CN115880745A - Data processing system for acquiring human face image characteristics - Google Patents
Data processing system for acquiring human face image characteristics Download PDFInfo
- Publication number
- CN115880745A CN115880745A CN202211089095.XA CN202211089095A CN115880745A CN 115880745 A CN115880745 A CN 115880745A CN 202211089095 A CN202211089095 A CN 202211089095A CN 115880745 A CN115880745 A CN 115880745A
- Authority
- CN
- China
- Prior art keywords
- image
- data processing
- preset
- target
- acquiring
- 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.)
- Granted
Links
Images
Classifications
-
- 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
- Collating Specific Patterns (AREA)
Abstract
The invention relates to a data processing system for acquiring human face image characteristics, which comprises: the image processing system comprises a database, a processor, a memory for storing computer programs, a first image data processing platform and N second image data processing platforms, wherein the database comprises a first face image list and a preset image information set, and when the computer programs are executed by the processor, the following steps are realized: and inputting the preset image information into a second image data processing platform for clustering, comparing the first face image with the preset image information to obtain a first target image feature set, obtaining a second target image feature set according to the first target image feature set and the first image data processing platform, and returning the second target image feature set to the first target user. The invention provides a novel method for acquiring human face image features, which is used for processing a human face image by associating a plurality of data query platforms, so that the accuracy of the acquired human face image features is higher.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a data processing system for acquiring human face image characteristics.
Background
With the rapid development of face image recognition technology, face recognition develops diversified industry applications in the fields of station and airport ticket checking systems, residential entrance and exit management, accurate commercial marketing, examination room identity verification and the like, at present, face image recognition methods are diversified, how to process face images is achieved, and therefore the face image recognition efficiency and accuracy are effectively improved, the mainstream research direction is achieved, the face image recognition speed and accuracy are continuously enhanced, and accurate services can be provided for various industries.
In the prior art, a method for acquiring a face image feature is known as follows: extracting the features of the face image, calculating the similarity between the face image to be searched and the face image vector in the target database by using the Euclidean distance or the cosine distance, and obtaining the face image features of which the similarity with the face feature to be searched meets the preset conditions, wherein the method for obtaining the face image features has the following problems:
on one hand, the number of the face images in the database is large, and the face images are compared with the feature vectors of each face image one by one, so that the searching and identifying range of the face images is expanded, and the real-time performance of face identification is reduced;
on the other hand, data information is stored in the same platform, and the data storage space is too large, so that information in the database is lost, and the accuracy of the acquired facial image features is low.
Disclosure of Invention
Aiming at the technical problem, the technical scheme adopted by the invention is as follows: a data processing system for obtaining features of an image of a human face, the system comprising: the image processing system comprises a database, a processor, a memory for storing computer programs, a first image data processing platform and N second image data processing platforms, wherein the database comprises a first face image list and a preset image information set, and when the computer programs are executed by the processor, the following steps are realized:
s100, acquiring a first face image list A = { A } from a database 1 ，……，A i ，……，A λ }，A i For the first face image corresponding to the ith first target user, i =1 … … λ, where λ is the number of first target users.
S200, acquiring a preset image information set M = { M } from a database 1 ，……，M s ，……，M v }，M s For the s-th preset image information in the database, s =1 … … v, v is the number of the preset image information.
S300, inputting M into N second image data processing platforms, and acquiring a target image vector list C corresponding to each second image data processing platform h ＝{C h0 ，……，C hk ，……，C hμ In which C hk H =1 … … N, C for the kth target image vector corresponding to the h-th second image data processing stage hk ＝M h+k*N ，M h+k*N For the (h + k × N) th preset image information in M, k =0 … … μ,
s400, for C hk Processing to obtain C h Corresponding target image vector type set C' h ＝{C' h1 ，……，C' hβ ，……，C' ha }，C' hβ ＝{C' 1 hβ ，……，C' u hβ ，……，C' w(β) hβ }，C' u hβ For the u-th target image vector in the beta-th class in the h-th second image data processing platform, u =1 … … w (beta), w (beta) is the number of target image vectors in the beta-th class, beta =1 … … a, and a is C' h Number of target image vector types.
S500, according to A and C' u hβ Acquiring a first target image feature set AC = { AC corresponding to a 1 ,……，AC i ,……，AC λ }，AC i Is A i A corresponding first list of target image features.
S600, according to AC i And a first image data processing platform for acquiring a second target image feature set AD = { corresponding to A 1 ,……，AD i ,……，AD λ }，AD i Is A i A corresponding second target image feature list.
S700, AD i And sending the information to the ith first target user.
Compared with the prior art, the data processing system for acquiring the human face image features has obvious beneficial effects, can achieve considerable technical progress and practicability by virtue of the technical scheme, has wide industrial utilization value, and at least has the following beneficial effects:
the invention provides a data processing system for acquiring human face image characteristics, which comprises: the system comprises a database, a processor, a memory for storing computer programs, a first image data processing platform and N second image data processing platforms, wherein the database comprises a first face image list and a preset image information set, and when the computer programs are executed by the processor, the following steps are realized: the method comprises the steps of obtaining a first face image list and a preset image information set from a database, inputting the preset image information set into N second image data processing platforms to obtain a target image vector list corresponding to each second image data processing platform, carrying out clustering processing on target image vectors in each second image data processing platform to obtain a target image vector type set corresponding to each second image data processing platform, obtaining a first target image feature set corresponding to a first face image according to the first face image list and the target image vectors, obtaining a second target image feature set corresponding to the first face image according to the first target image feature and the first image data processing platform, and sending the second target image feature set to a first target user. On one hand, the method does not compare the face images in the database one by one, compares the preset image information in the database with the preset image information type through clustering processing, and then compares the preset image information with the preset image information meeting the preset type, so that the searching and identifying range of the face images is reduced, and the real-time performance of face identification is improved; on the other hand, the multiple data query platforms are associated, the facial images are processed at the same time, the clustering number of the data in the data query platforms is dynamically adjusted, the data storage space is improved, and the accuracy of the acquired facial image features is higher under the condition that information in the database is not lost.
The foregoing description is only an overview of the technical solutions of the present invention, and in order to make the technical means of the present invention more clearly understood, the present invention may be implemented in accordance with the content of the description, and in order to make the above and other objects, features, and advantages of the present invention more clearly understood, the following preferred embodiments are specifically described below with reference to the accompanying drawings.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart of a computer program executed by a data processing system for obtaining facial image features according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Moreover, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
A data processing system for obtaining facial image features, the system comprising: a database, a processor, a memory storing a computer program, a first image data processing platform and N second image data processing platforms, wherein the database comprises a list of first face images and a set of pre-set image information, and when the computer program is executed by the processor, the following steps are implemented, as shown in fig. 1:
s100, acquiring a first face image list A = { A ] from a database 1 ，……，A i ，……，A λ }，A i For the first face image corresponding to the ith first target user, i =1 … … λ, where λ is the number of first target users.
Specifically, the database further includes an initial set of facial images of the target user, and when the computer program is executed by the processor, the first list of facial images is obtained by:
s110, acquiring a first face image list A and a second face image list B according to the initial face image set of the target user.
Specifically, the method further includes the following steps in S110:
s111, obtaining an initial face image set Q = { Q) of a target user from a database 1 ，……，Q g ，……，Q n }，Q g The image of the face of the g-th target user is g =1 … … n, and n is the number of the target users.
S112, according to Q, obtaining an initial time list T = { T) corresponding to Q 1 ，……，T g ，……，T n }，T g Is Q g Corresponding initial time node.
S113, when T is reached 0 -T g When T 'is less than or equal to T', the T is added g Corresponding Q g Inserting a first initial face image list A '= { A' 1 ，……，A' q ，……，A' b }，A' q A first initial corresponding to the qth first initial target userFace images, q =1 … … b, b is the number of first initial target users, where T 0 And T' is a preset first time threshold value for the current time node corresponding to Q.
Specifically, the value range of T 'is 90 days to 180 days, and those skilled in the art know that T' is selected according to actual requirements, and will not be described herein again.
S114, when T is 0 -T g When > T', let T g Corresponding Q g Inserting a second initial face image list B '= { B' 1 ，……，B' L ，……，B' t }，B' L And L =1 … … t is the second initial face image corresponding to the L-th second initial target user, and t is the number of the second initial target users.
S115, when b > lambda 0 When a = a 'and B = B' are obtained.
S116, when b is less than or equal to lambda 0 Then, T 'is replaced by T' 1 Wherein λ is 0 Is a preset image quantity threshold value, T' 1 Is a preset second time threshold.
In particular, λ is performed according to the amount of data that the second image data processing stage can store 0 The selection of (a) is known to those skilled in the art and will not be described herein in detail.
Concretely, T' 1 ＞T'。
By expanding the time range, more initial face images are distributed to the first initial face image set, the number of comparison with preset image information in a database can be reduced for follow-up, the face image searching and identifying range is narrowed, and the real-time performance of face identification is improved.
S107, when T is 0 -T L ≤T' 1 Then, mixing B' L Insert into A 'to obtain A = A', wherein T L Is B' L Corresponding time node.
S108, when T is reached 0 -T L ＞T' 1 Then, B 'is added' j Delete in B 'to obtain B = B'.
In the method, the initial face images of the target user are divided, are classified according to different time, and are dynamically adjusted in time according to the data size stored by the second image data processing platform, so that more face images are input into the second image data processing platform, the face image searching and identifying range can be narrowed, and the real-time performance of face identification is improved.
S200, acquiring a preset image information set M = { M ] from a database 1 ，……，M s ，……，M v }，M s For the s-th preset image information in the database, s =1 … … v, v is the number of the preset image information.
S300, inputting M into N second image data processing platforms, and acquiring a target image vector list C corresponding to each second image data processing platform h ＝{C h0 ，……，C hk ，……，C hμ In which C hk H =1 … … N, C for the kth target image vector corresponding to the h-th second image data processing stage hk ＝M h+k*N ，M h+k*N For the (h + k × N) th preset image information in M, k =0 … … μ,
according to the method, the preset image information is input into the N second image data processing platforms, a certain amount of preset image information is stored in each second image data processing platform, the data information is stored in different platforms, the relevance relation is established, the data storage space is reduced, and the accuracy of the acquired facial image features is high under the condition that database information is not lost.
S400, for C hk Processing to obtain C h Corresponding target image vector type set C' h ＝{C' h1 ，……，C' hβ ，……，C' ha }，C' hβ ＝{C' 1 hβ ，……，C' u hβ ，……，C' w(β) hβ }，C' u hβ For the u-th target image vector in the beta-th class in the h-th second image data processing platform, u =1 … … w (beta), w (beta) is the number of target image vectors in the beta-th class, and beta =1 … …a and a are C' h Number of target image vector types.
Specifically, the method in S400 further includes the following steps:
s401, from C h Obtaining intermediate image vector type set H h ＝{H h1 ，……，H hβ ，……，H ha H and h corresponding initial set of center points H 1 h ＝{H 1 h1 ，……，H 1 hβ ，……，H 1 ha }，H hβ For the beta-th intermediate image vector, H, in the H-th second image data processing stage 1 hβ From C when not initially clustered h The vector of the acquired beta-th sample, wherein the initial H hβ ＝Null。
S402, according to C h And H 1 h Obtaining C hk Corresponding target similarity list HC hk ＝{HC 1 hk ，……，HC β hk ，……，HC a hk }，HC β hk Is C hk And H 1 hβ The target similarity between the two models can be calculated by those skilled in the art according to any model, and will not be described herein again.
S403, when HC is present β hk Is HC hk At the time of middle minimum value, C is added hk Is inserted into H hβ To obtain C' hβ ＝H hβ 。
S404, according to H hβ Obtaining H h Corresponding set of intermediate center points H 2 h ＝{H 2 h1 ，……，H 2 hβ ，……，H 2 ha In which H 2 hβ The following conditions are met:
S405, repeatedly executing S402-S404 to obtain H h Corresponding target center point set H E h ＝{H E h1 ，……，H E hβ ，……，H E ha In which H E hβ Is the beta-th target central point obtained after the (E-1) th clustering.
Specifically, the above-mentioned H E hβ Obtaining method of (1) and 2 hβ the acquisition modes are consistent.
S406, when H is E hβ ＝H E-1 hβ Then, make C' hβ ＝H hβ 。
In the above, the preset image information in each second image data processing platform is clustered, similar images are clustered, and when the first face image is compared with the preset image information in the second image data processing platform subsequently, the face image recognition search range can be reduced, so that the real-time performance of face recognition is improved.
Specifically, in S400, a satisfies the following condition:
a＝r 1 (μ+1) 3 +r 2 (μ+1) 2 +r 3 (μ+1)+r 4 wherein r is 1 Is a preset first parameter, r 2 Is a preset second parameter, r 3 Is a preset third parameter, r 4 Is a preset fourth parameter.
Specifically, r 1 Has a value range of-8 e -13 ～-5e -12 Wherein e is the Euler constant.
Preferably, r 1 Is-4 e -12 。
Specifically, r 2 Has a value range of 2e -8 ～3e -6 。
Preferably, r 2 Is taken as 2e -7 。
Specifically, r 3 The value range of (A) is-1 to 1.
Preferably, r 3 Is-0.0068。
Specifically, r 4 The value range of (a) is 50 to 100.
Preferably, r 4 Is 99.025.
By adjusting the type number of the face images according to the data volume of the face images, the balance between the performance and the recall rate during data matching can be ensured, the efficiency of data operation and the comprehensiveness of data matching are ensured when the initial face image features are obtained, and the accuracy of the obtained face image features is higher.
S500 according to A and C' u hβ Acquiring a first target image feature set AC = { AC) corresponding to a 1 ,……，AC i ,……，AC λ }，AC i Is A i A corresponding first target image feature list.
Specifically, the method in S500 further includes the following steps:
s501, according to A i And C' h Obtaining A i Corresponding first similarity list F i h ＝{F i h1 ，……，F i hβ ，……，F i ha }，F i hβ Is A i And C' hβ The first similarity of (1).
Specifically, step S501 further includes the following steps:
s5011, obtaining a first face image vector list A corresponding to A according to A 0 ＝{A 0 1 ，……，A 0 i ，……，A 0 λ }，A 0 i Is A i A corresponding first face image vector.
Specifically, the model for converting the first face image into the first face image vector is selected by a person skilled in the art according to actual needs, and is not described herein again.
S5013 according to A 0 Obtaining A 0 i ＝(A 0 i1 ，……，A 0 ix ，……，A 0 ip ) Wherein A is 0 ix Is A 0 i The x-th bit value, x =1 … … p, p is the dimension of the first face image vector.
S5015 according to C' h Obtaining C' u hβ ＝(C' u1 hβ ，……，C' ux hβ ，……，C' up hβ ) Wherein, C' ux hβ Is C' u hβ The bit value of the x-th bit.
S5017 according to A 0 ix And C' ux hβ Obtaining a first similarity F i hβ Wherein F is i hβ The following conditions are met:
s503, when F i hβ ≥F 0 Then, obtain A i Corresponding second similarity list F i hβ ＝{F i1 hβ ，……，F iu hβ ，……，F iw(β) hβ }，F iu hβ Is A i And C' u hβ Second degree of similarity of (D), F 0 Is a preset first similarity threshold.
Specifically, F iu hβ And F i hβ The acquisition modes are consistent.
Specifically, F 0 The value range of (a) is 0.85-0.9, and a person skilled in the art sets a preset similarity threshold according to actual requirements, which are not described herein again.
S505, when F iu hβ ≥F 1 Get AC i In which F is 1 Is a preset second similarity threshold.
Specifically, F 1 The value range of (a) is 0.85-0.9, and a preset similarity threshold value is set by a person skilled in the art according to actual requirements, which is not described herein again.
Specifically, F 0 ＝F 1 。
Specifically, theThe second image data processing platform further comprises C h Corresponding first image feature list FC h ＝{FC h0 ,……，FC hk ，……，FC hμ }，FC hk ＝{FC 1 hk ，……，FC θ hk ，……，FC ζ hk }，FC θ hk Is C hk θ =1 … … ζ, ζ being the number of first fields, corresponding to the first record included in the θ -th first field in the first image feature list.
Specifically, step S505 further includes the following steps:
s501, when F iu hβ ≥F 1 Then, obtain F iu hβ Corresponding to C' u hβ 。
S503, when C' u hβ ＝C hk Get AC i ＝FC hk 。
The method comprises the steps of firstly comparing a first face image with a target image vector type in each second image data processing platform, selecting a target image vector type meeting conditions according to a preset similarity threshold, and then comparing the target image vector type meeting the conditions with the target image vector type meeting the conditions, so that the face image searching range can be reduced, the real-time performance of face recognition is improved, meanwhile, the data storage space is reduced, and the accuracy of the acquired face image features is higher under the condition that database information is not lost.
S600, according to AC i And a first image data processing platform for acquiring a second target image feature set AD = { corresponding to A 1 ,……，AD i ,……，AD λ }，AD i Is A i A corresponding second target image feature list.
Specifically, the first image data processing platform further includes a second image feature list M ' = { M ' corresponding to M ' 1 ，……，M' s ，……，M' v }，M' s ＝{M' s1 ，……，M' sε ，……，M' sη }，M' sε Is M s The epsilon-th image in the corresponding second image feature listThe second field contains a second record, ε =1 … … η, η being the number of second fields.
Specifically, S600 includes the following steps:
s601, when the theta first field is a preset field and the epsilon second field is a preset field, acquiring FC θ hk And M' sε Wherein the preset field is ID.
S603, when S = h + k × N, obtaining AD i ＝M' s 。
By associating the data in the first image data processing platform with the data in the second image data processing platform, the second image data processing platform with less stored data is subjected to the same ID matching with the first image data processing platform, so that all information of the first face image is acquired, the face image searching range is reduced, the real-time performance of face recognition is improved, and meanwhile, the accuracy of the acquired face image features is high under the condition that database information is not lost.
S700, AD i And sending the information to the ith first target user.
Specifically, the method further includes the following steps after S700:
s800, acquiring a second face image list B = { B ] from a database 1 ，……，B j ，……，B γ }，B j J =1 … … γ, where γ is the number of the second target users.
S900, inputting B and M into the first image data processing platform, and acquiring a second target image feature set BD = { BD = corresponding to B 1 ,……，BD j ,……，BD γ }，BD j Is B j A corresponding second target image feature list.
Specifically, the method in S900 further includes the following steps:
s901, according to B, obtaining a second face image vector list B corresponding to B 0 ＝{B 0 1 ，……，B 0 j ，……，B 0 γ }，B 0 j Is the jth second face image vector in B.
Specifically, the mode of acquiring the second facial image vector is consistent with the mode of acquiring the first facial image vector, and by ensuring the consistency of the acquisition modes, the situation that the accuracy of subsequent similarity calculation is influenced due to inconsistent acquired vector forms caused by different modes and the influence on the accuracy of the acquired facial image features can be avoided.
S903, obtaining B according to B and M j Corresponding third similarity list BM j ＝{BM 1 j ，……，BM s j ，……，BM v j }，BM s j Is B j And M s Third similarity.
Specifically, BM s j The obtaining method of (1), F iu hβ Acquisition mode and BM s j The acquisition modes are consistent.
S905, as BM s j ≥F 2 Then, acquire BD j Wherein, F 2 Is a preset third similarity threshold.
Specifically, F 2 The value range of (a) is 0.85-0.9, and a preset similarity threshold value is set by a person skilled in the art according to actual requirements, which is not described herein again.
Specifically, F 0 ＝F 1 ＝F 2 。
Specifically, step S905 further includes the following steps:
s9051, as BM s j ≥F 2 Then, obtain B j Corresponding M s 。
S9053 according to M s And M' s Acquire BD j ＝{M' s1 ，……，M' sε ，……，M' sη }。
S1000, using the BD j And sending the information to the jth second target user.
The invention provides a data processing system for acquiring human face image features, which comprises: the image processing system comprises a database, a processor, a memory for storing computer programs, a first image data processing platform and N second image data processing platforms, wherein the database comprises a first face image list and a preset image information set, and when the computer programs are executed by the processor, the following steps are realized: the method comprises the steps of obtaining a first face image list and a preset image information set from a database, inputting the preset image information set into N second image data processing platforms to obtain a target image vector list corresponding to each second image data processing platform, carrying out clustering processing on target image vectors in each second image data processing platform to obtain a target image vector type set corresponding to each second image data processing platform, obtaining a first target image characteristic set corresponding to a first face image according to the first face image list and the target image vectors, obtaining a second target image characteristic set corresponding to the first face image according to the first target image characteristic and the first image data processing platform, and sending the second target image characteristic set to a first target user. On one hand, the face image searching and identifying method does not compare the face images in the database one by one, compares the preset image information in the database with the preset image information type through clustering processing, and then compares the preset image information with the preset image information meeting the preset type, so that the face image searching and identifying range is reduced, and the real-time performance of face identification is improved; on the other hand, the multiple data query platforms are associated, the facial images are processed at the same time, the clustering number of the data in the data query platforms is dynamically adjusted, the data storage space is improved, and the accuracy of the acquired facial image features is high under the condition that information in a database is not lost.
Although some specific embodiments of the present invention have been described in detail by way of illustration, it should be understood by those skilled in the art that the above illustration is only for the purpose of illustration and is not intended to limit the scope of the invention. It will also be appreciated by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the invention. The scope of the invention is defined by the appended claims.
Claims (10)
1. A data processing system for obtaining features of an image of a human face, the system comprising: the system comprises a database, a processor, a memory storing a computer program, a first image data processing platform and N second image data processing platforms, wherein the database comprises a first face image list and a preset image information set, and when the computer program is executed by the processor, the following steps are realized:
s100, acquiring a first face image list A = { A } from a database 1 ，……，A i ，……，A λ }，A i For a first face image corresponding to the ith first target user, i =1 … … λ, where λ is the number of the first target users;
s200, acquiring a preset image information set M = { M } from a database 1 ，……，M s ，……，M v }，M s For the s-th preset image information in the database, s =1 … … v, where v is the number of the preset image information;
s300, inputting M into N second image data processing platforms, and acquiring a target image vector list C corresponding to each second image data processing platform h ＝{C h0 ，……，C hk ，……，C hμ In which C is hk H =1 … … N, C for the kth target image vector corresponding to the h-th second image data processing stage hk ＝M h+k*N ，M h+k*N For the (h + k × N) th preset image information in M, k =0 … … μ,
s400, for C hk Processing to obtain C h Corresponding target image vector type set C' h ＝{C' h1 ，……，C' hβ ，……，C' ha }，C' hβ ＝{C' 1 hβ ，……，C' u hβ ，……，C' w(β) hβ }，C' u hβ For the u-th target image vector in the beta-th class in the h-th second image data processing platform, u =1 … … w (beta), w (beta) is the number of target image vectors in the beta-th class, beta =1 … … a, and a is C' h The number of medium target image vector types;
s500, according to A and C' u hβ Acquiring a first target image feature set AC = { AC) corresponding to a 1 ,……，AC i ,……，AC λ }，AC i Is A i A corresponding first target image feature list;
s600, according to AC i And a first image data processing platform for acquiring a second target image feature set AD = { AD = (B) }corresponding to A 1 ,……，AD i ,……，AD λ }，AD i Is A i A corresponding second target image feature list;
s700, AD i And sending the information to the ith first target user.
2. The data processing system for acquiring facial image features as claimed in claim 1, further comprising the following steps in S400:
s401, from C h Obtaining intermediate image vector type set H h ＝{H h1 ，……，H hβ ，……，H ha H and h corresponding initial center point set H 1 h ＝{H 1 h1 ，……，H 1 hβ ，……，H 1 ha }，H hβ For the beta-th intermediate image vector, H, in the H-th second image data processing stage 1 hβ From C when not initially clustered h The vector of the acquired beta sample, wherein the initial H hβ ＝Null；
S402, according to C h And H 1 h Obtaining C hk Corresponding target similarity list HC hk ＝{HC 1 hk ，……，HC β hk ，……，HC a hk }，HC β hk Is C hk And H 1 hβ Target similarity between;
s403, when HC β hk Is HC hk At the time of middle minimum, C hk Is inserted into H hβ To obtain C' hβ ＝H hβ ；
S404, according to H hβ Obtaining H h Corresponding set of intermediate center points H 2 h ＝{H 2 h1 ，……，H 2 hβ ，……，H 2 ha In which H 2 hβ The following conditions are met:
s405, repeatedly executing S402-S404 to obtain H h Corresponding target center point set H E h ＝{H E h1 ，……，H E hβ ，……，H E ha In which H E hβ Is the beta-th target central point obtained after the (E-1) th clustering;
s406, when H is E hβ ＝H E-1 hβ Then, make C' hβ ＝H hβ 。
3. The data processing system for acquiring the features of the face image according to claim 2, wherein in S400, a satisfies the following condition:
a＝r 1 (μ+1) 3 +r 2 (μ+1) 2 +r 3 (μ+1)+r 4 wherein r is 1 Is a preset first parameter, r 2 Is a preset second parameter, r 3 Is a preset third parameter, r 4 Is a preset fourth parameter.
4. The data processing system for acquiring the features of the facial image according to claim 1, wherein in S500, the method further comprises the following steps:
s501, according to A i And C' h ObtainingA i Corresponding first similarity list F i h ＝{F i h1 ，……，F i hβ ，……，F i ha }，F i hβ Is A i And C' hβ A first similarity of;
s503, when F i hβ ≥F 0 When obtaining A i Corresponding second similarity list F i hβ ＝{F i1 hβ ，……，F iu hβ ，……，F iw (β) hβ }，F iu hβ Is A i And C' u hβ Second degree of similarity of (D), F 0 Is a preset first similarity threshold;
s505, when F iu hβ ≥F 0 Get AC i。
5. The data processing system for acquiring the features of the human face image according to claim 4, wherein in S501, the following steps are included:
s5011, obtaining a first face image vector list A corresponding to A according to A 0 ＝{A 0 1 ，……，A 0 i ，……，A 0 λ }，A 0 i Is A i A corresponding first face image vector;
s5013 according to A 0 Obtaining A 0 i ＝(A 0 i1 ，……，A 0 ix ，……，A 0 ip ) Wherein A is 0 ix Is A 0 i The x-th bit value, x =1 … … p, p is the dimension of the first face image vector;
s5015, according to C' h Obtaining C' u hβ ＝(C' u1 hβ ，……，C' ux hβ ，……，C' up hβ ) Wherein, C' ux hβ Is C' u hβ The bit value of the x-th bit;
s5017 according to A 0 ix And C' ux hβ Obtaining a first similarity F i hβ Wherein, F i hβ The following conditions are met:
6. the data processing system for acquiring the features of the human face image as claimed in claim 1, wherein the second image data processing platform further comprises C h Corresponding first image feature list FC h ＝{FC h0 ,……，FC hk ，……，FC hμ }，FC hk ＝{FC 1 hk ，……，FC θ hk ，……，FC ζ hk }，FC θ hk Is C hk Corresponding to the first record contained in the theta-th first field in the first image feature list, theta =1 … … ζ is the number of the first fields.
7. The data processing system for acquiring the features of the facial image according to claim 6, wherein in S505, the following steps are included:
s5051, when F iu hβ ≥F 0 Then, obtain F iu hβ Corresponding to C' u hβ ；
S5053 when C' u hβ ＝C hk Get AC i ＝FC hk 。
8. The data processing system for acquiring facial image features of claim 1, wherein the first image data processing platform further comprises a second image feature list M ' = { M = M ' corresponding to M ' 1 ，……，M' s ，……，M' v }，M' s ＝{M' s1 ，……，M' sε ，……，M' sη }，M' sε Is M s The epsilon second field in the corresponding second image feature listThe second record contained, ε =1 … … η, η being the number of second fields.
9. The data processing system for acquiring the features of the facial image as claimed in claim 8, wherein the fields contained in the first field belong to fields in the second field.
10. The data processing system for acquiring the features of the face image according to claim 8, wherein in S600, the following steps are included:
s601, when the theta-th first field is a preset field and the epsilon-th second field is a preset field, obtaining FC θ hk And M' sε Wherein the preset field is ID;
s603, when S = h + k × N, obtaining AD i ＝M' s 。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211089095.XA CN115880745B (en) | 2022-09-07 | 2022-09-07 | Data processing system for acquiring facial image characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211089095.XA CN115880745B (en) | 2022-09-07 | 2022-09-07 | Data processing system for acquiring facial image characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115880745A true CN115880745A (en) | 2023-03-31 |
| CN115880745B CN115880745B (en) | 2023-07-04 |
Family
ID=85769778
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211089095.XA Active CN115880745B (en) | 2022-09-07 | 2022-09-07 | Data processing system for acquiring facial image characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115880745B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015197029A1 (en) * | 2014-06-27 | 2015-12-30 | 北京奇虎科技有限公司 | Human face similarity recognition method and system |
| WO2017162076A1 (en) * | 2016-03-24 | 2017-09-28 | 北京握奇数据股份有限公司 | Face identification method and system |
| CN107679559A (en) * | 2017-09-15 | 2018-02-09 | 广东欧珀移动通信有限公司 | Image processing method, device, computer-readable recording medium and mobile terminal |
| CN109241852A (en) * | 2018-08-10 | 2019-01-18 | 广州杰赛科技股份有限公司 | Face identification method, device with supplementary features, computer equipment |
| WO2020006727A1 (en) * | 2018-07-05 | 2020-01-09 | 深圳前海达闼云端智能科技有限公司 | Face recognition method and device, and server |
| CN110825765A (en) * | 2019-10-23 | 2020-02-21 | 中国建设银行股份有限公司 | Face recognition method and device |
| WO2021082426A1 (en) * | 2019-10-29 | 2021-05-06 | 平安科技（深圳）有限公司 | Human face clustering method and apparatus, computer device, and storage medium |
| CN113377984A (en) * | 2021-06-08 | 2021-09-10 | 中山大学 | Database-based face searching method |
| CN114937306A (en) * | 2022-06-21 | 2022-08-23 | 合肥中科类脑智能技术有限公司 | Target tracking method and system based on face clustering |
| CN115019360A (en) * | 2022-04-29 | 2022-09-06 | 以萨技术股份有限公司 | Matching method and device, nonvolatile storage medium and computer equipment |
-
2022
- 2022-09-07 CN CN202211089095.XA patent/CN115880745B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015197029A1 (en) * | 2014-06-27 | 2015-12-30 | 北京奇虎科技有限公司 | Human face similarity recognition method and system |
| WO2017162076A1 (en) * | 2016-03-24 | 2017-09-28 | 北京握奇数据股份有限公司 | Face identification method and system |
| CN107679559A (en) * | 2017-09-15 | 2018-02-09 | 广东欧珀移动通信有限公司 | Image processing method, device, computer-readable recording medium and mobile terminal |
| WO2020006727A1 (en) * | 2018-07-05 | 2020-01-09 | 深圳前海达闼云端智能科技有限公司 | Face recognition method and device, and server |
| CN109241852A (en) * | 2018-08-10 | 2019-01-18 | 广州杰赛科技股份有限公司 | Face identification method, device with supplementary features, computer equipment |
| CN110825765A (en) * | 2019-10-23 | 2020-02-21 | 中国建设银行股份有限公司 | Face recognition method and device |
| WO2021082426A1 (en) * | 2019-10-29 | 2021-05-06 | 平安科技（深圳）有限公司 | Human face clustering method and apparatus, computer device, and storage medium |
| CN113377984A (en) * | 2021-06-08 | 2021-09-10 | 中山大学 | Database-based face searching method |
| CN115019360A (en) * | 2022-04-29 | 2022-09-06 | 以萨技术股份有限公司 | Matching method and device, nonvolatile storage medium and computer equipment |
| CN114937306A (en) * | 2022-06-21 | 2022-08-23 | 合肥中科类脑智能技术有限公司 | Target tracking method and system based on face clustering |
Non-Patent Citations (2)
| Title |
|---|
| 李振东;钟勇;张博言;曹冬平;: "基于深度特征聚类的海量人脸图像检索", 哈尔滨工业大学学报, no. 11 * |
| 杨之光;艾海舟;: "基于聚类的人脸图像检索及相关反馈", 自动化学报, no. 09 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115880745B (en) | 2023-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108804641B (en) | Text similarity calculation method, device, equipment and storage medium | |
| CN109086394B (en) | Search ranking method and device, computer equipment and storage medium | |
| CN110289050B (en) | Drug-target interaction prediction method based on graph convolution sum and word vector | |
| CN112732883A (en) | Fuzzy matching method and device based on knowledge graph and computer equipment | |
| CN103210368A (en) | Software application recognition | |
| CN110008397B (en) | Recommendation model training method and device | |
| CN110827924B (en) | Clustering method and device for gene expression data, computer equipment and storage medium | |
| CN113298197B (en) | Data clustering method, device, equipment and readable storage medium | |
| CN111382283A (en) | Resource category label labeling method and device, computer equipment and storage medium | |
| US20230021143A1 (en) | Resource selection determination using natural language processing based tiered clustering | |
| CN110609737A (en) | Associated data query method and device, computer equipment and storage medium | |
| CN112560444A (en) | Text processing method and device, computer equipment and storage medium | |
| CN111859004A (en) | Retrieval image acquisition method, device, equipment and readable storage medium | |
| CN110609952A (en) | Data acquisition method and system and computer equipment | |
| CN112231452A (en) | Question-answering method, device, equipment and storage medium based on natural language processing | |
| CN111177436A (en) | Face feature retrieval method, device and equipment | |
| CN111966842A (en) | Multimedia file pushing method, device and equipment | |
| CN110213660B (en) | Program distribution method, system, computer device and storage medium | |
| CN115880745B (en) | Data processing system for acquiring facial image characteristics | |
| CN115248837B (en) | Data processing system for obtaining geographic entity of text | |
| CN115909434B (en) | Data processing system for acquiring facial image characteristics | |
| CN115186138A (en) | Comparison method and terminal for power distribution network data | |
| CN115687579A (en) | Document tag generation and matching method and device and computer equipment | |
| CN115661472A (en) | Image duplicate checking method and device, computer equipment and storage medium | |
| CN111026935B (en) | Cross-modal retrieval reordering method based on adaptive measurement fusion |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |