CN114445925B - Facial recognition intelligent attendance system capable of being automatically loaded and deleted - Google Patents
Facial recognition intelligent attendance system capable of being automatically loaded and deleted Download PDFInfo
- Publication number
- CN114445925B CN114445925B CN202210373200.6A CN202210373200A CN114445925B CN 114445925 B CN114445925 B CN 114445925B CN 202210373200 A CN202210373200 A CN 202210373200A CN 114445925 B CN114445925 B CN 114445925B
- Authority
- CN
- China
- Prior art keywords
- module
- face recognition
- image
- data
- recognition image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000001815 facial effect Effects 0.000 title claims description 13
- 238000012217 deletion Methods 0.000 claims abstract description 25
- 230000037430 deletion Effects 0.000 claims abstract description 25
- 238000000605 extraction Methods 0.000 claims abstract description 22
- 238000010191 image analysis Methods 0.000 claims abstract description 15
- 238000013500 data storage Methods 0.000 claims abstract description 14
- 238000012706 support-vector machine Methods 0.000 claims description 37
- 238000000034 method Methods 0.000 claims description 30
- 238000012545 processing Methods 0.000 claims description 30
- 238000006243 chemical reaction Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000007726 management method Methods 0.000 claims description 9
- 230000003993 interaction Effects 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 3
- 238000013075 data extraction Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000011425 standardization method Methods 0.000 claims description 3
- 230000000737 periodic effect Effects 0.000 claims description 2
- 230000002093 peripheral effect Effects 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 20
- 238000010586 diagram Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 8
- 230000005284 excitation Effects 0.000 description 3
- 238000001914 filtration Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000016776 visual perception Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/10—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people together with the recording, indicating or registering of other data, e.g. of signs of identity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
- H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
- H04L9/3236—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials using cryptographic hash functions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
- H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
- H04L9/3247—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials involving digital signatures
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an intelligent attendance system capable of automatically loading and deleting faces, which comprises an image acquisition module, an image information extraction module, an image analysis module, a main control module, a data receiving interface, a data output interface, a data storage module, a deletion module and a block chain node, the output end of the image acquisition module is connected with the input end of the image information extraction module, the output end of the image information extraction module is connected with the output end of the image analysis module, the image analysis module is connected with the main control module, the main control module is also provided with a data receiving interface, a data output interface and a peripheral terminal, the output end of the data receiving interface is connected with the input end of the data storage module, the output end of the data storage module is connected with the input end of the deleting module, and the output end of the deleting module is provided with a block chain node. The invention can realize the identification of the user data information and the automatic deletion of the historical data information, and has high intelligent and automatic degree.
Description
Technical Field
The invention relates to the field of attendance checking, in particular to a face recognition intelligent attendance checking system capable of being automatically loaded and deleted.
Background
Attendance checking, as the name implies, refers to checking attendance, that is, obtaining attendance of students, employees or certain groups and individuals in a certain place and in a certain time period in a certain way, including attendance, late arrival, early departure, sick and fake, wedding and fake, mourning and fake, official break, working time, overtime and the like. The overall planning, arrangement and the like of the later stages are carried out through the research on the attendance conditions in the former stage and the present stage. The attendance checking is to maintain the normal work order of an enterprise, improve the work efficiency, seriously improve the enterprise discipline and enable the staff to voluntarily follow the working time and the labor discipline.
With the rapid development of artificial intelligence technology, attendance systems in the prior art become intelligent and automatic gradually. The face recognition technology is more and more commonly applied, and as the face recognition data information is applied day by day, the more face recognition data information is accumulated, the historical data information can be deleted in a manual mode, and therefore heavy burden is brought to manpower and labor. Therefore, how to realize the automatic loading and deletion of the intelligent attendance system becomes a problem to be solved urgently.
Disclosure of Invention
Aiming at the defects of the technology, the invention discloses a face recognition intelligent attendance system capable of automatically loading and deleting, which can realize the recognition of user data information and the automatic deletion of historical data information, wherein undeleted data information is transmitted to a block chain network through block chain nodes to realize the permanent storage of the data information.
The invention adopts the following technical scheme:
a facial recognition intelligent attendance system capable of automatic loading and deletion, comprising:
the image acquisition module is used for acquiring user face data information, outputting the acquired user face data information and realizing information interaction with other equipment, wherein the image acquisition module is compatible with a data acquisition module for video information acquisition and comprises an image acquisition module based on an FPGA control chip circuit and a TMS320DM8168 chip control circuit;
the face recognition image information extraction module is used for receiving the user face data characteristics output by the image acquisition module, and comprises an histogram of oriented gradients HOG module, a grayscale map conversion module, a classification module and an image output module, wherein the output end of the histogram of oriented gradients HOG module is connected with the input end of the grayscale map conversion module, the output end of the grayscale map conversion module is connected with the input end of the classification module, and the output end of the classification module is connected with the input end of the image output module;
the image analysis module is used for analyzing the user face data characteristics extracted by the face recognition image information extraction module so as to analyze the collected face recognition image information;
the main control module is used for controlling the acquisition of face recognition image information, and comprises an ARM single chip microcomputer chip and a circuit interface compatible with an FPGA control chip and a TMS320DM8168 chip control circuit interface; the external part of the main control module is also provided with an external terminal, and the external terminal is provided with a USB interface, an Ethernet interface or a wireless data terminal;
the data receiving interface is used for the face recognition image information received by the main control module;
the data output interface is used for outputting the face recognition image information refused to be received by the main control module;
the data storage module is used for storing the face recognition image information received by the main control module;
the deleting module is used for deleting the face recognition image information collected in the set historical time period; and
the block chain node is used for transmitting the face recognition image information refused to be received by the main control module and the face recognition image information acquired in the historical time period after the deletion module deletes the face recognition image information to a block chain network, and realizing the storage, management and application of data information through a data center management platform;
the output end of the image acquisition module is connected with the input end of the face recognition image information extraction module, the output end of the face recognition image information extraction module is connected with the output end of the image analysis module, the image analysis module is connected with the master control module, the master control module is further provided with a data receiving interface, a data output interface and a peripheral terminal, the output end of the data receiving interface is connected with the input end of the data storage module, the output end of the data storage module is connected with the input end of the deletion module, the output end of the deletion module is provided with a block chain node, the block chain node is used for realizing data information interaction with the data center management platform through a block chain network, and the output end of the data output interface is further provided with a block chain node.
As a further technical scheme of the invention, the FPGA control chip circuit is at least provided with 4 channels of data channel interfaces, and the TMS320DM8168 chip control circuit is externally connected with an ARM subsystem, a video processing subsystem, a coding and decoding subsystem and a DSP subsystem.
As a further technical scheme of the invention, a classification module in the face recognition image information extraction module is a Support Vector Machine (SVM) classifier, and the method for realizing data extraction by the face recognition image information extraction module comprises the following steps:
step 2, realizing gray scale conversion through a gray scale image conversion module, converting the data information of the face marked with date into a gray scale image, realizing data information conversion through a Gamma standardization method, and then expressing the output information content through a formula:
in the formula (1), wherein
Expressing the gray value of the face recognition image, and after the gray value of the face recognition image passes through the formula (1), realizing one-half processing of a Gamma value, namely processing input face recognition image information into one half, processing input one-half face recognition image information into one quarter, processing input one-quarter face recognition image information into one eighth, and further realizing the processing and calculation of the face recognition image information;
the face recognition image information adopts gradient calculation during calculation, and when the image is calculated through the gradient, the image is in pixels
The gradient values of the points are noted as:
in the formula (2), wherein
Represents the horizontal gradient values of the face recognition image,
represents the vertical gradient values of the face recognition image,
pixel values representing face recognition image information. At a pixel point
The gradient vector of (c) is taken as:
in formulas (3) and (4), wherein
Represents the gradient values of the face recognition image,
representing the gradient direction of the face recognition image;
by the method, a direction gradient histogram is then constructed, the image is divided into a plurality of modules, each module has 8 × 8 pixels, and the module gradient direction is divided into 9 blocks. Carrying out weighted projection on each pixel in the module in the histogram of the gradient direction, and then counting the histogram of the gradient direction of the module;
then carrying out module normalization processing on the segmented face recognition image information, combining several adjacent modules to further realize normalization processing, then generating a characteristic vector from the normalized face recognition image information, and further realizing the recognition of the face recognition image information of the face recognition image;
1. step 3, through a classification module of the SVM classifier, the method for realizing classification by the SVM classifier is as follows:
recording data information output by SVM classifier
Then, the face recognition image information recognition function of the classified face recognition image is:
in formula (5), where r represents a user face data information parameter input to the SVM classifier,
representing a variable of a user's face recognition image sample,
representing hidden variables input to the support vector machine SVM classifier,
the value space of the hidden variables of the SVM classifier is represented,
representing the description of the face recognition image sample in the SVM classifier, obtaining the optimal parameter r by applying the SVM classifier and data training through minimizing the objective function, and then obtaining the trained face recognition numberThe output according to the information function is:
in the formula (6), the first and second groups,
the representation of the objective function is shown as,
indicating the first in the face recognition process
The training task of the personal face identification data information is that after an optimized data objective function is solved, r is a positive sample of each face identification data information,
is shown as
The value influencing the data information acquisition during the face recognition,
is shown in
The data value of the face data information is acquired during face recognition,
indicating that the first image is obtained in the case of obtaining a positive sample of each piece of facial face recognition data information
And selecting the optimal hidden variable value to optimize the face and face identification information according to the data information value of the face identification.
Further selecting the optimal hidden variable value to optimize the face identification information;
and 4, outputting the data information through an image output module.
As a further embodiment of the present invention, the method for implementing data analysis by the image analysis module is to implement data information analysis by using a Retinex algorithm model, and the specific method is as follows:
assume that each image point in the acquired face recognition image information is taken as
And then recording the data information after image enhancement as:
in the formula (7), wherein
Representing the original face data information image input to the Retinex algorithm model,
the reflection component image of the face recognition image information received by the Retinex algorithm model is represented, L represents the brightness of the face recognition image in the dark environment, and the weight of the Retinex algorithm model is set as follows:
through the logarithm operation of formula (8), then
Representing the center-surrounding function of the face recognition image recognition information, and outputting an expression as:
in formula (9), where c represents a gaussian surrounding scale of the acquired face recognition image information, the face recognition image information output in a dark environment is:
wherein,
in order to output the image(s),
in the form of a gaussian filter function,
is a weight on the scale of the image,
(1) the programmable controller scans the face recognition image information acquired by adopting a sequential scanning mode through the date marked by the face recognition image information and periodically and circularly scans according to the instruction step number or the address number;
(2) when the set historical face recognition image information is retrieved, no jump instruction exists, and when the set historical face recognition image information is not retrieved, a jump instruction appears;
(3) when a jump instruction appears, sequentially executing the user program from the first instruction one by one until the program is finished, then returning to the first instruction again, starting the next scanning, and completing sampling and refreshing of the face recognition image information in each scanning process;
(4) and executing a deleting command, deleting the retrieved face recognition image information, and transmitting the face recognition image information which is not retrieved to the block chain network through the block chain nodes.
Has the positive and beneficial effects that:
the invention can realize the identification of the user data information and the automatic deletion of the historical data information, the undeleted data information is transmitted to the blockchain network through the blockchain node to realize the permanent storage of the data information, the intelligentization and automation degree is high, the intelligent processing of the face data information can be improved, the use is convenient, and the efficiency is high.
When the automatic loading is realized, the synchronous acquisition of face recognition image information acquisition and video screen data information acquisition is realized through the image acquisition module based on the FPGA control chip circuit and the TMS320DM8168 chip control circuit, wherein at least 4 data channel interfaces are arranged through the FPGA control chip circuit, and an ARM subsystem, a video processing subsystem, a coding and decoding subsystem and a DSP subsystem are connected and arranged outside the TMS320DM8168 chip control circuit, so that the automatic loading capacity of data information is improved.
When the facial face recognition image information is deleted, the face recognition image information acquired by a programmable controller in a sequential scanning mode is scanned by the date marked by the face recognition image information, and periodic cyclic scanning is carried out according to the instruction step number or the address number.
According to the method and the device, a block chain technology is introduced, and data information which cannot be deleted is transmitted to an upper data management center through a block chain node, so that permanent storage of the face data information is realized, the data storage capacity is improved, and the data information can be effectively prevented from being tampered.
The face recognition image information of the face recognition image in the dark environment can be obtained through the Retinex algorithm model, and the intelligent application capability of the attendance machine is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings are obtained by those skilled in the art without inventive exercise, wherein:
fig. 1 is an overall architecture schematic diagram of a facial recognition intelligent attendance system capable of automatic loading and deletion according to the invention;
fig. 2 is a schematic diagram of an image acquisition module in an intelligent attendance system with facial recognition function capable of automatic loading and deleting according to the present invention;
fig. 3 is a schematic diagram of a face recognition image information extraction module in an intelligent attendance system with face recognition function, which can be automatically loaded and deleted according to the present invention;
fig. 4 is a schematic diagram of a Retinex algorithm model in an intelligent attendance system with face recognition capable of automatic loading and deleting according to the invention;
fig. 5 is a schematic diagram of a method for extracting images by a face recognition image information extraction module in the intelligent face recognition attendance system capable of automatic loading and deletion according to the invention;
fig. 6 is a schematic diagram of the principle of deleting modules in the intelligent attendance system with facial recognition function, which can be automatically loaded and deleted according to the invention;
fig. 7 is a schematic diagram of deleting data information by a deleting module in the facial recognition intelligent attendance system capable of automatic loading and deleting according to the invention;
fig. 8 is a schematic diagram of a blockchain network system in which blockchain nodes in the face recognition intelligent attendance system can be automatically loaded and deleted according to the present invention.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, and it should be understood that the embodiments described herein are merely for the purpose of illustrating and explaining the present invention and are not intended to limit the present invention.
A face recognition intelligent attendance system capable of automatic loading and deleting, comprising:
the image acquisition module is used for acquiring user face data information, outputting the acquired user face data information and realizing information interaction with other equipment, wherein the image acquisition module is compatible with a data acquisition module for video information acquisition and comprises an image acquisition module based on an FPGA control chip circuit and a TMS320DM8168 chip control circuit;
the face recognition image information extraction module is used for receiving the user face data characteristics output by the image acquisition module and comprises an oriented gradient histogram HOG module, a grey-scale image conversion module, a classification module and an image output module, wherein the output end of the oriented gradient histogram HOG module is connected with the input end of the grey-scale image conversion module, the output end of the grey-scale image conversion module is connected with the input end of the classification module, and the output end of the classification module is connected with the input end of the image output module;
the image analysis module is used for analyzing the human face data characteristics of the user extracted by the human face recognition image information extraction module so as to analyze the collected human face recognition image information;
the main control module is used for controlling the acquisition of face recognition image information, and comprises an ARM single chip microcomputer chip and a circuit interface compatible with an FPGA control chip and a TMS320DM8168 chip control circuit interface; the external terminal is arranged outside the main control module and provided with a USB interface, an Ethernet interface or a wireless data terminal;
the data receiving interface is used for identifying the image information of the human face received by the main control module;
the data output interface is used for outputting the face recognition image information which is refused to be received by the main control module;
the data storage module is used for storing the face recognition image information received by the main control module;
the deleting module is used for deleting the face recognition image information collected in the set historical time period; and
the block chain node is used for transmitting the face recognition image information refused to be received by the main control module and the face recognition image information acquired in the historical time period after the deletion module deletes the face recognition image information to a block chain network, and realizing the storage, management and application of data information through a data center management platform;
the output end of the image acquisition module is connected with the input end of the face recognition image information extraction module, the output end of the face recognition image information extraction module is connected with the output end of the image analysis module, the image analysis module is connected with the master control module, the master control module is further provided with a data receiving interface, a data output interface and a peripheral terminal, the output end of the data receiving interface is connected with the input end of the data storage module, the output end of the data storage module is connected with the input end of the deletion module, the output end of the deletion module is provided with a block chain node, the block chain node is used for realizing data information interaction with the data center management platform through a block chain network, and the output end of the data output interface is further provided with a block chain node.
Through the embodiment, the method and the device can realize the identification of the user data information and the automatic deletion of the historical data information, the undeleted data information is transmitted to the block chain network through the block chain nodes to realize the permanent storage of the data information, the intelligentization and automation degree is high, the intelligent processing of the face data information can be improved, the use is convenient, and the efficiency is high.
In the invention, the FPGA control chip circuit is at least provided with 4 paths of data channel interfaces, and the TMS320DM8168 chip control circuit is externally connected with an ARM subsystem, a video processing subsystem, a coding and decoding subsystem and a DSP subsystem.
In the specific embodiment, the speed and the reliability of realizing face recognition by depending on a PC in the traditional scheme are improved by utilizing the pure hardware parallel processing characteristic of the FPGA, the system design is realized on a hardware platform taking the Cyclone III series EP3C40 of the Altera company as a core chip, and the user data information can be automatically recognized.
In a specific embodiment, the video acquisition module adopts multi-channel video multiplexing and can simultaneously convert four channels of analog video signals into digital video signals, acquired user face video data firstly needs to pass through an AD converter, then chrominance signals and luminance signals of video images are separated, the video signals are integrated through filtering, formats of the video signals are formulated through processing of a scaler, and then the digital video formats are sent to an input interface of a rear-end video processor according to the requirements of a system on the user face video images. The video acquisition module circuit supports 16 paths of digital video input and can receive digital video signals in two formats of ITU-R BT.656 and BT.1120. The multichannel video signal that TVP5158 chip's decoder input end was input is multiplexed at a road output after the digitization, improves the utilization ratio of main control chip VP mouth.
In the invention, the classification module is a Support Vector Machine (SVM) classifier, wherein:
a method for realizing data extraction by applying the face recognition image information extraction module in the embodiment comprises the following steps:
step 2, realizing gray scale conversion through a gray scale image conversion module, converting the data information of the face marked with date into a gray scale image, realizing data information conversion through a Gamma standardization method, and then expressing the output information content through a formula:
in the formula (1), wherein
The gray value of the face recognition image is represented, after the formula (1) is passed, half-processing of Gamma value can be realized, i.e. the input face recognition image information is processed into half, then the input half-processing of face recognition image information is processed into halfOne fourth, then processing the input one fourth of face recognition image information into one eighth, and further realizing the processing and calculation of the face recognition image information;
the face recognition image information adopts gradient calculation during calculation, and when the image is calculated through the gradient, the image is calculated in pixels
The gradient values of the points are noted as:
in the formula (2), wherein
Represents the horizontal gradient values of the face recognition image,
represents the vertical gradient values of the face recognition image,
pixel values representing face recognition image information. At a pixel point
The gradient vector of (c) is taken as:
in formulas (3) and (4), wherein
Representing the gradient values of the face recognition image,
representing the gradient direction of the face recognition image;
by the method, a direction gradient histogram is then constructed, the image is divided into a plurality of modules, each module has 8 × 8 pixels, and the module gradient direction is divided into 9 blocks. Carrying out weighted projection on each pixel in the module in the histogram of the gradient direction, and then counting the histogram of the gradient direction of the module;
then carrying out module normalization processing on the segmented face recognition image information, combining several adjacent modules to further realize normalization processing, then generating a characteristic vector from the normalized face recognition image information, and further realizing the recognition of the face recognition image information of the face recognition image;
recording data information output by SVM classifier
Then, the face recognition image information recognition function of the classified face recognition image is:
in equation (5), where r represents the input to the SVM classifier user facial data information parameter,
representing a user's face recognition image sample variable,
represents the hidden variables input to the support vector machine SVM classifier,
the branch of expressionThe value space of the hidden variables of the SVM classifier,
the face recognition image sample description in the SVM classifier is expressed, the SVM classifier is applied, data training is carried out through a minimized objective function to obtain an optimal parameter r, and the information function output of the trained face recognition data is as follows:
in the case of the formula (6),
the function of the object is represented by,
indicating the first in the face recognition process
The training task of the personal face identification data information is that after an optimized data objective function is solved, r is a positive sample of each face identification data information,
denotes the first
The value influencing the data information acquisition during the face recognition,
is shown in
Acquiring data values of face data information during face recognition,
is shown at the moment of obtaining each facetObtaining the first face identification data information under the condition of positive sample
And selecting the optimal hidden variable value to optimize the face and face identification information according to the data information value of the face identification.
Further selecting the optimal hidden variable value to optimize the face recognition information;
and 4, outputting the data information through an image output module.
In the above embodiment, the two classes are maximally separated by the support vector machine. And training and classifying the support vector machine by using the idea of classification interval. The classification interval is to ensure the minimum classification risk and the highest confidence after classification. The support vector machine classifier simplifies the classification problem and can eliminate a lot of redundant information.
A method for realizing data analysis by applying the image analysis module in the embodiment is to realize data information analysis by applying a Retinex algorithm model, and the method comprises the following steps:
assume that each image point in the acquired face recognition image information is taken as
And then recording the data information after image enhancement as:
in the formula (7), wherein
Representing the original face data information image input to the Retinex algorithm model,
a reflection component image representing the face recognition image information received by the Retinex algorithm model, and L represents the brightness of the face recognition image in the dark environmentAnd (3) setting the weight of the Retinex algorithm model as follows:
through the logarithm operation of formula (8), then
The center surrounding function of the face recognition image recognition information is represented, and the output expression is recorded as:
in formula (9), where c represents a gaussian surrounding scale of the acquired face recognition image information, the face recognition image information output in a dark environment is:
wherein,
in order to output the image(s),
is a function of a gaussian filter, and,
is a weight of the image scale and is,
In a particular embodiment, the Retinex algorithm uses the HSV color model. The HSV color space is established according to human visual perception, where V is independent of color information of an image, H and S are related to the way a person perceives color, and color and brightness separation is achieved using the HSV color space. The brightness of the image is analyzed by utilizing the guide filtering processing without destroying the edge information of the image, then the Gamma correction is used for measuring the light and dark regions of the image, further the proportion of the dark color part and the light color part is increased, the image contrast is more obvious, and finally the whole brightness of the image is calculated by linear stretching. Due to the fact that logarithmic transformation is carried out, the gray level of the image is reduced, the overall brightness of the image is reduced, the image enhancement method is adopted, the Sigmoid function is used for the transformed image, the self-adaptive weight is added, the dynamic range of the image brightness is improved, and meanwhile the details of the image are enhanced. In fig. 4, s represents one kind of input color information in the HSV color space, i.e., data information before enhancement, R represents data information after image brightness enhancement, L represents the length of an input image, and exp represents a summation output after data calculation.
In the invention, the deleting module comprises a programmable controller, face recognition image information with date information, an erasing module and a block chain node, wherein the face recognition image information is connected with the programmable controller, and the erasing module and the block chain node are connected with the programmable controller, and the method for erasing data by the deleting module comprises the following steps:
(1) the programmable controller acquires the face recognition image information in a sequential scanning mode, scans the face recognition image information by the date marked by the face recognition image information, and periodically and circularly scans according to the instruction step number or the address number;
(2) when the set historical face recognition image information is retrieved, no jump instruction exists, and when the set historical face recognition image information is not retrieved, a jump instruction appears;
(3) when a jump instruction appears, sequentially executing the user program from the first instruction one by one until the program is finished, then returning to the first instruction again, starting the next scanning, and completing sampling and refreshing of the face recognition image information in each scanning process;
(4) and executing a deleting command, deleting the retrieved face recognition image information, and transmitting the non-retrieved face recognition image information to the block chain network through the block chain nodes.
In a specific embodiment, the programmable controller is programmed according to user requirements to set data information to be erased by the erasing module, for example, when a user needs to erase face recognition data information in a certain historical period, a date is set on an acquired face recognition image, and the processing of the picture data information is realized by screening the dates in the period. Before erasing, in order to realize permanent storage, traceability and incapability of changing of data information, a block Chain node is arranged on a deletion module, a block Chain (authorized Chain) technology is introduced into the application, and the data information before erasing is transmitted to an upper management center through the block Chain (authorized Chain). The block chain is composed of a data layer, a network layer, a consensus layer, an excitation layer and an intelligent contract layer on the technical architecture. The data layer uses a Merkle tree for data storage, and is structurally connected in a chain manner through blocks. Because the data structure utilizes digital signature, hash function and asymmetric encryption technology, the security performance is higher in data interaction. The network layer is mainly formed by interweaving a plurality of network nodes, data communication and connection are realized by using a point-to-point technology through different network nodes, the technical weakness that a central server is used in the traditional technology is omitted, and node equipment in the network can be intercommunicated and interconnected. In a consensus layer in the network, the existence of a consensus mechanism can carry out consistency interaction on data set in the block chain network, so that the block chain network has better data consensus capability and data attack prevention capability. The excitation layer can provide excitation measures in the block chain, so that the action of the network node is sufficiently excited in the block chain, and the security verification can be carried out. In the intelligent contract layer, various program algorithms are fully utilized to execute and calculate the relationship between various data in the block chain network.
Although specific embodiments of the present invention have been described above, it will be understood by those skilled in the art that these specific embodiments are merely illustrative of and various omissions, substitutions and changes in the form and details of the methods and systems described above may be made by those skilled in the art without departing from the spirit and scope of the invention. For example, it is within the scope of the present invention to combine the steps of the methods described above to perform substantially the same function in substantially the same way to achieve substantially the same result. Accordingly, the scope of the invention is to be limited only by the following claims.
Claims (8)
1. The utility model provides a facial discernment intelligence attendance system that can automatic loading and delete which characterized in that: comprises the following steps:
the image acquisition module is used for acquiring user face data information, outputting the acquired user face data information and realizing information interaction with other equipment, wherein the image acquisition module is compatible with a data acquisition module for video information acquisition and comprises an image acquisition module based on an FPGA control chip circuit and a TMS320DM8168 chip control circuit;
the face recognition image extraction module is used for receiving the user face data information output by the image acquisition module, and comprises an histogram of oriented gradients HOG module, a grey-scale map conversion module, a classification module and an image output module, wherein the output end of the histogram of oriented gradients HOG module is connected with the input end of the grey-scale map conversion module, the output end of the grey-scale map conversion module is connected with the input end of the classification module, and the output end of the classification module is connected with the input end of the image output module;
the image analysis module is used for analyzing the user face data information extracted by the face recognition image extraction module so as to analyze the acquired face recognition image;
the main control module is used for controlling the acquisition of a face recognition image, and comprises an ARM single chip microcomputer chip and a circuit interface compatible with an FPGA control chip and a TMS320DM8168 chip control circuit interface; the external terminal is arranged outside the main control module and provided with a USB interface, an Ethernet interface or a wireless data terminal;
the data receiving interface is used for the face recognition image received by the main control module;
the data output interface is used for outputting the face recognition image which is refused to be received by the main control module;
the data storage module is used for storing the face recognition image received by the main control module;
the deleting module is used for deleting the face recognition image acquired in the set historical time period; and
the block chain node is used for transmitting the face recognition image refused to be received by the main control module and the face recognition image acquired in the historical time period after the deletion module deletes the face recognition image to a block chain network, and realizing the storage, management and application of data information through a data center management platform;
the output end of the image acquisition module is connected with the input end of the face recognition image extraction module, the output end of the face recognition image extraction module is connected with the output end of the image analysis module, the image analysis module is connected with the main control module, the main control module is further provided with a data receiving interface, a data output interface and an external terminal, the output end of the data receiving interface is connected with the input end of the data storage module, the output end of the data storage module is connected with the input end of the deletion module, the output end of the deletion module is provided with a block chain node, the block chain node realizes data information interaction with the data center management platform through a block chain network, and the output end of the data output interface is further provided with a block chain node.
2. The intelligent attendance system with automatic loading and deleting function as claimed in claim 1, wherein: the FPGA control chip circuit is at least provided with 4 paths of data channel interfaces, and an ARM subsystem, a video processing subsystem, a coding and decoding subsystem and a DSP subsystem are connected and arranged outside the TMS320DM8168 chip control circuit.
3. The intelligent attendance system capable of automatic loading and deletion according to claim 1, wherein: the method for realizing data extraction by the face recognition image extraction module comprises the following steps:
step 1, extracting face data information through a direction gradient histogram HOG module, selecting the acquired face data information, and marking the date of the face data information to be extracted;
step 2, realizing gray scale conversion through a gray scale image conversion module, converting the data information of the face marked with date into a gray scale image, realizing data information conversion through a Gamma standardization method, and then expressing the output information content through a formula:
in the formula (1), wherein
Expressing the gray value of the face recognition image, and after the formula (1) is passed, implementing one-half processing of Gamma value, i.e. processing the input face recognition image into one-half, then processing the input one-half face recognition image into one-quarter, then processing the input one-quarter face recognition image into one-eighth, and further implementing the processing and calculation of the face recognition image;
the face recognition image adopts gradient calculation during calculation, and when the image is calculated through the gradient, the image is calculated in pixels
The gradient values of the points are noted as:
in the formula (2), wherein
Represents the horizontal gradient value of the face recognition image,
represents the vertical gradient values of the face recognition image,
pixel values representing a face recognition image; at a pixel point
The gradient vector of (c) is taken as:
in formulas (3) and (4), wherein
Representing the gradient values of the face recognition image,
representing the gradient direction of the face recognition image;
by the method, a direction gradient histogram is constructed, the image is divided into a plurality of modules, each module has 8 pixels by 8 pixels, and the gradient direction of the modules is divided into 9 blocks; carrying out weighted projection on each pixel in the module in the histogram of the gradient direction, and then counting the histogram of the gradient direction of the module;
then carrying out module normalization processing on the segmented face recognition images, combining several adjacent modules to further realize normalization processing, then generating characteristic vectors from the normalized face recognition images, and further realizing the recognition of the face recognition images;
step 3, a classification module of a Support Vector Machine (SVM) classifier is used;
and 4, outputting the data information through an image output module.
4. The intelligent attendance system capable of automatic loading and deletion according to claim 1, wherein: and the classification module in the face recognition image extraction module is a Support Vector Machine (SVM) classifier.
5. The intelligent attendance system capable of automatic loading and deletion according to claim 4, wherein: the method for realizing classification by the SVM classifier comprises the following steps:
recording data information output by SVM classifier
And then the classified face recognition image recognition function is as follows:
in the formula (5), wherein
Representing the user facial data information parameters input to the SVM classifier,
representing a variable of a user's face recognition image sample,
representing hidden variables input to the support vector machine SVM classifier,
representing the value space of the hidden variables of the SVM classifier,
representing the description of the face recognition image sample in the SVM classifier, and obtaining the optimal parameter by applying the SVM classifier and data training through minimizing the objective function
Then the trained face recognition data information function output is:
in the case of the formula (6),
the representation of the objective function is shown as,
indicating first in face recognition process
After an optimized data objective function is solved, r is a positive sample of each face identification data information,
is shown as
The value influencing the data information acquisition during the face recognition,
is shown in
Acquiring data values of face data information during face recognition,
indicating that the first image is obtained in the case of obtaining a positive sample of each piece of facial face recognition data information
And selecting the optimal hidden variable value to optimize the face and face identification information according to the data information value of the face identification.
6. The intelligent attendance system with automatic loading and deleting function as claimed in claim 1, wherein: the method for realizing data analysis by the image analysis module is to realize data information analysis by applying a Retinex algorithm model, and the specific method is as follows:
assume that each image point in the acquired face recognition image is taken as
And recording the data information after image enhancement as:
in the formula (7), wherein
Representing the original face data information image input to the Retinex algorithm model,
the reflection component image of the face recognition image received by the Retinex algorithm model is represented, L represents the brightness of the face recognition image in the dark environment, and the weight of the Retinex algorithm model is set as follows:
through the logarithm operation of formula (8), then
The center surrounding function of the face recognition image recognition information is represented, and the output expression is recorded as:
in formula (9), where c represents a gaussian surrounding scale of the acquired face recognition image, the face recognition image output in a dark environment is:
wherein,
in order to output the image(s),
is a function of a gaussian filter, and,
is a weight on the scale of the image,
7. The intelligent attendance system with automatic loading and deleting function as claimed in claim 1, wherein: the deleting module comprises a programmable controller, a face recognition image with date information, an erasing module and a block chain node, wherein the face recognition image, the erasing module and the block chain node are connected with the programmable controller.
8. The intelligent attendance system with automatic loading and deleting function as claimed in claim 7, wherein: the method for the deletion module to erase the data comprises the following steps:
(1) the programmable controller scans the face recognition image acquired by adopting a sequential scanning mode through the date marked by the face recognition image and performs periodic cycle scanning according to the instruction step number or the address number;
(2) when the set historical face recognition image is retrieved, no jump instruction exists, and when the set historical face recognition image is not retrieved, a jump instruction appears;
(3) when a jump instruction appears, sequentially executing the user program from the first instruction one by one until the program is finished, then returning to the first instruction again, starting the next scanning, and completing the sampling and refreshing of the face recognition image in each scanning process;
(4) and executing a deleting command, deleting the retrieved face recognition image, and transmitting the non-retrieved face recognition image to the block chain network through the block chain link points.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210373200.6A CN114445925B (en) | 2022-04-11 | 2022-04-11 | Facial recognition intelligent attendance system capable of being automatically loaded and deleted |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210373200.6A CN114445925B (en) | 2022-04-11 | 2022-04-11 | Facial recognition intelligent attendance system capable of being automatically loaded and deleted |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114445925A CN114445925A (en) | 2022-05-06 |
| CN114445925B true CN114445925B (en) | 2022-07-22 |
Family
ID=81360409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210373200.6A Active CN114445925B (en) | 2022-04-11 | 2022-04-11 | Facial recognition intelligent attendance system capable of being automatically loaded and deleted |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114445925B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107066958A (en) * | 2017-03-29 | 2017-08-18 | 南京邮电大学 | A kind of face identification method based on HOG features and SVM multi-categorizers |
| CN108876294A (en) * | 2018-06-06 | 2018-11-23 | 曹婧月 | Attendance implementation method and Related product |
| CN109033975A (en) * | 2018-06-27 | 2018-12-18 | 山东大学 | Birds detection, identification and method for tracing and device in a kind of monitoring of seashore |
| CN110032915A (en) * | 2018-01-12 | 2019-07-19 | 杭州海康威视数字技术股份有限公司 | A kind of human face in-vivo detection method, device and electronic equipment |
| CN111078565A (en) * | 2019-12-19 | 2020-04-28 | 上海轻维软件有限公司 | Analysis method of software test result based on HOG feature extraction and SVM multi-classifier |
| CN111582209A (en) * | 2020-05-14 | 2020-08-25 | 重庆邮电大学 | Abnormal behavior supervision method for construction personnel in capital construction site |
| CN111641727A (en) * | 2020-06-16 | 2020-09-08 | 广东奥维信息科技有限公司 | Block chain system integration platform |
| CN111814738A (en) * | 2020-07-27 | 2020-10-23 | 深圳壹账通智能科技有限公司 | Human face recognition method, human face recognition device, computer equipment and medium based on artificial intelligence |
| CN112948355A (en) * | 2021-03-05 | 2021-06-11 | 湖北央中巨石信息技术有限公司 | Method, system, device and medium for destroying distributed network data |
| CN113469044A (en) * | 2021-06-30 | 2021-10-01 | 上海歆广数据科技有限公司 | Dining recording system and method |
| CN113642385A (en) * | 2021-07-01 | 2021-11-12 | 山东师范大学 | Deep learning-based facial nevus identification method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10944546B2 (en) * | 2017-07-07 | 2021-03-09 | Microsoft Technology Licensing, Llc | Blockchain object interface |
-
2022
- 2022-04-11 CN CN202210373200.6A patent/CN114445925B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107066958A (en) * | 2017-03-29 | 2017-08-18 | 南京邮电大学 | A kind of face identification method based on HOG features and SVM multi-categorizers |
| CN110032915A (en) * | 2018-01-12 | 2019-07-19 | 杭州海康威视数字技术股份有限公司 | A kind of human face in-vivo detection method, device and electronic equipment |
| CN108876294A (en) * | 2018-06-06 | 2018-11-23 | 曹婧月 | Attendance implementation method and Related product |
| CN109033975A (en) * | 2018-06-27 | 2018-12-18 | 山东大学 | Birds detection, identification and method for tracing and device in a kind of monitoring of seashore |
| CN111078565A (en) * | 2019-12-19 | 2020-04-28 | 上海轻维软件有限公司 | Analysis method of software test result based on HOG feature extraction and SVM multi-classifier |
| CN111582209A (en) * | 2020-05-14 | 2020-08-25 | 重庆邮电大学 | Abnormal behavior supervision method for construction personnel in capital construction site |
| CN111641727A (en) * | 2020-06-16 | 2020-09-08 | 广东奥维信息科技有限公司 | Block chain system integration platform |
| CN111814738A (en) * | 2020-07-27 | 2020-10-23 | 深圳壹账通智能科技有限公司 | Human face recognition method, human face recognition device, computer equipment and medium based on artificial intelligence |
| CN112948355A (en) * | 2021-03-05 | 2021-06-11 | 湖北央中巨石信息技术有限公司 | Method, system, device and medium for destroying distributed network data |
| CN113469044A (en) * | 2021-06-30 | 2021-10-01 | 上海歆广数据科技有限公司 | Dining recording system and method |
| CN113642385A (en) * | 2021-07-01 | 2021-11-12 | 山东师范大学 | Deep learning-based facial nevus identification method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114445925A (en) | 2022-05-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11263434B2 (en) | Fast side-face interference resistant face detection method | |
| CN103098078B (en) | Smile's detecting system and method | |
| CN108875821A (en) | The training method and device of disaggregated model, mobile terminal, readable storage medium storing program for executing | |
| CN109886153B (en) | Real-time face detection method based on deep convolutional neural network | |
| CN108777815A (en) | Method for processing video frequency and device, electronic equipment, computer readable storage medium | |
| CN112183468A (en) | Pedestrian re-identification method based on multi-attention combined multi-level features | |
| CN106845434B (en) | Image type machine room water leakage monitoring method based on support vector machine | |
| CN111583201B (en) | Transfer learning method for constructing super-resolution pathology microscope | |
| CN113435269A (en) | Improved water surface floating object detection and identification method and system based on YOLOv3 | |
| CN107786867A (en) | Image identification method and system based on deep learning architecture | |
| CN112036520A (en) | Panda age identification method and device based on deep learning and storage medium | |
| Diyasa et al. | Multi-face Recognition for the Detection of Prisoners in Jail using a Modified Cascade Classifier and CNN | |
| CN114037699B (en) | Pathological image classification method, equipment, system and storage medium | |
| CN116434012A (en) | Lightweight cotton boll detection method and system based on edge perception | |
| CN115482529A (en) | Method, equipment, storage medium and device for recognizing fruit image in near scene | |
| CN114445925B (en) | Facial recognition intelligent attendance system capable of being automatically loaded and deleted | |
| CN112084913B (en) | End-to-end human body detection and attribute identification method | |
| CN109766832A (en) | A kind of image pre-processing method, device, system, equipment and storage medium | |
| CN112613521B (en) | Multilevel data analysis system and method based on data conversion | |
| CN111950586B (en) | Target detection method for introducing bidirectional attention | |
| CN116824419A (en) | Dressing feature recognition method, recognition model training method and device | |
| CN111881803A (en) | Livestock face recognition method based on improved YOLOv3 | |
| CN116259091B (en) | Method and device for detecting silent living body | |
| CN110555342A (en) | image identification method and device and image equipment | |
| CN112712124B (en) | Multi-module cooperative object recognition system and method based on deep learning |
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 |