmitting data under control of the processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to implement the pedestrian identification method in the above embodiments.
In an alternative embodiment, the memory is used as a non-volatile computer-readable storage medium for storing non-volatile software programs, non-volatile computer-executable programs, and modules. The processor executes various functional applications and data processing of the device, namely, a method for identifying the identity of the pedestrian, by running the nonvolatile software program, the instructions and the modules stored in the memory.
The memory may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store a list of options, etc. Further, the memory may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, the memory optionally includes memory located remotely from the processor, and such remote memory may be connected to the external device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
One or more modules are stored in the memory and when executed by the one or more processors perform the method of pedestrian identification in any of the method embodiments described above.
The product can execute the method for identifying the identity of the pedestrian provided by the embodiment of the application, has corresponding functional modules and beneficial effects of the execution method, and can refer to the method for identifying the identity of the pedestrian provided by the embodiment of the application without detailed technical details described in the embodiment of the application.
The present disclosure also relates to a computer-readable storage medium for storing a computer-readable program for causing a computer to perform some or all of the above-described embodiments of a method for pedestrian identification.
That is, as can be understood by those skilled in the art, all or part of the steps in the method for implementing the embodiments described above may be implemented by a program instructing related hardware, where the program is stored in a storage medium and includes several instructions to enable a device (which may be a single chip, a chip, or the like) or a processor (processor) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the disclosure may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Furthermore, those of ordinary skill in the art will appreciate that while some embodiments described herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the disclosure and form different embodiments. For example, in the claims, any of the claimed embodiments may be used in any combination.
It will be understood by those skilled in the art that while the present disclosure has been described with reference to exemplary embodiments, various changes may be made and equivalents may be substituted for elements thereof without departing from the scope of the disclosure. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the disclosure without departing from the essential scope thereof. Therefore, it is intended that the disclosure not be limited to the particular embodiment disclosed, but that the disclosure will include all embodiments falling within the scope of the appended claims.
Claims (22)
1. A method of pedestrian identification, the method comprising:
acquiring a first real shooting whole body image and a corresponding first real shooting face image;
determining identity information of a person to which the first real shot face image belongs based on base database data, wherein the base database data comprises a plurality of base face images and corresponding identity information;
the first real shot whole-body image is correlated with the identity information of the person to which the first real shot face image belongs, and the first real shot whole-body image is stored in a bottom library as a bottom library whole-body image;
acquiring a second real-shot whole-body image of the pedestrian to be identified;
and determining the identity information of the pedestrian to be identified based on the first real-shot whole-body image and the second real-shot whole-body image.
2. The method of claim 1, wherein determining identity information of a person to whom the first live-shot facial image belongs based on underlying database data comprises:
extracting the characteristic vector of the first real shot facial image and the characteristic vector of the bottom base facial image as a first characteristic vector and a second characteristic vector;
calculating a first face similarity of the first feature vector and the second feature vector;
and determining the identity information of the person to which the first real shot face image belongs based on the first face similarity.
3. The method of claim 2, wherein determining identity information of a person to whom the first live-shot facial image belongs based on the first face similarity comprises:
comparing the first face similarity with a face similarity setting threshold;
and if the first face similarity is larger than or equal to the face similarity set threshold, determining the identity information of the person to which the first real shot face image belongs.
4. The method of any one of claims 1 to 3, wherein determining the identity information of the pedestrian to be identified based on the first live whole-body image and the second live whole-body image comprises:
extracting the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image as a third feature vector and a fourth feature vector;
calculating the whole-body similarity of the third feature vector and the fourth feature vector;
and determining the identity information of the pedestrian to be identified based on the whole-body similarity.
5. The method of claim 4, wherein the method further comprises:
acquiring a second real shooting face image corresponding to a second real shooting whole body image of the pedestrian to be identified;
extracting a feature vector of the second real shot face image and a feature vector of the first real shot face image as a fifth feature vector and a sixth feature vector;
calculating a second face similarity of the fifth feature vector and the sixth feature vector;
and displaying the identity information of the person to be recognized, the whole-body similarity and the second face similarity.
6. The method of claim 4, wherein determining the identity information of the pedestrian to be identified based on the whole-body similarity comprises:
comparing the whole-body similarity with a whole-body similarity setting threshold;
and if the whole-body similarity is larger than or equal to the whole-body similarity set threshold, determining the identity information of the pedestrian to be identified according to the identity information of the person to which the first real shot whole-body image belongs.
7. The method of claim 4, wherein the method further comprises:
determining whether to update the bottom library whole-body images in the bottom library based on the whole-body similarity.
8. The method of claim 7, wherein determining whether to update the bottom-library whole-body images in the bottom library based on the whole-body similarity comprises:
when the whole-body similarity is greater than or equal to the whole-body similarity set threshold, judging whether the whole-body similarity is greater than or equal to a first updating threshold;
if the whole-body similarity is larger than or equal to the first updating threshold, determining to update the whole-body image in the bottom library;
and taking the second real shot whole-body image as a new bottom base whole-body image to be stored in the bottom base, or fusing the characteristic vector of the first real shot whole-body image and the characteristic vector of the second real shot whole-body image to generate a new bottom base whole-body image, and storing the fused characteristic vector and the new bottom base whole-body image in the bottom base.
9. The method of claim 8, wherein storing the second live-shot whole-body image in the base as a new base whole-body image, or fusing the feature vector of the first live-shot whole-body image and the feature vector of the second live-shot whole-body image to generate a new base whole-body image, and storing the fused feature vector and the new base whole-body image in the base, comprises:
if the whole-body similarity is greater than or equal to a first updating threshold and less than or equal to a second updating threshold, determining to fuse the feature vector of the first real shot whole-body image and the feature vector of the second real shot whole-body image to generate a new whole-body image of the bottom base, and storing the fused feature vector and the new whole-body image of the bottom base into the bottom base;
and if the whole-body similarity is larger than a second updating threshold, determining that the second real-shot whole-body image is stored in the bottom base as a new bottom base whole-body image.
10. The method of claim 8, wherein fusing the feature vector of the first live-beat whole-body image and the feature vector of the second live-beat whole-body image to generate a new base whole-body image, and storing the fused feature vector and the new base whole-body image in the base, comprises:
respectively determining weights corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image according to the whole-body similarity;
and according to the weight corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image, performing weighted fusion on the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image into the bottom library.
11. An apparatus for pedestrian identification, the apparatus comprising:
the first real shooting module is used for acquiring a first real shooting whole-body image and a corresponding first real shooting face image;
the first real shooting verification module is used for determining the identity information of a person to which the first real shooting face image belongs based on bottom database data, wherein the bottom database data comprises a plurality of bottom database face images and corresponding identity information;
the first real shooting association module is used for associating the first real shooting whole-body image with the identity information of the person to which the first real shooting face image belongs and storing the first real shooting whole-body image as a bottom library whole-body image into a bottom library;
the second real shooting module is used for acquiring a second real shooting whole-body image of the pedestrian to be identified;
and the second real shooting verification module is used for determining the identity information of the pedestrian to be identified based on the first real shooting whole-body image and the second real shooting whole-body image.
12. The apparatus of claim 11, wherein the first tap verification module is to:
extracting the characteristic vector of the first real shot facial image and the characteristic vector of the bottom base facial image as a first characteristic vector and a second characteristic vector;
calculating a first face similarity of the first feature vector and the second feature vector;
and determining the identity information of the person to which the first real shot face image belongs based on the first face similarity.
13. The apparatus of claim 12, wherein determining identity information of a person to whom the first live facial image belongs based on the first facial similarity comprises:
comparing the first face similarity with a face similarity setting threshold;
and if the first face similarity is larger than or equal to the face similarity set threshold, determining the identity information of the person to which the first real shot face image belongs.
14. The apparatus of any one of claims 11-13, wherein the second tap verification module is to:
extracting the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image as a third feature vector and a fourth feature vector;
calculating the whole-body similarity of the third feature vector and the fourth feature vector;
and determining the identity information of the pedestrian to be identified based on the whole-body similarity.
15. The apparatus of claim 14, the apparatus further comprising:
the second real shooting face image acquisition module is used for acquiring a second real shooting face image corresponding to a second real shooting whole-body image of the pedestrian to be identified;
a feature vector extraction module, configured to extract a feature vector of the second real shot face image and a feature vector of the first real shot face image as a fifth feature vector and a sixth feature vector;
the similarity calculation module is used for calculating second face similarity of the fifth feature vector and the sixth feature vector;
and the display module is used for displaying the identity information of the person to be identified, the whole-body similarity and the second face similarity.
16. The apparatus of claim 14, wherein determining the identity information of the pedestrian to be identified based on the whole-body similarity comprises:
comparing the whole-body similarity with a whole-body similarity setting threshold;
and if the whole-body similarity is larger than or equal to the whole-body similarity set threshold, determining the identity information of the pedestrian to be identified according to the identity information of the person to which the first real shot whole-body image belongs.
17. The apparatus of claim 14, wherein the apparatus further comprises:
and the updating module is used for determining whether to update the whole-body image of the bottom library in the bottom library according to the whole-body similarity.
18. The apparatus of claim 17, wherein the update module is to:
when the whole-body similarity is greater than or equal to the whole-body similarity set threshold, judging whether the whole-body similarity is greater than or equal to a first updating threshold;
if the whole-body similarity is larger than or equal to the first updating threshold, determining to update the whole-body image in the bottom library;
and taking the second real shot whole-body image as a new bottom base whole-body image to be stored in the bottom base, or fusing the characteristic vector of the first real shot whole-body image and the characteristic vector of the second real shot whole-body image to generate a new bottom base whole-body image, and storing the fused characteristic vector and the new bottom base whole-body image in the bottom base.
19. The apparatus of claim 18, wherein storing the second live-shot whole-body image in the bottom library as a new bottom library whole-body image, or fusing a feature vector of the first live-shot whole-body image and a feature vector of the second live-shot whole-body image to generate a new bottom library whole-body image, and storing the fused feature vector and the new bottom library whole-body image in the bottom library comprises:
if the whole-body similarity is greater than or equal to a first updating threshold and less than or equal to a second updating threshold, determining to fuse the feature vector of the first real shot whole-body image and the feature vector of the second real shot whole-body image to generate a new whole-body image of the bottom base, and storing the fused feature vector and the new whole-body image of the bottom base into the bottom base;
and if the whole-body similarity is larger than a second updating threshold, determining that the second real-shot whole-body image is stored in the bottom base as a new bottom base whole-body image.
20. The apparatus of claim 19, wherein fusing the feature vector of the first live-beat whole-body image and the feature vector of the second live-beat whole-body image to generate a new base whole-body image, and storing the fused feature vector and the new base whole-body image in the base, comprises:
respectively determining weights corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image according to the whole-body similarity;
and according to the weight corresponding to the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image, performing weighted fusion on the feature vector of the second real-shot whole-body image and the feature vector of the first real-shot whole-body image to generate a new bottom-library whole-body image, and storing the fused feature vector and the new bottom-library whole-body image into the bottom library.
21. An electronic device comprising a memory and a processor, wherein the memory is configured to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method of any one of claims 1-10.
22. A computer-readable storage medium, on which a computer program is stored, the computer program being executable by a processor for implementing the method according to any one of claims 1-10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010725980.7A CN111898515A (en) | 2020-07-24 | 2020-07-24 | Method and device for identifying identity of pedestrian |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010725980.7A CN111898515A (en) | 2020-07-24 | 2020-07-24 | Method and device for identifying identity of pedestrian |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111898515A true CN111898515A (en) | 2020-11-06 |
Family
ID=73189957
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010725980.7A Pending CN111898515A (en) | 2020-07-24 | 2020-07-24 | Method and device for identifying identity of pedestrian |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111898515A (en) |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20050120339A (en) * | 2004-06-18 | 2005-12-22 | 엘지전자 주식회사 | Identity confirmation apparatus using mobile terminal's camera and the method of the same |
| KR20110129789A (en) * | 2010-05-26 | 2011-12-02 | 수원대학교산학협력단 | A security system and a method using multiplex biometrics of face and body |
| CN104794458A (en) * | 2015-05-07 | 2015-07-22 | 北京丰华联合科技有限公司 | Fuzzy video person identifying method |
| CN106874347A (en) * | 2016-12-26 | 2017-06-20 | 深圳市深网视界科技有限公司 | A kind of method and system for matching characteristics of human body and MAC Address |
| CN106991395A (en) * | 2017-03-31 | 2017-07-28 | 联想(北京)有限公司 | Information processing method, device and electronic equipment |
| CN107590452A (en) * | 2017-09-04 | 2018-01-16 | 武汉神目信息技术有限公司 | A kind of personal identification method and device based on gait and face fusion |
| US20180336687A1 (en) * | 2017-05-22 | 2018-11-22 | Creavision Technologies Ltd. | Systems and methods for user detection, identification, and localization within a defined space |
| CN108921008A (en) * | 2018-05-14 | 2018-11-30 | 深圳市商汤科技有限公司 | Portrait identification method, device and electronic equipment |
| CN109063691A (en) * | 2018-09-03 | 2018-12-21 | 武汉普利商用机器有限公司 | A kind of recognition of face bottom library optimization method and system |
| CN109376604A (en) * | 2018-09-25 | 2019-02-22 | 北京飞搜科技有限公司 | A kind of age recognition methods and device based on human body attitude |
| CN110598551A (en) * | 2019-08-09 | 2019-12-20 | 深圳力维智联技术有限公司 | Method, device, equipment and medium for improving pedestrian identity recognition efficiency |
| CN110705469A (en) * | 2019-09-30 | 2020-01-17 | 重庆紫光华山智安科技有限公司 | Face matching method and device and server |
| CN110765849A (en) * | 2019-09-09 | 2020-02-07 | 中国平安财产保险股份有限公司 | Identity information acquisition method and device based on micro expression and computer equipment |
| WO2020124710A1 (en) * | 2018-12-18 | 2020-06-25 | 深圳壹账通智能科技有限公司 | Auxiliary security inspection analysis method and apparatus, and computer device and storage medium |
-
2020
- 2020-07-24 CN CN202010725980.7A patent/CN111898515A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20050120339A (en) * | 2004-06-18 | 2005-12-22 | 엘지전자 주식회사 | Identity confirmation apparatus using mobile terminal's camera and the method of the same |
| KR20110129789A (en) * | 2010-05-26 | 2011-12-02 | 수원대학교산학협력단 | A security system and a method using multiplex biometrics of face and body |
| CN104794458A (en) * | 2015-05-07 | 2015-07-22 | 北京丰华联合科技有限公司 | Fuzzy video person identifying method |
| CN106874347A (en) * | 2016-12-26 | 2017-06-20 | 深圳市深网视界科技有限公司 | A kind of method and system for matching characteristics of human body and MAC Address |
| CN106991395A (en) * | 2017-03-31 | 2017-07-28 | 联想(北京)有限公司 | Information processing method, device and electronic equipment |
| US20180336687A1 (en) * | 2017-05-22 | 2018-11-22 | Creavision Technologies Ltd. | Systems and methods for user detection, identification, and localization within a defined space |
| CN107590452A (en) * | 2017-09-04 | 2018-01-16 | 武汉神目信息技术有限公司 | A kind of personal identification method and device based on gait and face fusion |
| CN108921008A (en) * | 2018-05-14 | 2018-11-30 | 深圳市商汤科技有限公司 | Portrait identification method, device and electronic equipment |
| CN109063691A (en) * | 2018-09-03 | 2018-12-21 | 武汉普利商用机器有限公司 | A kind of recognition of face bottom library optimization method and system |
| CN109376604A (en) * | 2018-09-25 | 2019-02-22 | 北京飞搜科技有限公司 | A kind of age recognition methods and device based on human body attitude |
| WO2020124710A1 (en) * | 2018-12-18 | 2020-06-25 | 深圳壹账通智能科技有限公司 | Auxiliary security inspection analysis method and apparatus, and computer device and storage medium |
| CN110598551A (en) * | 2019-08-09 | 2019-12-20 | 深圳力维智联技术有限公司 | Method, device, equipment and medium for improving pedestrian identity recognition efficiency |
| CN110765849A (en) * | 2019-09-09 | 2020-02-07 | 中国平安财产保险股份有限公司 | Identity information acquisition method and device based on micro expression and computer equipment |
| CN110705469A (en) * | 2019-09-30 | 2020-01-17 | 重庆紫光华山智安科技有限公司 | Face matching method and device and server |
Non-Patent Citations (1)
| Title |
|---|
| 黄玳;蔡晓东;胡月琳;曹艺;刘玉柱;: "基于特征融合的目标动态身份识别方法", 电视技术, no. 06, 15 June 2020 (2020-06-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110490078B (en) | Monitoring video processing method, device, computer equipment and storage medium | |
| US9743014B2 (en) | Image recognition system, image recognition apparatus, image recognition method, and computer program | |
| US9619708B2 (en) | Method of detecting a main subject in an image | |
| CN108038176B (en) | Method and device for establishing passerby library, electronic equipment and medium | |
| CN110705451A (en) | Face recognition method, face recognition device, terminal and server | |
| CN110688930B (en) | Face detection method and device, mobile terminal and storage medium | |
| CN107358219B (en) | Face recognition method and device | |
| CN109299658B (en) | Face detection method, face image rendering device and storage medium | |
| JP6941966B2 (en) | Person authentication device | |
| CN105007413B (en) | A kind of filming control method and user terminal | |
| CN110675426B (en) | Human body tracking method, device, equipment and storage medium | |
| US20130202162A1 (en) | Method of reconstructing three-dimensional facial shape | |
| JP7484985B2 (en) | Authentication system, authentication method, and program | |
| CN114625251A (en) | Interaction method and device based on VR, computer equipment and storage medium | |
| CN111353368A (en) | Pan-tilt camera, face feature processing method and device and electronic equipment | |
| CN113192164A (en) | Avatar follow-up control method and device, electronic equipment and readable storage medium | |
| CN110795592B (en) | Picture processing method, device and equipment | |
| CN107480580B (en) | Image recognition method and image recognition device | |
| CN110175553B (en) | Method and device for establishing feature library based on gait recognition and face recognition | |
| CN110557556A (en) | Multi-object shooting method and device | |
| CN112766065A (en) | Mobile terminal examinee identity authentication method, device, terminal and storage medium | |
| CN110633692A (en) | Pedestrian identification method and related device for unmanned aerial vehicle aerial photography | |
| CN111898515A (en) | Method and device for identifying identity of pedestrian | |
| CN110956098B (en) | Image processing method and related equipment | |
| CN111144240B (en) | Image processing method and related equipment |
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 |