CN111708907B - Target person query method, device, equipment and storage medium - Google Patents
Target person query method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111708907B CN111708907B CN202010529142.2A CN202010529142A CN111708907B CN 111708907 B CN111708907 B CN 111708907B CN 202010529142 A CN202010529142 A CN 202010529142A CN 111708907 B CN111708907 B CN 111708907B
- Authority
- CN
- China
- Prior art keywords
- value
- target
- value distribution
- target personnel
- determining
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000004458 analytical method Methods 0.000 claims description 18
- 238000001514 detection method Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims 1
- 239000003086 colorant Substances 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000003287 optical effect Effects 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- 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
- G06F16/5838—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Library & Information Science (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the invention discloses a query method, a query device, query equipment and a storage medium for target personnel. Wherein the method comprises the following steps: acquiring a target personnel image and determining a target personnel area in the target personnel image; determining the value of the pixel point of the target personnel area in an RGB space, and carrying out clustering treatment on the target personnel area according to the value in the RGB space to obtain at least two value distribution families; determining the value duty ratio of the target personnel area according to the at least two value distribution families; and storing the value duty ratio into a database so as to be convenient for inquiring the target personnel according to the data in the database. According to the embodiment of the invention, the color composition of the target personnel in the color space is determined by acquiring the target personnel area, and the target personnel is inquired according to the color composition. The problem of the low query accuracy that the prior art stores through face or single colour is solved, and the query efficiency and the query accuracy of target personnel are improved.
Description
Technical Field
The embodiment of the invention relates to a computer technology, in particular to a target person query method, a device, equipment and a storage medium.
Background
With the increase of outdoor activities of people, people finding are frequently met in public places such as airports, railway stations and bank business halls, the target query precision is improved, the time of people can be effectively saved, and the user satisfaction is improved.
In the prior art, most of the equipment such as a camera is used for target inquiry, and the camera is difficult to obtain a clear front face picture meeting the face recognition requirement due to the problems of arrangement positions, angles and the like of the camera, so that the efficiency and the accuracy of target personnel searching are lower.
Disclosure of Invention
The embodiment of the invention provides a query method, a query device, query equipment and a storage medium for target personnel, so as to improve the query precision and query efficiency of the target personnel.
In a first aspect, an embodiment of the present invention provides a method for querying a target person, where the method includes:
acquiring a target personnel image and determining a target personnel area in the target personnel image;
determining the value of the pixel point of the target personnel area in an RGB space, and carrying out clustering treatment on the target personnel area according to the value in the RGB space to obtain at least two value distribution families;
determining the value duty ratio of the target personnel area according to the at least two value distribution families;
and storing the value duty ratio into a database so as to be convenient for inquiring the target personnel according to the data in the database.
In a second aspect, an embodiment of the present invention further provides a query device for a target person, where the device includes:
the target area determining module is used for acquiring a target personnel image and determining a target personnel area in the target personnel image;
the value distribution family obtaining module is used for determining the value of the pixel point of the target personnel area in an RGB space, and carrying out clustering treatment on the target personnel area according to the value in the RGB space to obtain at least two value distribution families;
the duty ratio determining module is used for determining the value duty ratio of the target personnel area according to the at least two value distribution families;
and the target query module is used for storing the value duty ratio into a database so as to facilitate the query of target personnel according to the data in the database.
In a third aspect, an embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements a method for querying a target person according to any embodiment of the present application when the processor executes the program.
In a fourth aspect, embodiments of the present invention also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are for performing a method of querying a target person as described in any of the embodiments of the present application.
According to the embodiment of the invention, the target personnel area is obtained from the target personnel image, the target personnel area is clustered to obtain a plurality of value distribution families, the color value ratio of the target personnel is determined according to the pixel values of the value distribution families in the color space, and the target personnel is inquired according to the color value ratio. The problem of in prior art carry out personnel data storage through face or single color, the target personnel inquiry precision that causes is low is solved, through confirming the color composition, has improved the efficiency and the precision of target personnel inquiry.
Drawings
FIG. 1 is a flow chart of a method for querying a target person according to a first embodiment of the present invention;
FIG. 2 is a schematic view of a pedestrian frame in a first embodiment of the invention;
FIG. 3 is a flow chart of a method for querying a target person according to a second embodiment of the present invention;
FIG. 4 is a block diagram of a target person query device according to a third embodiment of the present invention;
fig. 5 is a schematic structural diagram of a computer device in a fourth embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings.
Example 1
Fig. 1 is a flow chart of a query method for a target person according to an embodiment of the present invention, where the embodiment is applicable to a case of querying the target person, and the method may be executed by a query device for the target person. As shown in fig. 1, the method specifically includes the following steps:
s110, acquiring a target personnel image, and determining a target personnel area in the target personnel image.
The target person image is an image containing target persons, and can be directly obtained from a plurality of images or extracted from a video. After the target personnel image is obtained, determining the area where the target personnel is located in the target personnel image, and selecting the target personnel area in a frame mode.
In this embodiment, optionally, acquiring the target person image includes: and extracting frames of video data of the target personnel according to a preset period to obtain images of the target personnel.
Specifically, video input is performed on a scene where the target person is located, and the video can include an action process of the target person. After the recorded video is obtained, the video can be frame-extracted according to a preset period, for example, one frame is extracted every five seconds in the preset period, and the extracted image is used as a target personnel image. The beneficial effects of setting up like this lie in, realize the automatic acquisition to target personnel's image, reduce the manual operation process, practice thrift inquiry time, improve target personnel's efficiency of inquiry.
In this embodiment, optionally, determining the target person area in the target person image includes: determining a target person area with the maximum target person-to-person frame, no shielding of the target person-to-person frame and/or random target person-to-person frame from the target person image; wherein the pedestrian frame is an image area including at least one pedestrian.
Specifically, the target person may have different target person areas in different target person images, and the target person may be blocked by other people during the course of action. The target person may be framed in the form of a pedestrian frame, which may be a square frame, including at least one pedestrian, for example, the pedestrian frame may only frame the target person. Fig. 2 is a schematic diagram of a pedestrian frame, and a dashed frame in fig. 2 is a pedestrian frame. When the target person acts, the distance between the target person and the camera can be changed, and the size of the pedestrian frame is changed. And selecting at least one target personnel image, and acquiring a target personnel area from the target personnel image. For example, three time target person images may be selected, which are the maximum time of the target person pedestrian frame, the least occlusion time of the target person pedestrian frame, and the random time, respectively. And when the target person area is determined, selecting the target person pedestrian frame with the maximum target person pedestrian frame, which is not shielded and/or random with one target person image, wherein the maximum moment of the target person pedestrian frame is the moment when the target person is closest to the camera. The method has the advantages that different target personnel areas are acquired for the same target personnel, the target personnel can be comprehensively known from multiple angles, the omission of image information of the target personnel is avoided, and the query precision of the target personnel is improved.
In this embodiment, optionally, after the target person image is acquired, the method further includes: based on the faster RCNN network and the FPN network structure, a target person detection network is obtained so as to detect and track target persons conveniently; the master RCNN network is used for detecting target personnel, and the FPN network is used for detecting target personnel of a small target.
Specifically, a plurality of target person images can be directly acquired, and a target person region is determined from the target person images. After the target person image is obtained, the target person in the image can be detected and tracked, the behavior track of the target person is determined, and then a plurality of target person images are extracted from the image of the behavior track of the target person to determine the target person area. The target person detection network can be obtained by adopting a master RCNN (faster Region Convolutional Neural Network, a convolutional neural network) network and an FPN (Feature Pyramid Networks, a characteristic pyramid network) network structure, so as to detect and track target persons. The master RCNN network and the FPN network can be used for detecting target personnel, wherein the FPN network greatly improves the performance of small target detection. For example, a strategy of capturing 5 frames per second may be employed to extract images from the video from which the target person is detected using a pre-trained network structure. And tracking the target personnel by adopting an SORT (tracking) algorithm, recording the behavior track of the target personnel, and extracting the target personnel area with the maximum target personnel pedestrian frame, which is not shielded and/or random by the target personnel pedestrian frame from the behavior track of the target personnel. The method has the advantages that the detection time is reduced by adopting the target tracking method, so that the real-time requirement is met, the target person images meeting the pedestrian frame requirement are prevented from being searched from the massive target person images, the manpower and time are saved, and the query efficiency of the target person is improved.
S120, determining the value of the pixel point of the target personnel area in the RGB space, and clustering the target personnel area according to the value in the RGB space to obtain at least two value distribution families.
After the target personnel area is obtained, the value of the pixel point in the target personnel area in the color space is determined, and an RGB (Red Green Blue) color space can be adopted. After obtaining the pixel values, clustering the pixel points with the same color by adopting a clustering algorithm to obtain at least two value distribution families, for example, extracting target personnel areas on the target personnel images at three moments, wherein the pixels of the target personnel areas at the three moments can be clustered by adopting an EM (Expectation Maximization, expected value and maximum value) clustering algorithm, and a plurality of value distribution families can be generated in the target personnel areas at each moment. The color composition of the target personnel can be obtained preliminarily through the RGB color space, so that the colors in a value distribution group are kept consistent.
S130, determining the value ratio of the target personnel area according to at least two value distribution families.
After the value distribution group is obtained, the value distribution group with similar colors may exist, the value distribution group with similar colors may be further adjusted, the value distribution groups with similar colors are combined into one distribution group, the types of the adjusted distribution groups are determined, the proportion of the distribution group colors to the overall colors of the target personnel is determined, and the color value ratio of the target personnel is determined.
In this embodiment, optionally, before determining the value ratio of the target personnel area according to at least two value distribution families, the method further includes: performing color difference analysis on at least two value distribution families based on LAB space to obtain an adjustment result of the value distribution families; correspondingly, determining the value ratio of the target personnel area according to at least two value distribution families comprises the following steps: and determining the value duty ratio of the target personnel area according to the adjustment result of the value distribution family.
Specifically, the obtained value distribution groups are obtained based on an RGB color space, and then an LAB (brightness, magenta-to-green range, yellow-to-blue range) color space is adopted to obtain the values of the pixels of a plurality of value distribution groups in the LAB color space, the average value of each value distribution group is calculated, and different value distribution groups are compared in pairs to perform color difference analysis. If the compared difference value is smaller than a preset threshold value, the two value distribution families are considered to be the same color or similar color, and the two value distribution families are combined; if the difference is greater than or equal to the preset threshold, no operation is performed, and the two value distribution families are considered to be two families irrelevant to the color. After adjustment, a new distribution group is obtained, and the color composition of the target personnel and the value ratio of each color are determined according to the new distribution group. For example, after obtaining the pixel values of each distribution group in L, A and B domains, the proportion of the pixel number of each distribution group in L, A and B domains is counted, so as to obtain the value duty ratio of each color. The beneficial effect that sets up like this is that through adopting LAB color space, can draw the multiple colour composition of target personnel, not only can handle single colour's clothes, and through confirming the colour value ratio of target personnel, effectively improve target personnel's inquiry precision.
And S140, storing the value duty ratio into a database so as to facilitate the inquiry of target personnel according to the data in the database.
The image to be queried is a photo which is uploaded in advance and is consistent with the current wearing of the target person. When the target person goes to the camera shooting field, relevant information of the target person is uploaded in advance, for example, the information such as a photo, a name, an identity card and the like which the target person currently wears can be included. After the value duty ratio of the target personnel is obtained, storing the data of the value duty ratio of the target personnel to the position corresponding to the target personnel in a database, and perfecting the data of the target personnel. When the target person needs to be inquired, the color data of the target person can be input, and the person information consistent with the color data can be searched from the database. For example, the database stores people wearing yellow, red, 50% green and 50% blue and 30% green and 70% blue, and currently, people wearing green and blue need to be searched, then green and blue are input to obtain people wearing 50% green and 50% blue and wearing 30% green and 70% blue, and people wearing 50% green and 50% blue and wearing 30% green and 70% blue are used for searching for the target people, so that the problem that only a single color needs to be input for searching for the target people in the prior art is solved, and the searching efficiency of the target people is improved.
According to the technical scheme, the target personnel area is obtained from the target personnel image, clustering is carried out on the target personnel area, a plurality of value distribution families are obtained, the color value ratio of the target personnel is determined according to the pixel values of the value distribution families in the color space, and the target personnel is inquired according to the value ratio. The problem of in prior art carry out personnel data's storage and inquiry through face or single color, the target personnel inquiry precision that causes is low is solved, through confirming the color composition, has improved target personnel's efficiency and precision of inquiry.
Example two
Fig. 3 is a flow chart of a query method for a target person according to a second embodiment of the present invention, which is further optimized based on the foregoing embodiment, and the method may be executed by a query device for a target person. As shown in fig. 3, the method specifically includes the following steps:
s310, acquiring a target person image, and determining a target person area in the target person image.
S320, determining the value of the pixel point of the target personnel area in the RGB space, and clustering the target personnel area according to the value in the RGB space to obtain at least two value distribution families.
S330, performing color difference analysis on at least two value distribution families based on LAB space to obtain an adjustment result of the value distribution families.
The method comprises the steps of determining the value of each value distribution group pixel in an LAB color space, calculating the pixel mean value of each value distribution group, performing color difference analysis according to the pixel mean value, and adjusting the value distribution groups according to the color difference analysis result. If the target personnel images at three moments are extracted, color difference analysis is carried out on the target personnel at three moments, and the adjustment precision of the value distribution family is improved by combining the color difference analysis results at three moments.
In this embodiment, optionally, performing color difference analysis on at least two value distribution families based on the LAB space to obtain an adjustment result of the value distribution family, including: determining the color difference value of at least two value distribution families according to the pixel mean value of at least two value distribution families in the LAB space and the picture blurring degree; and adjusting the two value distribution groups with the color difference value smaller than a preset threshold value into the distribution group with the same color.
Specifically, values of at least two value distribution family pixel points in an LAB color space are determined, a pixel mean value of the value distribution family is calculated, color difference analysis is carried out according to the pixel mean value and the picture blurring degree, and color difference values among the value distribution families are determined. If the color difference value is greater than or equal to a preset threshold value, the two value distribution families are considered to be two families irrelevant to the color; if the color difference is smaller than the preset threshold, combining the two value distribution groups into one distribution group, wherein the number of the distribution groups is smaller than or equal to that of the value distribution groups. The beneficial effects of setting up like this lie in, the picture of different degree of blurring can produce different influences to the pixel value, through considering the degree of blurring of picture, makes the image of different qualities have different weight influence factor to influence the calculation of colour difference, improves the accuracy of colour difference analysis, and then improves the inquiry precision of target personnel.
In this embodiment, optionally, determining the color difference of at least two value distribution families according to the pixel mean value of at least two value distribution families in the LAB space and the image blurring degree includes: determining pixel difference values of any two value distribution families on L, A and B domains of the LAB space according to pixel average values of at least two value distribution families; determining the color difference value of any two value distribution families according to the pixel difference value and the picture blurring degree; the color difference is calculated by the following formula:
wherein Y is the color difference between the two value distribution groups, Δl is the pixel difference between the two value distribution groups in the L domain, Δa is the pixel difference between the two value distribution groups in the a domain, Δb is the pixel difference between the two value distribution groups in the B domain, L, a and B are preset parameters, and α is the image blurring degree.
Specifically, after determining the pixel mean values of at least two value distribution families at a certain moment, determining the difference value of the pixel values of the two value distribution families in L, A and B domains respectively. And calculating the color difference values of the two value distribution families according to the pixel difference values of the two value distribution families on the three domains and the image blurring degree.
The calculation formula of the color difference value can be expressed as follows:
wherein Y is the color difference between the two value distribution groups, Δl is the pixel difference between the two value distribution groups in the L domain, Δa is the pixel difference between the two value distribution groups in the a domain, Δb is the pixel difference between the two value distribution groups in the B domain, L, a and B are preset parameters, and α is the image blurring degree. For example, l may be set to 1, a may be set to 1+0.045Δa, and b may be set to 1+0.015 Δb. The beneficial effect of setting like this lies in, can be according to the automatic parameter that presets of different colours, makes the parameter that presets have different values to different colours, improves the inquiry precision of target personnel.
S340, determining the value ratio of the target personnel area according to the adjustment result of the value distribution family.
And S350, storing the value duty ratio into a database so as to facilitate the inquiry of target personnel according to the data in the database.
According to the embodiment of the invention, the target personnel area is obtained from the target personnel image, the target personnel area is clustered to obtain a plurality of value distribution families, the color difference values among different value distribution families are calculated according to the pixel values of the value distribution families in the LAB color space, the value distribution families of the same color are combined to obtain the distribution family of the target personnel color distribution, so that the color value occupation ratio of the target personnel is determined, and the target personnel is inquired according to the value occupation ratio. The problem of in prior art carry out personnel inquiry through face or single color, the target personnel inquiry precision that causes is low is solved, through confirming the color composition, has improved the efficiency and the precision of target personnel inquiry.
Example III
Fig. 4 is a block diagram of a query device for a target person according to a third embodiment of the present invention, which can execute the query method for a target person according to any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method. As shown in fig. 4, the apparatus specifically includes:
the target area determining module 401 is configured to acquire a target person image, and determine a target person area in the target person image;
the value distribution family obtaining module 402 is configured to determine values of pixel points of the target personnel area in an RGB space, and perform clustering processing on the target personnel area according to the values in the RGB space to obtain at least two value distribution families;
a duty ratio determining module 403, configured to determine a value duty ratio of the target personnel area according to the at least two value distribution families;
and the target query module 404 is configured to store the value duty ratio in the database, so as to query the target personnel according to the data in the database.
Optionally, the apparatus further comprises:
the color difference analysis module is used for carrying out color difference analysis on at least two value distribution families based on LAB space to obtain an adjustment result of the value distribution families;
accordingly, the duty ratio determining module 403 is specifically configured to:
and determining the value duty ratio of the target personnel area according to the adjustment result of the value distribution family.
Optionally, the target area determining module 401 includes:
the image acquisition unit is used for extracting frames of video data input into the target personnel according to a preset period to obtain target personnel images.
Optionally, the target area determining module 401 further includes:
the pedestrian frame determining unit is used for determining a target person area of the target person pedestrian frame, which is the largest, not blocked and/or random from the target person image; wherein the pedestrian frame is an image area including at least one pedestrian.
Optionally, the color difference analysis module includes:
the color difference value determining unit is used for determining the color difference value of at least two value distribution families according to the pixel mean value of at least two value distribution families in the LAB space and the picture blurring degree;
and the distribution group adjusting unit is used for adjusting the two valued distribution groups with the color difference value smaller than a preset threshold value into the distribution groups with the same color.
Optionally, the color difference determining unit is specifically configured to:
determining pixel difference values of any two value distribution families on L, A and B domains of the LAB space according to pixel average values of at least two value distribution families;
determining the color difference value of any two value distribution families according to the pixel difference value and the picture blurring degree;
the color difference is calculated by the following formula:
wherein Y is the color difference between the two value distribution groups, Δl is the pixel difference between the two value distribution groups in the L domain, Δa is the pixel difference between the two value distribution groups in the a domain, Δb is the pixel difference between the two value distribution groups in the B domain, L, a and B are preset parameters, and α is the image blurring degree.
Optionally, the apparatus further comprises:
the target detection module is used for obtaining a target person detection network based on the master RCNN network and the FPN network structure so as to detect and track target persons conveniently; the master RCNN network is used for detecting target personnel, and the FPN network is used for detecting target personnel of a small target.
According to the embodiment of the invention, the target personnel area is obtained from the target personnel image, the target personnel area is clustered to obtain a plurality of value distribution families, the color value ratio of the target personnel is determined according to the pixel values of the value distribution families in the color space, and the target personnel is inquired according to the value ratio. The problem of in prior art carry out personnel inquiry through face or single color, the target personnel inquiry precision that causes is low is solved, through confirming the color composition, has improved the efficiency and the precision of target personnel inquiry.
Example IV
Fig. 5 is a schematic structural diagram of a computer device according to a fourth embodiment of the present invention. Fig. 5 illustrates a block diagram of an exemplary computer device 500 suitable for use in implementing embodiments of the invention. The computer device 500 shown in fig. 5 is merely an example, and should not be construed as limiting the functionality and scope of use of embodiments of the present invention.
As shown in fig. 5, the computer device 500 is in the form of a general purpose computing device. The components of computer device 500 may include, but are not limited to: one or more processors or processing units 501, a system memory 502, and a bus 503 that connects the various system components (including the system memory 502 and processing units 501).
Bus 503 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, micro channel architecture (MAC) bus, enhanced ISA bus, video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.
Computer device 500 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by computer device 500 and includes both volatile and nonvolatile media, removable and non-removable media.
The system memory 502 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 504 and/or cache memory 505. The computer device 500 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 506 may be used to read from or write to non-removable, nonvolatile magnetic media (not shown in FIG. 5, commonly referred to as a "hard disk drive"). Although not shown in fig. 5, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In such cases, each drive may be coupled to bus 503 through one or more data medium interfaces. Memory 502 may include at least one program product having a set (e.g., at least one) of program modules configured to carry out the functions of embodiments of the invention.
A program/utility 508 having a set (at least one) of program modules 507 may be stored, for example, in memory 502, such program modules 507 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. Program modules 507 typically perform the functions and/or methods of the described embodiments of the invention.
The computer device 500 may also communicate with one or more external devices 509 (e.g., keyboard, pointing device, display 510, etc.), one or more devices that enable a user to interact with the computer device 500, and/or any devices (e.g., network card, modem, etc.) that enable the computer device 500 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 511. Moreover, the computer device 500 may also communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet, through a network adapter 512. As shown, network adapter 512 communicates with other modules of computer device 500 via bus 503. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with computer device 500, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.
The processing unit 501 executes programs stored in the system memory 502 to perform various functional applications and data processing, for example, to implement a target person query method provided in an embodiment of the present invention, including:
acquiring a target personnel image and determining a target personnel area in the target personnel image;
determining the value of the pixel point of the target personnel area in an RGB space, and carrying out clustering treatment on the target personnel area according to the value in the RGB space to obtain at least two value distribution families;
determining the value duty ratio of the target personnel area according to at least two value distribution families;
and storing the value duty ratio into a database so as to conveniently inquire the target personnel according to the data in the database.
Example five
The fifth embodiment of the present invention further provides a storage medium containing computer executable instructions, where a computer program is stored, and when the program is executed by a processor, the method for querying a target person provided by the embodiment of the present invention includes:
acquiring a target personnel image and determining a target personnel area in the target personnel image;
determining the value of the pixel point of the target personnel area in an RGB space, and carrying out clustering treatment on the target personnel area according to the value in the RGB space to obtain at least two value distribution families;
determining the value duty ratio of the target personnel area according to at least two value distribution families;
and storing the value duty ratio into a database so as to conveniently inquire the target personnel according to the data in the database.
The computer storage media of embodiments of the invention may take the form of any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
Note that the above is only a preferred embodiment of the present invention and the technical principle applied. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, while the invention has been described in connection with the above embodiments, the invention is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the invention, which is set forth in the following claims.
Claims (7)
1. A method for querying a target person, comprising:
acquiring a target personnel image and determining a target personnel area in the target personnel image;
determining the value of the pixel point of the target personnel area in an RGB space, and carrying out clustering treatment on the target personnel area according to the value in the RGB space to obtain at least two value distribution families;
determining the value duty ratio of the target personnel area according to the at least two value distribution families;
storing the value duty ratio into a database so as to facilitate the inquiry of target personnel according to the data in the database;
wherein, before determining the value ratio of the target personnel area according to the at least two value distribution families, the method further comprises:
performing color difference analysis on the at least two value distribution families based on LAB space to obtain an adjustment result of the value distribution families;
correspondingly, determining the value duty ratio of the target personnel area according to the at least two value distribution families comprises the following steps:
determining the value duty ratio of the target personnel area according to the adjustment result of the value distribution family;
performing color difference analysis on the at least two value distribution families based on LAB space to obtain an adjustment result of the value distribution families, wherein the adjustment result comprises the following steps:
determining color difference values of the at least two value distribution families according to pixel mean values of the at least two value distribution families in an LAB space and the image blurring degree;
two value distribution families with the color difference value smaller than a preset threshold value are adjusted to be the distribution family of the same color;
wherein determining the color difference value of the at least two value distribution families according to the pixel mean value of the at least two value distribution families in the LAB space and the image blurring degree comprises:
determining pixel difference values of any two value distribution families on L, A and B domains of the LAB space according to the pixel average values of the at least two value distribution families;
determining the color difference value of any two value distribution families according to the pixel difference value and the picture blurring degree;
the color difference is calculated by the following formula:
wherein Y is the color difference between the two value distribution groups, Δl is the pixel difference between the two value distribution groups in the L domain, Δa is the pixel difference between the two value distribution groups in the a domain, Δb is the pixel difference between the two value distribution groups in the B domain, L, a and B are preset parameters, and α is the image blurring degree.
2. The method of claim 1, wherein acquiring the image of the target person comprises:
and performing frame extraction on video data input into a target person according to a preset period to obtain an image of the target person.
3. The method of claim 1, wherein determining a target person region in the target person image comprises:
determining the target personnel area of the target personnel pedestrian frame with the maximum target personnel pedestrian frame, which is not shielded and/or random; wherein the pedestrian frame is an image area including at least one pedestrian.
4. The method of claim 1, further comprising, after acquiring the image of the target person:
based on the faster RCNN network and the FPN network structure, the target person detection network is obtained so as to detect and track the target person; the master RCNN network is used for detecting the target personnel, and the FPN network is used for detecting the target personnel of the small target.
5. A target person querying device, comprising:
the target area determining module is used for acquiring a target personnel image and determining a target personnel area in the target personnel image;
the value distribution family obtaining module is used for determining the value of the pixel point of the target personnel area in an RGB space, and carrying out clustering treatment on the target personnel area according to the value in the RGB space to obtain at least two value distribution families;
the duty ratio determining module is used for determining the value duty ratio of the target personnel area according to the at least two value distribution families;
the target query module is used for storing the value duty ratio into a database so as to query target personnel according to data in the database;
the color difference analysis module is used for carrying out color difference analysis on the at least two value distribution families based on LAB space to obtain an adjustment result of the value distribution families;
the duty ratio determining module is specifically configured to determine a value duty ratio of the target personnel area according to an adjustment result of the value distribution family;
the color difference analysis module comprises:
the color difference value determining unit is used for determining the color difference value of the at least two value distribution families according to the pixel mean value of the at least two value distribution families in the LAB space and the picture blurring degree;
the distribution group adjusting unit is used for adjusting the two value distribution groups with the color difference value smaller than a preset threshold value into the distribution group with the same color;
the color difference value determining unit is specifically configured to:
determining pixel difference values of any two value distribution families on L, A and B domains of the LAB space according to the pixel average values of the at least two value distribution families;
determining the color difference value of any two value distribution families according to the pixel difference value and the picture blurring degree;
the color difference is calculated by the following formula:
wherein Y is the color difference between the two value distribution groups, Δl is the pixel difference between the two value distribution groups in the L domain, Δa is the pixel difference between the two value distribution groups in the a domain, Δb is the pixel difference between the two value distribution groups in the B domain, L, a and B are preset parameters, and α is the image blurring degree.
6. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of querying a target person according to any of claims 1-4 when executing the program.
7. A storage medium containing computer executable instructions which, when executed by a computer processor, are for performing the method of querying a target person as claimed in any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010529142.2A CN111708907B (en) | 2020-06-11 | 2020-06-11 | Target person query method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010529142.2A CN111708907B (en) | 2020-06-11 | 2020-06-11 | Target person query method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111708907A CN111708907A (en) | 2020-09-25 |
| CN111708907B true CN111708907B (en) | 2023-07-18 |
Family
ID=72540096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010529142.2A Active CN111708907B (en) | 2020-06-11 | 2020-06-11 | Target person query method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111708907B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112613452B (en) * | 2020-12-29 | 2023-10-27 | 广东电网有限责任公司清远供电局 | Personnel line-crossing identification method, device, equipment and storage medium |
| CN114943886B (en) * | 2021-02-08 | 2025-01-21 | 北京小米移动软件有限公司 | Image recognition method, device, terminal, and storage medium |
| CN117830953B (en) * | 2024-01-09 | 2025-01-28 | 北京首华建设经营有限公司 | Intelligent network color image tracking monitoring method, system, terminal and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268583A (en) * | 2014-09-16 | 2015-01-07 | 上海交通大学 | Pedestrian re-recognition method and system based on color area features |
| CN107248154A (en) * | 2017-05-27 | 2017-10-13 | 江苏理工学院 | A kind of cloth aberration real-time on-line detecting method |
| CN107358242A (en) * | 2017-07-11 | 2017-11-17 | 浙江宇视科技有限公司 | Target area color identification method, device and monitor terminal |
| WO2019095998A1 (en) * | 2017-11-15 | 2019-05-23 | 深圳云天励飞技术有限公司 | Image recognition method and device, computer device and computer-readable storage medium |
| CN110826418A (en) * | 2019-10-15 | 2020-02-21 | 深圳和而泰家居在线网络科技有限公司 | Face feature extraction method and device |
| CN111160169A (en) * | 2019-12-18 | 2020-05-15 | 中国平安人寿保险股份有限公司 | Face detection method, device, equipment and computer readable storage medium |
-
2020
- 2020-06-11 CN CN202010529142.2A patent/CN111708907B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268583A (en) * | 2014-09-16 | 2015-01-07 | 上海交通大学 | Pedestrian re-recognition method and system based on color area features |
| CN107248154A (en) * | 2017-05-27 | 2017-10-13 | 江苏理工学院 | A kind of cloth aberration real-time on-line detecting method |
| CN107358242A (en) * | 2017-07-11 | 2017-11-17 | 浙江宇视科技有限公司 | Target area color identification method, device and monitor terminal |
| WO2019095998A1 (en) * | 2017-11-15 | 2019-05-23 | 深圳云天励飞技术有限公司 | Image recognition method and device, computer device and computer-readable storage medium |
| CN110826418A (en) * | 2019-10-15 | 2020-02-21 | 深圳和而泰家居在线网络科技有限公司 | Face feature extraction method and device |
| CN111160169A (en) * | 2019-12-18 | 2020-05-15 | 中国平安人寿保险股份有限公司 | Face detection method, device, equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111708907A (en) | 2020-09-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109918971B (en) | Method and device for detecting number of people in monitoring video | |
| WO2020206850A1 (en) | Image annotation method and device employing high-dimensional image | |
| Han et al. | Fast saliency-aware multi-modality image fusion | |
| KR101500711B1 (en) | Method for human tracking using color histograms | |
| CN111708907B (en) | Target person query method, device, equipment and storage medium | |
| CN102509118B (en) | Method for monitoring video retrieval | |
| CN102609720B (en) | Pedestrian detection method based on position correction model | |
| CN103268480A (en) | A visual tracking system and method | |
| CN107133969A (en) | A kind of mobile platform moving target detecting method based on background back projection | |
| CN110096945B (en) | Indoor monitoring video key frame real-time extraction method based on machine learning | |
| CN107705254B (en) | City environment assessment method based on street view | |
| CN113516609B (en) | Split-screen video detection method and device, computer equipment and storage medium | |
| CN101582163B (en) | Method for capturing clearest human face in video monitor images | |
| CN111723656B (en) | Smog detection method and device based on YOLO v3 and self-optimization | |
| CN110544259A (en) | A computer vision-based detection method for camouflaged human objects in complex backgrounds | |
| CN111259763A (en) | Target detection method, device, electronic device and readable storage medium | |
| CN111460964A (en) | Moving target detection method under low-illumination condition of radio and television transmission machine room | |
| US7606414B2 (en) | Fusion of color space data to extract dominant color | |
| CN101364304A (en) | Shadow detection method based on color invariance and Gaussian model | |
| CN107292892B (en) | Video frame image segmentation method and device | |
| Miller et al. | Person tracking in UAV video | |
| CN104299234B (en) | The method and system that rain field removes in video data | |
| CN116596771A (en) | Weak light image enhancement method based on self-adaptive illumination initialization | |
| CN112926676B (en) | False target identification method and device and computer equipment | |
| CN106650639A (en) | Monitoring system pedestrian re-identification method suitable for security and protection |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220921 Address after: 25 Financial Street, Xicheng District, Beijing 100033 Applicant after: CHINA CONSTRUCTION BANK Corp. Address before: 25 Financial Street, Xicheng District, Beijing 100033 Applicant before: CHINA CONSTRUCTION BANK Corp. Applicant before: Jianxin Financial Science and Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |