WO2019042195A1 - 一种人体目标身份识别方法及装置 - Google Patents

一种人体目标身份识别方法及装置 Download PDF

Info

Publication number
WO2019042195A1
WO2019042195A1 PCT/CN2018/101665 CN2018101665W WO2019042195A1 WO 2019042195 A1 WO2019042195 A1 WO 2019042195A1 CN 2018101665 W CN2018101665 W CN 2018101665W WO 2019042195 A1 WO2019042195 A1 WO 2019042195A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
face information
image
identified
searched
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.)
Ceased
Application number
PCT/CN2018/101665
Other languages
English (en)
French (fr)
Inventor
浦世亮
申琳
沈林杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Hikvision Digital Technology Co Ltd
Original Assignee
Hangzhou Hikvision Digital Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hangzhou Hikvision Digital Technology Co Ltd filed Critical Hangzhou Hikvision Digital Technology Co Ltd
Priority to US16/643,221 priority Critical patent/US11126828B2/en
Priority to EP18851366.7A priority patent/EP3678047B1/en
Publication of WO2019042195A1 publication Critical patent/WO2019042195A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172—Classification, e.g. identification
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168—Feature extraction; Face representation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/179—Human faces, e.g. facial parts, sketches or expressions metadata assisted face recognition

Definitions

  • the present application relates to the field of image processing technologies, and in particular, to a human body target identification method and apparatus.
  • the existing identification scheme generally includes: determining a face area in an image, performing feature extraction on the face area, and identifying an identity of the human body object in the image according to the extracted face feature.
  • the purpose of the embodiments of the present application is to provide a method and device for identifying a human target to improve the accuracy of identity recognition.
  • the embodiment of the present application provides a human body target identification method, including: acquiring an image to be identified; extracting a target feature of the human target to be identified in the image to be identified, as a target feature to be searched; Corresponding relationship between the target feature and the face information, searching for the face information corresponding to the target feature to be searched; wherein, in the corresponding relationship, a target feature and the corresponding face information belong to the same human target; based on the obtained person The face information determines the identity of the human target to be identified.
  • the step of acquiring an image to be identified may include: receiving an image to be recognized input by a user; or acquiring an image to be recognized from the specified collection device.
  • the step of searching for the face information corresponding to the target feature to be searched based on the correspondence between the pre-established target feature and the face information may include: performing the correspondence between the target feature and the face feature based on the pre-established a step of: determining a face feature corresponding to the target feature to be searched; and determining, according to the obtained face information, the identity of the human target to be identified, comprising: determining the to-be-identified based on the obtained face feature The identity of the human target.
  • the step of searching for the face information corresponding to the target feature to be searched based on the correspondence between the pre-established target feature and the face information may include: corresponding to the face image based on the pre-established target feature a step of: determining a face image corresponding to the target feature to be searched; determining the identity of the human target to be identified based on the obtained face information, comprising: determining the to-be-identified based on the obtained face image The identity of the human target.
  • the step of searching for the face information corresponding to the hash value to be searched according to the correspondence between the pre-established hash value and the face information may include: separately calculating the pre-established hash value and the person a similarity between each hash value included in the correspondence relationship of the face information and the hash value to be searched; and face information corresponding to the hash value whose similarity satisfies the preset condition is determined.
  • the method further includes: determining an acquisition attribute of the image to be identified as a collection attribute to be searched; wherein the collection attribute includes a time for collecting the image to be identified And/or a location; the step of searching for the face information corresponding to the target feature to be searched based on the correspondence between the pre-established target feature and the face information may include: pre-establishing the target feature and the face information In the corresponding relationship, the target collection attribute that is smaller than the preset threshold is found. The face information corresponding to the target feature is searched for in the face information corresponding to the target collection attribute.
  • the method may further include: determining whether the image to be identified is There is a face region that satisfies the definition requirement; if present, the face information in the image to be recognized is extracted; if not, the target feature of the human target to be recognized in the image to be recognized is extracted, as a to-be-find The steps of the target feature.
  • the embodiment of the present application further provides a human body target identification device, including: an acquisition module, configured to acquire an image to be identified; and a first extraction module, configured to extract a human target to be identified in the image to be identified a target feature, as a target feature to be searched; a search module, configured to search for face information corresponding to the target feature to be searched based on a correspondence between the target feature and the face information, wherein the corresponding relationship is The target feature and the corresponding face information belong to the same human target; the first determining module is configured to determine the identity of the human target to be identified based on the obtained face information.
  • a human body target identification device including: an acquisition module, configured to acquire an image to be identified; and a first extraction module, configured to extract a human target to be identified in the image to be identified a target feature, as a target feature to be searched; a search module, configured to search for face information corresponding to the target feature to be searched based on a correspondence between the target feature and the face information, wherein the corresponding relationship
  • the acquiring module may be specifically configured to: receive an image to be recognized input by the user; or obtain an image to be identified from the specified collection device.
  • the searching module may be configured to: search for a face feature corresponding to the target feature to be searched based on a correspondence between a target feature and a face feature that is pre-established; the first determining module may specifically For determining an identity of the human target to be identified based on the obtained facial features.
  • the searching module may be configured to: search for a face image corresponding to the target feature to be searched based on a corresponding relationship between the target feature and the face image; the first determining module may specifically And configured to: determine an identity of the human target to be identified based on the obtained face image.
  • the first extraction module may be configured to: extract an original target feature of the human target to be identified in the image to be identified, and calculate a hash value of the original target feature as a hash value to be searched;
  • the searching module may be configured to: search for the face information corresponding to the hash value to be searched based on the correspondence between the previously established hash value and the face information.
  • the searching module may be specifically configured to: separately calculate a similarity between each hash value included in the correspondence between the pre-established hash value and the face information and the hash value to be searched And determining face information corresponding to the hash value whose similarity satisfies the preset condition.
  • the device may further include: a second determining module, configured to determine an acquisition attribute of the image to be identified, as a collection attribute to be searched; wherein the collection attribute includes a time when the image to be recognized is collected And the location module; the searching module may be configured to: in a correspondence between the target feature and the face information that is pre-established, search for a target collection attribute that is smaller than a preset threshold value; In the face information corresponding to the target collection attribute, the face information corresponding to the target feature to be searched is searched for.
  • a second determining module configured to determine an acquisition attribute of the image to be identified, as a collection attribute to be searched; wherein the collection attribute includes a time when the image to be recognized is collected
  • the location module the searching module may be configured to: in a correspondence between the target feature and the face information that is pre-established, search for a target collection attribute that is smaller than a preset threshold value; In the face information corresponding to the target collection attribute, the face information corresponding to the target feature to be searched is searched for.
  • the device may further include: a determining module, configured to determine whether there is a face region that meets the definition requirement in the image to be identified; if yes, trigger a second extraction module, if not, trigger the a first extraction module, configured to extract face information in the image to be identified.
  • a determining module configured to determine whether there is a face region that meets the definition requirement in the image to be identified; if yes, trigger a second extraction module, if not, trigger the a first extraction module, configured to extract face information in the image to be identified.
  • an embodiment of the present application further provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory complete communication with each other through a communication bus;
  • the computer program is stored; and the processor is configured to implement any of the above-mentioned human target identification methods when executing the program stored in the memory.
  • an embodiment of the present application further provides a computer readable storage medium, where the computer readable storage medium stores a computer program, and when the computer program is executed by the processor, implements any one of the above-mentioned human target identities. recognition methods.
  • an embodiment of the present application further provides an executable program code for being executed to implement any of the above-described human target identification methods.
  • FIG. 1 is a schematic flowchart diagram of a human body target identification method according to an embodiment of the present application
  • FIG. 2 is a schematic diagram of an application scenario provided by an embodiment of the present application.
  • FIG. 3 is a schematic structural diagram of a human body target identification device according to an embodiment of the present disclosure.
  • FIG. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • the embodiment of the present application provides a human body target identification method and device.
  • the method and device can be applied to a device having an image processing function, which is not limited.
  • the human body target identification method provided by the embodiment of the present application is first described in detail below.
  • FIG. 1 is a schematic flowchart of a human target identification according to an embodiment of the present application, including:
  • S101 Acquire an image to be identified.
  • S102 Extract a target feature of the human target to be identified in the image to be identified as a target feature to be searched.
  • S103 Search for the face information corresponding to the target feature to be searched based on the correspondence between the target feature and the face information.
  • a target feature in the correspondence relationship and the corresponding face information belong to the same human target.
  • S104 Determine an identity of the human target to be identified based on the obtained face information.
  • S101 Acquire an image to be identified.
  • S101 may include: receiving an image to be recognized input by a user; or, as another implementation manner, S101 may include: acquiring an image to be recognized from a specified collection device.
  • the user can input an image containing the human body target; or, the acquisition device that collects the human body target can be determined, and an image including the human body target is acquired from the collection device.
  • the image to be identified may be obtained in other manners, and is not limited.
  • S102 Extract a target feature of the human target to be identified in the image to be identified as a target feature to be searched.
  • the target features of the human target may include colors, textures, sizes, and the like, and may also include the characteristics of the worn clothing, such as whether the type of the backpack or the pants, or the height and shape of the human target, and are not limited. .
  • S103 Search for the face information corresponding to the target feature to be searched based on the correspondence between the target feature and the face information.
  • a target feature in the correspondence relationship and the corresponding face information belong to the same human target.
  • the first type acquiring an image acquired by the collecting device; extracting a target feature of the human target in the image, and a facial feature of the face region satisfying the definition requirement in the image; establishing the target feature and the face information Corresponding relationship; wherein the target feature belongs to the same human target as the face information, the face information includes the face feature, or the face information includes the face feature and the image.
  • the image collected by the collecting device includes both a clear human body region and a clear human face region, and the two regions are aimed at the same human body target, the target features and faces of the same human target can be acquired in the image. feature.
  • the face feature may be stored as the face information corresponding to the target feature, or the image and the face feature may be stored as face information corresponding to the target feature.
  • one or more acquisition devices can be communicatively coupled to a server that transmits the acquired images to the server, the server extracts target features and facial features of the same human target in the image, and the target Features and face features are stored in a server local database or in a database connected to the server.
  • the server may store the target feature, the face feature, and the image together in a server local database or in a database connected to the server.
  • the execution entity of the embodiment of the present application and the server may be the same device or different devices.
  • multiple acquisition devices in the scenario of FIG. 2 can be communicatively connected to the same server.
  • the images collected by the collection devices are relatively clear, and the server can extract and store target features and facial features in the image.
  • the S103 may include: searching for the facial features corresponding to the target features to be searched based on the correspondence between the pre-established target features and the facial features.
  • S103 may include: searching for a face feature and an image corresponding to the target feature to be searched based on a correspondence relationship between the target feature and the face feature and the image, where the image includes a face region that satisfies the definition requirement.
  • the second type acquiring an image collected by the collecting device, where the image includes a face region satisfying the definition requirement; extracting a target feature of the human target in the image; establishing a correspondence between the target feature and the face information;
  • the face information includes the image.
  • the second method is different from the first method in that the face information in the second method does not include a face feature and only includes an image, and the image includes a face region. It can be understood that an image containing a face region can also be identified as face information.
  • the S103 may include: searching for an image corresponding to the target feature to be searched based on the correspondence between the target feature and the face image that is pre-established, where the image includes a person meeting the clarity requirement Face area.
  • the third type acquiring a face image and a human body target image of the same human target; extracting a face feature of the same human target in the face image, and a target feature of the same human target in the human target image; Corresponding relationship between the target feature of the human target and the face information, the face information includes the face feature, or the face information includes the face feature and the face image.
  • the target feature and the face feature of the same human target can be extracted in the same image, and in the third mode and the fourth mode below, the image can be in different images. Extract target features and face features of the same human target.
  • the S103 may include: searching for the facial features corresponding to the target features to be searched based on the correspondence between the pre-established target features and the facial features.
  • S103 may include: searching for a face feature and a face image corresponding to the target feature to be searched based on a correspondence between the target feature and the face feature and the face image.
  • a fourth type acquiring a face image and a human body target image of the same human target; extracting a target feature of the same human target in the human target image; establishing a correspondence between the target feature and the face information of the human target,
  • the face information includes the face image.
  • the fourth method is different from the third method in that the face information in the third mode includes a face feature, and the face information in the fourth mode does not include a face feature, and only the image is included. Contains the face area. It can be understood that an image containing a face region can also be identified as face information.
  • the S103 may include: searching for a face image corresponding to the target feature to be searched based on the correspondence between the target feature and the face image.
  • the target feature of the human target can be represented by a hash value.
  • the target feature extracted in S102 is a hash value, and the target feature in the pre-established correspondence relationship is also a hash value.
  • the S103 may include: extracting an original target feature of the human target to be identified in the image to be identified, and calculating a hash value of the original target feature as a hash value to be searched.
  • S104 may include: searching for face information corresponding to the hash value to be searched based on a correspondence between the previously established hash value and the face information.
  • the hash feature is used to represent the target feature, and the search efficiency can be improved.
  • the searching for the face information corresponding to the hash value to be searched based on the correspondence between the previously established hash value and the face information may include:
  • the similarity between hash values there are many ways to calculate the similarity. For example, you can use the Hamming distance between hash values to calculate the similarity between hash values.
  • the hash values in the correspondence may be arranged according to the order of similarity from high to low, and then the preset number of hashes are selected as the hash value whose similarity satisfies the preset condition, and the selected hash value is selected.
  • the corresponding face information is used as the face information corresponding to the target feature to be searched.
  • the hash value with the largest similarity may be used as the hash value whose similarity satisfies the preset condition; or, the hash value whose similarity is greater than the preset threshold may be used as the hash whose similarity satisfies the preset condition. Value, etc., is not limited.
  • the method further includes: determining an acquisition attribute of the image to be identified as an acquisition attribute to be searched.
  • the collection attribute includes a time and/or a location at which the image to be identified is collected.
  • S103 includes: in a corresponding relationship between the target feature and the face information, the target acquisition attribute that is smaller than the preset threshold is found in the face information corresponding to the target collection attribute; And searching for face information corresponding to the target feature to be searched.
  • the target features that are close to the acquisition time and/or the collection location of the image to be identified are searched to narrow the search range, and then further searched in the narrowed search range.
  • the movement trajectory of the human target to be identified is generally continuous, and the acquisition time and/or the acquisition image has a higher probability of containing the same human target in the similar image. Therefore, the application of the present embodiment is more accurate.
  • S104 Determine an identity of the human target to be identified based on the obtained face information.
  • the correspondence between the face information and the identity may be pre-stored, and the identity corresponding to the face information obtained in S103 may be searched according to the correspondence.
  • the correspondence between the target feature and the face information, and the correspondence between the face information and the identity may be stored in the same device, or may be stored in different devices.
  • the two types of correspondences may be stored in the execution body of the embodiment of the present application, and the two types of correspondences may be searched for in other devices.
  • the method may include: determining whether there is a face region that satisfies the definition requirement in the image to be identified; if present, extracting face information in the image to be recognized; If it does not exist, execute S104.
  • the face region is directly extracted as face information, and the face information is used for identity recognition, if there is no clearness in the image to be identified
  • the higher-level face area is used for identification using the embodiment shown in FIG.
  • the person X passes through the collection device A and the collection device E, and the image of the person X collected by the device A is captured, and the face region has a higher definition, and the person X collected by the collection device E In the image, the sharpness of the face area is poor.
  • the collecting device A transmits the image with higher definition to the server, and the server extracts and stores the target feature and the face feature of the human target in the image.
  • the unclear image collected by the collection device A is used as the acquired image to be recognized, and the target feature of the human target to be identified in the image is extracted as the target feature to be searched, and the target to be searched is searched for in the corresponding relationship stored by the server.
  • the face feature corresponding to the feature It can be understood that the target feature to be searched and the target feature stored by the server are the target features of the person X, and the two can be matched successfully. Therefore, the face feature stored by the server, that is, the face feature of the person X is found. .
  • the server searches the database for the identity corresponding to the face feature of the person X.
  • the human body can be determined by the corresponding relationship between the target feature and the face information and the target feature of the human target in the image to be recognized.
  • the face information of the target and then determining the identity of the human target based on the face information.
  • the embodiment of the present application further provides a human body target identification device.
  • FIG. 3 is a schematic structural diagram of a human body target identification device according to an embodiment of the present disclosure, including: an obtaining module 301, configured to acquire an image to be identified; and a first extraction module 302, configured to extract, to be identified, an image to be identified a target feature of the human target as the target feature to be searched; the search module 303 is configured to search for face information corresponding to the target feature to be searched based on the correspondence between the target feature and the face information that is pre-established; wherein, the corresponding A target feature in the relationship belongs to the same human target as the corresponding face information.
  • the first determining module 304 is configured to determine the identity of the human target to be identified based on the obtained face information.
  • the acquiring module 301 may be specifically configured to: receive an image to be recognized input by a user; or obtain an image to be recognized from the specified collection device.
  • the searching module 303 may be specifically configured to: search for a face feature corresponding to the target feature to be searched based on a correspondence between the target feature and the face feature that is pre-established; the first determining module 304 may specifically For determining an identity of the human target to be identified based on the obtained facial features.
  • the searching module 303 is specifically configured to: search for a face image corresponding to the target feature to be searched based on the correspondence between the target feature and the face image that is pre-established; the first determining module 304 may specifically And configured to: determine an identity of the human target to be identified based on the obtained face image.
  • the first extraction module 302 may be specifically configured to: extract an original target feature of the human target to be identified in the image to be identified, and calculate a hash value of the original target feature as a hash value to be searched.
  • the searching module 303 is specifically configured to: search for the face information corresponding to the hash value to be searched based on the correspondence between the hash value and the face information.
  • the searching module 303 may be specifically configured to: separately calculate a similarity between each hash value included in the correspondence between the pre-established hash value and the face information and the hash value to be searched. The degree information corresponding to the hash value whose similarity satisfies the preset condition is determined.
  • the device may further include: a second determining module (not shown), configured to determine an collection attribute of the image to be identified as an acquisition attribute to be searched; wherein the collection attribute includes The time and/or location of the to-be-identified image is collected; the searching module 303 is specifically configured to: in the correspondence between the target feature and the face information that is pre-established, the difference between the search and the to-be-searched collection attribute is less than The target acquisition attribute of the threshold is set; and the face information corresponding to the target feature to be searched is searched for in the face information corresponding to the target collection attribute.
  • a second determining module (not shown), configured to determine an collection attribute of the image to be identified as an acquisition attribute to be searched; wherein the collection attribute includes The time and/or location of the to-be-identified image is collected
  • the searching module 303 is specifically configured to: in the correspondence between the target feature and the face information that is pre-established, the difference between the search and the to-be-searched collection attribute is less than The target acquisition attribute of the threshold
  • the device may further include: a determining module and a second extracting module (not shown), wherein the determining module is configured to determine whether there is a person in the image to be identified that meets the definition requirement The face region; if present, triggers the second extraction module, if not, triggers the first extraction module 302; and the second extraction module is configured to extract face information in the image to be recognized.
  • a determining module is configured to determine whether there is a person in the image to be identified that meets the definition requirement The face region; if present, triggers the second extraction module, if not, triggers the first extraction module 302; and the second extraction module is configured to extract face information in the image to be recognized.
  • the embodiment of the present application further provides an electronic device, as shown in FIG. 4, including a processor 401, a communication interface 402, a memory 403, and a communication bus 404, wherein the processor 401, the communication interface 402, and the memory 403 pass through the communication bus 404.
  • the memory 403 is configured to store the computer program
  • the processor 401 is configured to: when the program stored on the memory 403 is executed, the following steps are performed: acquiring the image to be identified; and extracting the human body to be identified in the image to be identified Target feature of the target, as a target feature to be searched; searching for face information corresponding to the target feature to be searched based on a correspondence between the target feature and the face information; wherein, a target feature in the correspondence relationship The corresponding face information belongs to the same human target; and based on the obtained face information, the identity of the human target to be identified is determined.
  • the step of acquiring an image to be identified includes: receiving an image to be recognized input by a user; or acquiring an image to be recognized from a specified collection device.
  • the step of searching for the face information corresponding to the target feature to be searched based on the correspondence between the pre-established target feature and the face information includes: pre-establishing the target feature and the face feature Corresponding relationship, searching for a face feature corresponding to the target feature to be searched; determining the identity of the human target to be identified based on the obtained face information, comprising: determining the to-be based on the obtained face feature Identify the identity of a human target.
  • the step of searching for the face information corresponding to the target feature to be searched based on the correspondence between the target feature and the face information including: the pre-established target feature and the face image Corresponding relationship, searching for a face image corresponding to the target feature to be searched; determining the identity of the human target to be identified based on the obtained face information, comprising: determining the to-be-based based on the obtained face image Identify the identity of a human target.
  • the step of extracting a target feature of the human target to be identified in the image to be identified as a target feature to be searched includes: extracting an original target feature of the human target to be recognized in the image to be recognized, and calculating And the step of searching for the face information corresponding to the target feature to be searched, and the step of: searching for the hash information corresponding to the target feature to be searched according to the correspondence between the target feature and the face information, including: The face information corresponding to the hash value to be searched is searched for based on the correspondence between the previously established hash value and the face information.
  • the step of searching for the face information corresponding to the hash value to be searched according to the correspondence between the pre-established hash value and the face information includes: separately calculating a pre-established hash value and The similarity between each hash value included in the correspondence relationship of the face information and the hash value to be searched; the face information corresponding to the hash value whose similarity satisfies the preset condition is determined.
  • the processor 401 is further configured to: after the step of acquiring an image to be identified, determining an collection attribute of the image to be identified as a collection attribute to be searched; wherein the collection attribute includes The step of collecting the time and/or the location of the image to be identified; the step of searching for the face information corresponding to the target feature to be searched based on the correspondence between the pre-established target feature and the face information, including: pre-established In the corresponding relationship between the target feature and the face information, searching for a target collection attribute whose difference from the to-be-searched collection attribute is less than a preset threshold; and searching for the target to be searched in the face information corresponding to the target collection attribute Face information corresponding to the feature.
  • the processor 401 is further configured to: after the step of acquiring the image to be identified, before the step of determining the identity of the human target to be identified based on the obtained face information, Determining, in the image to be recognized, whether there is a face region that satisfies the definition requirement; if yes, extracting face information in the image to be recognized; if not, performing extracting the human body to be recognized in the image to be recognized The target feature of the target as a step to find the target feature.
  • the communication bus mentioned in the above electronic device may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus.
  • PCI Peripheral Component Interconnect
  • EISA Extended Industry Standard Architecture
  • the communication bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.
  • the communication interface is used for communication between the above electronic device and other devices.
  • the memory may include a random access memory (RAM), and may also include a non-volatile memory (NVM), such as at least one disk storage.
  • RAM random access memory
  • NVM non-volatile memory
  • the memory may also be at least one storage device located away from the aforementioned processor.
  • the above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; or may be a digital signal processing (DSP), dedicated integration.
  • CPU central processing unit
  • NP network processor
  • DSP digital signal processing
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array
  • the embodiment of the present application provides a computer readable storage medium, where the computer readable storage medium stores a computer program, and when the computer program is executed by the processor, the following steps are performed: acquiring an image to be identified; and extracting the image to be recognized
  • the target feature of the target to be identified is used as the target feature to be searched; and the face information corresponding to the target feature to be searched is searched based on the correspondence between the target feature and the face information; wherein the corresponding relationship is
  • the target feature belongs to the same human target as the corresponding face information; and based on the obtained face information, the identity of the target to be identified is determined.
  • the step of acquiring an image to be identified includes: receiving an image to be recognized input by a user; or acquiring an image to be recognized from a specified collection device.
  • the step of searching for the face information corresponding to the target feature to be searched based on the correspondence between the pre-established target feature and the face information includes: pre-establishing the target feature and the face feature Corresponding relationship, searching for a face feature corresponding to the target feature to be searched; determining the identity of the human target to be identified based on the obtained face information, comprising: determining the to-be based on the obtained face feature Identify the identity of a human target.
  • the step of searching for the face information corresponding to the target feature to be searched based on the correspondence between the target feature and the face information including: the pre-established target feature and the face image Corresponding relationship, searching for a face image corresponding to the target feature to be searched; determining the identity of the human target to be identified based on the obtained face information, comprising: determining the to-be-based based on the obtained face image Identify the identity of a human target.
  • the step of extracting a target feature of the human target to be identified in the image to be identified as a target feature to be searched includes: extracting an original target feature of the human target to be recognized in the image to be recognized, and calculating And the step of searching for the face information corresponding to the target feature to be searched, and the step of: searching for the hash information corresponding to the target feature to be searched according to the correspondence between the target feature and the face information, including: The face information corresponding to the hash value to be searched is searched for based on the correspondence between the previously established hash value and the face information.
  • the step of searching for the face information corresponding to the hash value to be searched according to the correspondence between the pre-established hash value and the face information includes: separately calculating a pre-established hash value and The similarity between each hash value included in the correspondence relationship of the face information and the hash value to be searched; the face information corresponding to the hash value whose similarity satisfies the preset condition is determined.
  • the following steps may be further implemented: after the step of acquiring the image to be identified, determining an collection attribute of the image to be identified as an acquisition attribute to be searched;
  • the collecting attribute includes a time and/or a location for collecting the image to be identified;
  • the target collection attribute whose difference from the to-be-searched collection attribute is less than a preset threshold is searched; in the face information corresponding to the target collection attribute, the search center The face information corresponding to the search target feature is described.
  • the following steps may be further implemented: after the step of acquiring an image to be identified, determining, according to the obtained face information, the target of the human body to be identified Before the step of identifying, determining whether there is a face region satisfying the definition requirement in the image to be identified; if present, extracting face information in the image to be recognized; if not, performing the extracting the to-be-identified
  • the target feature of the human target to be identified in the image as a step of finding the target feature.
  • the embodiment of the present application also discloses an executable program code for being executed to implement any of the above-described human target identification methods.
  • the various embodiments in the present specification are described in a related manner, and the same or similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments.
  • the apparatus embodiment shown in FIG. 3, the electronic device embodiment shown in FIG. 4, the above computer readable storage medium embodiment, and the above executable program code embodiment are basically similar to FIG.
  • the method embodiment shown in 2 so the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment shown in FIG. 1-2.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Image Analysis (AREA)
  • Collating Specific Patterns (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本申请实施例提供了一种人体目标身份识别方法及装置，方法包括：提取图像中待识别人体目标的目标特征，作为待查找目标特征，基于预先建立的目标特征与人脸信息的对应关系，查找待查找目标特征对应的人脸信息；基于得到的人脸信息，确定待识别人体目标的身份；可见，本方案中，不需要提取图像中的人脸特征，即使图像中的人脸区域不清晰，或者被其他物体遮挡，也不会降低身份识别的准确性；因此，应用本方案，提高了身份识别的准确性。

Description

一种人体目标身份识别方法及装置
本申请要求于2017年8月31日提交中国专利局、申请号为201710769677.5、发明名称为“一种人体目标身份识别方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及图像处理技术领域，特别是涉及一种人体目标身份识别方法及装置。
背景技术
在视频监控过程中，通常需要对监控图像中出现的人体目标进行身份识别。现有的识别方案通常包括：确定图像中的人脸区域，对人脸区域进行特征提取，根据所提取的人脸特征，识别图像中人体目标的身份。
上述方案中，如果图像中的人脸区域不清晰，或者被其他物体遮挡，则确定出的身份不准确。
发明内容
本申请实施例的目的在于提供一种人体目标身份识别方法及装置，以提高身份识别的准确性。
为达到上述目的，本申请实施例提供了一种人体目标身份识别方法，包括：获取待识别图像；提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征；基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息；其中，所述对应关系中一份目标特征与对应的人脸信息属于同一人体目标；基于得到的人脸信息，确定所述待识别人体目标的身份。
可选的，所述获取待识别图像的步骤，可以包括：接收用户输入的待识别图像；或者，从指定采集设备中获取待识别图像。
可选的，所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，可以包括：基于预先建立的目标特征与人脸特征的对应关系，查找所述待查找目标特征对应的人脸特征；所 述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤，包括：基于得到的人脸特征，确定所述待识别人体目标的身份。
可选的，所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，可以包括：基于预先建立的目标特征与人脸图像的对应关系，查找所述待查找目标特征对应的人脸图像；所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤，包括：基于得到的人脸图像，确定所述待识别人体目标的身份。
可选的，所述提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征的步骤，可以包括：提取所述待识别图像中待识别人体目标的原始目标特征，计算所述原始目标特征的哈希值，作为待查找哈希值；所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，可以包括：基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息。
可选的，所述基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息的步骤，可以包括：分别计算预先建立的哈希值与人脸信息的对应关系中所包括的各哈希值与所述待查找哈希值之间的相似度；确定相似度满足预设条件的哈希值对应的人脸信息。
可选的，在所述获取待识别图像的步骤之后，还可以包括：确定所述待识别图像的采集属性，作为待查找采集属性；其中，所述采集属性包含采集所述待识别图像的时刻和/或地点；所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，可以包括：在预先建立的目标特征与人脸信息的对应关系中，查找与所述待查找采集属性的差值小于预设阈值的目标采集属性；在所述目标采集属性对应的人脸信息中，查找所述待查找目标特征对应的人脸信息。
可选的，在所述获取待识别图像的步骤之后，在所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤之前，还可以包括：判断所述待识别图像中是否存在满足清晰度要求的人脸区域；如果存在，提取所述待识别图像中的人脸信息；如果不存在，执行所述提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征的步骤。
为达到上述目的，本申请实施例还提供了一种人体目标身份识别装置，包括：获取模块，用于获取待识别图像；第一提取模块，用于提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征；查找模块，用于基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息；其中，所述对应关系中一份目标特征与对应的人脸信息属于同一人体目标；第一确定模块，用于基于得到的人脸信息，确定所述待识别人体目标的身份。
可选的，所述获取模块，具体可以用于：接收用户输入的待识别图像；或者，从指定采集设备中获取待识别图像。
可选的，所述查找模块，具体可以用于：基于预先建立的目标特征与人脸特征的对应关系，查找所述待查找目标特征对应的人脸特征；所述第一确定模块，具体可以用于：基于得到的人脸特征，确定所述待识别人体目标的身份。
可选的，所述查找模块，具体可以用于：基于预先建立的目标特征与人脸图像的对应关系，查找所述待查找目标特征对应的人脸图像；所述第一确定模块，具体可以用于：基于得到的人脸图像，确定所述待识别人体目标的身份。
可选的，所述第一提取模块，具体可以用于：提取所述待识别图像中待识别人体目标的原始目标特征，计算所述原始目标特征的哈希值，作为待查找哈希值；所述查找模块，具体可以用于：基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息。
可选的，所述查找模块，具体可以用于：分别计算预先建立的哈希值与人脸信息的对应关系中所包括的各哈希值与所述待查找哈希值之间的相似度；确定相似度满足预设条件的哈希值对应的人脸信息。
可选的，所述装置还可以包括：第二确定模块，用于确定所述待识别图像的采集属性，作为待查找采集属性；其中，所述采集属性包含采集所述待识别图像的时刻和/或地点；所述查找模块，具体可以用于：在预先建立的目标特征与人脸信息的对应关系中，查找与所述待查找采集属性的差值小于预 设阈值的目标采集属性；在所述目标采集属性对应的人脸信息中，查找所述待查找目标特征对应的人脸信息。
可选的，所述装置还可以包括：判断模块，用于判断所述待识别图像中是否存在满足清晰度要求的人脸区域；如果存在，触发第二提取模块，如果不存在，触发所述第一提取模块；第二提取模块，用于提取所述待识别图像中的人脸信息。
为达到上述目的，本申请实施例还提供了一种电子设备，包括处理器、通信接口、存储器和通信总线，其中，处理器，通信接口，存储器通过通信总线完成相互间的通信；存储器，用于存放计算机程序；处理器，用于执行存储器上所存放的程序时，实现上述任一种人体目标身份识别方法。
为达到上述目的，本申请实施例还提供了一种计算机可读存储介质，所述计算机可读存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现上述任一种人体目标身份识别方法。
为达到上述目的，本申请实施例还提供了一种可执行程序代码，所述可执行程序代码用于被运行以实现上述任一种人体目标身份识别方法。
应用本申请所示实施例，提取图像中待识别人体目标的目标特征，作为待查找目标特征，基于预先建立的目标特征与人脸信息的对应关系，查找待查找目标特征对应的人脸信息；基于得到的人脸信息，确定待识别人体目标的身份；可见，本方案中，不需要提取图像中的人脸特征，即使图像中的人脸区域不清晰，或者被其他物体遮挡，也不会降低身份识别的准确性；因此，应用本方案，提高了身份识别的准确性。
当然，实施本申请的任一产品或方法并不一定需要同时达到以上所述的所有优点。
附图说明
为了更清楚地说明本申请实施例和现有技术的技术方案，下面对实施例和现有技术中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的一种人体目标身份识别方法的流程示意图；
图2为本申请实施例提供的一种应用场景示意图；
图3为本申请实施例提供的一种人体目标身份识别装置的结构示意图；
图4为本申请实施例提供的一种电子设备的结构示意图。
具体实施方式
为使本申请的目的、技术方案、及优点更加清楚明白，以下参照附图并举实施例，对本申请进一步详细说明。显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
为了解决上述技术问题，本申请实施例提供了一种人体目标身份识别方法及装置。该方法及装置可以应用于具有图像处理功能的设备，具体不做限定。
下面首先对本申请实施例提供的一种人体目标身份识别方法进行详细说明。
图1为本申请实施例提供的一种人体目标身份识别的流程示意图，包括：
S101：获取待识别图像。
S102：提取该待识别图像中待识别人体目标的目标特征，作为待查找目标特征。
S103：基于预先建立的目标特征与人脸信息的对应关系，查找该待查找目标特征对应的人脸信息。其中，该对应关系中一份目标特征与对应的人脸信息属于同一人体目标。
S104：基于得到的人脸信息，确定该待识别人体目标的身份。
应用本申请图1所示实施例，提取图像中待识别人体目标的目标特征，作为待查找目标特征，基于预先建立的目标特征与人脸信息的对应关系，查找待查找目标特征对应的人脸信息；基于得到的人脸信息，确定待识别人体 目标的身份；可见，本方案中，不需要提取图像中的人脸特征，即使图像中的人脸区域不清晰，或者被其他物体遮挡，也不会降低身份识别的准确性；因此，应用本方案，提高了身份识别的准确性。
下面对图1所示实施进行详细说明：
S101：获取待识别图像。
作为一种实施方式，S101可以包括：接收用户输入的待识别图像；或者，作为另一种实施方式，S101可以包括：从指定采集设备中获取待识别图像。
可以理解，当需要识别某人体目标的身份时，用户可以输入包含该人体目标的图像；或者，可以确定采集到该人体目标的采集设备，从该采集设备中获取包含该人体目标的图像。
或者，也可以采用其他方式获取待识别图像，具体不做限定。
S102：提取该待识别图像中待识别人体目标的目标特征，作为待查找目标特征。
人体目标的目标特征可以包含颜色、纹理、尺寸等特征，也可以包含所穿戴的服饰特征，比如，是否背包、衣裤类型等，也可以包含人体目标的身高、体型等特征，具体不做限定。
在图像中提取目标特征的方式有很多，比如，利用边缘检测算法，检测图像中的人体目标区域，再提取该区域的图像特征；或者，利用预先训练得到的神经网络提取图像中人体目标的目标特征，等等，具体不做限定。
S103：基于预先建立的目标特征与人脸信息的对应关系，查找该待查找目标特征对应的人脸信息。其中，该对应关系中一份目标特征与对应的人脸信息属于同一人体目标。
建立该对应关系的方式有多种，下面介绍几种具体方式：
第一种：获取采集设备采集的图像；提取所述图像中人体目标的目标特征、以及所述图像中满足清晰度要求的人脸区域的人脸特征；建立所述目标特征与人脸信息的对应关系；其中，所述目标特征与所述人脸信息属于同一人体目标，所述人脸信息包括所述人脸特征，或者，所述人脸信息包括所述 人脸特征及所述图像。
可以理解，如果采集设备采集的图像中既包含清晰的人体区域又包含清晰的人脸区域，且两个区域针对同一人体目标，则可以在该图像中获取到同一人体目标的目标特征及人脸特征。可以将该人脸特征作为人脸信息与该目标特征对应存储，也可以将该图像及该人脸特征作为人脸信息与该目标特征对应存储。
举例来说，一台或多台采集设备可以与一台服务器通信连接，这些采集设备将采集的图像发送给该服务器，服务器提取图像中同一人体目标的目标特征及人脸特征，并将该目标特征及人脸特征存储至服务器本地数据库中，或者存储至与服务器相连的数据库中。
或者，服务器也可以将该目标特征、人脸特征及该图像一并存储至服务器本地数据库中，或者存储至与服务器相连的数据库中。本申请实施例的执行主体与该服务器可以为同一设备，也可以为不同设备。
比如，图2场景中的多台采集设备可以与同一服务器通信连接，这些采集设备采集到的图像较清晰，该服务器可以提取并存储图像中的目标特征及人脸特征。
如果应用第一种方式建立对应关系，则S103可以包括：基于预先建立的目标特征与人脸特征的对应关系，查找所述待查找目标特征对应的人脸特征。或者，S103可以包括：基于预先建立的目标特征与人脸特征及图像的对应关系，查找所述待查找目标特征对应的人脸特征及图像，该图像中包含满足清晰度要求的人脸区域。
第二种：获取采集设备采集的图像，所述图像中包含满足清晰度要求的人脸区域；提取所述图像中人体目标的目标特征；建立所述目标特征与人脸信息的对应关系；所述人脸信息包括所述图像。
第二种方式与第一种方式的不同之处在于，第二种方式中的人脸信息不包含人脸特征，仅包含图像，该图像中包含人脸区域。可以理解，包含人脸区域的图像也可以作为人脸信息进行身份识别。
如果应用第二种方式建立对应关系，则S103可以包括：基于预先建立的 目标特征与人脸图像的对应关系，查找所述待查找目标特征对应的图像，该图像中包含满足清晰度要求的人脸区域。
第三种：获取同一人体目标的人脸图像及人体目标图像；提取所述人脸图像中该同一人体目标的人脸特征、以及所述人体目标图像中该同一人体目标的目标特征；建立所述人体目标的目标特征与人脸信息的对应关系，所述人脸信息包括所述人脸特征，或者，所述人脸信息包括所述人脸特征及所述人脸图像。
上述第一种方式与第二种方式中，可以在同一张图像中提取到同一人体目标的目标特征及人脸特征，而在第三种方式及下面第四种方式中，可以在不同图像中提取同一人体目标的目标特征及人脸特征。
如果应用第三种方式建立对应关系，则S103可以包括：基于预先建立的目标特征与人脸特征的对应关系，查找所述待查找目标特征对应的人脸特征。或者，S103可以包括：基于预先建立的目标特征与人脸特征及人脸图像的对应关系，查找所述待查找目标特征对应的人脸特征及人脸图像。
第四种：获取同一人体目标的人脸图像及人体目标图像；提取所述人体目标图像中该同一人体目标的目标特征；建立所述人体目标的目标特征与人脸信息的对应关系，所述人脸信息包括所述人脸图像。
第四种方式与第三种方式的不同之处在于，第三中方式中的人脸信息包含人脸特征，第四种方式中的人脸信息不包含人脸特征，仅包含图像，该图像中包含人脸区域。可以理解，包含人脸区域的图像也可以作为人脸信息进行身份识别。
如果应用第四种方式建立对应关系，则S103可以包括：基于预先建立的目标特征与人脸图像的对应关系，查找所述待查找目标特征对应的人脸图像。
作为一种实施方式，人体目标的目标特征可以用哈希值来表示。这种实施方式中，S102中提取的目标特征为哈希值，预先建立的对应关系中的目标特征也为哈希值。
具体的，S103可以包括：提取所述待识别图像中待识别人体目标的原始目标特征，计算所述原始目标特征的哈希值，作为待查找哈希值。
S104可以包括：基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息。
应用本实施方式，用哈希值来表示目标特征，可以提高查找效率。
在本实施方式中，基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息，可以包括：
分别计算预先建立的哈希值与人脸信息的对应关系中所包括的各哈希值与所述待查找哈希值之间的相似度；确定相似度满足预设条件的哈希值对应的人脸信息。
计算相似度的方式有很多，比如可以利用哈希值之间的汉明距离，计算哈希值之间的相似度。可以按照相似度由高到低的顺序，排列该对应关系中的各哈希值，然后选取前预设数量个哈希值作为相似度满足预设条件的哈希值，将选取的哈希值对应的人脸信息作为待查找目标特征对应的人脸信息。
或者，也可以仅将相似度最大的哈希值作为相似度满足预设条件的哈希值；或者，也可以将相似度大于预设阈值的哈希值作为相似度满足预设条件的哈希值，等等，具体不做限定。
作为一种实施方式，在S101之后，还可以包括：确定所述待识别图像的采集属性，作为待查找采集属性。其中，所述采集属性包含采集所述待识别图像的时刻和/或地点。
S103包括：在预先建立的目标特征与人脸信息的对应关系中，查找与所述待查找采集属性的差值小于预设阈值的目标采集属性；在所述目标采集属性对应的人脸信息中，查找所述待查找目标特征对应的人脸信息。
在本实施方式中，先在所建立的对应关系中查找与待识别图像的采集时间和/或采集地点较相近的目标特征，以缩小查找范围，然后在缩小后的查找范围中作进一步的查找。
可以理解，如果采用计算哈希值相似度的查找方式，本实施方式中，不需要计算待查找哈希值与对应关系中所有哈希值的相似度，而是先根据采集属性过滤掉一部分哈希值，仅计算待查找哈希值与剩余部分哈希值的相似度， 降低了计算量，进一步提高了查找效率。
再者，待识别人体目标的移动轨迹一般是连续的，采集时间和/或采集地点较相近的图像中包含同一人体目标的概率较大，因此，应用本实施方式查找更准确。
S104：基于得到的人脸信息，确定所述待识别人体目标的身份。
作为一种实施方式，可以预先存储人脸信息与身份的对应关系，根据该对应关系，查找S103中得到的人脸信息对应的身份。
上述目标特征与人脸信息的对应关系、与该人脸信息与身份的对应关系可以存储为同一设备中，也可以存储为不同设备中。本申请实施例的执行主体中可以存储这两类对应关系，也可以到其他设备中查找这两类对应关系。
作为一种实施方式，在S101之后、S104之前，可以包括：判断所述待识别图像中是否存在满足清晰度要求的人脸区域；如果存在，提取所述待识别图像中的人脸信息；如果不存在，执行S104。
在本实施方式中，如果待识别图像中存在清晰度较高的人脸区域，则直接提取该人脸区域作为人脸信息，利用该人脸信息进行身份识别，如果待识别图像中不存在清晰度较高的人脸区域，再利用图1所示实施例进行身份识别。
下面介绍一个具体的实施例：
假设在图2所示场景中，人员X经过采集设备A和采集设备E，采集设备A采集到的人员X的图像中，人脸区域清晰度较高，而采集设备E采集到的人员X的图像中，人脸区域清晰度较差。采集设备A将该清晰度较高的图像发送给服务器，服务器提取并存储该图像中的人体目标的目标特征及人脸特征。
假设将采集设备A采集到的不清晰的图像作为获取到的待识别图像，提取该图像中待识别人体目标的目标特征，作为待查找目标特征，在服务器存储的对应关系中查找该待查找目标特征对应的人脸特征。可以理解，该待查找目标特征与上述服务器存储的目标特征均为人员X的目标特征，二者可以匹配成功，因此，便查找到了上述服务器存储的人脸特征，也即人员X的人脸特征。
假设一数据库中存储了人脸特征与身份的对应关系，服务器在该数据库中查找上述人员X的人脸特征对应的身份。
可见，在本方案中，即使待识别图像中的人脸区域不清晰，仍可以通过预先建立的目标特征与人脸信息的对应关系、以及待识别图像中的人体目标的目标特征确定出该人体目标的人脸信息，再根据该人脸信息确定出该人体目标的身份。
尤其是在能够实现全景细节的大范围监控场景中，对于单台采集设备来说，其恰好采集到目标正脸的图像、并且该图像清晰度较高的概率较低；但是，对于同一场景中的多台采集设备来说，存在一台采集设备恰好采集到目标正脸的图像、并且该图像清晰度较高的概率则较高；应用本申请实施例，根据该包含目标正脸、且清晰度较高的图像，建立目标特征与人脸信息的对应关系，再利用该对应关系，识别待识别图像中人体目标的身份，即使待识别图像不清晰，也不会降低身份识别的准确性。
应用本申请图1所示实施例，提取图像中待识别人体目标的目标特征，作为待查找目标特征，基于预先建立的目标特征与人脸信息的对应关系，查找待查找目标特征对应的人脸信息；基于得到的人脸信息，确定待识别人体目标的身份；可见，本方案中，不需要提取图像中的人脸特征，即使图像中的人脸区域不清晰，或者被其他物体遮挡，也不会降低身份识别的准确性；因此，应用本方案，提高了身份识别的准确性。
与上述方法实施例相对应，本申请实施例还提供一种人体目标身份识别装置。
图3为本申请实施例提供的一种人体目标身份识别装置的结构示意图，包括：获取模块301，用于获取待识别图像；第一提取模块302，用于提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征；查找模块303，用于基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息；其中，所述对应关系中一份目标特征与对应的人脸信息属于同一人体目标；第一确定模块304，用于基于得到的人脸信息，确定所述待识别人体目标的身份。
作为一种实施方式，获取模块301，具体可以用于：接收用户输入的待识别图像；或者，从指定采集设备中获取待识别图像。
作为一种实施方式，查找模块303，具体可以用于：基于预先建立的目标特征与人脸特征的对应关系，查找所述待查找目标特征对应的人脸特征；第一确定模块304，具体可以用于：基于得到的人脸特征，确定所述待识别人体目标的身份。
作为一种实施方式，查找模块303，具体可以用于：基于预先建立的目标特征与人脸图像的对应关系，查找所述待查找目标特征对应的人脸图像；第一确定模块304，具体可以用于：基于得到的人脸图像，确定所述待识别人体目标的身份。
作为一种实施方式，第一提取模块302，具体可以用于：提取所述待识别图像中待识别人体目标的原始目标特征，计算所述原始目标特征的哈希值，作为待查找哈希值；查找模块303，具体可以用于：基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息。
作为一种实施方式，查找模块303，具体可以用于：分别计算预先建立的哈希值与人脸信息的对应关系中所包括的各哈希值与所述待查找哈希值之间的相似度；确定相似度满足预设条件的哈希值对应的人脸信息。
作为一种实施方式，所述装置还可以包括：第二确定模块(图中未示出)，用于确定所述待识别图像的采集属性，作为待查找采集属性；其中，所述采集属性包含采集所述待识别图像的时刻和/或地点；查找模块303，具体可以用于：在预先建立的目标特征与人脸信息的对应关系中，查找与所述待查找采集属性的差值小于预设阈值的目标采集属性；在所述目标采集属性对应的人脸信息中，查找所述待查找目标特征对应的人脸信息。
作为一种实施方式，所述装置还可以包括：判断模块和第二提取模块(图中未示出)，其中，判断模块，用于判断所述待识别图像中是否存在满足清晰度要求的人脸区域；如果存在，触发第二提取模块，如果不存在，触发第一提取模块302；第二提取模块，用于提取所述待识别图像中的人脸信息。
应用本申请图3所示实施例，提取图像中待识别人体目标的目标特征，作 为待查找目标特征，基于预先建立的目标特征与人脸信息的对应关系，查找待查找目标特征对应的人脸信息；基于得到的人脸信息，确定待识别人体目标的身份；可见，本方案中，不需要提取图像中的人脸特征，即使图像中的人脸区域不清晰，或者被其他物体遮挡，也不会降低身份识别的准确性；因此，应用本方案，提高了身份识别的准确性。
本申请实施例还提供了一种电子设备，如图4所示，包括处理器401、通信接口402、存储器403和通信总线404，其中，处理器401，通信接口402，存储器403通过通信总线404完成相互间的通信，存储器403，用于存放计算机程序；处理器401，用于执行存储器403上所存放的程序时，实现如下步骤：获取待识别图像；提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征；基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息；其中，所述对应关系中一份目标特征与对应的人脸信息属于同一人体目标；基于得到的人脸信息，确定所述待识别人体目标的身份。
作为一种实施方式，所述获取待识别图像的步骤，包括：接收用户输入的待识别图像；或者，从指定采集设备中获取待识别图像。
作为一种实施方式，所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：基于预先建立的目标特征与人脸特征的对应关系，查找所述待查找目标特征对应的人脸特征；所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤，包括：基于得到的人脸特征，确定所述待识别人体目标的身份。
作为一种实施方式，所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：基于预先建立的目标特征与人脸图像的对应关系，查找所述待查找目标特征对应的人脸图像；所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤，包括：基于得到的人脸图像，确定所述待识别人体目标的身份。
作为一种实施方式，所述提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征的步骤，包括：提取所述待识别图像中待识别人体目标的原始目标特征，计算所述原始目标特征的哈希值，作为待查找哈希 值；所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息。
作为一种实施方式，所述基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息的步骤，包括：分别计算预先建立的哈希值与人脸信息的对应关系中所包括的各哈希值与所述待查找哈希值之间的相似度；确定相似度满足预设条件的哈希值对应的人脸信息。
作为一种实施方式，处理器401还用于实现如下步骤：在所述获取待识别图像的步骤之后，确定所述待识别图像的采集属性，作为待查找采集属性；其中，所述采集属性包含采集所述待识别图像的时刻和/或地点；所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：在预先建立的目标特征与人脸信息的对应关系中，查找与所述待查找采集属性的差值小于预设阈值的目标采集属性；在所述目标采集属性对应的人脸信息中，查找所述待查找目标特征对应的人脸信息。
作为一种实施方式，处理器401还用于实现如下步骤：在所述获取待识别图像的步骤之后，在所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤之前，判断所述待识别图像中是否存在满足清晰度要求的人脸区域；如果存在，提取所述待识别图像中的人脸信息；如果不存在，执行所述提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征的步骤。
上述电子设备提到的通信总线可以是外设部件互连标准(Peripheral Component Interconnect，PCI)总线或扩展工业标准结构(Extended Industry Standard Architecture，EISA)总线等。该通信总线可以分为地址总线、数据总线、控制总线等。为便于表示，图中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
通信接口用于上述电子设备与其他设备之间的通信。
存储器可以包括随机存取存储器(Random Access Memory，RAM)，也可以包括非易失性存储器(Non-Volatile Memory，NVM)，例如至少一个磁盘存 储器。可选的，存储器还可以是至少一个位于远离前述处理器的存储装置。
上述的处理器可以是通用处理器，包括中央处理器(Central Processing Unit，CPU)、网络处理器(Network Processor，NP)等；还可以是数字信号处理器(Digital Signal Processing，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。
应用本申请图4所示实施例，提取图像中待识别人体目标的目标特征，作为待查找目标特征，基于预先建立的目标特征与人脸信息的对应关系，查找待查找目标特征对应的人脸信息；基于得到的人脸信息，确定待识别人体目标的身份；可见，本方案中，不需要提取图像中的人脸特征，即使图像中的人脸区域不清晰，或者被其他物体遮挡，也不会降低身份识别的准确性；因此，应用本方案，提高了身份识别的准确性。
本申请实施例提供一种计算机可读存储介质，所述计算机可读存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现如下步骤：获取待识别图像；提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征；基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息；其中，所述对应关系中一份目标特征与对应的人脸信息属于同一人体目标；基于得到的人脸信息，确定所述待识别人体目标的身份。
作为一种实施方式，所述获取待识别图像的步骤，包括：接收用户输入的待识别图像；或者，从指定采集设备中获取待识别图像。
作为一种实施方式，所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：基于预先建立的目标特征与人脸特征的对应关系，查找所述待查找目标特征对应的人脸特征；所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤，包括：基于得到的人脸特征，确定所述待识别人体目标的身份。
作为一种实施方式，所述基于预先建立的目标特征与人脸信息的对应关 系，查找所述待查找目标特征对应的人脸信息的步骤，包括：基于预先建立的目标特征与人脸图像的对应关系，查找所述待查找目标特征对应的人脸图像；所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤，包括：基于得到的人脸图像，确定所述待识别人体目标的身份。
作为一种实施方式，所述提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征的步骤，包括：提取所述待识别图像中待识别人体目标的原始目标特征，计算所述原始目标特征的哈希值，作为待查找哈希值；所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息。
作为一种实施方式，所述基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息的步骤，包括：分别计算预先建立的哈希值与人脸信息的对应关系中所包括的各哈希值与所述待查找哈希值之间的相似度；确定相似度满足预设条件的哈希值对应的人脸信息。
作为一种实施方式，所述计算机程序被处理器执行时还可以实现如下步骤：在所述获取待识别图像的步骤之后，确定所述待识别图像的采集属性，作为待查找采集属性；其中，所述采集属性包含采集所述待识别图像的时刻和/或地点；所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：
在预先建立的目标特征与人脸信息的对应关系中，查找与所述待查找采集属性的差值小于预设阈值的目标采集属性；在所述目标采集属性对应的人脸信息中，查找所述待查找目标特征对应的人脸信息。
作为一种实施方式，所述计算机程序被处理器执行时还可以实现如下步骤：在所述获取待识别图像的步骤之后，在所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤之前，判断所述待识别图像中是否存在满足清晰度要求的人脸区域；如果存在，提取所述待识别图像中的人脸信息；如果不存在，执行所述提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征的步骤。
本申请实施例还公开了一种可执行程序代码，所述可执行程序代码用于被运行以实现上述任一种人体目标身份识别方法。
需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
本说明书中的各个实施例均采用相关的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于图3所示的装置实施例、图4所示的电子设备实施例、上述计算机可读存储介质实施例、以及上述可执行程序代码实施例而言，由于其基本相似于图1-2所示的方法实施例，所以描述的比较简单，相关之处参见图1-2所示的方法实施例的部分说明即可。
以上所述仅为本申请的较佳实施例而已，并非用于限定本申请的保护范围。凡在本申请的精神和原则之内所作的任何修改、等同替换、改进等，均包含在本申请的保护范围内。

Claims (24)

  1. 一种人体目标身份识别方法，其特征在于，包括：
    获取待识别图像；
    提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征；
    基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息；其中，所述对应关系中一份目标特征与对应的人脸信息属于同一人体目标；
    基于得到的人脸信息，确定所述待识别人体目标的身份。
  2. 根据权利要求1所述的方法，其特征在于，所述获取待识别图像的步骤，包括：
    接收用户输入的待识别图像；
    或者，从指定采集设备中获取待识别图像。
  3. 根据权利要求1所述的方法，其特征在于，所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：
    基于预先建立的目标特征与人脸特征的对应关系，查找所述待查找目标特征对应的人脸特征；
    所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤，包括：
    基于得到的人脸特征，确定所述待识别人体目标的身份。
  4. 根据权利要求1所述的方法，其特征在于，所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：
    基于预先建立的目标特征与人脸图像的对应关系，查找所述待查找目标特征对应的人脸图像；
    所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤，包括：
    基于得到的人脸图像，确定所述待识别人体目标的身份。
  5. 根据权利要求1所述的方法，其特征在于，所述提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征的步骤，包括：
    提取所述待识别图像中待识别人体目标的原始目标特征，计算所述原始目标特征的哈希值，作为待查找哈希值；
    所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：
    基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息。
  6. 根据权利要求5所述的方法，其特征在于，所述基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息的步骤，包括：
    分别计算预先建立的哈希值与人脸信息的对应关系中所包括的各哈希值与所述待查找哈希值之间的相似度；
    确定相似度满足预设条件的哈希值对应的人脸信息。
  7. 根据权利要求1所述的方法，其特征在于，在所述获取待识别图像的步骤之后，还包括：
    确定所述待识别图像的采集属性，作为待查找采集属性；其中，所述采集属性包含采集所述待识别图像的时刻和/或地点；
    所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：
    在预先建立的目标特征与人脸信息的对应关系中，查找与所述待查找采集属性的差值小于预设阈值的目标采集属性；
    在所述目标采集属性对应的人脸信息中，查找所述待查找目标特征对应 的人脸信息。
  8. 根据权利要求1所述的方法，其特征在于，在所述获取待识别图像的步骤之后，在所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤之前，还包括：
    判断所述待识别图像中是否存在满足清晰度要求的人脸区域；
    如果存在，提取所述待识别图像中的人脸信息；
    如果不存在，执行所述提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征的步骤。
  9. 一种人体目标身份识别装置，其特征在于，包括：
    获取模块，用于获取待识别图像；
    第一提取模块，用于提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征；
    查找模块，用于基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息；其中，所述对应关系中一份目标特征与对应的人脸信息属于同一人体目标；
    第一确定模块，用于基于得到的人脸信息，确定所述待识别人体目标的身份。
  10. 根据权利要求9所述的装置，其特征在于，所述获取模块，具体用于：
    接收用户输入的待识别图像；
    或者，从指定采集设备中获取待识别图像。
  11. 根据权利要求9所述的装置，其特征在于，所述查找模块，具体用于：
    基于预先建立的目标特征与人脸特征的对应关系，查找所述待查找目标特征对应的人脸特征；
    所述第一确定模块，具体用于：
    基于得到的人脸特征，确定所述待识别人体目标的身份。
  12. 根据权利要求9所述的装置，其特征在于，所述查找模块，具体用于：
    基于预先建立的目标特征与人脸图像的对应关系，查找所述待查找目标特征对应的人脸图像；
    所述第一确定模块，具体用于：
    基于得到的人脸图像，确定所述待识别人体目标的身份。
  13. 根据权利要求9所述的装置，其特征在于，所述第一提取模块，具体用于：
    提取所述待识别图像中待识别人体目标的原始目标特征，计算所述原始目标特征的哈希值，作为待查找哈希值；
    所述查找模块，具体用于：
    基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息。
  14. 根据权利要求13所述的装置，其特征在于，所述查找模块，具体用于：
    分别计算预先建立的哈希值与人脸信息的对应关系中所包括的各哈希值与所述待查找哈希值之间的相似度；确定相似度满足预设条件的哈希值对应的人脸信息。
  15. 根据权利要求9所述的装置，其特征在于，所述装置还包括：
    第二确定模块，用于确定所述待识别图像的采集属性，作为待查找采集属性；其中，所述采集属性包含采集所述待识别图像的时刻和/或地点；
    所述查找模块，具体用于：
    在预先建立的目标特征与人脸信息的对应关系中，查找与所述待查找采集属性的差值小于预设阈值的目标采集属性；在所述目标采集属性对应的人脸信息中，查找所述待查找目标特征对应的人脸信息。
  16. 根据权利要求9所述的装置，其特征在于，所述装置还包括：
    判断模块，用于判断所述待识别图像中是否存在满足清晰度要求的人脸区域；如果存在，触发第二提取模块，如果不存在，触发所述第一提取模块；
    第二提取模块，用于提取所述待识别图像中的人脸信息。
  17. 一种电子设备，其特征在于，包括处理器、通信接口、存储器和通信总线，其中，处理器，通信接口，存储器通过通信总线完成相互间的通信；存储器，用于存放计算机程序；处理器，用于执行存储器上所存放的程序时，实现如下步骤：
    获取待识别图像；
    提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征；
    基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息；其中，所述对应关系中一份目标特征与对应的人脸信息属于同一人体目标；
    基于得到的人脸信息，确定所述待识别人体目标的身份。
  18. 根据权利要求17所述的设备，其特征在于，所述获取待识别图像的步骤，包括：
    接收用户输入的待识别图像；
    或者，从指定采集设备中获取待识别图像。
  19. 根据权利要求17所述的设备，其特征在于，所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：
    基于预先建立的目标特征与人脸特征的对应关系，查找所述待查找目标特征对应的人脸特征；
    所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤，包括：
    基于得到的人脸特征，确定所述待识别人体目标的身份。
  20. 根据权利要求17所述的设备，其特征在于，所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：
    基于预先建立的目标特征与人脸图像的对应关系，查找所述待查找目标特征对应的人脸图像；
    所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤，包括：
    基于得到的人脸图像，确定所述待识别人体目标的身份。
  21. 根据权利要求17所述的设备，其特征在于，所述提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征的步骤，包括：
    提取所述待识别图像中待识别人体目标的原始目标特征，计算所述原始目标特征的哈希值，作为待查找哈希值；
    所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：
    基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息。
  22. 根据权利要求21所述的设备，其特征在于，所述基于预先建立的哈希值与人脸信息的对应关系，查找所述待查找哈希值对应的人脸信息的步骤，包括：
    分别计算预先建立的哈希值与人脸信息的对应关系中所包括的各哈希值与所述待查找哈希值之间的相似度；
    确定相似度满足预设条件的哈希值对应的人脸信息。
  23. 根据权利要求17所述的设备，其特征在于，所述处理器还用于实现如下步骤：
    在所述获取待识别图像的步骤之后，确定所述待识别图像的采集属性，作为待查找采集属性；其中，所述采集属性包含采集所述待识别图像的时刻 和/或地点；
    所述基于预先建立的目标特征与人脸信息的对应关系，查找所述待查找目标特征对应的人脸信息的步骤，包括：
    在预先建立的目标特征与人脸信息的对应关系中，查找与所述待查找采集属性的差值小于预设阈值的目标采集属性；
    在所述目标采集属性对应的人脸信息中，查找所述待查找目标特征对应的人脸信息。
  24. 根据权利要求17所述的设备，其特征在于，所述处理器还用于实现如下步骤：
    在所述获取待识别图像的步骤之后，在所述基于得到的人脸信息，确定所述待识别人体目标的身份的步骤之前，判断所述待识别图像中是否存在满足清晰度要求的人脸区域；
    如果存在，提取所述待识别图像中的人脸信息；
    如果不存在，执行所述提取所述待识别图像中待识别人体目标的目标特征，作为待查找目标特征的步骤。
PCT/CN2018/101665 2017-08-31 2018-08-22 一种人体目标身份识别方法及装置 Ceased WO2019042195A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US16/643,221 US11126828B2 (en) 2017-08-31 2018-08-22 Method and device for recognizing identity of human target
EP18851366.7A EP3678047B1 (en) 2017-08-31 2018-08-22 Method and device for recognizing identity of human target

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710769677.5A CN109426785B (zh) 2017-08-31 2017-08-31 一种人体目标身份识别方法及装置
CN201710769677.5 2017-08-31

Publications (1)

Publication Number Publication Date
WO2019042195A1 true WO2019042195A1 (zh) 2019-03-07

Family

ID=65504729

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/101665 Ceased WO2019042195A1 (zh) 2017-08-31 2018-08-22 一种人体目标身份识别方法及装置

Country Status (4)

Country Link
US (1) US11126828B2 (zh)
EP (1) EP3678047B1 (zh)
CN (1) CN109426785B (zh)
WO (1) WO2019042195A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112001932A (zh) * 2020-09-01 2020-11-27 腾讯科技（深圳）有限公司 人脸识别方法、装置、计算机设备和存储介质
US12136324B2 (en) 2019-03-27 2024-11-05 Nec Corporation Monitoring device, suspicious object detecting method, and recording medium

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109426785B (zh) * 2017-08-31 2021-09-10 杭州海康威视数字技术股份有限公司 一种人体目标身份识别方法及装置
CN111666786B (zh) * 2019-03-06 2024-05-03 杭州海康威视数字技术股份有限公司 图像处理方法、装置、电子设备及存储介质
CN112686085A (zh) * 2019-10-18 2021-04-20 晋城三赢精密电子有限公司 应用于摄像装置中的智能识别方法、摄像装置及存储介质
CN111046831B (zh) * 2019-12-20 2023-06-30 上海信联信息发展股份有限公司 家禽识别方法、装置及服务器
CN113033266A (zh) * 2019-12-25 2021-06-25 杭州海康威视数字技术股份有限公司 人员运动轨迹追踪方法、装置、系统及电子设备
CN111538861B (zh) * 2020-04-22 2023-08-15 浙江大华技术股份有限公司 基于监控视频进行图像检索的方法、装置、设备及介质
CN111680654B (zh) * 2020-06-15 2023-10-13 杭州海康威视数字技术股份有限公司 一种基于物品取放事件的人员信息获取方法、装置及设备
CN112541384B (zh) * 2020-07-30 2023-04-28 深圳市商汤科技有限公司 可疑对象查找方法及装置、电子设备及存储介质
CN112232210B (zh) * 2020-10-16 2024-06-28 京东方科技集团股份有限公司 一种人员流量分析方法和系统、电子设备和可读存储介质
CN112380941B (zh) * 2020-11-11 2025-04-15 哈尔滨海邻科信息技术有限公司 身份识别方法、装置、终端设备及存储介质
CN115376160B (zh) * 2022-07-15 2026-04-10 新瑞鹏宠物医疗集团有限公司 宠物身份识别方法、装置、电子设备及存储介质
CN116824654A (zh) * 2022-11-30 2023-09-29 慧之安信息技术股份有限公司 基于边缘计算的人脸面部识别监测方法
CN120954058B (zh) * 2025-10-17 2026-02-10 深圳市快瞳科技有限公司 宠物身份识别方法、装置、设备及存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101021870A (zh) * 2007-03-20 2007-08-22 北京中星微电子有限公司 一种图片查询方法及系统
CN106295504A (zh) * 2016-07-26 2017-01-04 车广为 人脸识别基础上的增强显示方法
CN106845385A (zh) * 2017-01-17 2017-06-13 腾讯科技（上海）有限公司 视频目标跟踪的方法和装置

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2418312A (en) * 2004-09-18 2006-03-22 Hewlett Packard Development Co Wide area tracking system
WO2010006367A1 (en) * 2008-07-16 2010-01-21 Imprezzeo Pty Ltd Facial image recognition and retrieval
US8891880B2 (en) * 2009-10-16 2014-11-18 Nec Corporation Person clothing feature extraction device, person search device, and processing method thereof
CN103793443A (zh) * 2012-11-05 2014-05-14 腾讯科技（深圳）有限公司 控制应用的方法和装置
CN103942563A (zh) * 2014-03-31 2014-07-23 北京邮电大学 一种多模态行人再辨识技术
CN104463148B (zh) * 2014-12-31 2017-07-28 南京信息工程大学 基于图像重构和哈希算法的人脸识别方法
JP6402653B2 (ja) * 2015-03-05 2018-10-10 オムロン株式会社 物体認識装置、物体認識方法、およびプログラム
CN104850828B (zh) * 2015-04-29 2018-06-12 小米科技有限责任公司 人物识别方法及装置
CN105160295B (zh) * 2015-07-14 2019-05-17 东北大学 一种面向大规模人脸数据库的快速高效人脸检索方法
CN105447466B (zh) * 2015-12-01 2019-07-23 深圳市图灵机器人有限公司 一种基于Kinect传感器的身份综合识别方法
CN105808709B (zh) * 2016-03-04 2019-10-29 智慧眼科技股份有限公司 人脸识别快速检索方法及装置
CN106446816B (zh) * 2016-09-14 2019-12-27 北京旷视科技有限公司 人脸识别方法及装置
CN106778474A (zh) * 2016-11-14 2017-05-31 深圳奥比中光科技有限公司 3d人体识别方法及设备
CN106991395B (zh) * 2017-03-31 2020-05-26 联想(北京)有限公司 信息处理方法、装置及电子设备
EP3418944B1 (en) * 2017-05-23 2024-03-13 Canon Kabushiki Kaisha Information processing apparatus, information processing method, and program
CN109426785B (zh) * 2017-08-31 2021-09-10 杭州海康威视数字技术股份有限公司 一种人体目标身份识别方法及装置
CN110516083B (zh) * 2019-08-30 2022-07-12 京东方科技集团股份有限公司 相册管理方法、存储介质及电子设备

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101021870A (zh) * 2007-03-20 2007-08-22 北京中星微电子有限公司 一种图片查询方法及系统
CN106295504A (zh) * 2016-07-26 2017-01-04 车广为 人脸识别基础上的增强显示方法
CN106845385A (zh) * 2017-01-17 2017-06-13 腾讯科技（上海）有限公司 视频目标跟踪的方法和装置

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
See also references of EP3678047A4
SUN WEI: "Research and Implementation of Search and Track Algorithm Based on People Attributes", CHINA MASTER'S THESIS, no. 6, 15 June 2013 (2013-06-15), XP009519593 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12136324B2 (en) 2019-03-27 2024-11-05 Nec Corporation Monitoring device, suspicious object detecting method, and recording medium
US12175846B2 (en) 2019-03-27 2024-12-24 Nec Corporation Monitoring device, suspicious object detecting method, and recording medium
US12190692B2 (en) 2019-03-27 2025-01-07 Nec Corporation Monitoring device, suspicious object detecting method, and recording medium
US12406560B2 (en) * 2019-03-27 2025-09-02 Nec Corporation Monitoring device, suspicious object detecting method, and recording medium
US12412460B2 (en) * 2019-03-27 2025-09-09 Nec Corporation Monitoring device, suspicious object detecting method, and recording medium
CN112001932A (zh) * 2020-09-01 2020-11-27 腾讯科技（深圳）有限公司 人脸识别方法、装置、计算机设备和存储介质
CN112001932B (zh) * 2020-09-01 2023-10-31 腾讯科技（深圳）有限公司 人脸识别方法、装置、计算机设备和存储介质

Also Published As

Publication number Publication date
US20200193146A1 (en) 2020-06-18
EP3678047A1 (en) 2020-07-08
US11126828B2 (en) 2021-09-21
EP3678047A4 (en) 2020-07-29
CN109426785A (zh) 2019-03-05
EP3678047B1 (en) 2023-07-19
CN109426785B (zh) 2021-09-10

Similar Documents

Publication Publication Date Title
WO2019042195A1 (zh) 一种人体目标身份识别方法及装置
US11157720B2 (en) Method and device for determining path of human target
CN110866466B (zh) 一种人脸识别方法、装置、存储介质和服务器
US10402627B2 (en) Method and apparatus for determining identity identifier of face in face image, and terminal
CN109145742B (zh) 一种行人识别方法及系统
CN112016353B (zh) 一种基于视频的人脸图像进行身份识别方法及装置
WO2018099032A1 (zh) 一种目标跟踪方法及装置
CN111079757B (zh) 服饰属性识别方法、装置及电子设备
CN111507200A (zh) 体温检测方法、体温检测装置、及双光相机
JP2018505495A (ja) 指紋重複領域の面積算出方法、それを行う電子機器、コンピュータプログラム、及び、記録媒体
WO2019076187A1 (zh) 视频遮蔽区域选取方法、装置、电子设备及系统
CN112733814B (zh) 一种基于深度学习的行人徘徊滞留检测方法、系统及介质
TWI704505B (zh) 人臉辨識系統、建立人臉辨識之資料之方法及其人臉辨識之方法
WO2020108075A1 (zh) 结合人脸与外观的两阶段行人搜索方法
CN114048344A (zh) 一种相似人脸搜索方法、装置、设备和可读存储介质
CN110706247A (zh) 一种目标跟踪方法、装置及系统
WO2018121287A1 (zh) 目标再识别方法和装置
CN110717357B (zh) 预警方法、装置、电子设备及存储介质
WO2020147346A1 (zh) 图像识别方法、系统及装置
US10659680B2 (en) Method of processing object in image and apparatus for same
CN110298318A (zh) 人头人体联合检测方法、装置和电子设备
CN113657434A (zh) 人脸人体关联方法、系统以及计算机可读存储介质
CN108171138B (zh) 一种生物特征信息获取方法和装置
WO2013075295A1 (zh) 低分辨率视频的服装识别方法及系统
CN111814510A (zh) 一种遗留物主体检测方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18851366

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2018851366

Country of ref document: EP

Effective date: 20200331