CN113554006A - Face prosthesis system construction method, electronic device and storage medium - Google Patents
Face prosthesis system construction method, electronic device and storage medium Download PDFInfo
- Publication number
- CN113554006A CN113554006A CN202111095752.7A CN202111095752A CN113554006A CN 113554006 A CN113554006 A CN 113554006A CN 202111095752 A CN202111095752 A CN 202111095752A CN 113554006 A CN113554006 A CN 113554006A
- Authority
- CN
- China
- Prior art keywords
- prosthesis
- type
- face
- prostheses
- attack
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010276 construction Methods 0.000 title claims abstract description 10
- 238000000034 method Methods 0.000 claims abstract description 35
- 230000007613 environmental effect Effects 0.000 claims description 36
- 230000003068 static effect Effects 0.000 claims description 15
- 230000006798 recombination Effects 0.000 claims description 8
- 238000005215 recombination Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 230000001815 facial effect Effects 0.000 claims description 4
- 238000005286 illumination Methods 0.000 claims description 3
- 230000008521 reorganization Effects 0.000 claims 1
- 210000003128 head Anatomy 0.000 description 23
- 210000000887 face Anatomy 0.000 description 9
- 230000006872 improvement Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 4
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 2
- 210000004709 eyebrow Anatomy 0.000 description 2
- 239000004744 fabric Substances 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 239000004677 Nylon Substances 0.000 description 1
- 239000006004 Quartz sand Substances 0.000 description 1
- NIXOWILDQLNWCW-UHFFFAOYSA-N acrylic acid group Chemical group C(C=C)(=O)O NIXOWILDQLNWCW-UHFFFAOYSA-N 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 238000005034 decoration Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000011152 fibreglass Substances 0.000 description 1
- 239000006260 foam Substances 0.000 description 1
- -1 full-color sandstone Substances 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 239000010440 gypsum Substances 0.000 description 1
- 229910052602 gypsum Inorganic materials 0.000 description 1
- 239000004816 latex Substances 0.000 description 1
- 229920000126 latex Polymers 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229920001778 nylon Polymers 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000004033 plastic Substances 0.000 description 1
- 229920003023 plastic Polymers 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000007639 printing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 239000011347 resin Substances 0.000 description 1
- 229920005989 resin Polymers 0.000 description 1
- 239000000741 silica gel Substances 0.000 description 1
- 229910002027 silica gel Inorganic materials 0.000 description 1
Images
Landscapes
- Image Processing (AREA)
Abstract
The embodiment of the invention relates to the field of image processing, and discloses a face prosthesis system construction method, electronic equipment and a storage medium, wherein the method comprises the following steps: constructing different types of prostheses containing target faces; attacking a plurality of face recognition systems by adopting the prosthesis to obtain recognition results; determining the type attack level of each prosthesis of the type according to the identification result; the higher the type attack level is, the higher the probability that the corresponding type of false body is recognized as the target face by the face recognition system is; and storing the information of each prosthesis of the type and the information of the corresponding attack level of the type in an information base in an associated manner. According to the scheme, a plurality of face recognition systems are used as attack objects of various types of prostheses, the attack capability of different types of prostheses is obtained according to the recognition result after attack, and the type attack level of the corresponding type of prostheses is determined, so that the face prosthesis system capable of comprehensively evaluating the attack capability of various types of prostheses is constructed.
Description
Technical Field
The present invention relates to the field of image processing, and in particular, to a method for constructing a human face prosthesis system, an electronic device, and a storage medium.
Background
The method for carrying out prosthesis attack aiming at the face recognition system means that a face prosthesis which contains a target face but is not the person is input into the face recognition system, so that the face recognition system can recognize the prosthesis as the person as far as possible. Once the face recognition system identifies the prosthesis as principal, it indicates that the attack was successful. The reality of the prosthesis attack is the similarity between the prosthesis and the real person.
The existing construction of a prosthesis system does not form a complete scale system, related personnel only manufacture a prosthesis according to experience to attack a face recognition system, and the constructed prosthesis is relatively simple, so that the safety and the recognition effect of the face recognition system cannot be comprehensively and effectively verified.
Disclosure of Invention
The embodiment of the invention aims to provide a face prosthesis system construction method, electronic equipment and a storage medium, wherein different types of face prostheses are constructed to attack a plurality of face recognition systems, and the recognition results under different indexes are obtained to comprehensively evaluate the attack strengths of the different types of face prostheses, so that face prosthesis systems with different attack strength grades are constructed.
In order to solve the above technical problem, an embodiment of the present invention provides a method for constructing a human face prosthesis system, including:
constructing different types of prostheses containing target faces;
attacking a plurality of face recognition systems by adopting the prosthesis to obtain recognition results;
determining the type attack level of each prosthesis of the type according to the identification result; the higher the type attack level is, the higher the probability that the corresponding type of false body is recognized as the target face by the face recognition system is;
and storing the information of each prosthesis of the type and the information of the corresponding attack level of the type in an information base in an associated manner.
An embodiment of the present invention also provides an electronic device, including:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of facial prosthesis architecture described above.
Embodiments of the present invention also provide a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program implements the facial prosthesis system construction method as described above.
Compared with the prior art, the embodiment of the invention has the advantages that different types of prostheses containing target human faces are constructed; attacking a plurality of face recognition systems by adopting a prosthesis to obtain recognition results; determining the type attack level of each type of prosthesis according to the identification result; and storing the information of each type of prosthesis and the information of the corresponding type attack level in an information base in a correlation manner so as to construct a human face prosthesis system. According to the scheme, a plurality of face recognition systems are used as attack objects of various types of prostheses, the attack capability of different types of prostheses is obtained according to the recognition result after attack, and the type attack level of the corresponding type of prostheses is determined, so that the face prosthesis system capable of comprehensively evaluating the attack capability of various types of prostheses is constructed.
Drawings
FIG. 1 is a first flowchart of a method for constructing a human face prosthesis system according to an embodiment of the present invention;
FIG. 2 is a specific flowchart II of a face prosthesis architecture construction method according to an embodiment of the present invention;
FIG. 3 is a third flowchart of a method for constructing a human face prosthesis system according to an embodiment of the present invention;
FIG. 4 is a detailed flowchart IV of a face prosthesis architecture construction method according to an embodiment of the present invention;
FIG. 5 is a detailed flowchart of a method for constructing a human face prosthesis system according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, it will be appreciated by those of ordinary skill in the art that numerous technical details are set forth in order to provide a better understanding of the present application in various embodiments of the present invention. However, the technical solution claimed in the present application can be implemented without these technical details and various changes and modifications based on the following embodiments.
An embodiment of the present invention relates to a method for constructing a human face prosthesis system, and as shown in fig. 1, the method for constructing a human face prosthesis system provided in this embodiment includes the following steps.
Step 101: different types of prostheses are constructed that contain the target face.
The prosthesis in this embodiment is defined as an objective existence body in all forms of a living body of a non-target person constructed for a target face image stored in a face recognition system. These prostheses are constructed for the purpose of attacking the face recognition system, so that the face recognition system determines whether the prosthesis is a target face image stored in the face recognition system by performing face recognition on the prosthesis.
Specifically, when the prosthesis is constructed, the constructed prosthesis can be classified into different types to obtain different types of prostheses. The dividing manner of the prosthesis type is not limited in this embodiment. For example, the prosthesis may comprise: the two-dimensional static paper image comprises printing paper, matte photographic paper, highlight photographic paper, suede photographic paper, matte powder and bright copper; a two-dimensional non-paper-like prosthesis comprising: cloth, silk, acrylic, portrait cloth, etc.; the two-dimensional static electronic image comprises a mobile phone, a tablet computer and a computer; the two-dimensional dynamic electronic image comprises a recorded video and a synthesized video; the three-dimensional mask comprises a plastic mask, a 3D paper mask, a latex mask and a silica gel mask; the three-dimensional head die comprises foam, resin, full-color sandstone, quartz sand, nylon, glass fiber reinforced plastic, gypsum and the like.
Step 102: and attacking a plurality of face recognition systems by adopting the prosthesis to obtain recognition results.
Specifically, the essence of the process of attacking the face recognition system by the prosthesis is that the face recognition system scans the prosthesis to obtain a face image, and compares the face image obtained by scanning with the face images stored in the face image library to judge whether the faces in the two face images are the same person. The judgment as to whether or not the same person is present can be determined by recognizing the similarity between the faces in the two face images. In this embodiment, the recognition result output by the face recognition system is a similarity value between a face included in a face image corresponding to the prosthesis and a face in a face image in the face image library.
Step 103: determining the type attack level of each type of prosthesis according to the identification result; the higher the type attack level is, the higher the probability that the corresponding type prosthesis is recognized as the target face by the face recognition system is.
Specifically, for each type of prosthesis, multiple face recognition systems are used as attack objects, and a recognition result, that is, a similarity value between a face included in a face image corresponding to the prosthesis and a face in a face image library is obtained. According to the size of the similarity value, the attack strength value of different types of false body attacking face recognition systems can be obtained in a quantification mode (the attack strength value is in proportion to the similarity value). And carrying out grade division on the attack strength value according to a preset interval division rule to obtain a plurality of attack grades. The classification is carried out according to the prosthesis type dimension, and different types of prostheses have different types of attack levels. The higher the type attack level is, the higher the probability that the corresponding type prosthesis is recognized as the target face by the face recognition system is.
In one example, this step may be implemented as follows.
And comprehensively scoring the recognition results obtained after the same type of prosthesis attacks a plurality of face recognition systems, and determining the type attack grade of each type of prosthesis according to the scoring condition.
Specifically, the plurality of face recognition systems are respectively attacked to the same type of prosthesis, so that a plurality of recognition results can be obtained, and the recognition results represent similarity values between faces contained in face images corresponding to the type of prosthesis and faces in face images in a face image library to different degrees. By comprehensively scoring the recognition results, for example, calculating a weighted average of a plurality of similarity values, the overall attack capability of the prosthesis attacking a plurality of face recognition systems can be obtained. According to the scoring situation, such as the weighted average value of a plurality of similarity values, the type attack level of the corresponding type of prosthesis can be accurately determined.
Step 104: and storing the information of each type of prosthesis and the information of the corresponding type of attack level in an information base in an associated manner.
The information of the prosthesis may include an ontology of the prosthesis, a paper/electronic document (such as an image and a video) recording the ontology, and storage location information for storing the ontology of the prosthesis and the paper/electronic document.
Compared with the related art, the embodiment constructs different types of prostheses containing the target human face; attacking a plurality of face recognition systems by adopting a prosthesis to obtain recognition results; determining the type attack level of each type of prosthesis according to the identification result; and storing the information of each type of prosthesis and the information of the corresponding type attack level in an information base in a correlation manner so as to construct a human face prosthesis system. According to the scheme, a plurality of face recognition systems are used as attack objects of various types of prostheses, the attack capability of different types of prostheses is obtained according to the recognition result after attack, and the type attack level of the corresponding type of prostheses is determined, so that the face prosthesis system capable of comprehensively evaluating the attack capability of various types of prostheses is constructed.
Another embodiment of the invention relates to a method for constructing a human face prosthesis system, which is an improvement of the method shown in fig. 1. As shown in fig. 2, the improvement is that step 101 may include the following sub-steps.
Substep 1011: for the same type of prosthesis, multiple prosthesis samples of different finenesses are constructed.
Wherein the fineness of the prosthesis can be described in terms of the clarity, coarseness, and the like of the prosthesis. For example, for two-dimensional static paper images, two-dimensional static electronic images, two-dimensional dynamic electronic image type prostheses, the finesse may include: the resolution, definition, and smooth playing degree of continuous frame images; for three-dimensional mask and three-dimensional head model type prostheses, the fineness may include: the shape, texture, degree of quality of color, etc. of the prosthesis.
Specifically, when constructing multiple prosthesis samples with different fineness, an original sample of the same type of prosthesis may be constructed first, and then the original sample may be adjusted in terms of the content included in the fineness to obtain multiple derivative samples. The original sample and the derivative sample are collectively used as a prosthesis sample. When the original sample is adjusted on the fine content, the original sample can be adjusted wholly, or the original sample can be adjusted locally, for example, the resolution and the definition of a two-dimensional static paper image, a two-dimensional static electronic image, and a two-dimensional dynamic electronic image type prosthesis sample, and the playing fluency of a continuous frame image are adjusted locally, or the quality of the shape, the texture, and the color of a three-dimensional mask and a three-dimensional head model type prosthesis sample is adjusted locally.
Accordingly, step 103 may include the following sub-steps.
Substep 1031: for the prosthesis of the same type, obtaining the fineness attack grades of the prosthesis with different fineness of the prosthesis under the prosthesis of the type according to the corresponding recognition results of the prosthesis samples with different fineness under the prosthesis of the type; wherein, the higher the fineness attack level is, the higher the probability that the prosthesis with corresponding fineness is recognized as the target face by the face recognition system is.
Specifically, after prosthesis samples with different fineness in different prosthesis types are formed, the prosthesis samples with different fineness in different prosthesis types can be adopted to attack the face recognition system, so that the face recognition system compares a face image corresponding to the scanned prosthesis sample with a stored face image in the face image library, judges whether the faces in the two face images are the same person, and outputs a corresponding recognition result, namely a similarity value between the face contained in the face image corresponding to the prosthesis and the face in the face image library. The attack strength values of the prosthesis samples with different finenesses in different types can be quantized according to the size of the similarity value (the attack strength values are in proportion to the similarity value). And carrying out grade division on the attack strength value according to a preset interval division rule to obtain a plurality of attack grades. The classification is carried out according to the fineness dimensionality of the prosthesis under each prosthesis type, and the prostheses with different fineness under different types have different fineness attack grades. Wherein, the higher the fineness attack level is, the higher the probability that the prosthesis representing the corresponding fineness is recognized as the target face by the face recognition system is.
Compared with the related technology, the embodiment continues to refine the prosthesis under each type from the aspect of fineness to construct the prosthesis samples with different fineness on the basis of the prosthesis types, and determines the fineness attack levels of the prosthesis samples with different fineness under different prosthesis types according to the recognition results obtained after the prosthesis samples attack the face recognition system, so that the attack strength of the prosthesis can be evaluated more finely, and the face prosthesis system with different attack strength levels is constructed.
Another embodiment of the invention relates to a method for constructing a human face prosthesis system, which is an improvement of the method shown in fig. 1. As shown in fig. 3, the improvement is that, in step 101, when the prosthesis type is a three-dimensional mask or a three-dimensional head model, the following steps may be included after step 101.
Step 105: and (3) shading and/or decorating the key points of the face on the prosthesis.
Specifically, after the three-dimensional mask or three-dimensional head model type prosthesis is formed, for example, after the head model is formed, the key points of the face on the head model, such as the key points of eyebrows, eyes, nose, mouth, etc., may be blocked by other objects, or these positions may be decorated, for example, by wearing glasses, a mask, a wig, etc., so as to expand the existence forms of more prostheses.
Accordingly, step 102 may include the following sub-steps.
Substep 1021: and attacking a plurality of face recognition systems by adopting the shielded and/or decorated prosthesis to obtain recognition results. Wherein the information of the prosthesis comprises: face keypoints on the prosthesis are occluded and/or decorated.
Specifically, the prosthesis after occlusion and/or decoration still belongs to a prosthesis of a three-dimensional mask or three-dimensional head model type, and therefore, after a corresponding recognition result is obtained subsequently, the type attack level obtained based on the recognition result still belongs to the type attack level of the prosthesis of the three-dimensional mask or three-dimensional head model type. Accordingly, the information of the prosthesis stored in the information base includes: face keypoints on the prosthesis are occluded and/or decorated.
Compared with the related technology, the embodiment aims at the prostheses of the three-dimensional mask or three-dimensional head model type, the prostheses are shielded and/or decorated, the processed prostheses are adopted to attack the face recognition system, the form sample of the prostheses of the type recognized by the face recognition system is expanded, so that the aggressivity of the prostheses of the three-dimensional mask or three-dimensional head model type is more comprehensively evaluated, and the face prosthesis systems with different attack intensity levels can be more perfectly constructed.
Another embodiment of the invention relates to a method for constructing a human face prosthesis system, which is an improvement of the method shown in fig. 1. As shown in fig. 4, the improvement is that, in step 101, when the prosthesis type is a three-dimensional mask or a three-dimensional head model, the following steps may be included after step 101.
Step 106: adjusting the environmental parameters of the prosthesis; wherein the environmental parameters include: at least one of temperature, humidity, illumination intensity.
Specifically, after the three-dimensional mask or three-dimensional head model type prosthesis is formed, for example, after the head model is formed, the head model may be placed in different specified environmental states to obtain the prosthesis under different environmental parameters. The environmental state may be described by environmental parameters including, for example, at least one of temperature, humidity, and illumination intensity.
On this basis, step 102 may comprise the following sub-steps.
Substep 1022: and attacking a plurality of face recognition systems by adopting the prosthesis under different environmental parameters to obtain recognition results.
In particular, the prostheses under different environmental parameters still belong to the respective prosthesis type, i.e. the three-dimensional mask or three-dimensional head model type. Therefore, after the corresponding recognition result is obtained subsequently, the type attack level obtained based on the recognition result still belongs to the type attack level of the three-dimensional mask or the three-dimensional head model.
To embody the refinement of the type attack level by different environmental parameters, step 103 may further include the following sub-steps.
Substep 1032: obtaining the environmental attack levels of the prosthesis under different environmental parameters according to the corresponding recognition results of the prosthesis under different environmental parameters; the higher the environmental attack level is, the higher the probability that the prosthesis under the corresponding environmental parameters is recognized as a target face by the face recognition system is; the information of the prosthesis includes: information of the environmental parameters in which the prosthesis is located.
Specifically, the environmental parameters of the three-dimensional mask or three-dimensional head model type prosthesis are adjusted, the prosthesis under different environmental parameters is adopted to attack the face recognition system, and a recognition result is obtained, namely, the similarity value between the face contained in the face image corresponding to the prosthesis and the face in the face image library. The attack strength value of the prosthesis under different environmental parameters can be obtained quantitatively according to the size of the similarity value (the attack strength value is in proportion to the similarity value). And carrying out grade division on the attack strength value according to a preset interval division rule to obtain a plurality of attack grades. The false body is divided according to the dimension of the environmental parameter where the false body is located, and the false body under different environmental parameters has different environmental attack levels. The higher the environmental attack level is, the higher the probability that the prosthesis representing the corresponding environmental parameters is recognized as the target face by the face recognition system is.
Compared with the related technology, the embodiment continues to refine the prosthesis under the prosthesis type of the three-dimensional mask or the three-dimensional head model from different angles of the environmental parameters on the basis of the prosthesis type to construct the prosthesis under different environmental parameters, and determines the environmental attack levels of the prosthesis under different environmental parameters under the three-dimensional mask or the three-dimensional head model according to the recognition result output after the face recognition system recognizes the prosthesis, so that the attack strength of the prosthesis can be evaluated more finely, and the face prosthesis system with different attack strength levels is constructed.
Another embodiment of the invention relates to a method for constructing a human face prosthesis system, which is an improvement of the method shown in fig. 1. As shown in fig. 5, the improvement is that after step 101, the following steps may be included.
Step 107: and performing partial region recombination on the human faces in the different types of prostheses to obtain at least one mixed type of prosthesis.
Specifically, because the different types of prostheses exist in different forms, when the different types of prostheses are recombined, the forms of the prostheses can be adaptively adjusted and then recombined.
For example, a certain image frame in the two-dimensional dynamic electronic image can be extracted to be used as a static image and recombined with the two-dimensional static electronic image; for another example, the images in the two-dimensional static electronic image and the two-dimensional dynamic electronic image can be printed, and the printed paper image and the two-dimensional static paper image are recombined; for another example, these printed paper images or two-dimensional static paper images are recombined with a three-dimensional mask and a three-dimensional head model. During recombination, the face part areas in the prosthesis can be recombined in modes of interchange, superposition and the like to obtain a mixed type prosthesis, and the mixed type prosthesis covers the characteristics of different types of prostheses, so that the safety level of the face recognition system can be more comprehensively evaluated.
In one example, this step may be implemented as follows.
For different types of prostheses which attack too many individual face recognition systems, partial region recombination is carried out on the faces in the different types of prostheses to obtain mixed types of prostheses based on the attack levels of the different types of prostheses.
Specifically, the two or more types of prostheses that are recombined in the present embodiment may be prostheses that have already completed attacking the face recognition system and obtained a type attack level. According to the corresponding attack level of the types of the prostheses, the different types of prostheses can be flexibly and purposefully recombined. For example, two types of prostheses with higher type attack levels may be reassembled, two types of prostheses with lower type attack levels may be reassembled, or two types of prostheses with higher type attack levels may be reassembled.
In one example, the partial region reorganizing of the human face among different types of prostheses to obtain a mixed type of prosthesis may include: and carrying out prosthesis recombination on the positions of the key points of the human face in the different types of prostheses to obtain mixed type prostheses.
For example, the positions of key points of a human face in a two-dimensional static paper image, a two-dimensional static electronic image and a two-dimensional dynamic electronic image, such as eyebrows, eyes, a nose, a mouth and the like, can be cut off and pasted to a three-dimensional mask and a three-dimensional head model type prosthesis, such as a human head model, corresponding key points of the human face to carry out face region recombination, and a mixed type prosthesis is obtained.
On this basis, step 102 may comprise the following sub-steps.
Substep 1023: and attacking a plurality of face recognition systems by adopting mixed type prosthesis to obtain recognition results.
In particular, a mixed type prosthesis may be used as a special type of prosthesis, so that after a subsequent corresponding recognition result is obtained, a type attack level obtained based on the recognition result belongs to the type attack level of the corresponding mixed type prosthesis.
Compared with the related technology, the embodiment continuously constructs mixed type prosthesis enriched prosthesis types from the heavy angles of the face regions among the multiple types of prostheses on the basis of the prosthesis types, and determines the type attack level of the prostheses under the mixed type according to the identification results obtained after the prostheses attack the face identification system, so that the attack strength of the prostheses can be more comprehensively evaluated, and a more complete face prosthesis system with different attack strength levels is constructed.
Another embodiment of the invention relates to an electronic device, as shown in FIG. 6, comprising at least one processor 202; and a memory 201 communicatively coupled to the at least one processor 202; wherein the memory 201 stores instructions executable by the at least one processor 202, the instructions being executable by the at least one processor 202 to enable the at least one processor 202 to perform any of the method embodiments described above.
Where the memory 201 and the processor 202 are coupled in a bus, the bus may comprise any number of interconnected buses and bridges that couple one or more of the various circuits of the processor 202 and the memory 201 together. The bus may also connect various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface provides an interface between the bus and the transceiver. The transceiver may be one element or a plurality of elements, such as a plurality of receivers and transmitters, providing a means for communicating with various other apparatus over a transmission medium. The data processed by the processor 202 is transmitted over a wireless medium through an antenna, which further receives the data and transmits the data to the processor 202.
The processor 202 is responsible for managing the bus and general processing and may also provide various functions including timing, peripheral interfaces, voltage regulation, power management, and other control functions. And memory 201 may be used to store data used by processor 202 in performing operations.
Another embodiment of the present invention relates to a computer-readable storage medium storing a computer program. The computer program realizes any of the above-described method embodiments when executed by a processor.
That is, as can be understood by those skilled in the art, all or part of the steps in the method for implementing the embodiments described above may be implemented by a program instructing related hardware, where the program is stored in a storage medium and includes several instructions to enable a device (which may be a single chip, a chip, or the like) or a processor (processor) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
It will be understood by those of ordinary skill in the art that the foregoing embodiments are specific examples for carrying out the invention, and that various changes in form and details may be made therein without departing from the spirit and scope of the invention in practice.
Claims (12)
1. A human face prosthesis system construction method is characterized by comprising the following steps:
constructing different types of prostheses containing target faces;
attacking a plurality of face recognition systems by adopting the prosthesis to obtain recognition results;
determining the type attack level of each prosthesis of the type according to the identification result; the higher the type attack level is, the higher the probability that the corresponding type of false body is recognized as the target face by the face recognition system is;
and storing the information of each prosthesis of the type and the information of the corresponding attack level of the type in an information base in an associated manner.
2. The method of claim 1, wherein the prosthesis type comprises: at least one of a two-dimensional static paper image, a two-dimensional static electronic image, a two-dimensional dynamic electronic image, a three-dimensional mask, and a three-dimensional head model.
3. The method of claim 1, wherein constructing different types of prostheses including a target face comprises:
constructing a plurality of prosthesis samples with different fineness for the same type of prosthesis;
determining the type attack level of each prosthesis according to the identification result, comprising:
for the same type of prosthesis, obtaining the fineness attack grades of the prosthesis with different fineness of the prosthesis under the type of prosthesis according to the identification result corresponding to the prosthesis sample with different fineness under the type of prosthesis;
wherein the higher the fineness attack level, the higher the probability that a prosthesis of corresponding fineness is recognized as the target face by the face recognition system.
4. The method of claim 2, wherein when the prosthesis type is the three-dimensional mask or the three-dimensional head model, after the constructing a different type of prosthesis containing a target face, further comprising:
shielding and/or decorating the key points of the face on the prosthesis;
the system for attacking a plurality of human face recognition systems by adopting the prosthesis comprises:
attacking said plurality of face recognition systems with said occluded and/or decorated prosthesis;
wherein the information of the prosthesis comprises: the face key points on the prosthesis are occluded and/or decorated.
5. The method of claim 2, wherein when the prosthesis type is the three-dimensional mask or the three-dimensional head model, after the constructing a different type of prosthesis containing a target face, further comprising:
adjusting the environmental parameters of the prosthesis;
the system for attacking a plurality of human face recognition systems by adopting the prosthesis comprises:
attacking the plurality of face recognition systems with the prosthesis under different environmental parameters;
wherein the environmental parameters include: at least one of temperature, humidity, illumination intensity.
6. The method of claim 5, wherein said determining a type attack level for each of said types of prostheses based on said identification comprises:
obtaining the environmental attack levels of the prosthesis under different environmental parameters according to the identification results corresponding to the prosthesis under different environmental parameters;
the higher the environmental attack level is, the higher the probability that the prosthesis under the corresponding environmental parameters is recognized as the target face by the face recognition system is; the information of the prosthesis comprises: information of an environmental parameter in which the prosthesis is located.
7. The method of claim 1, wherein after constructing the different types of prostheses including the target face, the method further comprises:
carrying out partial region recombination on the human faces in the different types of prostheses to obtain at least one mixed type of prosthesis;
the system for attacking a plurality of human face recognition systems by adopting the prosthesis comprises:
and attacking the plurality of face recognition systems using the mixed type prosthesis.
8. The method according to claim 7, wherein the partial region reorganizing is performed on the human face in the different types of prostheses to obtain a mixed type of prosthesis, and the method comprises the following steps:
and for different types of prostheses which attack the plurality of face recognition systems, carrying out partial region recombination on the faces in the different types of prostheses to obtain the mixed type of prostheses based on the attack grades of the types of the prostheses.
9. The method according to claim 8, wherein the performing partial region reorganization on the human face in the different types of prostheses to obtain the mixed type of prostheses comprises:
and carrying out prosthesis recombination on the positions of the key points of the human face in the different types of prostheses to obtain the mixed type prosthesis.
10. The method of claim 1, wherein said determining a type attack level for each of said types of prostheses based on said identification comprises:
and comprehensively scoring the recognition results obtained after the same type of prosthesis attacks the plurality of face recognition systems, and determining the type attack level of each type of prosthesis according to the scoring condition.
11. An electronic device, comprising:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of constructing a facial prosthesis architecture of any one of claims 1 to 10.
12. A computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the facial prosthesis architecture construction method of any one of claims 1 to 10.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111095752.7A CN113554006A (en) | 2021-09-18 | 2021-09-18 | Face prosthesis system construction method, electronic device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111095752.7A CN113554006A (en) | 2021-09-18 | 2021-09-18 | Face prosthesis system construction method, electronic device and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113554006A true CN113554006A (en) | 2021-10-26 |
Family
ID=78134444
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111095752.7A Pending CN113554006A (en) | 2021-09-18 | 2021-09-18 | Face prosthesis system construction method, electronic device and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113554006A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020231249A1 (en) * | 2019-05-16 | 2020-11-19 | Cybersecurity Malaysia | Method and system to determine risks associated with spoof attacks |
CN112668510A (en) * | 2020-12-31 | 2021-04-16 | 公安部第三研究所 | Method, system, device, processor and storage medium for realizing performance test of three-dimensional face recognition equipment |
US20210150240A1 (en) * | 2019-11-14 | 2021-05-20 | Nec Laboratories America, Inc. | Face spoofing detection using a physical-cue-guided multi-source multi-channel framework |
CN113312965A (en) * | 2021-04-14 | 2021-08-27 | 重庆邮电大学 | Method and system for detecting unknown face spoofing attack living body |
-
2021
- 2021-09-18 CN CN202111095752.7A patent/CN113554006A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020231249A1 (en) * | 2019-05-16 | 2020-11-19 | Cybersecurity Malaysia | Method and system to determine risks associated with spoof attacks |
US20210150240A1 (en) * | 2019-11-14 | 2021-05-20 | Nec Laboratories America, Inc. | Face spoofing detection using a physical-cue-guided multi-source multi-channel framework |
CN112668510A (en) * | 2020-12-31 | 2021-04-16 | 公安部第三研究所 | Method, system, device, processor and storage medium for realizing performance test of three-dimensional face recognition equipment |
CN113312965A (en) * | 2021-04-14 | 2021-08-27 | 重庆邮电大学 | Method and system for detecting unknown face spoofing attack living body |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US12039454B2 (en) | Microexpression-based image recognition method and apparatus, and related device | |
CN106303233B (en) | A kind of video method for secret protection based on expression fusion | |
CN101558431B (en) | Face authentication device | |
CN110088776A (en) | For training the computer equipment of deep neural network | |
CN110458185A (en) | Image-recognizing method and device, storage medium, computer equipment | |
CN108198130B (en) | Image processing method, image processing device, storage medium and electronic equipment | |
WO2023040679A1 (en) | Fusion method and apparatus for facial images, and device and storage medium | |
CN110288513B (en) | Method, apparatus, device and storage medium for changing face attribute | |
CN113449704B (en) | Face recognition model training method and device, electronic equipment and storage medium | |
CN113420731B (en) | Model training method, electronic device and computer-readable storage medium | |
CN111967319B (en) | Living body detection method, device, equipment and storage medium based on infrared and visible light | |
CN109711384A (en) | A kind of face identification method based on depth convolutional neural networks | |
CN111062314A (en) | Image selection method and device, computer readable storage medium and electronic equipment | |
CN109886873A (en) | A kind of simulated portrait generation method and device based on deep learning | |
CN112272295A (en) | Method for generating video with three-dimensional effect, method for playing video, device and equipment | |
CN111080754B (en) | Character animation production method and device for connecting characteristic points of head and limbs | |
CN114973349A (en) | Face image processing method and training method of face image processing model | |
CN109299702A (en) | A kind of Human bodys' response method and system based on depth space-time diagram | |
CN109543593A (en) | Detection method, electronic equipment and the computer readable storage medium of replay attack | |
CN113392769A (en) | Face image synthesis method and device, electronic equipment and storage medium | |
Gonzalez-Soler et al. | Semi-synthetic data generation for tattoo segmentation | |
CN113554006A (en) | Face prosthesis system construction method, electronic device and storage medium | |
CN111652878A (en) | Image detection method, image detection device, computer equipment and storage medium | |
CN112221133A (en) | Game picture customizing method, cloud server, terminal and storage medium | |
CN108288034B (en) | A kind of method for evaluating quality and system of game design |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211026 |
|
RJ01 | Rejection of invention patent application after publication |