CN110176039A - A kind of video camera adjusting process and system for recognition of face - Google Patents
A kind of video camera adjusting process and system for recognition of face Download PDFInfo
- Publication number
- CN110176039A CN110176039A CN201910327395.9A CN201910327395A CN110176039A CN 110176039 A CN110176039 A CN 110176039A CN 201910327395 A CN201910327395 A CN 201910327395A CN 110176039 A CN110176039 A CN 110176039A
- Authority
- CN
- China
- Prior art keywords
- face
- video camera
- brightness
- qualified
- clarity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 230000001815 facial effect Effects 0.000 claims abstract description 107
- 238000004458 analytical method Methods 0.000 claims abstract description 68
- 238000001514 detection method Methods 0.000 claims abstract description 68
- 238000012797 qualification Methods 0.000 claims description 10
- 238000011426 transformation method Methods 0.000 claims description 3
- 238000009434 installation Methods 0.000 abstract description 37
- 230000000694 effects Effects 0.000 abstract description 26
- 238000013441 quality evaluation Methods 0.000 abstract description 15
- 230000002452 interceptive effect Effects 0.000 abstract description 3
- 210000001747 pupil Anatomy 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 241000288673 Chiroptera Species 0.000 description 1
- 206010057855 Hypotelorism of orbit Diseases 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
Abstract
The present invention discloses a kind of video camera adjusting process and system for recognition of face, wherein, video camera adjusting process is based on Face datection model algorithm and carries out facial image detection to original image, and characteristic point calibration is carried out to the facial image monitored, according to the facial image and its characteristic point further to local luminance, clarity, interpupillary distance, one of parameters such as attitude angle a variety of are analyzed, and corresponding video camera adjustment is fed back based on the analysis results and is suggested, operator suggests carrying out adjustment according to adjustment, realize dynamically interactive type adjustment, preferable installation effect is reached.Video camera adjustment system uses above-mentioned video camera adjusting process, the analysis result for the face picture that can be acquired according to video camera, the quality evaluation of digitization is carried out to the effect of video camera installation, adjustment, and gives the installation feedback that adjustment personnel can interact to reach preferable installation effect.
Description
Technical field
The present invention relates to information technology field more particularly to a kind of video camera adjusting process for recognition of face and it is
System.
Background technique
Since video camera needs to carry out intelligent recognition, installation site, angle and the week of video camera to the video content of shooting
Collarette border is again varied, it is understood that there may be environment light is too dark or too bright, installation space is narrow, mounting height is too high or too low, peace
Fill hypotelorism or too far situations such as, to video camera content of shooting identification accuracy produce bigger effect.On the market to video
The installation and adjustment of content Intelligent identification camera often rely on the personal experience of construction personnel or the personal warp of research staff
Test, adjust video camera deflection angle or upper and lower pitch angle, so that video camera is substantially aligned with target area.
Currently, the face recognition technology based on video camera identification technology it is more and more common be applied to safety monitoring etc. neck
Domain, to illumination, block, the indexs such as clarity, integrality, posture it is more demanding, rely only on engineering staff or research staff
Experience not can guarantee installation effect and installation quality, need to install video camera, the quality of the effect of adjustment progress digitization is commented
Estimate, and gives the installation feedback that adjustment personnel can interact.
Summary of the invention
It, can be according to taking the photograph the purpose of the present invention is to provide a kind of video camera adjusting process and system for recognition of face
The analysis of the face picture of camera acquisition is as a result, carry out the quality evaluation of digitization to the effect of video camera installation, adjustment, and give
The feedback that can interact of installation adjustment personnel is given to reach preferable installation effect.
To achieve the goals above, the invention provides the following technical scheme:
A kind of video camera adjusting process for recognition of face, comprising steps of
S1, a frame original image is obtained, the original image is converted to the gray level image in rectangular coordinate system xOy, and
Grey level histogram is calculated according to the gray level image;
S2, the brightness score value L that the original image is calculated using the grey level histogram, when the brightness score value L reaches
When qualified luminance threshold, step S3 is executed, when the brightness score value L is not up to qualified luminance threshold, terminates detection, output is aobvious
Show brightness score value L and operator is prompted to carry out brightness adjusting;
S3, facial image detection is carried out to the original image based on Face datection model algorithm, if the original graph
Include facial image as in, then characteristic point calibration is carried out to the facial image, step S4 is executed, if in the original image
Not comprising facial image, then terminate to detect;
S4, Parameter analysis is carried out to the facial image, and feeds back corresponding video camera adjustment based on the analysis results and suggests;
Wherein, the parameter includes one of local luminance, clarity, interpupillary distance, attitude angle or a variety of, and corresponding packet is suggested in the adjustment
Include brightness adjustment suggestion, Focussing suggestion, mounting distance adjustment suggestion, setting angle adjustment suggestion.
Preferably, step S2 is specifically included:
S201, the brightness score value L that the original image is calculated using the grey level histogram:
In formula (2), Q(n)For the pixel number of each gray value n, W is the width of the gray level image, and H is the ash
Spend the height of image;
S202, when the brightness score value L reaches qualified luminance threshold, determine that the original image brightness is qualified and executes
In next step;When the brightness score value L is less than qualified luminance threshold, terminates adjustment and export display brightness score value L, wherein institute
Qualified luminance threshold is stated to be arranged between 0.5-0.7.
Specifically, in step S3, the specific method for carrying out characteristic point calibration to the facial image is to be calculated based on MTCNN
Method identifies eyes, nose and two this five key feature points of the corners of the mouth and by the characteristic point from the facial image
It calibrates and.
Preferably, the step S4 includes:
S401, local luminance point is carried out for the facial image as new original image using the method for step S1, S2
Analysis, obtains face gray level image and brightness score value M in rectangular coordinate system xOy, closes when the brightness score value M is greater than or equal to
When lattice luminance threshold, determine that the facial image brightness is qualified and performs the next step;When the brightness score value M is less than qualified brightness
When threshold value, determines the unqualified output display brightness score value M of the facial image brightness and prompt operator's progress video camera bright
Spend adjustment;
S402, clarity analysis is carried out to the face gray level image using Laplacian algorithm, calculates Laplace and calculates
The mean value Clarity of son determines the face gray scale if the mean value Clarity is greater than or equal to qualified clarity threshold
Image definition is qualified and performs the next step;If the mean value Clarity is less than qualified clarity threshold, the face is determined
Gray level image clarity is unqualified, and output display mean value Clarity simultaneously prompts operator to carry out focal length of camera adjustment;
S403, the coordinate for obtaining eyes in the face gray level image, and pass through the pupil of Euclidean distance formula calculating eyes
Away from d, when the interpupillary distance d is greater than or equal to qualified interpupillary distance threshold value, then determine that the face gray level image size is qualified and executes
In next step;When the interpupillary distance d is less than qualified interpupillary distance threshold value, determines that the face gray level image is too small, export the interpupillary distance d simultaneously
Operator is prompted to carry out the adjustment of video camera mounting distance;
S404, the appearance that the facial image is calculated using five characteristic points in MTCNN algorithm and the facial image
State angle determines that the video camera setting angle closes when the absolute value of the attitude angle is both less than or is equal to qualified angle threshold
Lattice simultaneously terminate adjustment;When the absolute value of the attitude angle is greater than qualified angle threshold, the video camera setting angle is determined not
Qualification exports the attitude angle and operator is prompted to carry out the adjustment of video camera setting angle.
Preferably, the step S402 is specifically included:
S4021, using Laplacian algorithm to the face gray level image in rectangular coordinate system x-axis, y-axis both direction
Second differnce operation is carried out respectively, obtains the difference form of Laplace operator:
S4022, the mean value Clarity, the mean value Clarity for calculating Laplace operator are used to characterize the face ash
Spend the clarity of image:
In above-mentioned formula 4-3, formula 4-4, f (x, y) indicates gray value of the facial image at coordinate (x, y), and W is described
The width of gray level image, H are the height of the gray level image;
If S4023, the mean value Clarity reach qualified clarity threshold, determine that the face gray level image is clear
Degree is qualified and performs the next step;If the mean value Clarity is less than qualified clarity threshold, the face gray level image is determined
Clarity is unqualified, and output display mean value Clarity simultaneously prompts operator to carry out focal length of camera adjustment, wherein the conjunction
Lattice clarity threshold is arranged between 35-45.
Preferably, step S404 is specifically included:
S4041, three-dimensional system of coordinate O-xyz is established, utilizes five spies in MTCNN algorithm and the face gray level image
Point is levied, the attitude angle for calculating the facial image using Y-axis as the coordinate transformation method of the Eulerian angles system of main shaft, the appearance are used
State angle includes pitch angle Pitch, deflection angle Yaw and rotation angle Roll;
S4042, when the absolute value of the attitude angle is both less than or when being equal to qualified angle threshold, determine the video camera peace
Dress angle is qualified and terminates adjustment;When the absolute value of the attitude angle is greater than qualified angle threshold, the video camera peace is determined
It is unqualified to fill angle, export the attitude angle and operator is prompted to carry out the adjustment of video camera setting angle.
Specifically, the pitch angle Pitch qualified angle threshold be arranged between 18 ° -22 °, the deflection angle Yaw and
The qualified angle threshold of the rotation angle Roll is arranged between 8 ° -12 °.
Specifically, the qualified interpupillary distance threshold value is arranged between 50-70 pixel.
A kind of video camera adjustment system for recognition of face, using the above-mentioned video camera adjustment side for recognition of face
Method, the video camera adjustment system include video camera link block, picture brightness analysis module, face recognition module and face point
Analyse module, wherein
The video camera link block is used to obtain a frame original image by the video camera;
The picture brightness analysis module includes brightness score value computing unit and brightness judging unit, the brightness score value meter
It calculates unit to be used to the original image being converted to the gray level image in rectangular coordinate system xOy, be calculated according to the gray level image
Grey level histogram, and calculate using the grey level histogram brightness score value L of the original image;The brightness judging unit
For judging the size relation of the brightness score value L and qualified luminance threshold, when the brightness score value L reaches qualified luminance threshold
When, the brightness judging unit issues the first signal control face recognition module and works on;When the brightness score value L not
When reaching qualified luminance threshold, the brightness judging unit issues the first signal control face recognition module and stops working,
And it exports display brightness score value L and operator is prompted to carry out brightness adjusting;
The face recognition module includes recognition unit and calibration unit, and the recognition unit is calculated based on Face datection model
Method MTCNN carries out facial image detection to the original image, if in the original image including facial image, the calibration
Unit is used to carry out characteristic point calibration to the facial image;If not including facial image, the knowledge in the original image
Other unit issues second control signal and controls the human face analysis module from service;
The human face analysis module is used to carry out Parameter analysis to the facial image, and feedback corresponds to based on the analysis results
Video camera adjust suggest;Wherein, the parameter includes one of local luminance, clarity, interpupillary distance, attitude angle or a variety of,
It includes brightness adjustment suggestion that the adjustment, which is suggested corresponding to, Focussing suggestion, mounting distance adjustment is suggested, setting angle adjustment is built
View.
Preferably, the calibration unit be based on MTCNN algorithm, identified from the facial image eyes, nose and
The characteristic point simultaneously is calibrated by two this five key feature points of the corners of the mouth.
Specifically, the human face analysis module includes face brightness detection unit, and the face brightness detection unit is from institute
It states calibration unit and obtains the facial image, and be sent to the brightness score value for the facial image as new original image
Computing unit and the brightness judging unit carry out local luminance analysis, obtain the face grayscale image in rectangular coordinate system xOy
Picture and brightness score value M;When the brightness judging unit determines the facial image brightness qualification, the face brightness detection is single
Member output shows that the facial image brightness is qualified;When the brightness judging unit determines that the facial image brightness is unqualified
When, the face brightness detection unit exports display brightness score value M and operator is prompted to carry out video camera brightness adjusting.
Further, the human face analysis module further includes face clarity detection unit, the face clarity detection
Unit is used to carry out clarity analysis to the face gray level image using Laplacian algorithm, calculates the equal of Laplace operator
Value Clarity;If the mean value Clarity is greater than or equal to qualified clarity threshold, the face clarity detection unit is defeated
Show that the face gray level image clarity is qualified out, if the mean value Clarity is less than qualified clarity threshold, the face
Clarity detection unit output display mean value Clarity simultaneously prompts operator to carry out focal length of camera adjustment.
Preferably, the human face analysis module further includes face interpupillary distance detection unit, the face interpupillary distance detection unit is used
In the coordinate for obtaining eyes in the face gray level image, and by the interpupillary distance d of Euclidean distance formula calculating eyes, when the pupil
When reaching qualified interpupillary distance threshold value away from d, the face interpupillary distance detection unit output shows that the face gray level image size is qualified;When
When the interpupillary distance d is less than qualified interpupillary distance threshold value, the face interpupillary distance detection unit output display interpupillary distance d and prompt operator into
Row video camera mounting distance adjustment.
Preferably, the human face analysis module further includes facial angle detection unit, and the facial angle detection unit is used
In the attitude angle for calculating the facial image using five characteristic points in MTCNN algorithm and the facial image, when described
When the absolute value of attitude angle is both less than or is equal to qualified angle threshold, the facial angle detection unit determines the video camera peace
Filling the output display of angle qualification terminates adjustment;When the absolute value of the attitude angle is greater than qualified angle threshold, the face angle
It spends detection unit and determines that the video camera setting angle is unqualified, export the attitude angle and operator is prompted to carry out video camera
Setting angle adjustment.
Compared with prior art, a kind of video camera adjusting process and system for recognition of face provided by the invention has
Below the utility model has the advantages that
A kind of video camera adjusting process for recognition of face provided by the invention calculates original graph using grey level histogram
The brightness score value L of picture, and the brightness score value L is compared with preset qualified luminance threshold, it is tied if brightness is unqualified
Beam detection, while display brightness score value L, prompt operator's progress adjustment are exported, brightness is also Real-time Feedback in calibration procedures
To operator, to reach preferable installation effect.Original image is carried out in addition, this method is based on Face datection model algorithm
Facial image detection, and characteristic point calibration is carried out to the facial image monitored, one is clicked through according to the facial image and its feature
Step to one of parameters such as local luminance, clarity, interpupillary distance, attitude angle or it is a variety of analyze, and it is anti-based on the analysis results
It presents corresponding video camera adjustment to suggest, operator suggests carrying out adjustment according to adjustment, the corresponding ginseng of Real-time Feedback in calibration procedures
Several analyses are as a result, realization dynamically interactive type adjustment, has reached preferable installation effect.
The present invention also provides a kind of video camera adjustment system for recognition of face, which is directed to face using above-mentioned
The video camera adjusting process of identification, can be according to the analysis for the face picture that video camera acquires as a result, to video camera installation, adjustment
Effect carry out the quality evaluation of digitization, and give the feedback that can interact of installation adjustment personnel to reach preferably installation effect
Fruit.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes a part of the invention, this hair
Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In the accompanying drawings:
Fig. 1 is a kind of video camera adjusting process flow chart for recognition of face in the embodiment of the present invention;
Fig. 2 is facial image and its characteristic point schematic diagram in the embodiment of the present invention;
Fig. 3 is the angle of pitch schematic diagram installed in the embodiment of the present invention;
Fig. 4 is the schematic diagram of attitude angle under three-dimensional system of coordinate O-xyz in the embodiment of the present invention.
Appended drawing reference:
11- left eye characteristic point, 12- right eye characteristic point;
2- nose characteristic point, the left corners of the mouth characteristic point of 31-;
The right corners of the mouth characteristic point of 32-.
Specific embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, implement below in conjunction with the present invention
Attached drawing in example, technical scheme in the embodiment of the invention is clearly and completely described.Obviously, described embodiment
Only a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, the common skill in this field
Art personnel all other embodiment obtained without creative labor belongs to the model that the present invention protects
It encloses.
Embodiment one
Referring to Fig. 1, a kind of video camera adjusting process for recognition of face provided in this embodiment, comprising steps of
S1, a frame original image is obtained, original image is converted to the gray level image in rectangular coordinate system xOy, and according to
Gray level image calculates grey level histogram.
S2, the brightness score value L that original image is calculated using grey level histogram, when brightness score value L is bright more than or equal to qualified
When spending threshold value, step S3 is executed, when brightness score value L is less than qualified luminance threshold, terminates detection, exports display brightness score value L
And operator is prompted to carry out brightness adjusting.
S3, facial image detection is carried out to original image based on Face datection model algorithm, if including in original image
Facial image then carries out characteristic point calibration to facial image, executes step S4, if not including facial image in original image,
Then terminate to detect.
S4, Parameter analysis is carried out to facial image respectively, and feeds back corresponding video camera adjustment based on the analysis results and suggests;
Wherein, parameter includes one of local luminance, clarity, interpupillary distance, attitude angle or a variety of, and adjustment suggests that corresponding includes brightness tune
Whole suggestion, Focussing suggestion, mounting distance adjustment is suggested, setting angle adjustment is suggested.
A kind of video camera adjusting process for recognition of face provided in an embodiment of the present invention, is calculated using grey level histogram
The brightness score value L of original image, and the brightness score value L is compared with preset qualified luminance threshold, if brightness does not conform to
Lattice then terminate to detect, while exporting display brightness score value L, prompt operator's progress adjustment, and brightness is also real in calibration procedures
When feed back to operator, to reach preferable installation effect.In addition, this method is based on Face datection model algorithm to original graph
Characteristic point calibration is carried out as carrying out facial image detection, and to the facial image monitored, according to the facial image and its feature
Point further to one of parameters such as local luminance, clarity, interpupillary distance, attitude angle or it is a variety of analyze, and according to analysis
As a result it feeds back corresponding video camera adjustment to suggest, operator suggests carrying out adjustment, Real-time Feedback in calibration procedures according to adjustment
The analysis of corresponding parameter is as a result, realization dynamically interactive type adjustment, has reached preferable installation effect.
It should be noted that the related algorithm of recognition of face is replaced using the camera adjusting process of the embodiment of the present invention
Other recognizers such as bar code identification, two dimensional code identification, picture recognition, commodity identification are changed to, express delivery cabinet, self-service are applied to
The camera adjustment in the places such as supermarket, Self-help vending machine reaches the bats such as brightness is suitable, clear and legible, angle is appropriate, size is suitable
Effect is taken the photograph, also within the scope of the present invention.
In a kind of video camera adjusting process for recognition of face provided in an embodiment of the present invention, step S1 is according to grayscale image
As calculating grey level histogram method particularly includes:
The pixel number Q of each gray value n is calculated first(n), then calculate each gray value and occur in gray level image
Frequency, and draw frequency-gray value grey level histogram.
In formula (1), W is the width of gray level image, and H is the height of gray level image, and P (w, h) is the pixel at coordinate (w, h)
Gray value.
The method that step S2 carries out original image Luminance Analysis specifically includes:
S201, the brightness score value L that original image is calculated using grey level histogram:
In formula (2), Q(n)For the pixel number of each gray value n, W is the width of gray level image, and H is gray level image
Highly;Gray value interval is used in formula (2) accounts for the ratio of whole image as brightness score value L for the pixel number of (60,230),
It will be appreciated by persons skilled in the art that adjusting gray scale according to specific camera operation environment is corresponding between (0,255)
It is worth the upper limit and/or lower limit in section, both falls within the protection scope of the application.
S202, when brightness score value L reaches qualified luminance threshold, determine that original image brightness is qualified and simultaneously perform the next step;
When brightness score value L is less than qualified luminance threshold, terminates adjustment and export display brightness score value L, wherein qualified luminance threshold is set
It sets between 0.5-0.7, it is preferable that qualified luminance threshold is set as 0.6.
Specifically, when brightness score value L is less than qualified luminance threshold, when operator carries out brightness debugging, calculating is worked as in real time
Preceding brightness score value L and output feeds back to operator, and operator does the brightness of video camera according to real-time brightness score value L
Corresponding adjustment out realizes the brightness to original image and carries out digitization until brightness score value L is greater than qualified luminance threshold
Quality evaluation, and the installation feedback that adjustment personnel can interact is given to reach preferable installation effect.
Referring to Fig. 2, the specific method for carrying out characteristic point calibration to facial image is to be based on MTCNN algorithm in step S3
Or other mature face recognition algorithms, identify that left eye characteristic point 11, right eye characteristic point 12, nose are special from facial image
Characteristic point simultaneously is calibrated by sign point 2, left corners of the mouth characteristic point 31 and this five key feature points of right corners of the mouth characteristic point 32.
Please refer to Fig. 2, Fig. 3 or Fig. 4 a kind of video camera adjusting process for recognition of face provided in an embodiment of the present invention
In, step S4 is specifically included:
S401, local luminance analysis is carried out for facial image as new original image using the method for step S1, S2, obtained
To the face gray level image and brightness score value M in rectangular coordinate system xOy, when brightness score value M is greater than or equal to qualified luminance threshold
When value, determine that facial image brightness is qualified and performs the next step;When brightness score value M is less than qualified luminance threshold, face is determined
The unqualified output display brightness score value M of brightness of image simultaneously prompts operator to carry out video camera brightness adjusting;Wherein, qualified brightness
Threshold value is arranged between 0.5-0.7, it is preferable that qualified luminance threshold is set as 0.6.It is identical as step S202, as brightness score value M
Less than qualified luminance threshold, when operator carries out brightness debugging, current brightness score value M is calculated in real time and output feeds back to behaviour
Make personnel, operator makes corresponding adjustment to the brightness of video camera according to real-time brightness score value M, until brightness score value M
Greater than qualified luminance threshold, realizes the brightness to facial image and carry out the quality evaluation of digitization, and give installation adjustment people
The feedback that member can interact is to reach preferable video camera installation effect.
S402, face gray level image is carried out clearly using Laplacian algorithm or other Edge-Detection Algorithms
Degree analysis, calculates the mean value Clarity of Laplace operator, if mean value Clarity is greater than or equal to qualified clarity threshold,
Determine that face gray level image clarity is qualified and performs the next step;If mean value Clarity is less than qualified clarity threshold, determine
Face gray level image clarity is unqualified, and output display mean value Clarity simultaneously prompts operator to carry out focal length of camera adjustment;
Wherein, qualified clarity threshold is arranged between 35-45, it is preferable that qualified clarity threshold is set as 40.Work as mean value
Clarity is less than qualified clarity threshold and calculates current mean value Clarity in real time when operator carries out clarity debugging
And export and feed back to operator, operator makes corresponding tune to the focal length of video camera according to real-time mean value Clarity
It is whole, until mean value Clarity is greater than qualified clarity threshold, realize the quality that digitization is carried out to the clarity of facial image
Assessment, that is, realize the quality evaluation for being arranged to focal length of camera and carrying out digitization, and give operator's dynamic that can be interacted
Adjustment feedback, to reach preferable video camera installation effect.
S403, the coordinate for obtaining eyes in face gray level image, the i.e. coordinate of left eye characteristic point 11 and right eye characteristic point 12,
And the interpupillary distance d of eyes, i.e. the distance between left eye characteristic point 11 and right eye characteristic point 12 are calculated by Euclidean distance formula, work as pupil
When being greater than or equal to qualified interpupillary distance threshold value away from d, then determine that face gray level image size is qualified and performs the next step;When interpupillary distance d is small
When qualified interpupillary distance threshold value, determine that face gray level image is too small, exports interpupillary distance d and operator is prompted to carry out video camera locating distance
From adjustment, wherein qualified interpupillary distance threshold value is arranged between 50-70 pixel, it is preferable that qualified interpupillary distance threshold value is set as 60
Pixel.Current interpupillary distance d is calculated in real time when operator carries out mounting distance debugging when interpupillary distance d is less than qualified interpupillary distance threshold value
And export and feed back to operator, operator makes corresponding adjustment to the mounting distance of video camera according to real-time interpupillary distance d,
Until interpupillary distance d is greater than qualified interpupillary distance threshold value, the quality evaluation that digitization is carried out to the size of facial image is realized, that is, realize
The quality evaluation of digitization is carried out to video camera mounting distance, and gives operator's dynamic adjustment feedback that can be interacted, with
Reach preferable video camera installation effect.
S404, the attitude angle that facial image is calculated using five characteristic points in MTCNN algorithm and facial image, work as appearance
When the absolute value at state angle is both less than or is equal to qualified angle threshold, determining that video camera setting angle is qualified simultaneously terminates adjustment;Work as appearance
When the absolute value at state angle is greater than qualified angle threshold, determine that video camera setting angle is unqualified, exports attitude angle and prompt to operate
Personnel carry out the adjustment of video camera setting angle.When the absolute value of attitude angle is greater than qualified angle threshold, operator is installed
When angle is debugged, current attitude angle is calculated in real time and output feeds back to operator, operator is according to real-time attitude angle
Corresponding adjustment is made to the setting angle of video camera, until the absolute value of attitude angle is less than or equal to qualified angle threshold, reality
The quality evaluation that digitization is carried out to the attitude angle of facial image is showed, that is, realize and data are carried out to video camera setting angle
The quality evaluation of change, and operator's dynamic adjustment feedback that can be interacted is given, to reach preferable video camera installation effect.
Wherein, the method that step S402 carries out the analysis adjusting of video camera clarity based on recognition of face specifically includes:
S4021, using Laplacian algorithm to the face gray level image in rectangular coordinate system xOy, in x, y both direction
Second differnce operation is carried out respectively:
Obtain the difference form of Laplace operator:
S4022, the mean value Clarity for calculating Laplace operator, mean value Clarity are used to characterize face gray level image
Clarity:
In above-mentioned formula (4-1)~formula (4-4), f (x, y) indicates that gray value of the facial image at coordinate (x, y), W are ash
The width of image is spent, H is the height of gray level image.
If S4023, mean value Clarity reach qualified clarity threshold, face gray level image clarity qualification is determined simultaneously
It performs the next step;If mean value Clarity is less than qualified clarity threshold, determine that face gray level image clarity is unqualified, it is defeated
Mean value Clarity is shown out and operator is prompted to carry out focal length of camera adjustment, wherein qualified clarity threshold setting exists
Between 35-45, it is preferable that qualified clarity threshold is set as 40.
Specifically, when mean value Clarity is less than qualified clarity threshold, and operator carries out clarity debugging, in real time
It calculates current mean value Clarity and exports and feed back to operator, operator is according to real-time mean value Clarity to camera shooting
The focal length of machine makes corresponding adjustment, until mean value Clarity is greater than qualified clarity threshold, realizes to the clear of facial image
Clear degree carries out the quality evaluation of digitization, that is, realizes the quality evaluation for being arranged to focal length of camera and carrying out digitization, and gives
The dynamic adjustment feedback that operator can interact, to reach preferable video camera installation effect.
It is specifically wrapped referring to Fig. 2, step S403 carries out the method that the analysis of video camera mounting distance is adjusted based on recognition of face
It includes:
S4031, the coordinate for obtaining eyes in face gray level image, i.e. the seat of left eye characteristic point 11 and right eye characteristic point 12
Mark, and pass through the interpupillary distance d of Euclidean distance formula calculating eyes:
(x1, y1) in formula (4-5), (x2, y2) are respectively the coordinate of left eye characteristic point 11 and right eye characteristic point 12.
S4032, when interpupillary distance d is greater than or equal to qualified interpupillary distance threshold value, then determine that face gray level image size is qualified and holds
Row is in next step;When interpupillary distance d is less than qualified interpupillary distance threshold value, determine that face gray level image is too small, exports interpupillary distance d and prompt operator
Member's carry out video camera mounting distance adjustment, wherein qualified interpupillary distance threshold value is arranged between 50-70 pixel, it is preferable that qualified
Interpupillary distance threshold value is set as 60 pixels.
Specifically, when interpupillary distance d is less than qualified interpupillary distance threshold value, when operator carries out mounting distance debugging, calculating is worked as in real time
Preceding interpupillary distance d is simultaneously exported and is fed back to operator, and operator makes pair the mounting distance of video camera according to real-time interpupillary distance d
The adjustment answered, until interpupillary distance d is greater than qualified interpupillary distance threshold value, the quality for realizing the size progress digitization to facial image is commented
Estimate, that is, realize the quality evaluation for carrying out digitization to video camera mounting distance, and give operator's dynamic tune that can be interacted
School feedback, to reach preferable video camera installation effect.
It is specifically included referring to Fig. 4, step S404 carries out the method that the analysis of video camera attitude angle is adjusted based on recognition of face:
S4041, three-dimensional system of coordinate O-xyz is established, utilizes five features in MTCNN algorithm and face gray level image
Point, uses the attitude angle that facial image is calculated using Y-axis as the coordinate transformation method of the Eulerian angles system of main shaft, and attitude angle includes bowing
Elevation angle Pitch, deflection angle Yaw and rotation angle Roll;
S4042, when the absolute value of attitude angle is both less than or when being equal to qualified angle threshold, determine that video camera setting angle closes
Lattice simultaneously terminate adjustment;When attitude angle is greater than qualified angle threshold, determine that video camera setting angle is unqualified, output attitude angle is simultaneously
Operator is prompted to carry out the adjustment of video camera setting angle.Pitch angle Pitch qualified angle threshold setting 18 ° -22 ° it
Between, deflection angle Yaw and rotation angle Roll qualified angle threshold be arranged between 8 ° -12 °, it is preferable that pitch angle Pitch's
Qualified angle threshold is set as 20 °, and the qualified angle threshold of the rotation angle Roll of deflection angle Yaw sum is set as 10 °, also, advises
Determine the direction deflection angle Yaw: Zuo Zheng, right negative, the direction pitch angle Pitch: it is upper it is negative, lower just, the direction rotation angle Roll: a left side is negative, it is right just.
Referring to Fig. 3, illustrating the calibration procedures of video camera attitude angle by taking the debugging of pitch angle Pitch as an example, in practical tune
During examination, needs the personnel of assistant adjustment to stand before video camera, cooperated with nature, pass through the detection face figure of MTCNN
Left eye characteristic point 11, right eye characteristic point 12, nose characteristic point 2, left corners of the mouth characteristic point 31 and the right corners of the mouth characteristic point 32 of picture this
Five key feature points, and it is general using Y-axis as the coordinate transform side of corner (Eulerian angles) system of main shaft using domestic contrast
Method calculates the pitch angle Pitch of facial image, the pitch angle Pitch of the facial image detected and the installation pitch angle of video camera
Pitch is at corresponding relationship.When the pitch angle Pitch of facial image is greater than qualified pitch angle threshold value, operator carries out video camera
When pitch angle debugging is installed, current pitch angle Pitch is calculated in real time and output feeds back to operator, operator is according to reality
When pitch angle Pitch corresponding adjustment is made to the installation pitch angle of video camera, until pitch angle Pitch is less than or equal to
Qualified pitch angle threshold value, for example, when qualified pitch angle threshold value is arranged is 20 °, when face pitch angle Pitch > 20 °, video camera to
Upper lift, if Pitch < -20 °, video camera is pressed downward, and until detecting that face pitch angle reaches requirement, is realized to people
The pitch angle of face image carries out the quality evaluation of digitization, that is, realizes the matter that digitization is carried out to video camera installation pitch angle
Amount assessment, and operator's dynamic adjustment feedback that can be interacted is given, to reach preferable video camera pitching installation effect, similarly
Adjust other two angles.
It is worth noting that, in attached drawing of the embodiment of the present invention 2, attached drawing 4 coordinate system to establish mode only be that one of which can
Can, the constraint for specifically establishing coordinate system is not done, and those skilled in the art build according to the habit of oneself and the size of image etc.
It founds suitable coordinate system and both falls within the scope of the present invention.In addition, the present embodiment is based on face for the processing of facial image
What gray level image carried out, for part treatment process, it will be appreciated by persons skilled in the art that carrying out people using RGB image
It includes one of local luminance, clarity, interpupillary distance, attitude angle or many kinds of parameters analysis knot that face image processing, which obtains in this method,
Fruit, and feed back corresponding video camera adjustment based on the analysis results and suggest that operator suggests carrying out adjustment according to adjustment, to reach
Preferable installation effect also falls into the scope of the present invention.
Embodiment two
Fig. 2, Fig. 3 or Fig. 4 are please referred to, the embodiment of the present invention provides a kind of video camera adjustment system for recognition of face,
Using the video camera adjusting process for recognition of face provided in above-described embodiment, video camera adjustment system includes that video camera connects
Connection module, picture brightness analysis module, face recognition module and human face analysis module, the input terminal connection of video camera link block
Video camera, output end are separately connected picture brightness analysis module and face recognition module, and the input terminal of face recognition module also connects
It is connected to the output end of picture brightness analysis module, the output end of face recognition module connects human face analysis module.
Video camera link block is used to acquire real-time video picture by video camera, and frame original is obtained from video pictures
Beginning image is sent to picture brightness analysis module and face recognition module.
Picture brightness analysis module includes sequentially connected brightness score value computing unit and brightness judging unit, brightness score value
Computing unit is used to original image being converted to the gray level image in rectangular coordinate system xOy, and it is straight to calculate gray scale according to gray level image
Fang Tu, and utilize the brightness score value L of grey level histogram calculating original image;Brightness judging unit is for judging brightness score value L
With the size relation of qualified luminance threshold, when brightness score value L reaches qualified luminance threshold, brightness judging unit issues the first letter
Number control face recognition module work on;When brightness score value L is not up to qualified luminance threshold, brightness judging unit issues the
One signal control face recognition module stops working, and exports display brightness score value L and operator is prompted to carry out brightness tune
School.
Face recognition module includes sequentially connected recognition unit and calibration unit, and recognition unit is based on Face datection model
Algorithm MTCNN to original image carry out facial image detection, if in original image include facial image, calibration unit for pair
Facial image carries out characteristic point calibration;If not including facial image in original image, recognition unit issues second control signal
Control human face analysis module from service.It demarcates unit and is based on MTCNN algorithm, left eye characteristic point is identified from facial image
11, this five key feature points of right eye characteristic point 12, nose characteristic point 2, left corners of the mouth characteristic point 31 and right corners of the mouth characteristic point 32
And characteristic point is calibrated to come.
Human face analysis module for facial image is carried out respectively local luminance analysis, clarity analysis, interpupillary distance calculate with
And attitude angle operation, respectively correspond the output brightness adjustment suggestion of video camera, Focussing suggestion, mounting distance adjustment suggest with
And setting angle adjustment is suggested.
Preferably, human face analysis module includes face brightness detection unit, face brightness detection unit and calibration unit, bright
Degree score value computing unit and brightness judging unit communicate to connect respectively;Face brightness detection unit obtains face figure from calibration unit
Picture, and using facial image as new original image be sent to brightness score value computing unit and brightness judging unit to carry out part bright
Degree analysis, obtains face gray level image and brightness score value M in rectangular coordinate system xOy;When brightness judging unit determines face
When brightness of image qualification, the output display facial image brightness of face brightness detection unit is qualified;When brightness judging unit determines people
When face image brightness is unqualified, face brightness detection unit output display brightness score value M simultaneously prompts operator to carry out video camera
Brightness adjusting.
Preferably, human face analysis module further includes face clarity detection unit, the input of face clarity detection unit
End connection face brightness detection unit simultaneously obtains face gray level image;Face clarity detection unit is used to utilize Laplacian
Algorithm carries out clarity analysis to face gray level image, calculates the mean value Clarity of Laplace operator;If mean value Clarity is big
In or equal to qualified clarity threshold, face clarity detection unit output display face gray level image clarity is qualified, if
Value Clarity is less than qualified clarity threshold, and face clarity detection unit output display mean value Clarity simultaneously prompts operator
Member carries out focal length of camera adjustment.
Further, human face analysis module further includes face interpupillary distance detection unit, the input terminal of face interpupillary distance detection unit
Connection face brightness detection unit simultaneously obtains face gray level image;Face interpupillary distance detection unit is for obtaining in face gray level image
The coordinate of eyes, and pass through the interpupillary distance d of Euclidean distance formula calculating eyes, when interpupillary distance d reaches qualified interpupillary distance threshold value, face pupil
It is qualified that display face gray level image size is exported away from detection unit;When interpupillary distance d is less than qualified interpupillary distance threshold value, the detection of face interpupillary distance
Unit output display interpupillary distance d simultaneously prompts operator to carry out the adjustment of video camera mounting distance.
Preferably, human face analysis module further includes facial angle detection unit, and the input terminal of facial angle detection unit connects
It connects calibration unit and obtains facial image;Facial angle detection unit is used to utilize five in MTCNN algorithm and facial image
A characteristic point calculates the attitude angle of facial image, when the absolute value of attitude angle is both less than or is equal to qualified angle threshold, face
Angle detection unit determines that the output display of video camera setting angle qualification terminates adjustment;When attitude angle is greater than qualified angle threshold
When, facial angle detection unit determines that video camera setting angle is unqualified, exports attitude angle and operator is prompted to image
Machine setting angle adjustment.
A kind of video camera adjustment system for recognition of face provided in an embodiment of the present invention, can acquire according to video camera
Face picture analysis as a result, to video camera installation, adjustment effect carry out digitization quality evaluation, and give installation adjust
The feedback that school personnel can interact is to reach preferable installation effect.
In the description of above embodiment, specific features, structure or feature can be real in any one or more
Applying can be combined in any suitable manner in example or example.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (14)
1. a kind of video camera adjusting process for recognition of face, which is characterized in that comprising steps of
S1, a frame original image is obtained, the original image is converted to the gray level image in rectangular coordinate system xOy, and according to
The gray level image calculates grey level histogram;
S2, the brightness score value L that the original image is calculated using the grey level histogram, when the brightness score value L reaches qualified
When luminance threshold, step S3 is executed, when the brightness score value L is not up to qualified luminance threshold, terminates detection, output display is bright
Degree score value L simultaneously prompts operator to carry out brightness adjusting;
S3, facial image detection is carried out to the original image based on Face datection model algorithm, if in the original image
Comprising facial image, then characteristic point calibration is carried out to the facial image, step S4 is executed, if do not wrapped in the original image
Containing facial image, then terminate to detect;
S4, Parameter analysis is carried out to the facial image, and feeds back corresponding video camera adjustment based on the analysis results and suggests;Its
In, the parameter includes one of local luminance, clarity, interpupillary distance, attitude angle or a variety of, and the adjustment suggests that correspondence includes
Brightness adjustment suggestion, Focussing suggestion, mounting distance adjustment is suggested, setting angle adjustment is suggested.
2. the video camera adjusting process according to claim 1 for recognition of face, which is characterized in that step S2 is specifically wrapped
It includes:
S201, the brightness score value L that the original image is calculated using the grey level histogram:
In formula (2), Q(n)For the pixel number of each gray value n, W is the width of the gray level image, and H is the grayscale image
The height of picture;
S202, when the brightness score value L reaches qualified luminance threshold, determine that the original image brightness is qualified and executes next
Step;When the brightness score value L is less than qualified luminance threshold, terminates adjustment and export display brightness score value L, wherein the conjunction
Lattice luminance threshold is arranged between 0.5-0.7.
3. being directed to the video camera adjusting process of recognition of face according to claim 1, which is characterized in that in step S3, to institute
Stating facial image and carrying out the specific method of characteristic point calibration is to be based on MTCNN algorithm, is identified from the facial image double
The characteristic point simultaneously is calibrated by eye, nose and two this five key feature points of the corners of the mouth.
4. the video camera adjusting process according to claim 3 for recognition of face, which is characterized in that the step S4 packet
It includes:
S401, local luminance analysis is carried out for the facial image as new original image using the method for step S1, S2, obtained
To the face gray level image and brightness score value M in rectangular coordinate system xOy, when the brightness score value M is bright more than or equal to qualified
When spending threshold value, determine that the facial image brightness is qualified and performs the next step;When the brightness score value M is less than qualified luminance threshold
When, determine the unqualified output display brightness score value M of the facial image brightness and operator is prompted to carry out video camera brightness tune
School;
S402, clarity analysis is carried out to the face gray level image using Laplacian algorithm, calculates Laplace operator
Mean value Clarity determines the face gray level image if the mean value Clarity is greater than or equal to qualified clarity threshold
Clarity is qualified and performs the next step;If the mean value Clarity is less than qualified clarity threshold, the face gray scale is determined
Image definition is unqualified, and output display mean value Clarity simultaneously prompts operator to carry out focal length of camera adjustment;
S403, the coordinate for obtaining eyes in the face gray level image, and pass through the interpupillary distance d that Euclidean distance formula calculates eyes,
When the interpupillary distance d is greater than or equal to qualified interpupillary distance threshold value, then determine that the face gray level image size is qualified and executes next
Step;When the interpupillary distance d is less than qualified interpupillary distance threshold value, determines that the face gray level image is too small, export the interpupillary distance d and prompt
Operator carries out the adjustment of video camera mounting distance;
S404, the attitude angle that the facial image is calculated using five characteristic points in MTCNN algorithm and the facial image,
When the absolute value of the attitude angle is both less than or is equal to qualified angle threshold, determine that the video camera setting angle is qualified and ties
Beam adjustment;When the absolute value of the attitude angle is greater than qualified angle threshold, determine that the video camera setting angle is unqualified, it is defeated
Out the attitude angle and prompt operator carry out the adjustment of video camera setting angle.
5. the video camera adjusting process according to claim 4 for recognition of face, which is characterized in that the step S402
It specifically includes:
S4021, the face gray level image is distinguished in rectangular coordinate system x-axis, y-axis both direction using Laplacian algorithm
Second differnce operation is carried out, the difference form of Laplace operator is obtained:
S4022, the mean value Clarity, the mean value Clarity of Laplace operator are calculated for characterizing the face grayscale image
The clarity of picture:
In above-mentioned formula 4-3, formula 4-4, f (x, y) indicates that gray value of the facial image at coordinate (x, y), W are the gray scale
The width of image, H are the height of the gray level image;
If S4023, the mean value Clarity reach qualified clarity threshold, determine that the face gray level image clarity is closed
Lattice simultaneously perform the next step;If the mean value Clarity is less than qualified clarity threshold, determine that the face gray level image is clear
Spend unqualified, output display mean value Clarity simultaneously prompts operator to carry out focal length of camera adjustment, wherein described qualified clear
Clear degree threshold value is arranged between 35-45.
6. the video camera adjusting process according to claim 4 for recognition of face, which is characterized in that the step S404
It specifically includes:
S4041, three-dimensional system of coordinate O-xyz is established, utilizes five features in MTCNN algorithm and the face gray level image
Point uses the attitude angle that the facial image is calculated using Y-axis as the coordinate transformation method of the Eulerian angles system of main shaft, the posture
Angle includes pitch angle Pitch, deflection angle Yaw and rotation angle Roll;
S4042, when the absolute value of the attitude angle is both less than or when being equal to qualified angle threshold, determine the video camera established angle
Degree is qualified and terminates adjustment;When the absolute value of the attitude angle is greater than qualified angle threshold, the video camera established angle is determined
Spend it is unqualified, export the attitude angle and prompt operator carry out the adjustment of video camera setting angle.
7. the video camera adjusting process according to claim 6 for recognition of face, which is characterized in that the pitch angle
The qualified angle threshold of Pitch is arranged between 18 ° -22 °, the qualified angle of the deflection angle Yaw and the rotation angle Roll
Threshold value is spent to be arranged between 8 ° -12 °.
8. the video camera adjusting process according to claim 4 for recognition of face, which is characterized in that the qualification interpupillary distance
Threshold value is arranged between 50-70 pixel.
9. a kind of video camera adjustment system for recognition of face, which is characterized in that using any power in the claims 1-8
Benefit requires the video camera adjusting process for recognition of face, and the video camera adjustment system includes video camera connection mould
Block, picture brightness analysis module, face recognition module and human face analysis module, wherein
The video camera link block is used to obtain a frame original image by the video camera;
The picture brightness analysis module includes brightness score value computing unit and brightness judging unit, and the brightness score value calculates single
Member calculates gray scale for the original image to be converted to the gray level image in rectangular coordinate system xOy, according to the gray level image
Histogram, and calculate using the grey level histogram brightness score value L of the original image;The brightness judging unit is used for
The size relation for judging the brightness score value L and qualified luminance threshold, when the brightness score value L reaches qualified luminance threshold,
The brightness judging unit issues the first signal control face recognition module and works on;When the brightness score value L does not reach
When to qualified luminance threshold, the brightness judging unit, which issues the first signal and controls the face recognition module, to stop working, and
And it exports display brightness score value L and operator is prompted to carry out brightness adjusting;
The face recognition module includes recognition unit and calibration unit, and the recognition unit is based on Face datection model algorithm
MTCNN carries out facial image detection to the original image, if in the original image including facial image, the calibration is single
Member is for carrying out characteristic point calibration to the facial image;If not including facial image, the identification in the original image
Unit issues second control signal and controls the human face analysis module from service;
The human face analysis module is used to carry out Parameter analysis to the facial image, and feeds back corresponding take the photograph based on the analysis results
Camera adjustment is suggested;Wherein, the parameter includes one of local luminance, clarity, interpupillary distance, attitude angle or a variety of, described
It includes brightness adjustment suggestion that adjustment suggestion is corresponding, Focussing suggestion, mounting distance adjustment is suggested, setting angle adjusts and suggests.
10. the video camera adjustment system according to claim 9 for recognition of face, which is characterized in that the calibration is single
Member is based on MTCNN algorithm, and eyes, nose and two corners of the mouth this five key feature points are identified from the facial image simultaneously
The characteristic point is calibrated to come.
11. the video camera adjustment system according to claim 10 for recognition of face, which is characterized in that the face point
Analysing module includes face brightness detection unit, and the face brightness detection unit obtains the face figure from the calibration unit
Picture, and the brightness score value computing unit and brightness judgement list are sent to using the facial image as new original image
Member carries out local luminance analysis, obtains face gray level image and brightness score value M in rectangular coordinate system xOy;When the brightness
When judging unit determines the facial image brightness qualification, the face brightness detection unit output shows that the facial image is bright
Degree is qualified;When the brightness judging unit determines that the facial image brightness is unqualified, the face brightness detection unit is defeated
Out display brightness score value M and prompt operator carry out video camera brightness adjusting.
12. the video camera adjustment system according to claim 10 for recognition of face, which is characterized in that the face point
Analysis module further includes face clarity detection unit, and the face clarity detection unit is used to utilize Laplacian algorithm pair
The face gray level image carries out clarity analysis, calculates the mean value Clarity of Laplace operator;If the mean value Clarity
More than or equal to qualified clarity threshold, the face clarity detection unit output shows the face gray level image clarity
Qualification, if the mean value Clarity is less than qualified clarity threshold, the face clarity detection unit output display mean value
Clarity simultaneously prompts operator to carry out focal length of camera adjustment.
13. the video camera adjustment system according to claim 10 for recognition of face, which is characterized in that the face point
Analysis module further includes face interpupillary distance detection unit, and the face interpupillary distance detection unit is double in the face gray level image for obtaining
The coordinate of eye, and by the interpupillary distance d of Euclidean distance formula calculating eyes, it is described when the interpupillary distance d reaches qualified interpupillary distance threshold value
The output of face interpupillary distance detection unit shows that the face gray level image size is qualified;When the interpupillary distance d is less than qualified interpupillary distance threshold value
When, the face interpupillary distance detection unit output display interpupillary distance d simultaneously prompts operator to carry out the adjustment of video camera mounting distance.
14. the video camera adjustment system according to claim 10 for recognition of face, which is characterized in that the face point
Analysing module further includes facial angle detection unit, and the facial angle detection unit is used to utilize MTCNN algorithm and the people
Five characteristic points in face image calculate the attitude angle of the facial image, when the absolute value of the attitude angle is both less than or is equal to
When qualified angle threshold, the facial angle detection unit determines that the video camera setting angle qualification output display terminates to adjust
School;When the absolute value of the attitude angle is greater than qualified angle threshold, the facial angle detection unit determines the video camera
Setting angle is unqualified, exports the attitude angle and operator is prompted to carry out the adjustment of video camera setting angle.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910327395.9A CN110176039A (en) | 2019-04-23 | 2019-04-23 | A kind of video camera adjusting process and system for recognition of face |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910327395.9A CN110176039A (en) | 2019-04-23 | 2019-04-23 | A kind of video camera adjusting process and system for recognition of face |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110176039A true CN110176039A (en) | 2019-08-27 |
Family
ID=67689885
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910327395.9A Pending CN110176039A (en) | 2019-04-23 | 2019-04-23 | A kind of video camera adjusting process and system for recognition of face |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110176039A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110830717A (en) * | 2019-11-12 | 2020-02-21 | 维沃移动通信有限公司 | Parameter value acquisition method and electronic equipment |
CN110910491A (en) * | 2019-11-27 | 2020-03-24 | 广西科技大学 | Three-dimensional human body modeling optimization method and system |
CN112929612A (en) * | 2021-03-01 | 2021-06-08 | 西藏宁算科技集团有限公司 | Camera mounting and debugging tool and method for mounting camera by using same |
CN113206994A (en) * | 2021-03-30 | 2021-08-03 | 华为技术有限公司 | Camera automatic engineering survey method and device, electronic equipment, chip and storage medium |
CN113973201A (en) * | 2020-07-24 | 2022-01-25 | 华为技术有限公司 | Method and device for evaluating erection of shooting equipment |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101661557A (en) * | 2009-09-22 | 2010-03-03 | 中国科学院上海应用物理研究所 | Face recognition system and face recognition method based on intelligent card |
CN105654463A (en) * | 2015-11-06 | 2016-06-08 | 乐视移动智能信息技术(北京)有限公司 | Image processing method applied to continuous shooting process and apparatus thereof |
CN105791709A (en) * | 2015-12-29 | 2016-07-20 | 福建星网锐捷通讯股份有限公司 | Automatic exposure processing method and apparatus with back-light compensation |
CN106030614A (en) * | 2014-04-22 | 2016-10-12 | 史內普艾德有限公司 | System and method for controlling a camera based on processing an image captured by other camera |
CN206260049U (en) * | 2016-08-09 | 2017-06-16 | 擎中科技(上海)有限公司 | A kind of bore hole 3D display devices |
CN106911921A (en) * | 2017-05-12 | 2017-06-30 | 安徽大学 | The infrared touch-control of projecting apparatus and Atomatic focusing method based on single camera |
CN107895150A (en) * | 2016-11-30 | 2018-04-10 | 奥瞳系统科技有限公司 | Face datection and head pose angle based on the small-scale convolutional neural networks module of embedded system are assessed |
CN107977648A (en) * | 2017-12-20 | 2018-05-01 | 武汉大学 | A kind of method of discrimination and system of the identity card clarity based on recognition of face |
CN108111818A (en) * | 2017-12-25 | 2018-06-01 | 北京航空航天大学 | Moving target active perception method and apparatus based on multiple-camera collaboration |
CN108965729A (en) * | 2018-08-13 | 2018-12-07 | Oppo广东移动通信有限公司 | Control method, device, electronic equipment and computer readable storage medium |
CN108961242A (en) * | 2018-07-04 | 2018-12-07 | 北京临近空间飞行器系统工程研究所 | A kind of fluorescent staining image CTC intelligent identification Method |
CN108986094A (en) * | 2018-07-20 | 2018-12-11 | 南京开为网络科技有限公司 | For the recognition of face data automatic update method in training image library |
CN109360530A (en) * | 2018-10-30 | 2019-02-19 | 武汉华星光电技术有限公司 | Liquid crystal display device and its method for controlling backlight thereof |
-
2019
- 2019-04-23 CN CN201910327395.9A patent/CN110176039A/en active Pending
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101661557A (en) * | 2009-09-22 | 2010-03-03 | 中国科学院上海应用物理研究所 | Face recognition system and face recognition method based on intelligent card |
CN106030614A (en) * | 2014-04-22 | 2016-10-12 | 史內普艾德有限公司 | System and method for controlling a camera based on processing an image captured by other camera |
CN105654463A (en) * | 2015-11-06 | 2016-06-08 | 乐视移动智能信息技术(北京)有限公司 | Image processing method applied to continuous shooting process and apparatus thereof |
CN105791709A (en) * | 2015-12-29 | 2016-07-20 | 福建星网锐捷通讯股份有限公司 | Automatic exposure processing method and apparatus with back-light compensation |
CN206260049U (en) * | 2016-08-09 | 2017-06-16 | 擎中科技(上海)有限公司 | A kind of bore hole 3D display devices |
CN107895150A (en) * | 2016-11-30 | 2018-04-10 | 奥瞳系统科技有限公司 | Face datection and head pose angle based on the small-scale convolutional neural networks module of embedded system are assessed |
CN106911921A (en) * | 2017-05-12 | 2017-06-30 | 安徽大学 | The infrared touch-control of projecting apparatus and Atomatic focusing method based on single camera |
CN107977648A (en) * | 2017-12-20 | 2018-05-01 | 武汉大学 | A kind of method of discrimination and system of the identity card clarity based on recognition of face |
CN108111818A (en) * | 2017-12-25 | 2018-06-01 | 北京航空航天大学 | Moving target active perception method and apparatus based on multiple-camera collaboration |
CN108961242A (en) * | 2018-07-04 | 2018-12-07 | 北京临近空间飞行器系统工程研究所 | A kind of fluorescent staining image CTC intelligent identification Method |
CN108986094A (en) * | 2018-07-20 | 2018-12-11 | 南京开为网络科技有限公司 | For the recognition of face data automatic update method in training image library |
CN108965729A (en) * | 2018-08-13 | 2018-12-07 | Oppo广东移动通信有限公司 | Control method, device, electronic equipment and computer readable storage medium |
CN109360530A (en) * | 2018-10-30 | 2019-02-19 | 武汉华星光电技术有限公司 | Liquid crystal display device and its method for controlling backlight thereof |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110830717A (en) * | 2019-11-12 | 2020-02-21 | 维沃移动通信有限公司 | Parameter value acquisition method and electronic equipment |
CN110830717B (en) * | 2019-11-12 | 2021-06-25 | 维沃移动通信有限公司 | Parameter value acquisition method and electronic equipment |
CN110910491A (en) * | 2019-11-27 | 2020-03-24 | 广西科技大学 | Three-dimensional human body modeling optimization method and system |
CN113973201A (en) * | 2020-07-24 | 2022-01-25 | 华为技术有限公司 | Method and device for evaluating erection of shooting equipment |
CN112929612A (en) * | 2021-03-01 | 2021-06-08 | 西藏宁算科技集团有限公司 | Camera mounting and debugging tool and method for mounting camera by using same |
CN113206994A (en) * | 2021-03-30 | 2021-08-03 | 华为技术有限公司 | Camera automatic engineering survey method and device, electronic equipment, chip and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110176039A (en) | A kind of video camera adjusting process and system for recognition of face | |
CN107045863B (en) | The GTG method of adjustment and device of a kind of display panel | |
CN104268591B (en) | A kind of facial critical point detection method and device | |
Zhu et al. | Image quality assessment by visual gradient similarity | |
CN104137028B (en) | Control the device and method of the rotation of displayed image | |
CN106782268A (en) | Display system and the driving method for display panel | |
CN110277058A (en) | The luminance compensation method and its device of organic light emitting display panel | |
CN110032271A (en) | Contrast control device and its method, virtual reality device and storage medium | |
CN102609684B (en) | Human body posture detection method and device | |
CN107194316A (en) | A kind of evaluation method of mood satisfaction, apparatus and system | |
CN105791709A (en) | Automatic exposure processing method and apparatus with back-light compensation | |
CN107798318A (en) | The method and its device of a kind of happy micro- expression of robot identification face | |
CN107609516A (en) | Adaptive eye moves method for tracing | |
WO2016150112A1 (en) | Method and apparatus for adjusting brightness of display screen | |
CN110379355A (en) | A kind of control system and control method of large screen display wall | |
CN110348496A (en) | A kind of method and system of facial image fusion | |
CN108921209A (en) | Image identification method, device and electronic equipment | |
US20090167840A1 (en) | Video instant messaging system and method thereof | |
CN111563449A (en) | Real-time classroom attention detection method and system | |
CN110032970A (en) | Biopsy method, device, computer equipment and the storage medium of high-accuracy | |
WO2019127319A1 (en) | Distortion measurement method and system for head-mounted display device | |
CN112756840A (en) | Welding quality detection system | |
CN107944424A (en) | Front end human image collecting and Multi-angle human are distributed as comparison method | |
CN111462105A (en) | Method for measuring sawtooth degree of edge pixel | |
Mikhailiuk et al. | The effect of display brightness and viewing distance: a dataset for visually lossless image compression |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190827 |