CN107464322A - A kind of method based on recognition of face security monitoring personnel's access - Google Patents
A kind of method based on recognition of face security monitoring personnel's access Download PDFInfo
- Publication number
- CN107464322A CN107464322A CN201710673279.3A CN201710673279A CN107464322A CN 107464322 A CN107464322 A CN 107464322A CN 201710673279 A CN201710673279 A CN 201710673279A CN 107464322 A CN107464322 A CN 107464322A
- Authority
- CN
- China
- Prior art keywords
- image
- equipment
- personnel
- pixel
- value
- 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.)
- Withdrawn
Links
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
- G07C9/00—Individual registration on entry or exit
- G07C9/30—Individual registration on entry or exit not involving the use of a pass
- G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
- G07C9/37—Individual registration on entry or exit not involving the use of a pass in combination with an identity check using biometric data, e.g. fingerprints, iris scans or voice recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
Abstract
The present invention relates to a kind of method based on recognition of face security monitoring personnel's access, this method includes:There is provided and run the intelligent access system with face identification functions, the intelligent access system includes test carpet backing and installation system, personnel inspection equipment, apparatus for controlling access and real-time display equipment, the apparatus for controlling access, the personnel inspection equipment and the real-time display equipment are all disposed within the front of the test carpet backing and installation system, the personnel inspection equipment is used to carry out personnel's detection based on image recognition to that will enter personnel to obtain and export target name, the apparatus for controlling access is connected with the personnel inspection equipment, for receiving the target name, and target name is matched with each name for allowing access into tester, the match is successful, then export exact p-value personnel's signal, it fails to match, then export non-test personnel's signal.By means of the invention it is possible to the working environment of effective maintenance test system.
Description
Technical field
The present invention relates to carpet applications, more particularly to a kind of method based on recognition of face security monitoring personnel's access.
Background technology
Carpet be advantageous in that it is following some:
1st, shatter-resistant is safer.The sclerotin of old man and child are more crisp, on rigid ground material such as ceramic tile, floor, are easier
Slip, especially southern wet season, once fall down, it is easy to fracture.The carpet of soft pile is then entirely different.Just
As a vase, if dropped on rigid ground material, crush in a moment, but if on carpet, then can be safe and sound.It is safely
The basis of family, a safety, comfortable domestic environment are built to household, beyond all else is important.
2nd, it is nontoxic more healthy.Carpet is textile, and it is well below state to be practically free of the pernicious gas such as formaldehyde, other VOC
Family's standard.This is more than a large amount of other hard materials safety using glue, it is well known that formaldehyde is colorless and odorless, is waved
Hair was for up to 10 years.Formaldehyde can injure the immune system of human body, cause nausea and vomiting, cell carcinogenesis is caused when serious, we
Household how is endured to live in this environmentCarpet is selected, you are without fear of an attack from the rear again.
3rd, except mite effect is good.Many consumers may be unaware that acarid is ubiquitous.It is main to understand that method is heating and taken out
Inhale.On rigid ground material, acarid not only flies upward, can not disappeared at all by sweeping the floor and mopping floor in surface flotation everywhere.Ai Jia carpets
By the anti-mite special processing of insect prevention, the production environment for not possessing acarid is allowed to, even if there is a small amount of presence, with dust catcher can
Easily clean.
4th, not airborne dust is purified.Domestic many places are very big compared to the environmental quality gap of European countries, we from it is small just
There is such experience.In the sun, dust is clearly seen unfortunately ceaselessly flying.Rigid ground material can not adsorb dust, walk and
Slight vibrations can kick up dust, and breathing problem is triggered in suction thoracic cavity.And carpet enriches pile and can effectively adsorbed
Dust, general family once can all be exhausted for two to three days using dust catcher dust suction.In other words carpet act as inhaling temporarily
Dirt device.
5th, energy-conservation saves more money.Rigid ground material is the good conductor of heat, and cold wind or warm wind directly can flow away from ground.Carpet
It is the non-conductor of heat, just as cotton-wadded quilt, can effectively obstructs the loss of cold and hot energy, power saving is more than 30%.
The content of the invention
Currently, when carpet is tested, only there is the test of default height ranges, weight range and predetermined human body type
Personnel could meet testing requirement, therefore, it is necessary to establish tester's mechanism of permitting the entrance.And lack such machine in the prior art
System.
In order to solve the above problems, the invention provides a kind of method based on recognition of face security monitoring personnel's access,
Personnel inspection equipment is used to be used to carry out personnel's detection based on image recognition to personnel will be entered to obtain and export target
Name, it is connected using apparatus for controlling access with personnel inspection equipment, for receiving the target name, and by target name and respectively
The individual name for allowing access into tester is matched, and the match is successful, then exports exact p-value personnel's signal, it fails to match, then
Non-test personnel's signal is exported, particular, it is important that when to personnel's image recognition, by the headwear subgraph at original image
Separate to obtain the higher images to be recognized of reference value.
According to an aspect of the present invention, there is provided a kind of method based on recognition of face security monitoring personnel's access, the party
Method includes:
1) a kind of intelligent access system with face identification functions is provided, including test carpet backing and installation system, personnel's detection are set
Standby, apparatus for controlling access and real-time display equipment, the apparatus for controlling access, the personnel inspection equipment and the real-time display
Equipment is all disposed within the front of the test carpet backing and installation system, and the personnel inspection equipment is used to be based on to that will enter personnel
The personnel of image recognition are detected to obtain and export target name, and the apparatus for controlling access connects with the personnel inspection equipment
Connect, matched for receiving the target name, and by target name with each name for allowing access into tester, matched
Success, then exact p-value personnel's signal is exported, it fails to match, then exports non-test personnel's signal;And
2) the intelligent access system is run.
More specifically, described with the intelligent access system of face identification functions:The apparatus for controlling access
With success when, it is determined that simultaneously output matching to the name for allowing access into tester.
More specifically, described with the intelligent access system of face identification functions:The real-time display equipment and institute
Apparatus for controlling access connection is stated, for receiving simultaneously exact p-value personnel signal described in real-time display or non-test personnel's letter
Number;Wherein, the real-time display equipment is additionally operable to when receiving the exact p-value personnel signal, receives simultaneously real-time display
What is be fitted on allows access into the name of tester.
More specifically, described with the intelligent access system of face identification functions, in addition to:Speech play chip,
It is connected with the apparatus for controlling access, for when receiving non-test personnel's signal, sending and the non-test personnel
Audio alert information corresponding to signal;Access control equipment, enter in the on-state, being put into personnel on test carpet backing and installation system
Side, it is additionally operable in the closed state, forbidding personnel to enter above test carpet backing and installation system;Wherein, the speech play chip and institute
State the front that access control equipment is all disposed within the test carpet backing and installation system;The access control equipment is connected with the apparatus for controlling access,
For when receiving exact p-value personnel's signal, the open mode to be converted to from the closed mode, it is additionally operable to receiving
During to non-test personnel's signal, the closed mode is converted to from the open mode.
More specifically, described with the intelligent access system of face identification functions, in addition to:
Histogram treatment equipment, it is connected with image averaging equipment, optimizes image for receiving, the optimization image is performed
Grey level histogram processing, to obtain the grey level histogram of the optimization image;Go to disturb processing equipment, with the histogram treatment
Equipment connects, for obtaining multiple peak values on the ordinate direction in the grey level histogram, amplitude in multiple peak values is small
In the peak value equal to default noise amplitudes threshold value all as interference peak value, gone from multiple peak values unless each Interference Peaks value to obtain
One or more refers to peak value;
Threshold value selects equipment, goes to disturb processing equipment to be connected with described, for being asked with reference to peak value one or more
Average value processing is to obtain equalization peak value, and by the grey level histogram, value on ordinate direction is close to the equalization
The tonal gradation on abscissa corresponding to peak value is as binary-state threshold;Binary conversion treatment equipment, respectively with the histogram
Processing equipment connects with threshold value selection equipment, for optimizing the pixel value of each pixel in image, pixel value is big
In equal to the binary-state threshold, using its pixel value as 255, pixel value is less than the binary-state threshold, by its picture
Element value is used as 0, to obtain binary image;Multiple ccd sensors, for simultaneously to the front personnel of the test carpet backing and installation system
Image data acquiring is carried out to obtain multiple alternative images.
More specifically, described with the intelligent access system of face identification functions, in addition to:
Image removal equipment, it is connected respectively with the multiple ccd sensor, for the deterioration to the multiple alternative image
Degree is detected, and will reject each alternative image after degradation is less than each alternative image of default degradation threshold value
Exported as each image to be averaged;Image averaging equipment, it is connected with described image removal equipment, for receiving described each treat
The average image, and each pending image is averaged to obtain and export optimization image;
Noise analysis equipment, it is connected with binary conversion treatment equipment, for receiving binary image, to the binary image
Noise type analysis is carried out to determine the maximum noise type of the noise amplitude in the binary image to be used as main noise
Type exports;Stencil-chosen equipment, it is connected with the noise analysis equipment, for receiving the main noise type, and is based on
The main noise type determines medium filtering template;Contour detecting equipment, it is connected with the noise analysis equipment, for judging
Objective contour in the binary image;
Differentiation filter apparatus, it is connected respectively with the stencil-chosen equipment and the contour detecting equipment, for group
Into each contour pixel of the objective contour, based on the medium filtering template according to the medium filtering window centered on it
Intraoral pixel distribution determines different filtering strategies, it is described based on the medium filtering template according to the intermediate value centered on it
Pixel distribution in filter window determines that different filtering strategies include:When the object pixel quantity in medium filtering window is more than
During equal to non-targeted pixel quantity in medium filtering window, the average of pixel value of each object pixel is taken as the profile
The pixel value of pixel, when the object pixel quantity in medium filtering window is less than the non-targeted pixel quantity in medium filtering window
When, take pixel value of the average as the contour pixel of the pixel value of each non-targeted pixel;The differentiation filter apparatus
Each non-contour pixel to being not belonging to the objective contour in the binary image is additionally operable to, based on the medium filtering
Template is used as the non-contour pixel according to using the average of the pixel value of all pixels in the medium filtering window centered on it
Pixel value;The differentiation filter apparatus exports differentiation filtering image;
Gradual enhancing equipment, is connected with differentiation filter apparatus, for receiving differentiation filtering image, based on default mesh
Mark gray threshold scope determines whether each pixel in the differentiation filtering image belongs to object pixel, by the difference
Change all object pixels composition preliminary aim region in filtering image, improve preliminary region in the differentiation filtering image
The gray value grade of all pixels improves image to obtain contrast, strengthens the highlights region in the contrast raising image,
The dark portion region in the contrast raising image is reduced simultaneously, to obtain targets improvement image, to the targets improvement image
Picture smooth treatment is carried out to obtain gradual enhancing image;
Facial recognition device, it is connected with gradual enhancing equipment, for receiving gradual enhancing image, from described gradual
Identified in enhancing image and be partitioned into facial subgraph;Headwear identification equipment, for receiving the facial subgraph, from the face
Identified in portion's subgraph and be partitioned into headwear subgraph;Image separation equipment, respectively with the facial recognition device and the head
Identification equipment connection is adornd, for receiving the facial subgraph and the headwear subgraph, and will from the facial subgraph
Headwear subgraph separation is will separate the facial subgraph after the headwear subgraph as going the output of headwear subgraph;People
Member's detection device, be connected with described image separation equipment, for removing headwear subgraph described in receiving, and is based on the headwear subgraph
Personnel's name as target name to export as corresponding to determining.
More specifically, described with the intelligent access system of face identification functions:The multiple ccd sensor is arranged side by side
Set.
More specifically, described with the intelligent access system of face identification functions:Making an uproar in the binary image
Sound type includes internal noise caused by sensitive components, the interference noise of transmission channel, electrical machinery and moves the shake brought
Grain noise caused by noise and photosensitive material.
Brief description of the drawings
Embodiment of the present invention is described below with reference to accompanying drawing, wherein:
Fig. 1 is the structure square frame of the intelligent access system with face identification functions according to embodiment of the present invention
Figure.
Reference:1 test carpet backing and installation system;2 personnel inspection equipments;3 apparatus for controlling access;4 real-time display equipment
Embodiment
The embodiment of the intelligent access system with face identification functions of the present invention is carried out below with reference to accompanying drawings
Describe in detail.
Carpet and have the following disadvantages:
1. because the carpet surface of carpet is intensive pile construction, there is very strong dirt ability of receiving.This aspect relatively drops
Dustiness in low air, but on the other hand also cause to adhere to too many dust on carpet.Some carpets are due to cleaning simultaneously
Not in time, easy breed bacteria, acarid, long-term use can trigger children's bronchitis and breathing problem etc. for maintenance.
2. the cleaning maintenance difficulty of carpet is big.First, the easy dust suction of carpet is dirty, it is necessary to frequent dedusting, secondly as ground
Blanket concrete dynamic modulus, easily shelters evil people and countenance evil practices, and cleans out relatively difficult.
But existing carpet test system lacks mechanism of permitting the entrance, causes carpet test accuracy not high.It is above-mentioned in order to overcome
Deficiency, the present invention have built a kind of method based on recognition of face security monitoring personnel's access, and this method includes:1) provide a kind of
Intelligent access system with face identification functions;And 2) run the intelligent access system.Fig. 1 is to be implemented according to the present invention
The block diagram of the intelligent access system with face identification functions shown in scheme, described device include test carpet knot
Structure, personnel inspection equipment, apparatus for controlling access and real-time display equipment;
Wherein, the apparatus for controlling access, the personnel inspection equipment and the real-time display equipment are all disposed within described
Test the front of carpet backing and installation system;
Wherein, the personnel inspection equipment is used to carry out personnel's detection based on image recognition to that will enter personnel to obtain
Obtaining and export target name, the apparatus for controlling access is connected with the personnel inspection equipment, for receiving the target name,
And matched target name with each name for allowing access into tester, the match is successful, then exports exact p-value personnel
Signal, it fails to match, then exports non-test personnel's signal.
Then, continue to carry out further the concrete structure of the intelligent access system with face identification functions of the present invention
Explanation.
Described with the intelligent access system of face identification functions:
The apparatus for controlling access when the match is successful, it is determined that simultaneously output matching to the surname for allowing access into tester
Name.
Described with the intelligent access system of face identification functions:
The real-time display equipment is connected with the apparatus for controlling access, for receiving simultaneously exact p-value described in real-time display
Personnel's signal or non-test personnel's signal;
Wherein, the real-time display equipment is additionally operable to when receiving the exact p-value personnel signal, is received and real-time
Show the name for allowing access into tester matched.
Described with the intelligent access system of face identification functions, in addition to:
Speech play chip, it is connected with the apparatus for controlling access, for when receiving non-test personnel's signal,
Send audio alert information corresponding with non-test personnel's signal;
Access control equipment, enter in the on-state, being put into personnel above test carpet backing and installation system, be additionally operable to closing shape
Under state, personnel are forbidden to enter above test carpet backing and installation system;
Wherein, the speech play chip and the access control equipment are all disposed within the front of the test carpet backing and installation system;
Wherein, the access control equipment is connected with the apparatus for controlling access, for receiving exact p-value personnel's signal
When, the open mode is converted to from the closed mode, is additionally operable to when receiving non-test personnel's signal, from the opening
State Transferring is the closed mode.
Described with the intelligent access system of face identification functions, in addition to:
Histogram treatment equipment, it is connected with image averaging equipment, optimizes image for receiving, the optimization image is performed
Grey level histogram processing, to obtain the grey level histogram of the optimization image;
Go to disturb processing equipment, be connected with the histogram treatment equipment, it is vertical in the grey level histogram for obtaining
Multiple peak values on coordinate direction, amplitude in multiple peak values is less than or equal to the peak value of default noise amplitudes threshold value all as interference
Peak value, gone from multiple peak values unless each Interference Peaks value one or more refer to peak value to obtain;
Threshold value selects equipment, goes to disturb processing equipment to be connected with described, for being asked with reference to peak value one or more
Average value processing is to obtain equalization peak value, and by the grey level histogram, value on ordinate direction is close to the equalization
The tonal gradation on abscissa corresponding to peak value is as binary-state threshold;
Binary conversion treatment equipment, it is connected respectively with the histogram treatment equipment and threshold value selection equipment, for excellent
Change the pixel value of each pixel in image, pixel value be more than or equal to the binary-state threshold, using its pixel value as
255, pixel value is less than the binary-state threshold, using its pixel value as 0, to obtain binary image;
Multiple ccd sensors, for simultaneously to it is described test carpet backing and installation system front personnel carry out image data acquiring with
Obtain multiple alternative images.
More specifically, described with the intelligent access system of face identification functions, in addition to:
Image removal equipment, it is connected respectively with the multiple ccd sensor, for the deterioration to the multiple alternative image
Degree is detected, and will reject each alternative image after degradation is less than each alternative image of default degradation threshold value
Exported as each image to be averaged;
Image averaging equipment, it is connected with described image removal equipment, for receiving each image to be averaged, and to institute
Each pending image is stated to be averaged to obtain and export optimization image;
Noise analysis equipment, it is connected with binary conversion treatment equipment, for receiving binary image, to the binary image
Noise type analysis is carried out to determine the maximum noise type of the noise amplitude in the binary image to be used as main noise
Type exports;
Stencil-chosen equipment, it is connected with the noise analysis equipment, for receiving the main noise type, and is based on institute
State main noise type and determine medium filtering template;
Contour detecting equipment, it is connected with the noise analysis equipment, for judging the target wheel in the binary image
It is wide;
Differentiation filter apparatus, it is connected respectively with the stencil-chosen equipment and the contour detecting equipment, for group
Into each contour pixel of the objective contour, based on the medium filtering template according to the medium filtering window centered on it
Intraoral pixel distribution determines different filtering strategies, it is described based on the medium filtering template according to the intermediate value centered on it
Pixel distribution in filter window determines that different filtering strategies include:When the object pixel quantity in medium filtering window is more than
During equal to non-targeted pixel quantity in medium filtering window, the average of pixel value of each object pixel is taken as the profile
The pixel value of pixel, when the object pixel quantity in medium filtering window is less than the non-targeted pixel quantity in medium filtering window
When, take pixel value of the average as the contour pixel of the pixel value of each non-targeted pixel;The differentiation filter apparatus
Each non-contour pixel to being not belonging to the objective contour in the binary image is additionally operable to, based on the medium filtering
Template is used as the non-contour pixel according to using the average of the pixel value of all pixels in the medium filtering window centered on it
Pixel value;The differentiation filter apparatus exports differentiation filtering image;
Gradual enhancing equipment, is connected with differentiation filter apparatus, for receiving differentiation filtering image, based on default mesh
Mark gray threshold scope determines whether each pixel in the differentiation filtering image belongs to object pixel, by the difference
Change all object pixels composition preliminary aim region in filtering image, improve preliminary region in the differentiation filtering image
The gray value grade of all pixels improves image to obtain contrast, strengthens the highlights region in the contrast raising image,
The dark portion region in the contrast raising image is reduced simultaneously, to obtain targets improvement image, to the targets improvement image
Picture smooth treatment is carried out to obtain gradual enhancing image;
Facial recognition device, it is connected with gradual enhancing equipment, for receiving gradual enhancing image, from described gradual
Identified in enhancing image and be partitioned into facial subgraph;
Headwear identification equipment, for receiving the facial subgraph, identify and split from the facial subgraph and lift one's head
Adorn subgraph;
Image separation equipment, it is connected respectively with the facial recognition device and the headwear identification equipment, for receiving
Facial subgraph and the headwear subgraph are stated, and separates the headwear subgraph that will separate from the facial subgraph
Facial subgraph after the headwear subgraph, which is used as, goes headwear subgraph to export;
Personnel inspection equipment, be connected with described image separation equipment, for receive it is described remove headwear subgraph, and be based on institute
Personnel's name corresponding to the determination of headwear subgraph is stated to be used as target name to export.
More specifically, described with the intelligent access system of face identification functions:
The multiple ccd sensor is set up in parallel.
More specifically, described with the intelligent access system of face identification functions:
Noise type in the binary image includes the interference of internal noise, transmission channel caused by sensitive components
Grain noise caused by the jittering noise and photosensitive material that noise, electrical machinery motion are brought.
In addition, personnel inspection equipment can be realized using dsp chip.The inside of dsp chip is separated using program and data
Harvard structure, there is special hardware multiplier, widely used pile line operation, there is provided special DSP instructions, can be used for
Quickly realize various digital signal processing algorithms.
According to the requirement of Digital Signal Processing, dsp chip typically has some following main features:(1) in a finger
A multiplication and a sub-addition can be completed by making in the cycle.(2) program and data space is separated, can simultaneously access instruction sum
According to.(3) there is quick RAM in piece, can be generally accessed simultaneously in two pieces by independent data/address bus.(4) there is low overhead
Or without overhead loop and the hardware supported redirected.(5) quickly interrupt processing and Hardware I/O are supported.(6) have within the monocycle
Multiple hardware address generators of operation.(7) multiple operations can be performed parallel.(8) pile line operation is supported, makes fetching, translate
The operation such as code and execution can be with Overlapped Execution.
The data format that is worked according to dsp chip is classified.The dsp chip that data are worked with fixed point format is referred to as pinpointing
Dsp chip, TMS320C1X/C2X, TMS320C2XX/C5X, TMS320C54X/C62XX such as TI companies are serial, AD companies
ADSP21XX series, the DSP16/16A of AT & T Corp., the MC56000 of Motolora companies etc..What is worked in a floating-point format is referred to as
Float-point DSP, such as the TMS320C3X/C4X/C8X of TI companies, the ADSP21XXX of AD companies is serial, AT & T Corp.
MC96002 of DSP32/32C, Motolora company etc..
Not exclusively, some dsp chips use customized floating-point to floating-point format used by different float-point DSPs
Form, such as TMS320C3X, and some dsp chips then use IEEE standard floating-point format, such as motorola inc
ZR35325 of MB86232 and ZORAN companies of MC96002, FUJITSU company etc..
Using the intelligent access system with face identification functions of the present invention, it is difficult to for the test of carpet in the prior art
Differentiate the technical problem of target detection personnel, by introducing at various high accuracy, targetedly image capture device and images
Discriminating of the equipment completion to tester is managed, wherein headwear mechanism is additionally used to ensure the accuracy of identification result, so as to
The legal tester is only allowed to participate in test, the strong guarantee accuracy of test result.
It is understood that although the present invention is disclosed as above with preferred embodiment, but above-described embodiment and it is not used to
Limit the present invention.For any those skilled in the art, without departing from the scope of the technical proposal of the invention,
Many possible changes and modifications are all made to technical solution of the present invention using the technology contents of the disclosure above, or are revised as
With the equivalent embodiment of change.Therefore, every content without departing from technical solution of the present invention, the technical spirit pair according to the present invention
Any simple modifications, equivalents, and modifications made for any of the above embodiments, still fall within the scope of technical solution of the present invention protection
It is interior.
Claims (8)
1. a kind of method based on recognition of face security monitoring personnel's access, this method include:
1) a kind of intelligent access system with face identification functions is provided, including tests carpet backing and installation system, personnel inspection equipment, enter
Enter control device and real-time display equipment, the apparatus for controlling access, the personnel inspection equipment and the real-time display equipment
The front of the test carpet backing and installation system is all disposed within, the personnel inspection equipment is used to carry out being based on image to that will enter personnel
The personnel of identification are detected to obtain and export target name, and the apparatus for controlling access is connected with the personnel inspection equipment, are used
In the reception target name, and target name is matched with each name for allowing access into tester, the match is successful,
Exact p-value personnel's signal is then exported, it fails to match, then exports non-test personnel's signal;And
2) the intelligent access system is run.
2. the method as described in claim 1, it is characterised in that:
The apparatus for controlling access when the match is successful, it is determined that simultaneously output matching to the name for allowing access into tester.
3. method as claimed in claim 2, it is characterised in that:
The real-time display equipment is connected with the apparatus for controlling access, for receiving simultaneously exact p-value personnel described in real-time display
Signal or non-test personnel's signal;
Wherein, the real-time display equipment is additionally operable to when receiving the exact p-value personnel signal, receives simultaneously real-time display
What is matched allows access into the name of tester.
4. method as claimed in claim 3, it is characterised in that the intelligent access system also includes:
Speech play chip, it is connected with the apparatus for controlling access, for when receiving non-test personnel's signal, sending
Audio alert information corresponding with non-test personnel's signal;
Access control equipment, enter in the on-state, being put into personnel above test carpet backing and installation system, be additionally operable in off position
Under, forbid personnel to enter above test carpet backing and installation system;
Wherein, the speech play chip and the access control equipment are all disposed within the front of the test carpet backing and installation system;
Wherein, the access control equipment is connected with the apparatus for controlling access, for when receiving exact p-value personnel's signal, from
The closed mode is converted to the open mode, is additionally operable to when receiving non-test personnel's signal, from the open mode
Be converted to the closed mode.
5. method as claimed in claim 4, it is characterised in that the intelligent access system also includes:
Histogram treatment equipment, it is connected with image averaging equipment, optimizes image for receiving, gray scale is performed to the optimization image
Histogram treatment, to obtain the grey level histogram of the optimization image;
Go to disturb processing equipment, be connected with the histogram treatment equipment, for obtaining the ordinate in the grey level histogram
Multiple peak values on direction, amplitude in multiple peak values is less than or equal to the peak value of default noise amplitudes threshold value all as Interference Peaks
Value, gone from multiple peak values unless each Interference Peaks value one or more refer to peak value to obtain;
Threshold value selects equipment, goes to disturb processing equipment to be connected with described, for being averaging to one or more with reference to peak value
Value processing is to obtain equalization peak value, and by the grey level histogram, value on ordinate direction is close to the equalization peak value
Tonal gradation on corresponding abscissa is as binary-state threshold;
Binary conversion treatment equipment, it is connected with the histogram treatment equipment and threshold value selection equipment, schemes for optimizing respectively
The pixel value of each pixel as in, pixel value is more than or equal to the binary-state threshold, using its pixel value as 255,
Pixel value is less than the binary-state threshold, using its pixel value as 0, to obtain binary image;
Multiple ccd sensors, for carrying out image data acquiring to the front personnel of the test carpet backing and installation system to obtain simultaneously
Multiple alternative images.
6. method as claimed in claim 5, it is characterised in that the intelligent access system also includes:
Image removal equipment, it is connected respectively with the multiple ccd sensor, for the degradation to the multiple alternative image
Detected, will reject degradation be less than each alternative image after each alternative image of default degradation threshold value as
Each image output to be averaged;
Image averaging equipment, it is connected with described image removal equipment, for receiving each image to be averaged, and to described each
Individual pending image is averaged to obtain and export optimization image;
Noise analysis equipment, it is connected with binary conversion treatment equipment, for receiving binary image, the binary image is carried out
Noise type is analyzed to determine the maximum noise type of the noise amplitude in the binary image to be used as main noise type
Output;
Stencil-chosen equipment, it is connected with the noise analysis equipment, for receiving the main noise type, and is based on the master
Noise type is wanted to determine medium filtering template;
Contour detecting equipment, it is connected with the noise analysis equipment, for judging the objective contour in the binary image;
Differentiation filter apparatus, it is connected respectively with the stencil-chosen equipment and the contour detecting equipment, for forming institute
Each contour pixel of objective contour is stated, based on the medium filtering template according in the medium filtering window centered on it
Pixel distribution determine different filtering strategies, it is described based on the medium filtering template according to the medium filtering centered on it
Pixel distribution in window determines that different filtering strategies include:When the object pixel quantity in medium filtering window is more than or equal to
During non-targeted pixel quantity in medium filtering window, the average of pixel value of each object pixel is taken as the contour pixel
Pixel value, when the object pixel quantity in medium filtering window be less than medium filtering window in non-targeted pixel quantity when,
Take pixel value of the average of the pixel value of each non-targeted pixel as the contour pixel;The differentiation filter apparatus is also used
Each non-contour pixel of the objective contour is not belonging in the binary image, based on the medium filtering template
According to the picture using the average of the pixel value of all pixels in the medium filtering window centered on it as the non-contour pixel
Element value;The differentiation filter apparatus exports differentiation filtering image;
Gradual enhancing equipment, is connected with differentiation filter apparatus, for receiving differentiation filtering image, based on goal-selling ash
Degree threshold range determines whether each pixel in the differentiation filtering image belongs to object pixel, and the differentiation is filtered
All object pixels composition preliminary aim region in ripple image, that improves preliminary region in the differentiation filtering image owns
The gray value grade of pixel improves image to obtain contrast, strengthens the highlights region in the contrast raising image, simultaneously
The dark portion region in the contrast raising image is reduced, to obtain targets improvement image, the targets improvement image is carried out
Picture smooth treatment is to obtain gradual enhancing image;
Facial recognition device, it is connected with gradual enhancing equipment, for receiving gradual enhancing image, from the gradual enhancing
Identified in image and be partitioned into facial subgraph;
Headwear identification equipment, for receiving the facial subgraph, identified from the facial subgraph and be partitioned into headwear
Image;
Image separation equipment, it is connected respectively with the facial recognition device and the headwear identification equipment, for receiving the face
Portion's subgraph and the headwear subgraph, and separate the headwear subgraph with by described in separation from the facial subgraph
Facial subgraph after headwear subgraph, which is used as, goes headwear subgraph to export;
Personnel inspection equipment, be connected with described image separation equipment, for receive it is described remove headwear subgraph, and be based on the head
Personnel's name corresponding to subgraph determination is adornd to be used as target name to export.
7. method as claimed in claim 6, it is characterised in that:
The multiple ccd sensor is set up in parallel.
8. the method as described in claim 6-7 is any, it is characterised in that:
Noise type in the binary image is made an uproar including the interference of internal noise, transmission channel caused by sensitive components
Grain noise caused by the jittering noise and photosensitive material that sound, electrical machinery motion are brought.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710673279.3A CN107464322A (en) | 2017-08-09 | 2017-08-09 | A kind of method based on recognition of face security monitoring personnel's access |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710673279.3A CN107464322A (en) | 2017-08-09 | 2017-08-09 | A kind of method based on recognition of face security monitoring personnel's access |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107464322A true CN107464322A (en) | 2017-12-12 |
Family
ID=60548669
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710673279.3A Withdrawn CN107464322A (en) | 2017-08-09 | 2017-08-09 | A kind of method based on recognition of face security monitoring personnel's access |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107464322A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108682424A (en) * | 2018-07-13 | 2018-10-19 | 广州势必可赢网络科技有限公司 | A kind of audio collecting device and method |
CN108960121A (en) * | 2018-06-28 | 2018-12-07 | 王忠亮 | Channel control platform based on face recognition |
CN109297116A (en) * | 2018-12-04 | 2019-02-01 | 宁波锦觅网络科技有限公司 | Automate dust-falling mechanism |
CN110097662A (en) * | 2018-01-29 | 2019-08-06 | 特斯联(北京)科技有限公司 | A kind of electronic identity authentication method and Verification System |
CN112818644A (en) * | 2020-01-09 | 2021-05-18 | 赵国良 | Real-time conversion system for writing revision format |
-
2017
- 2017-08-09 CN CN201710673279.3A patent/CN107464322A/en not_active Withdrawn
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110097662A (en) * | 2018-01-29 | 2019-08-06 | 特斯联(北京)科技有限公司 | A kind of electronic identity authentication method and Verification System |
CN108960121A (en) * | 2018-06-28 | 2018-12-07 | 王忠亮 | Channel control platform based on face recognition |
CN108960121B (en) * | 2018-06-28 | 2019-05-21 | 南京成君电子科技有限公司 | Channel control platform based on face recognition |
CN108682424A (en) * | 2018-07-13 | 2018-10-19 | 广州势必可赢网络科技有限公司 | A kind of audio collecting device and method |
CN109297116A (en) * | 2018-12-04 | 2019-02-01 | 宁波锦觅网络科技有限公司 | Automate dust-falling mechanism |
CN112818644A (en) * | 2020-01-09 | 2021-05-18 | 赵国良 | Real-time conversion system for writing revision format |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107452113A (en) | Intelligent access system with face identification functions | |
CN107464322A (en) | A kind of method based on recognition of face security monitoring personnel's access | |
CN101738394B (en) | Method and system for detecting indoor smog | |
CN105178758B (en) | For automatic door operator in ATM protective cabin for the specific crowd and control method | |
CN106358021A (en) | Smart community monitoring system | |
CN102302370A (en) | Method and device for detecting tumbling | |
CN111104910B (en) | Garbage delivery behavior supervision method and related products | |
CN107272083B (en) | Intelligent safety inspection method in a kind of home life | |
CN105151938B (en) | A kind of elevator video monitoring device and method | |
CN108806142A (en) | A kind of unmanned security system, method and sweeping robot | |
CN108492517A (en) | A kind of fire monitoring system for building | |
CN109345524A (en) | A kind of bearing open defect detection system of view-based access control model | |
CN108487818A (en) | A kind of security window control platform | |
CN106485217A (en) | A kind of method and system of identification visit scenic spot stream of people's saturation degree | |
CN107374209A (en) | The shared sleeping apparatus that slide fastener automatically controls | |
CN107897012A (en) | One kind experiment mouse cage | |
CN108338748A (en) | A kind of automatic cleaning method, device and clean robot based on video monitoring | |
CN110151058A (en) | Smart home sweeping robot | |
CN111192332A (en) | Smoke detection-based smoke machine control method and smoke machine | |
CN107397407A (en) | The device for protecting pregnant woman to fall | |
CN105913524A (en) | ATM protective cabin door electric system having human body detection with voice-controlled sensitivity and achievement method thereof | |
CN102542304A (en) | Region segmentation skin-color algorithm for identifying WAP (Wireless Application Protocol) mobile porn image | |
CN109717792B (en) | Motor noise elimination platform | |
Cheong et al. | Reducing airborne indoor fungi and fine particulates in carpeted Australian homes using intensive, high efficiency HEPA vacuuming | |
CN105913523A (en) | A cabin door electronic system for protecting a specific person from being trapped in an ATM machine protective cabin and an achieving method thereof |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20171212 |