CN112307979A - Personnel attribute identification method and device and computer equipment - Google Patents
Personnel attribute identification method and device and computer equipment Download PDFInfo
- Publication number
- CN112307979A CN112307979A CN202011198597.7A CN202011198597A CN112307979A CN 112307979 A CN112307979 A CN 112307979A CN 202011198597 A CN202011198597 A CN 202011198597A CN 112307979 A CN112307979 A CN 112307979A
- Authority
- CN
- China
- Prior art keywords
- person
- attribute
- image
- feature vector
- same
- 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 65
- 238000012545 processing Methods 0.000 claims abstract description 32
- 230000015654 memory Effects 0.000 claims description 14
- 238000011176 pooling Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 7
- 238000007418 data mining Methods 0.000 abstract description 3
- 238000001514 detection method Methods 0.000 abstract description 3
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000011840 criminal investigation Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
The invention relates to the technical field of identification, and discloses a personnel attribute identification method, a personnel attribute identification device and computer equipment. The personnel attribute identification method comprises the following steps: acquiring a first image acquired by first image acquisition equipment and a second image acquired by second image acquisition equipment; processing the first image and the second image to obtain a first feature vector of the first person in the first image and a second feature vector of the second person in the second image; determining whether the first person and the second person are the same person according to the first feature vector and the second feature vector; and if the person is the same person, combining the attribute characteristics of the first person with the attribute characteristics of the second person. The personnel attribute identification method can more accurately and comprehensively realize identification of the attributes of the pedestrians, and provides more effective support for video monitoring and detection, safe traffic and big data mining.
Description
Technical Field
The invention belongs to the technical field of identification, and particularly relates to a personnel attribute identification method, a personnel attribute identification device and a computer readable storage medium.
Background
The person attribute identification is to intelligently describe the multi-dimensional characteristics of pedestrians through a computer vision method, wherein the characteristics comprise face attributes, hair styles, sexes, clothing categories, clothing colors, backpacks, mobile phones, hats and the like. In practical application, the attribute identification of personnel has important significance, for example, a basic portrait of a user can be obtained through the attribute identification of personnel, effective clues can be provided for public security organs in criminal investigation, and the investigation range is narrowed; in the safety traffic requirement, the safety traffic safety inspection system can provide detected safety violation information and eliminate hidden dangers as far as possible; meanwhile, the attribute identification information of the personnel can provide personnel attribute labels and stay tracks for the big data service, and has important significance for mining commercial values.
At present, a person attribute identification method generally adopts a camera to acquire image information, and then identifies a person attribute in the image information. However, due to the limited shooting angle of a single camera, the categories of the attributes of the persons in the image information obtained by the single camera are limited, and the reliability of attribute identification is not high.
Disclosure of Invention
The invention aims to provide a personnel attribute identification method, a personnel attribute identification device and computer equipment, which are used for solving one or more technical problems in the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
in a first aspect, the present invention provides a method for identifying a person attribute, the method comprising the steps of:
acquiring a first image acquired by first image acquisition equipment and a second image acquired by second image acquisition equipment, wherein the first image acquisition equipment and the second image acquisition equipment are arranged at different positions of the same spatial region;
processing the first image and the second image to obtain a first feature vector of a first person in the first image and a second feature vector of a second person in the second image;
determining whether the first person and the second person are the same person according to the first feature vector and the second feature vector;
and if the person is the same person, combining the attribute characteristics of the first person with the attribute characteristics of the second person.
According to the personnel attribute identification method, the first image and the second image which are respectively collected by the two image collection devices at different positions of the same space region are received, whether the same person exists in the first image and the second image is judged, if the same person exists, the attribute characteristics of the person in the first image and the person in the second image are combined, so that the identification of the attributes of the pedestrian can be more accurately and comprehensively realized, and more effective support is provided for video monitoring and detection, safe passing and big data mining.
In a first possible implementation of the first aspect, determining whether the first person and the second person are the same person based on the first feature vector and the second feature vector includes:
determining whether a person is present in the first image and the second image within a cross-coverage area;
and if so, determining whether the first person and the second person are the same person according to the first feature vector of the first person in the cross coverage area and the second feature vector of the second person in the cross coverage area.
The embodiment described above can reduce the amount of data processing and improve the efficiency of attribute identification by first determining whether there is a person in the cross coverage area in the first image and the second image before determining whether the first person and the second person are the same person, and when there is a person in the cross coverage area, determining whether the first person and the second person are the same person based on the first feature vector of the first person in the cross coverage area and the second feature vector of the second person in the cross coverage area.
In a second possible implementation manner of the first aspect, the person attribute identification method further includes:
processing the first image and the second image by using an attribute identification network to obtain attribute characteristics of the first person and attribute characteristics of the second person, wherein the attribute identification network comprises a first convolution layer, a first maximum pooling layer, a second convolution layer, a second maximum pooling layer, an Iception layer and full-connection layers, the number of the full-connection layers is the same as the number of the attribute characteristics, and each full-connection layer in the full-connection layers corresponds to one loss function.
The attribute identification network adopted by the embodiment can realize multi-attribute classification identification at the same time, and effectively improves the comprehensiveness and accuracy of attribute identification.
In a third possible implementation of the first aspect, the first person and the second person are the same person, and the method of merging the attribute features of the first person with the attribute features of the second person includes:
comparing the attribute features of the first person with the attribute features of the second person one by one, and judging whether the attribute features belonging to the same attribute category exist or not;
and if not, combining the attribute features of the first person with the attribute features of the second person.
According to the embodiment, the attribute characteristics of the first person are compared with the attribute characteristics of the second person one by one, so that mistakes and omissions can be effectively avoided, and the accuracy of attribute identification is improved.
In a fourth possible implementation of the first aspect, the first person and the second person are the same person, and the method of merging the attribute features of the first person with the attribute features of the second person further includes:
if the attribute features belonging to the same attribute category exist, judging whether the attribute features belonging to the same attribute category are consistent;
and if the attribute features are consistent, selecting and reserving the attribute features belonging to the same attribute category.
According to the embodiment, the same attribute characteristics are reserved in an alternative reservation mode, so that the subsequent data processing amount can be reduced, and the attribute identification efficiency is improved.
In a fifth possible implementation manner of the first aspect, the person attribute identification method further includes:
acquiring a third image acquired by the first image acquisition equipment before acquiring the first image, wherein the first image and the third image are adjacent image frames;
processing the third image to obtain a third feature vector including a third person in the third image;
determining whether the first person and the third person are the same person according to the third feature vector and the first feature vector;
and if the first person is the same person, giving the ID corresponding to the third person to the first person.
According to the embodiment, the identification number ID of the person appearing in the third image is given to the person still appearing in the first image, so that the use of the identification number ID can be effectively reduced, and the efficiency of subsequent processing is improved.
In a sixth possible implementation manner of the first aspect, the person attribute identification method further includes:
and giving the ID of the first person to the second person.
According to the embodiment, the second person is given the identification number ID of the first person in the first image, so that the use of the identification number ID can be further reduced, and the efficiency of subsequent processing is further improved.
In a second aspect, the present invention further provides a person attribute identification apparatus, including:
the device comprises an acquisition unit, a processing unit and a display unit, wherein the acquisition unit is used for acquiring a first image acquired by first image acquisition equipment and a second image acquired by second image acquisition equipment, and the first image acquisition equipment and the second image acquisition equipment are arranged at different positions of the same spatial region;
the processing unit is used for processing the first image and the second image to obtain a first feature vector of the first person in the first image and a second feature vector of the second person in the second image;
a judging unit, configured to determine whether the first person and the second person are the same person according to the first feature vector and the second feature vector;
a merging unit, configured to merge the attribute feature of the first person and the attribute feature of the second person when the first person and the second person are the same person.
In a first possible implementation manner of the second aspect, the determination unit includes:
a first discrimination module for determining whether a person in a cross-coverage area exists in the first image and the second image;
and a second judging module, configured to determine whether the first person and the second person are the same person according to a first feature vector of a first person in the cross coverage area and a second feature vector of a second person in the cross coverage area when the first image and the second image have the person in the cross coverage area.
In a second possible implementation of the second aspect, the person attribute identification apparatus further includes:
and the attribute identification unit is used for processing the first image and the second image by utilizing an attribute identification network to obtain the attribute characteristics of the first person and the attribute characteristics of the second person, wherein the attribute identification network comprises a first convolution layer, a first maximum pooling layer, a second convolution layer, a second maximum pooling layer, an Iception layer and full-connection layers, the number of the full-connection layers is the same as the number of the attribute characteristics, and each full-connection layer in the full-connection layers corresponds to a loss function.
In a third possible embodiment of the second aspect, the merging unit comprises:
the third judging module is used for comparing the attribute characteristics of the first person with the attribute characteristics of the second person one by one and judging whether the attribute characteristics belonging to the same attribute category exist or not;
and the first merging module is used for merging the attribute feature of the first person with the attribute feature of the second person when the attribute features belonging to the same attribute category do not exist.
In a fourth possible embodiment of the second aspect, the merging unit further comprises:
the fourth judging module is used for judging whether the attribute features belonging to the same attribute category are consistent or not when the attribute features belonging to the same attribute category exist;
and the second merging module is used for selecting and reserving the attribute features belonging to the same attribute category when the attribute features of the same attribute category are consistent.
In a fifth possible implementation of the second aspect, the person attribute identification apparatus further includes:
the acquisition unit is used for acquiring a third image acquired by the first image acquisition device before acquiring the first image, wherein the first image and the third image are adjacent image frames;
the processing unit is further configured to process the third image to obtain a third feature vector including a third person in the third image;
the judging unit is further configured to determine whether the first person and the third person are the same person according to the third feature vector and the first feature vector;
and the identity mark setting unit is used for giving the identity identification number ID corresponding to the third person to the first person if the first person and the third person are the same person.
In a sixth possible embodiment of the second aspect, the identification flag setting unit is further configured to assign the identification number ID of the first person to the second person.
In a third aspect, the present invention further provides a computer device, including a memory and a processor, which are communicatively connected, where the memory is used to store a computer program, and the processor is used to read the computer program and execute the person attribute identification method according to the first aspect or any one of the possible embodiments of the first aspect.
In a fourth aspect, the present invention further provides a computer-readable storage medium, having stored thereon instructions that, when executed on a computer, perform a person attribute identification method as described in the first aspect or any one of the possible embodiments of the first aspect.
In a fifth aspect, the present invention provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform a person attribute identification method as described in the first aspect or any one of the possible embodiments of the first aspect.
Drawings
In order to more clearly illustrate the embodiments of the present invention 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 obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a method for identifying a person attribute according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of the cross coverage area of the first image acquisition device and the second image acquisition device in the embodiment of the present invention.
Fig. 3 is a schematic structural diagram of an attribute identification network in an embodiment of the present invention.
Fig. 4 is a block diagram of a structure of a person attribute identification apparatus according to an embodiment of the present invention.
Fig. 5 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and specific embodiments. It should be noted that the description of the embodiments is provided to help understanding of the present invention, but the present invention is not limited thereto. Specific structural and functional details disclosed herein are merely illustrative of example embodiments of the invention. This invention may, however, be embodied in many alternate forms and should not be construed as limited to the embodiments set forth herein.
It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments of the present invention.
It should be understood that, for the term "and/or" as may appear herein, it is merely an associative relationship that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists alone, B exists alone, and A and B exist at the same time; for the term "/and" as may appear herein, which describes another associative object relationship, it means that two relationships may exist, e.g., a/and B, may mean: a exists independently, and A and B exist independently; in addition, for the character "/" that may appear herein, it generally means that the former and latter associated objects are in an "or" relationship.
It will be understood that when an element is referred to herein as being "connected," "connected," or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may be present. Conversely, if a unit is referred to herein as being "directly connected" or "directly coupled" to another unit, it is intended that no intervening units are present. In addition, other words used to describe the relationship between elements should be interpreted in a similar manner (e.g., "between … …" versus "directly between … …", "adjacent" versus "directly adjacent", etc.).
It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments of the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises," "comprising," "includes" and/or "including," when used herein, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, numbers, steps, operations, elements, components, and/or groups thereof.
It should also be noted that, in some alternative designs, the functions/acts noted may occur out of the order noted in the figures. For example, two figures shown in succession may, in fact, be executed substantially concurrently, or the figures may sometimes be executed in the reverse order, depending upon the functionality/acts involved.
It should be understood that specific details are provided in the following description to facilitate a thorough understanding of example embodiments. However, it will be understood by those of ordinary skill in the art that the example embodiments may be practiced without these specific details. For example, systems may be shown in block diagrams in order not to obscure the examples in unnecessary detail. In other instances, well-known processes, structures and techniques may be shown without unnecessary detail in order to avoid obscuring example embodiments.
As shown in fig. 1, a first aspect of the embodiments of the present invention provides a method for identifying a person attribute, where the method includes the following steps:
step S101: acquiring a first image acquired by a first image acquisition device and a second image acquired by a second image acquisition device, wherein the first image acquisition device and the second image acquisition device are arranged at different positions of the same spatial region.
In a specific implementation process, it should be noted that, in the embodiment of the present invention, the first image capturing device and the second image capturing device may be, but are not limited to, cameras or devices with cameras arranged in places such as shopping malls, hotels, airports, exhibition halls, elevators, and the like, where the devices with cameras may be, but are not limited to, advertising screens with cameras, for example: the elevator is characterized in that the place is an elevator car, a camera is arranged in the center of the top of the elevator car, the shooting angle of the camera is downward and covers the bottom surface of the whole elevator car, an advertisement display screen with the camera is arranged on one side of the elevator, and the shooting angle of the camera of the advertisement display screen is horizontal and covers the opposite side surfaces of the elevator car; another example is: the place is the market, and the entrance of market sets up first camera, and one side wall top of market sets up the second camera, first camera with the second camera all just is the interior region of market.
Step S102: and processing the first image and the second image to obtain a first feature vector including a first person in the first image and a second feature vector including a second person in the second image.
In a specific implementation process, it should be noted that the first feature vector and the second feature vector in the embodiment of the present invention may be, but are not limited to, an attribute feature vector, a posture feature vector, and the like, and a manner of obtaining the first feature vector and the second feature vector in the embodiment of the present invention is a conventional manner.
Step S103: and determining whether the first person and the second person are the same person according to the first feature vector and the second feature vector.
In a specific implementation process, there are many ways to determine whether the first person and the second person are the same person according to the first feature vector and the second feature vector, for example, comparing the first feature vector and the second feature vector, determining similarity between the first person and the second person, if the similarity is greater than or equal to a threshold value, such as 75% or 80%, it indicates that the first person and the second person are the same person, and if the similarity is less than the threshold value, it indicates that the first person and the second person are different persons. Of course, the determination analysis may be performed in other manners besides the determination based on the similarity between the first feature vector and the second feature vector, which is not limited herein.
In a specific implementation process, since the first image capturing device and the second image capturing device are disposed at different positions of the same spatial area, there may be cross coverage in the image capturing areas of the first image capturing device and the second image capturing device, and therefore, in a possible implementation, the method for determining whether the first person and the second person are the same person according to the first feature vector and the second feature vector in the embodiment of the present invention includes:
determining whether a person is present in the first image and the second image within a cross-coverage area;
and if so, determining whether the first person and the second person are the same person according to the first feature vector of the first person in the cross coverage area and the second feature vector of the second person in the cross coverage area.
The embodiment described above can reduce the amount of data processing and improve the efficiency of attribute identification by first determining whether there is a person in the cross coverage area in the first image and the second image before determining whether the first person and the second person are the same person, and when there is a person in the cross coverage area, determining whether the first person and the second person are the same person based on the first feature vector of the first person in the cross coverage area and the second feature vector of the second person in the cross coverage area.
It should be noted that, in the embodiment of the present invention, after the first image capturing device and the second image capturing device are installed and fixed, the cross coverage area is predetermined by the capturing parameters and the spatial position of each other, which is inconvenient to fix, as shown in fig. 2.
Step S104: and if the person is the same person, combining the attribute characteristics of the first person with the attribute characteristics of the second person.
In a specific implementation process, there are many methods for identifying attribute features, and in a feasible implementation scheme, the method for identifying attribute features adopted in the embodiment of the present invention specifically includes:
processing the first image and the second image by using an attribute identification network to obtain attribute features of the first person and attribute features of the second person, wherein the attribute identification network comprises a first convolution layer, a first maximum pooling layer, a second convolution layer, a second maximum pooling layer, an Iception layer and full-connection layers, as shown in FIG. 3, the number of the full-connection layers is the same as the number of the attribute features, and each full-connection layer in the full-connection layers corresponds to a loss function.
According to the embodiment, the number of the layers of the full-connection layers is set to be the same as the number of the attribute features, and each layer of the full-connection layer corresponds to one loss function, so that multi-attribute classification identification can be realized at the same time, and the comprehensiveness and accuracy of attribute identification are effectively improved.
In a specific implementation process, due to different positions of the first image capturing device and the second image capturing device, when a person appears in a cross coverage area of the first image capturing device and the second image capturing device, an attribute of the person in a first image acquired by the first image capturing device and an attribute of the person in a second image acquired by the second image capturing device may also be different, and therefore, in a possible implementation, in an embodiment of the present invention, when the first person and the second person are the same person, a method for combining an attribute feature of the first person and an attribute feature of the second person includes:
comparing the attribute features of the first person with the attribute features of the second person one by one, and judging whether the attribute features belonging to the same attribute category exist or not;
and if not, combining the attribute features of the first person with the attribute features of the second person.
For example: the first person's attribute features include white hat, red T-shirt, mechanical watch, black pants, and white shoes, and the second person's attribute features include white hat, red T-shirt, black pants, white shoes, and yellow backpack.
Comparing the attribute features of the first person with the attribute features of the second person one by one, and judging whether the attribute features belonging to the same attribute category exist or not, namely: and comparing the white hat, the red T-shirt, the mechanical watch, the black trousers and the white shoes of the first person with the white hat, the red T-shirt, the black trousers, the white shoes and the yellow backpack of the second person one by one, and judging whether attribute characteristics belonging to the same attribute category exist.
After comparing and judging one by one, it is found that the mechanical watch in the attribute features of the first person does not exist in the attribute features of the second person, and the yellow backpack in the attribute features of the second person does not exist in the attribute features of the first person, at this time, the attribute features of the first person and the attribute features of the second person are merged, and the merged attribute features are as follows: white hat, red T-shirt, mechanical watch, black pants, white shoes, and yellow backpack.
According to the embodiment, the attribute characteristics of the first person are compared with the attribute characteristics of the second person one by one, so that mistakes and omissions can be effectively avoided, and the accuracy and comprehensiveness of attribute identification are improved.
In a specific implementation, since part of the attribute feature can be obtained from any angle, such as a hat worn on the head, a piece of clothing worn on the body, and the like, in a possible implementation, in an embodiment of the present invention, when the first person and the second person are the same person, the method for combining the attribute feature of the first person and the attribute feature of the second person further includes:
if the attribute features belonging to the same attribute category exist, judging whether the attribute features belonging to the same attribute category are consistent;
and if the attribute features are consistent, selecting and reserving the attribute features belonging to the same attribute category.
It should be noted that, in the embodiments of the present invention, the attribute category may be, but is not limited to, hat, clothes, trousers, shoes, earrings, watches, satchel, etc., and the attribute feature of the attribute category may be, but is not limited to, style, color, length, etc. For example: the attribute feature of the first person comprises a white hat, then hat is an attribute category of the first person, and white is a color attribute feature of the attribute category of hat; another example is: the attribute-feature of the first person comprises a red T-shirt, the garment is an attribute-category of the first person, and the red belongs to a color attribute-feature of the attribute-category of garments to which the T-shirt belongs.
For ease of understanding, the above example is continued for detailed explanation:
the method for judging whether the white hat in the attribute feature of the first person and the white hat in the attribute feature of the second person belong to the same attribute class of hat comprises the following steps: and comparing the color attribute feature of white in the hat attribute category of the first person with the color attribute feature of white in the hat attribute category of the second person, judging whether the color attribute features of the first person and the hat attribute features of the second person are consistent, and selecting and retaining the attribute features belonging to the same attribute category if the color attribute features of the first person and the hat attribute features of the second person are consistent.
The red T-shirt in the attribute features of the first person and the red T-shirt in the attribute features of the second person belong to the same attribute category of clothes, and at the moment, the method for judging whether the attribute features belonging to the same attribute category are consistent is as follows: and comparing the color attribute feature of red in the clothes attribute category of the first person with the color attribute feature of red in the clothes attribute category of the second person, judging whether the color attribute features of the first person and the second person are consistent, meanwhile, comparing the style attribute feature of the T-shirt in the clothes attribute category of the first person with the style attribute feature of the T-shirt in the clothes attribute category of the second person, judging whether the style attribute features of the first person and the second person are consistent, and selecting to keep the attribute features belonging to the same attribute category if the color attribute features and the style attribute features are consistent.
According to the embodiment, the accuracy of attribute identification can be effectively improved by judging whether the attribute features belonging to the same attribute category are consistent or not; the same attribute characteristics are reserved in an alternative reservation mode, so that the subsequent data processing amount can be reduced, and the attribute identification efficiency is improved.
In a specific implementation process, since the angle at which the first image capturing device acquires the first image is different from the angle at which the second image capturing device acquires the second image, the result of identifying the attribute features of the same attribute class, such as the color of a hat, may be different due to the different angles and different light conditions, so that the color of the hat in the first image acquired by the first image capturing device is purple, and the color of the hat in the second image acquired by the second image capturing device is gray, in order to further improve the accuracy of attribute identification, in a possible implementation scheme, in an embodiment of the present invention, when the attribute features of the same attribute class are not consistent, the embodiment of the present invention selects an image acquired by an image capturing device with a high identification accuracy, where the identification accuracy is specifically to the same attribute during attribute feature identification training, and the identification accuracy of the attribute in the first image acquired by the first image acquisition device and the second image acquired by the second image acquisition device.
In a specific implementation process, in order to facilitate management of attribute characteristics of each person, in the embodiment of the present invention, each person who appears in an acquisition range of the first image acquisition device and the second image acquisition device is assigned with an identity identifier ID, but because a person stays in the first image acquisition device for a long time, at this time, a continuous multi-frame image acquired by the first image acquisition device includes the person, and if the person is assigned with an independent identity identifier ID in each frame of image, the person not only wastes the identity identifier ID, but also causes data bloat and affects subsequent processing efficiency, and also makes attribute identification data unreliable, and cannot provide effective support for video monitoring and detection, safe traffic, big data mining, and the like. Therefore, in order to avoid the above situation, in a possible implementation, the method for identifying a person attribute provided by the embodiment of the present invention further includes:
acquiring a third image acquired by the first image acquisition equipment before acquiring the first image, wherein the first image and the third image are adjacent image frames;
processing the third image to obtain a third feature vector including a third person in the third image;
determining whether the first person and the third person are the same person according to the third feature vector and the first feature vector;
and if the first person is the same person, giving the ID corresponding to the third person to the first person.
According to the embodiment, the identification numbers ID of the personnel appearing in the third image are given to the personnel still appearing in the first image, so that each personnel only has one independent identification number ID, the use of the identification numbers ID can be effectively reduced, the subsequent processing efficiency is improved, and the reliability of attribute identification can be guaranteed.
In a specific implementation process, when there is a cross coverage area in the acquisition areas of the first image acquisition device and the second image acquisition device, in order to avoid assigning identification numbers ID to persons appearing in the cross coverage area, in a feasible implementation, the method for identifying person attributes provided in the embodiment of the present invention further includes:
and giving the ID of the first person to the second person.
According to the embodiment, the situation that personnel appearing in the cross coverage area are respectively endowed with the ID can be effectively avoided, the use of the ID is further reduced, the efficiency of subsequent processing is improved, and the reliability of attribute identification is guaranteed.
Based on the above personnel attribute identification method, a second aspect of the embodiments of the present invention further provides a personnel attribute identification apparatus, as shown in fig. 4, where the personnel attribute identification apparatus includes:
the device comprises an acquisition unit, a processing unit and a display unit, wherein the acquisition unit is used for acquiring a first image acquired by first image acquisition equipment and a second image acquired by second image acquisition equipment, and the first image acquisition equipment and the second image acquisition equipment are arranged at different positions of the same spatial region;
the processing unit is used for processing the first image and the second image to obtain a first feature vector of the first person in the first image and a second feature vector of the second person in the second image;
a judging unit, configured to determine whether the first person and the second person are the same person according to the first feature vector and the second feature vector;
a merging unit, configured to merge the attribute feature of the first person and the attribute feature of the second person when the first person and the second person are the same person.
In a first possible implementation manner of the second aspect, the determination unit includes:
a first discrimination module for determining whether a person in a cross-coverage area exists in the first image and the second image;
and a second judging module, configured to determine whether the first person and the second person are the same person according to a first feature vector of a first person in the cross coverage area and a second feature vector of a second person in the cross coverage area when the first image and the second image have the person in the cross coverage area.
In a second possible implementation of the second aspect, the person attribute identification apparatus further includes:
and the attribute identification unit is used for processing the first image and the second image by utilizing an attribute identification network to obtain the attribute characteristics of the first person and the attribute characteristics of the second person, wherein the attribute identification network comprises a first convolution layer, a first maximum pooling layer, a second convolution layer, a second maximum pooling layer, an Iception layer and full-connection layers, the number of the full-connection layers is the same as the number of the attribute characteristics, and each full-connection layer in the full-connection layers corresponds to a loss function.
In a third possible embodiment of the second aspect, the merging unit comprises:
the third judging module is used for comparing the attribute characteristics of the first person with the attribute characteristics of the second person one by one and judging whether the attribute characteristics belonging to the same attribute category exist or not;
and the first merging module is used for merging the attribute feature of the first person with the attribute feature of the second person when the attribute features belonging to the same attribute category do not exist.
In a fourth possible embodiment of the second aspect, the merging unit further comprises:
the fourth judging module is used for judging whether the attribute features belonging to the same attribute category are consistent or not when the attribute features belonging to the same attribute category exist;
and the second merging module is used for selecting and reserving the attribute features belonging to the same attribute category when the attribute features of the same attribute category are consistent.
In a fifth possible implementation of the second aspect, the person attribute identification apparatus further includes:
the acquisition unit is used for acquiring a third image acquired by the first image acquisition device before acquiring the first image, wherein the first image and the third image are adjacent image frames;
the processing unit is further configured to process the third image to obtain a third feature vector including a third person in the third image;
the judging unit is further configured to determine whether the first person and the third person are the same person according to the third feature vector and the first feature vector;
and the identity mark setting unit is used for giving the identity identification number ID corresponding to the third person to the first person if the first person and the third person are the same person.
In a sixth possible embodiment of the second aspect, the identification flag setting unit is further configured to assign the identification number ID of the first person to the second person.
The working process, the working details and the technical effects of the foregoing apparatus provided in the embodiments of the present invention may refer to the person attribute identification method described in the first aspect or any one of the feasible embodiments of the first aspect, and are not described herein again.
The third aspect of the present invention further provides a computer device for executing the person attribute identification method according to any one of the possible embodiments of the first aspect or the first aspect, as shown in fig. 5, where the computer device includes a memory and a processor, which are communicatively connected, where the memory is used for storing a computer program, and the processor is used for reading the computer program and executing the person attribute identification method according to any one of the possible designs of the first aspect or the first aspect. For example, the Memory may include, but is not limited to, a Random-Access Memory (RAM), a Read-Only Memory (ROM), a Flash Memory (Flash Memory), a First-in First-out (FIFO), and/or a First-in Last-out (FILO), and the like; the processor may not be limited to the microprocessor of the model number employing the STM32F105 family. In addition, the computer device may also include, but is not limited to, a power module, a display screen, and other necessary components.
For the working process, the working details, and the technical effects of the computer device provided in the embodiments of the present invention, reference may be made to the person attribute identification method described in the first aspect or any one of the feasible embodiments of the first aspect, which is not described herein again.
A fourth aspect of the embodiments of the present invention further provides a computer-readable storage medium storing instructions including the method for identifying a person attribute according to any one of the first aspect or any one of the possible implementations of the first aspect, that is, the computer-readable storage medium stores instructions that, when executed on a computer, perform the method for identifying a person attribute according to any one of the first aspect or any one of the possible implementations of the first aspect. The computer-readable storage medium refers to a carrier for storing data, and may include, but is not limited to, floppy disks, optical disks, hard disks, flash memories, flash disks and/or Memory sticks (Memory sticks), etc., and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
A fifth aspect of embodiments of the present invention provides a computer program product comprising instructions, which when run on a computer, cause the computer to perform a method for identifying a person attribute as described in the first aspect or any one of the possible implementations of the first aspect. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable devices.
The embodiments described above are merely illustrative, and may or may not be physically separate, if referring to units illustrated as separate components; if reference is made to a component displayed as a unit, it may or may not be a physical unit, and may be located in one place or distributed over a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: modifications may be made to the embodiments described above, or equivalents may be substituted for some of the features described. And such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Finally, it should be noted that the present invention is not limited to the above alternative embodiments, and that various other forms of products can be obtained by anyone in light of the present invention. The above detailed description should not be taken as limiting the scope of the invention, which is defined in the claims, and which the description is intended to be interpreted accordingly.
Claims (10)
1. A person attribute identification method, characterized in that the method comprises the steps of:
acquiring a first image acquired by first image acquisition equipment and a second image acquired by second image acquisition equipment, wherein the first image acquisition equipment and the second image acquisition equipment are arranged at different positions of the same spatial region;
processing the first image and the second image to obtain a first feature vector of a first person in the first image and a second feature vector of a second person in the second image;
determining whether the first person and the second person are the same person according to the first feature vector and the second feature vector;
and if the person is the same person, combining the attribute characteristics of the first person with the attribute characteristics of the second person.
2. The method of claim 1, wherein determining whether the first person and the second person are the same person based on the first feature vector and the second feature vector comprises:
determining whether a person is present in the first image and the second image within a cross-coverage area;
and if so, determining whether the first person and the second person are the same person according to the first feature vector of the first person in the cross coverage area and the second feature vector of the second person in the cross coverage area.
3. The method of claim 1, further comprising:
processing the first image and the second image by using an attribute identification network to obtain attribute characteristics of the first person and attribute characteristics of the second person, wherein the attribute identification network comprises a first convolution layer, a first maximum pooling layer, a second convolution layer, a second maximum pooling layer, an Iception layer and full-connection layers, the number of the full-connection layers is the same as the number of the attribute characteristics, and each full-connection layer in the full-connection layers corresponds to one loss function.
4. The person attribute identification method according to claim 1, wherein merging the attribute feature of the first person and the attribute feature of the second person if the persons are the same person comprises:
comparing the attribute features of the first person with the attribute features of the second person one by one, and judging whether the attribute features belonging to the same attribute category exist or not;
and if not, combining the attribute features of the first person with the attribute features of the second person.
5. The method of claim 4, further comprising:
if yes, judging whether the attribute features belonging to the same attribute category are consistent;
and if the attribute features are consistent, selecting and reserving the attribute features belonging to the same attribute category.
6. The method of claim 1, further comprising:
acquiring a third image acquired by the first image acquisition equipment before acquiring the first image, wherein the first image and the third image are adjacent image frames;
processing the third image to obtain a third feature vector including a third person in the third image;
determining whether the first person and the third person are the same person according to the third feature vector and the first feature vector;
and if the first person is the same person, giving the ID corresponding to the third person to the first person.
7. The person attribute identification method of claim 6, further comprising:
and giving the ID of the first person to the second person.
8. A person attribute identification apparatus, comprising:
the device comprises an acquisition unit, a processing unit and a display unit, wherein the acquisition unit is used for acquiring a first image acquired by first image acquisition equipment and a second image acquired by second image acquisition equipment, and the first image acquisition equipment and the second image acquisition equipment are arranged at different positions of the same spatial region;
the processing unit is used for processing the first image and the second image to obtain a first feature vector of the first person in the first image and a second feature vector of the second person in the second image;
a judging unit, configured to determine whether the first person and the second person are the same person according to the first feature vector and the second feature vector;
a merging unit, configured to merge the attribute feature of the first person and the attribute feature of the second person when the first person and the second person are the same person.
9. A computer device comprising a memory and a processor communicatively coupled, wherein the memory is configured to store a computer program and the processor is configured to read the computer program and perform the method of any of claims 1 to 7.
10. A computer-readable storage medium having stored thereon instructions which, when executed on a computer, perform the method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011198597.7A CN112307979A (en) | 2020-10-31 | 2020-10-31 | Personnel attribute identification method and device and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011198597.7A CN112307979A (en) | 2020-10-31 | 2020-10-31 | Personnel attribute identification method and device and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112307979A true CN112307979A (en) | 2021-02-02 |
Family
ID=74332323
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011198597.7A Pending CN112307979A (en) | 2020-10-31 | 2020-10-31 | Personnel attribute identification method and device and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112307979A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114792451A (en) * | 2022-06-22 | 2022-07-26 | 深圳市海清视讯科技有限公司 | Information processing method, device, and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107657232A (en) * | 2017-09-28 | 2018-02-02 | 南通大学 | A kind of pedestrian's intelligent identification Method and its system |
| CN107832672A (en) * | 2017-10-12 | 2018-03-23 | 北京航空航天大学 | A kind of pedestrian's recognition methods again that more loss functions are designed using attitude information |
| CN109271888A (en) * | 2018-08-29 | 2019-01-25 | 汉王科技股份有限公司 | Personal identification method, device, electronic equipment based on gait |
| CN109325412A (en) * | 2018-08-17 | 2019-02-12 | 平安科技（深圳）有限公司 | Pedestrian recognition method, device, computer equipment and storage medium |
| CN109948709A (en) * | 2019-03-21 | 2019-06-28 | 南京斯玛唯得智能技术有限公司 | A kind of multitask Attribute Recognition system of target object |
| CN110175587A (en) * | 2019-05-30 | 2019-08-27 | 黄岩 | A kind of video frequency tracking method based on recognition of face and Algorithm for gait recognition |
| CN110378931A (en) * | 2019-07-10 | 2019-10-25 | 成都数之联科技有限公司 | A kind of pedestrian target motion track acquisition methods and system based on multi-cam |
| CN110598716A (en) * | 2019-09-09 | 2019-12-20 | 北京文安智能技术股份有限公司 | Personnel attribute identification method, device and system |
| CN111160312A (en) * | 2020-01-02 | 2020-05-15 | 重庆紫光华山智安科技有限公司 | Target identification method and device and electronic equipment |
| CN111291612A (en) * | 2019-12-20 | 2020-06-16 | 深圳久凌软件技术有限公司 | Pedestrian re-identification method and device based on multi-person multi-camera tracking |
| CN111325198A (en) * | 2018-12-13 | 2020-06-23 | 北京地平线机器人技术研发有限公司 | Video object feature extraction method and device and video object matching method and device |
-
2020
- 2020-10-31 CN CN202011198597.7A patent/CN112307979A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107657232A (en) * | 2017-09-28 | 2018-02-02 | 南通大学 | A kind of pedestrian's intelligent identification Method and its system |
| CN107832672A (en) * | 2017-10-12 | 2018-03-23 | 北京航空航天大学 | A kind of pedestrian's recognition methods again that more loss functions are designed using attitude information |
| CN109325412A (en) * | 2018-08-17 | 2019-02-12 | 平安科技（深圳）有限公司 | Pedestrian recognition method, device, computer equipment and storage medium |
| CN109271888A (en) * | 2018-08-29 | 2019-01-25 | 汉王科技股份有限公司 | Personal identification method, device, electronic equipment based on gait |
| CN111325198A (en) * | 2018-12-13 | 2020-06-23 | 北京地平线机器人技术研发有限公司 | Video object feature extraction method and device and video object matching method and device |
| CN109948709A (en) * | 2019-03-21 | 2019-06-28 | 南京斯玛唯得智能技术有限公司 | A kind of multitask Attribute Recognition system of target object |
| CN110175587A (en) * | 2019-05-30 | 2019-08-27 | 黄岩 | A kind of video frequency tracking method based on recognition of face and Algorithm for gait recognition |
| CN110378931A (en) * | 2019-07-10 | 2019-10-25 | 成都数之联科技有限公司 | A kind of pedestrian target motion track acquisition methods and system based on multi-cam |
| CN110598716A (en) * | 2019-09-09 | 2019-12-20 | 北京文安智能技术股份有限公司 | Personnel attribute identification method, device and system |
| CN111291612A (en) * | 2019-12-20 | 2020-06-16 | 深圳久凌软件技术有限公司 | Pedestrian re-identification method and device based on multi-person multi-camera tracking |
| CN111160312A (en) * | 2020-01-02 | 2020-05-15 | 重庆紫光华山智安科技有限公司 | Target identification method and device and electronic equipment |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114792451A (en) * | 2022-06-22 | 2022-07-26 | 深圳市海清视讯科技有限公司 | Information processing method, device, and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10049283B2 (en) | Stay condition analyzing apparatus, stay condition analyzing system, and stay condition analyzing method | |
| CN107358146B (en) | Method for processing video frequency, device and storage medium | |
| JP6276519B2 (en) | Person counting device and human flow line analyzing device | |
| CN108369770B (en) | Litter monitoring arrangement and the Litter monitoring system and Litter monitoring method for having the Litter monitoring arrangement | |
| CN103440475B (en) | A kind of ATM user face visibility judge system and method | |
| JP5569990B2 (en) | Attribute determination method, attribute determination apparatus, program, recording medium, and attribute determination system | |
| CN105187785B (en) | A kind of across bayonet pedestrian's identifying system and method based on choice of dynamical notable feature | |
| CN113759431B (en) | Security data association method and device and X-ray security system | |
| KR20120019008A (en) | Method and devices for tracking multiple object | |
| CN107615298A (en) | Face identification method and system | |
| KR101448391B1 (en) | People counting system | |
| CN109961472B (en) | Method, system, storage medium and electronic device for generating 3D thermodynamic diagram | |
| CN105740379A (en) | Photo classification management method and apparatus | |
| Nambiar et al. | A multi-camera video dataset for research on high-definition surveillance | |
| CN109241877A (en) | A kind of track identification system, method, apparatus and its computer storage medium | |
| CN105989174A (en) | Extraction device and extraction method of area of interest | |
| JP2014016968A (en) | Person retrieval device and data collection device | |
| JP2022191288A (en) | Information processing device | |
| JP3459950B2 (en) | Face detection and face tracking method and apparatus | |
| KR20080057500A (en) | System and method for tracking movement images of people | |
| CN112307979A (en) | Personnel attribute identification method and device and computer equipment | |
| CN105955451A (en) | Travel prompt method based on user image | |
| CN110717428A (en) | Identity recognition method, device, system, medium and equipment fusing multiple features | |
| CN108664908A (en) | Face identification method, equipment and computer readable storage medium | |
| Islam et al. | Correlating belongings with passengers in a simulated airport security checkpoint |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210202 |
|
| RJ01 | Rejection of invention patent application after publication |