CN106874871B - Living body face double-camera identification method and identification device - Google Patents
Living body face double-camera identification method and identification device Download PDFInfo
- Publication number
- CN106874871B CN106874871B CN201710081460.5A CN201710081460A CN106874871B CN 106874871 B CN106874871 B CN 106874871B CN 201710081460 A CN201710081460 A CN 201710081460A CN 106874871 B CN106874871 B CN 106874871B
- Authority
- CN
- China
- Prior art keywords
- image
- face
- white
- black
- camera
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000000605 extraction Methods 0.000 claims abstract description 48
- 210000000056 organ Anatomy 0.000 claims abstract description 48
- 244000144972 livestock Species 0.000 claims abstract description 7
- 230000003068 static effect Effects 0.000 claims description 15
- 230000003287 optical effect Effects 0.000 claims description 7
- 241001465754 Metazoa Species 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 6
- 230000009286 beneficial effect Effects 0.000 claims description 4
- 238000005265 energy consumption Methods 0.000 claims description 4
- 230000003595 spectral effect Effects 0.000 claims description 4
- 201000008217 Aggressive systemic mastocytosis Diseases 0.000 claims description 3
- 238000005286 illumination Methods 0.000 claims description 3
- 239000013589 supplement Substances 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 238000010438 heat treatment Methods 0.000 claims description 2
- 230000004044 response Effects 0.000 claims description 2
- 238000000513 principal component analysis Methods 0.000 claims 2
- 238000005516 engineering process Methods 0.000 description 6
- 230000009977 dual effect Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 210000004709 eyebrow Anatomy 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002329 infrared spectrum Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
Images
Classifications
-
- 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
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- 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
-
- 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
- G06V40/40—Spoof detection, e.g. liveness detection
- G06V40/45—Detection of the body part being alive
Landscapes
- Engineering & Computer Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
- Studio Devices (AREA)
Abstract
The invention discloses a living body face double-camera identification method, which comprises the following steps: acquiring a black-and-white image and an infrared image through a configured black-and-white camera with double cameras; a color camera acquires a color image; extracting two-dimensional state organ feature points from the black and white image and the face part in the near-infrared image by combining a feature extraction algorithm; and forming organ feature points in a three-dimensional state, identifying the features of the living human face through a human face and livestock algorithm, and judging whether the human face image is the living human face. The invention also discloses a living human face double-camera recognition device. The method comprises the steps of collecting a color image and a near-infrared image through two cameras, obtaining a background-removed image, namely a face part, extracting organ feature points through a feature extraction algorithm, identifying the features of a living body face through a face feature algorithm, and judging whether the face image is the living body face. The identification method of the invention has high reliability, convenience, practicability and low realization cost.
Description
Technical Field
The invention relates to the technical field of living body face recognition, in particular to a living body face double-camera recognition method and a living body face double-camera recognition device.
Background
With the continuous updating of security technologies, face recognition technologies are also more and more widely applied in life. Especially in government departments, frontiers and financial industries, the intelligent security monitoring system plays an irreplaceable role in security protection. The face recognition technology is mature day by day, the commercial application is wider, however, the face is very easy to copy in the modes of photos, videos and the like, so the face recognition, especially the living body face recognition authentication system, is an important threat to the counterfeit of the face of a legal user. In recent years, although the living body face detection technology has been advanced, the safety reliability and cost performance of the conventional method cannot be well balanced in practical application.
The existing living body face recognition technology mainly detects whether the face characteristics are met through a common camera, and is still easy to be cheated by fake plastic and other entity head images. Still others are made to see the blood vessel distribution inside the living face by scanning fine biological features of the living face through a professional grade infrared bolometric imaging lens. However, such devices are very expensive, which results in a device that is suitable only for certain specific applications and is not widely used.
Disclosure of Invention
The invention aims to provide a living human face double-camera identification method and a living human face double-camera identification device which are high in reliability, convenient and practical, and aim to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme:
a method for recognizing a human face by two cameras is characterized by comprising the following steps:
the method comprises the following steps that through two configured cameras, the two cameras are a black-and-white camera and a color camera, the black-and-white camera obtains a black-and-white image generated under the condition of natural light or white light and obtains another near-infrared image generated under the condition of near-infrared light; the color camera acquires a color image generated under the condition of natural light or white light;
extracting two-dimensional organ feature points from the black-and-white image obtained by the black-and-white camera in the double cameras and the face part in the near-infrared image by combining a feature extraction algorithm;
and extracting organ feature points from the face part of the color image acquired by the color camera in the double cameras by combining a feature extraction algorithm and extracting organ feature points from the near-infrared image acquired by the black-and-white camera to form three-dimensional organ feature points, identifying the features of the face of the living body by the face and livestock algorithm, and judging whether the face image is the face of the living body.
The identification method further comprises the following steps:
when a living human face is judged, acquiring a complete human face feature picture and a data value from the extracted two-dimensional state organ feature points, and matching with a comparison feature database by combining a feature matching algorithm;
and outputting display or control after matching is finished.
The black-and-white image and the near infrared image are obtained through the configured black-and-white cameras with the double cameras, and the method comprises the following steps:
carrying out difference operation on the black-and-white image and the near-infrared image, wherein the difference operation comprises the following steps:
taking a black-and-white image as a static image of the current environment, taking a near-infrared image as an active light source image, carrying out difference operation on the static image and the active light source image to obtain a difference image, and obtaining a complete background-removed active light source image according to the characteristics of the active light source.
The method for identifying the characteristics of the living human face through the human face and livestock algorithm comprises the following steps:
and (3) performing an optical flow field estimation model on the color image acquired by the color camera in the double cameras and the near-infrared image acquired by the black-and-white camera, estimating differential distribution, judging whether the image is a living human face, if so, extracting organ feature points in a two-dimensional state by combining an information acquisition module and a feature extraction algorithm, and if not, finishing.
A two camera recognition device of live body face, recognition device includes:
the image acquisition module is used for acquiring a black-and-white image generated under the condition of natural light or white light and acquiring a near-infrared image generated under the condition of near-infrared light by the black-and-white camera; the color camera acquires a color image generated under the condition of natural light or white light;
the information acquisition module is used for extracting two-dimensional state organ feature points from the black and white images acquired by the black and white cameras in the double cameras and the face part in the near infrared image by combining a feature extraction algorithm;
and the judging module is used for extracting organ feature points from the face part of the color image acquired by the color cameras in the double cameras by combining a feature extraction algorithm and near-infrared images acquired by the black and white cameras to form organ feature points in a three-dimensional state, identifying the features of the living face by a face and livestock algorithm and judging whether the face image is the living face.
The identification device further comprises:
the matching module is used for obtaining a complete face feature picture and a data value from the extracted two-dimensional state organ feature points when a living body face is judged, and then matching with the comparison feature database by combining a feature matching algorithm;
and the display or control module is used for outputting display or control after the matching is finished.
The image acquisition module is also used for acquiring black-and-white images and near-infrared images through configured black-and-white cameras with double cameras, and comprises:
carrying out difference operation on the black-and-white image and the near-infrared image, wherein the difference operation comprises the following steps:
taking a black-and-white image as a static image of the current environment, taking a near-infrared image as an active light source image, carrying out difference operation on the static image and the active light source image to obtain a difference image, and obtaining a complete background-removed active light source image according to the characteristics of the active light source.
The judging module is also used for carrying out an optical flow field estimation model on the color image acquired by the color camera in the double cameras and the near-infrared image acquired by the black and white camera, estimating difference distribution, judging whether the image is a living human face, if so, extracting organ feature points in a two-dimensional state by combining the information acquiring module and a feature extraction algorithm, and if not, ending the process.
The utility model provides a two camera recognition device of live body face, includes:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to:
the method comprises the following steps that through two configured cameras, the two cameras are a black-and-white camera and a color camera, the black-and-white camera obtains a black-and-white image generated under the condition of natural light or white light and obtains another near-infrared image generated under the condition of near-infrared light; the color camera acquires a color image generated under the condition of natural light or white light;
extracting two-dimensional organ feature points from the black-and-white image obtained by the black-and-white camera in the double cameras and the face part in the near-infrared image by combining a feature extraction algorithm;
and extracting organ feature points from the face part of the color image acquired by the color camera in the double cameras by combining a feature extraction algorithm and extracting organ feature points from the near-infrared image acquired by the black-and-white camera to form three-dimensional organ feature points, identifying the features of the face of the living body by the face and livestock algorithm, and judging whether the face image is the face of the living body.
The two cameras are respectively provided with an infrared LED and a white light LED.
The invention has the beneficial effects that: the method comprises the steps of collecting a color image and a near-infrared image through two cameras, obtaining a background-removed image, namely a face part, extracting organ feature points through a feature extraction algorithm, identifying the features of a living body face through a face feature algorithm, and judging whether the face image is the living body face. The identification method of the invention has high reliability, convenience, practicability and low realization cost.
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
Drawings
Fig. 1 is a flowchart of a living human face dual-camera recognition method of embodiment 1;
fig. 2 is a flowchart of the face recognition algorithm of step 102 of example 1 for recognizing characteristics of a live face;
fig. 3 is a block diagram of a living human face dual-camera recognition apparatus of embodiment 1;
fig. 4 is a schematic structural diagram of the living human face dual cameras of embodiment 1;
fig. 5 is a hardware block diagram of the living human face dual-camera recognition apparatus according to embodiment 1.
FIG. 6 is a flowchart of a method for identifying a living human face by two cameras according to embodiment 2;
fig. 7 is a flow chart of a living human face dual-camera recognition method according to embodiment 2;
fig. 8 is a block diagram of a living body face double-camera recognition apparatus of embodiment 2;
in the figure, 1, a lens, 2, a light homogenizing plate, 3, an infrared emission tube, 4, a lens base, 5, a mainboard and lamp board connector, 6, a USB connector, 7, a power connector, 8, a mainboard and Sensor board connector, 9, a lamp board and mainboard connector, 10, a main control board, 11, a Sensor board, 12 and a lamp board.
Detailed Description
wherein, carry out background difference operation to black and white image and near-infrared image, background difference operation includes: taking a black-and-white image as a static image of the current environment, taking a near-infrared image as an active light source image, carrying out differential operation on the static image and the active light source image to obtain a differential image, and obtaining a complete background-removed active light source image according to the characteristics of the active light source;
the basic process of background difference operation is as follows: the active light source partial image in the image is extracted by adopting pixel difference-based closed-value method between two adjacent frames of the image (the 1 st frame is a black-and-white image which is a static background image of the current environment, and the 2 nd frame is an active light source image which is a near-infrared image). Firstly, subtracting corresponding pixel values of adjacent frame images to obtain a differential image, and if the change of the corresponding pixel values is smaller than a predetermined threshold value, regarding the differential image as a background pixel, wherein if the pixel values of image areas change greatly, the areas are marked as foreground pixels because of an active light source, and the positions of active light source targets in the image can be determined by utilizing the marked pixel areas. Because the time interval between two adjacent frames is very short, the previous frame image is used as the background model of the current frame, so that the real-time performance is better, the background is not accumulated, the updating speed is high, the algorithm is simple, and the calculated amount is small;
102, extracting organ feature points from a face part of a color image acquired by a color camera in the double cameras by combining a feature extraction algorithm and near-infrared images acquired by a black-and-white camera to form organ feature points in a three-dimensional state, identifying the features of a living body face through a face and animal algorithm, and judging whether the face image is the living body face;
referring to fig. 2, the identification of the characteristics of the living human face through the human face algorithm includes: and converting a color image acquired by a color camera in the double cameras into a black-and-white image, performing an optical flow field estimation model with a near-infrared image acquired by the black-and-white camera, estimating differential distribution, judging whether the image is a living human face, if so, extracting organ feature points by combining an information acquisition module and a feature extraction algorithm, and if not, ending.
In step 103, the process of extracting the organ feature points by combining the information acquisition module and the feature extraction algorithm comprises two steps of visual geometric feature extraction and shape and texture feature extraction:
visual geometric feature extraction: extracting representative parts of the face of the human face, such as eyebrow, eye, nose, mouth and human face outline, and extracting an intuitive geometric figure by utilizing the characteristic relation of each part;
shape and texture feature extraction: the shape feature extraction is to extract the coordinate vectors of edges, outlines or some key points of the face image, is a binary feature and has strong capability of resisting illumination change; the extraction of the texture features is the gray value of the pixels of the human face image and is beneficial supplement of the shape features, and the ASMs/AAMs model utilizes the shape and the texture features to carry out statistical modeling through PCA.
Referring to fig. 3, the present embodiment further provides a living human face dual-camera recognition apparatus, where the recognition apparatus includes:
the image acquisition module 301 is used for acquiring a black-and-white image generated under the condition of natural light or white light and acquiring a near-infrared image generated under the condition of near-infrared light by using the two cameras, wherein the two cameras are a black-and-white camera and a color camera; the color camera acquires a color image generated under the condition of natural light or white light;
the image obtaining module 301 is further configured to perform a background difference operation on two face images, where the background difference operation includes: taking a black-and-white image as a static image of the current environment, taking a near-infrared image as an active light source image, carrying out difference operation on the static image and the active light source image to obtain a difference image, and obtaining a complete background-removed active light source image according to the characteristics of the active light source.
The information acquisition module 302 is configured to extract feature points of a two-dimensional organ by combining a black-and-white image acquired by a black-and-white camera in the two cameras with a human face part in a near-infrared image according to a feature extraction algorithm;
the judging module 303 is configured to extract organ feature points from the face part of the color image acquired by the color camera in the dual cameras by combining a feature extraction algorithm with the near-infrared image acquired by the black-and-white camera to form organ feature points in a three-dimensional state, identify features of a living body face through a face and animal algorithm, and judge whether the face image is the living body face;
the judging module 303 is further configured to convert a color image acquired by a color camera in the dual cameras into a black-and-white image, perform an optical flow field estimation model with a near-infrared image acquired by the black-and-white camera, estimate difference distribution, judge whether the image is a living human face, extract organ feature points in a two-dimensional state through the information acquiring module in combination with a feature extraction algorithm if the image is the living human face, and end if the image is not the living human face.
This implementation still provides a two camera recognition device of live body face, includes:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to:
the method comprises the following steps that through two configured cameras, the two cameras are a black-and-white camera and a color camera, the black-and-white camera obtains a black-and-white image generated under the condition of natural light or white light and obtains another near-infrared image generated under the condition of near-infrared light; the color camera acquires a color image generated under the condition of natural light or white light;
extracting two-dimensional organ feature points from the black-and-white image obtained by the black-and-white camera in the double cameras and the face part in the near-infrared image by combining a feature extraction algorithm;
and extracting organ feature points from the face part of the color image acquired by the color camera in the double cameras by combining a feature extraction algorithm and extracting organ feature points from the near-infrared image acquired by the black-and-white camera to form three-dimensional organ feature points, identifying the features of the face of the living body by the face and livestock algorithm, and judging whether the face image is the face of the living body.
The double cameras are respectively provided with an infrared LED and a white LED,
referring to fig. 5, a Sensor1 is a near infrared spectrum image, an infrared LED provides a light source, which may be a 850nm or 940nm light source, and a black and white camera acquires the image;
the Sensor2 is a color image, a white light LED provides a light source, and a color camera acquires an image;
the DSP controls the LED driver to control the on and off or flicker of the LED, provides the best light source state for the Sensor1 and the Sensor2, and can add a light source light homogenizing plate above the LED lamp to uniformly diffuse light.
Dps controls the properties (resolution, frame rate, exposure time, gain, brightness, etc.) of Sensor1 and Sensor2, respectively;
the DSP precisely controls the current of the LED, the LED lighting time and the properties of the Sensor1 and the Sensor2 according to the external environment light source, so that the optimal image can be obtained.
Referring to fig. 4, in this embodiment, the energy comparison between the active infrared light source collected by the living body face dual-camera recognition device and the corresponding wavelength in the sunlight or white light LED lamp is a core factor affecting the performance of the living body face dual-camera recognition device, and the whole living body face dual-camera recognition device is designed based on the improvement of the ratio.
Within a possible range, the energy output and the utilization efficiency of the infrared light source should be increased as much as possible, and meanwhile, the energy interference of sunlight is reduced as much as possible;
selecting an LED infrared light source with concentrated spectral energy to work under the conditions of high output power and short pulse time so as to improve the energy efficiency ratio;
the infrared light of the LED is concentrated and uniformly illuminated in an image acquisition area through a proper light homogenizing design, so that the light energy loss is reduced as much as possible;
the lens adopts a band-pass filter to intercept the spectral energy outside the output range of the LED from entering so as to reduce the interference of sunlight;
the image sensor takes the response in the wavelength range of the LED as sensitive as possible as priority to keep the exposure time as short as possible, thereby reducing the working pulse time of the LED and reducing the energy consumption of the system;
in the aspect of the constitution of the living body face double-camera recognition device, the living body face double-camera recognition device comprises a white light image acquisition channel and an infrared image acquisition channel, wherein the white light provides image preview and keeps the images smooth at a high frame rate; the infrared image acquisition is based on the requirement of satisfying algorithm, and the frame rate output as less as possible is selected to reduce the overall energy consumption and reduce the heating.
wherein, carry out background difference operation to black and white image and near-infrared image, background difference operation includes: taking a black-and-white image as a static image of the current environment, taking a near-infrared image as an active light source image, carrying out difference operation on the static image and the active light source image to obtain a difference image, and obtaining a complete background-removed active light source image according to the characteristics of the active light source, namely, an image of only the face part is left;
102, extracting two-dimensional state organ feature points from a black-and-white image acquired by a black-and-white camera in the double cameras and a face part in a near-infrared image by combining a feature extraction algorithm;
103, extracting organ feature points from the face part of the color image acquired by the color camera in the double cameras by combining a feature extraction algorithm and near-infrared images acquired by the black and white camera to form organ feature points in a three-dimensional state, identifying the features of the living face by a face and animal algorithm, and judging whether the face image is the living face;
wherein, the characteristics of discernment live body face through people's face draught animal algorithm includes: and converting a color image acquired by a color camera in the double cameras into a black-and-white image, performing an optical flow field estimation model with a near-infrared image acquired by the black-and-white camera, estimating differential distribution, judging whether the image is a living human face, if so, extracting organ feature points by combining an information acquisition module and a feature extraction algorithm, and if not, ending.
104, when the living body face is judged, obtaining a complete face feature picture and a data value from the extracted two-dimensional state organ feature points, and matching with a comparison feature database by combining a feature matching algorithm;
and 105, outputting display or control after matching is finished.
In step 103, the process of extracting the organ feature points by combining the information acquisition module and the feature extraction algorithm comprises two steps of visual geometric feature extraction and shape and texture feature extraction:
visual geometric feature extraction: extracting representative parts of the face of the human face, such as eyebrow, eye, nose, mouth and human face outline, and extracting an intuitive geometric figure by utilizing the characteristic relation of each part;
shape and texture feature extraction: the shape feature extraction is to extract the coordinate vectors of edges, outlines or some key points of the face image, is a binary feature and has strong capability of resisting illumination change; the extraction of the texture features is the gray value of the pixels of the human face image and is beneficial supplement of the shape features, and the ASMs/AAMs model utilizes the shape and the texture features to carry out statistical modeling through PCA.
Referring to fig. 7, fig. 7 is a flow chart of a living body face dual-camera identification method, an upper computer (which may be any operating system such as Windows, MAC OS, iOS, Android, Linux, and the like) acquires a near-infrared image and a color image through a USB2.0 dual-camera identification device, the image is preprocessed (such as sharpening, binarization, background removal, and the like) first, feature points are extracted, features of a face part are extracted, features and anti-counterfeiting (such as face photos, close body features, and the like) of a living body face are identified through a face beast algorithm, an accurate and complete face feature image and data values are obtained, and then a feature database is compared through a feature matching algorithm, wherein the feature database may be a local database, a cloud database, or newly-built cloud database data. And after the matching is finished, the display and control (which can be any action or equipment such as unlocking, attendance checking, identity recognition, visitor record and the like) are output.
Referring to fig. 8, the present embodiment further provides a living human face dual-camera recognition apparatus, where the recognition apparatus includes:
the image acquisition module 301 is used for acquiring a black-and-white image generated under the condition of natural light or white light and acquiring a near-infrared image generated under the condition of near-infrared light by using the two cameras, wherein the two cameras are a black-and-white camera and a color camera; the color camera acquires a color image generated under the condition of natural light or white light;
the image obtaining module 301 is further configured to perform a background difference operation on two face images, where the background difference operation includes: taking a black-and-white image as a static image of the current environment, taking a near-infrared image as an active light source image, carrying out difference operation on the static image and the active light source image to obtain a difference image, and obtaining a complete background-removed active light source image according to the characteristics of the active light source.
The information acquisition module 302 is configured to extract feature points of a two-dimensional organ by combining a black-and-white image acquired by a black-and-white camera in the two cameras with a human face part in a near-infrared image according to a feature extraction algorithm;
the judging module 303 is configured to extract organ feature points from the face part of the color image acquired by the color camera in the dual cameras by combining a feature extraction algorithm with the near-infrared image acquired by the black-and-white camera to form organ feature points in a three-dimensional state, identify features of a living body face through a face and animal algorithm, and judge whether the face image is the living body face;
the judging module 303 is further configured to convert a color image acquired by a color camera in the dual cameras into a black-and-white image, perform an optical flow field estimation model with a near-infrared image acquired by the black-and-white camera, estimate difference distribution, judge whether the image is a living human face, extract organ feature points in a two-dimensional state through the information acquiring module in combination with a feature extraction algorithm if the image is the living human face, and end if the image is not the living human face;
the matching module 304 is used for obtaining an accurate and complete human face feature picture and a data value when a living human face is judged, and matching the human face feature picture and the data value with a comparison feature database by combining a feature matching algorithm;
and a display or control module 305 for outputting display or control after the matching is completed.
The present invention is not limited to the above embodiments, and other methods and devices for recognizing a living human face by using the same or similar methods or devices as those of the above embodiments of the present invention are within the scope of the present invention.
Claims (1)
1. A method for recognizing a human face by two cameras is characterized by comprising the following steps:
101, acquiring a black-and-white image generated under the condition of natural light or white light LED and a near-infrared image generated under the condition of infrared LED by using two configured cameras, namely a black-and-white camera and a color camera, wherein the two cameras are respectively configured with the infrared LED and the white-light LED; the color camera acquires a color image generated under the condition of a natural light or white light LED;
selecting an LED infrared light source with concentrated spectral energy to work under the conditions of high output power and short pulse time so as to improve the energy efficiency ratio; the infrared light of the LED is concentrated and uniformly illuminated in an image acquisition area through a proper light homogenizing design, so that the light energy loss is reduced as much as possible; the lens adopts a band-pass filter to intercept the spectral energy outside the output range of the LED from entering so as to reduce the interference of sunlight; the image sensor takes the response in the wavelength range of the LED as sensitive as possible as priority to keep the exposure time as short as possible, thereby reducing the working pulse time of the LED and reducing the energy consumption of the system; in the aspect of the constitution of the living body face double-camera recognition device, the living body face double-camera recognition device comprises a white light image acquisition channel and an infrared image acquisition channel, wherein the white light provides image preview and keeps the images smooth at a high frame rate; the infrared image acquisition selects the frame rate output as less as possible on the basis of meeting the algorithm requirement so as to reduce the overall energy consumption and reduce the heating;
wherein, carry out background difference operation to black and white image and near-infrared image, background difference operation includes: taking a black-and-white image as a static image of the current environment, taking a near-infrared image as an active light source image, carrying out differential operation on the static image and the active light source image to obtain a differential image, and obtaining a complete background-removed active light source image according to the characteristics of the active light source; the basic process of background difference operation is as follows: extracting an active light source partial image in the image by adopting pixel difference-based closed-value method between two adjacent frames of the image; if the pixel value change of the image area is greater than the threshold value, which is caused by the active light source, the areas are marked as foreground pixels, and the marked pixel area is used for determining the position of the active light source target in the image;
102, extracting two-dimensional state organ feature points from a black-and-white image acquired by a black-and-white camera in the double cameras and a face part in a near-infrared image by combining a feature extraction algorithm;
extracting organ feature points from the face part of the color image acquired by the color camera in the double cameras by combining a feature extraction algorithm and extracting organ feature points from the near-infrared image acquired by the black-and-white camera to form three-dimensional organ feature points, identifying the features of the face of the living body by the face and livestock algorithm, and judging whether the face image is the face of the living body;
wherein, the characteristics of discernment live body face through people's face draught animal algorithm includes: converting a color image acquired by a color camera in the double cameras into a black-and-white image, performing an optical flow field estimation model with a near-infrared image acquired by the black-and-white camera, estimating differential distribution, judging whether the image is a living human face, if so, extracting organ feature points by combining an information acquisition module and a feature extraction algorithm, and if not, ending;
the process of extracting the organ feature points by combining the information acquisition module with the feature extraction algorithm comprises two steps of visual geometric feature extraction and shape and texture feature extraction:
visual geometric feature extraction: extracting representative parts of the face of the human face, and extracting an intuitive geometric figure by using the characteristic relation of each part;
shape and texture feature extraction: the shape feature extraction is to extract the coordinate vectors of edges, outlines or some key points of the face image, is a binary feature and has strong capability of resisting illumination change; the extraction of the texture features is the gray value of the pixels of the human face image and is beneficial supplement of the shape features, and the ASMs/AAMs model is statistically modeled by PCA (principal component analysis) by utilizing the shape and texture features;
the identification method further comprises the following steps:
when a living human face is judged, acquiring a complete human face feature picture and a data value from the extracted two-dimensional state organ feature points, and matching with a comparison feature database by combining a feature matching algorithm;
and outputting display or control after matching is finished.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710081460.5A CN106874871B (en) | 2017-02-15 | 2017-02-15 | Living body face double-camera identification method and identification device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710081460.5A CN106874871B (en) | 2017-02-15 | 2017-02-15 | Living body face double-camera identification method and identification device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106874871A CN106874871A (en) | 2017-06-20 |
CN106874871B true CN106874871B (en) | 2020-06-05 |
Family
ID=59167378
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710081460.5A Expired - Fee Related CN106874871B (en) | 2017-02-15 | 2017-02-15 | Living body face double-camera identification method and identification device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106874871B (en) |
Families Citing this family (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107688775A (en) * | 2017-07-11 | 2018-02-13 | 浙江新再灵科技股份有限公司 | Binocular camera recognition of face right discriminating system and method based on elevator scene |
CN107633198A (en) * | 2017-07-25 | 2018-01-26 | 百度在线网络技术(北京)有限公司 | Biopsy method, device, equipment and storage medium |
CN109426762B (en) * | 2017-08-22 | 2021-03-23 | 深圳荆虹科技有限公司 | Biological recognition system, method and biological recognition terminal |
CN107657248A (en) * | 2017-10-26 | 2018-02-02 | 广州云从信息科技有限公司 | A kind of infrared binocular In vivo detections of Android based on recognition of face certification |
CN108280418A (en) * | 2017-12-12 | 2018-07-13 | 北京深醒科技有限公司 | The deception recognition methods of face image and device |
CN108304789A (en) * | 2017-12-12 | 2018-07-20 | 北京深醒科技有限公司 | Recognition algorithms and device |
CN108197586B (en) * | 2017-12-12 | 2020-04-21 | 北京深醒科技有限公司 | Face recognition method and device |
CN108344738B (en) * | 2018-01-22 | 2021-07-27 | 翰飞骏德(北京)医疗科技有限公司 | Imaging method for hydroxyapatite and device thereof |
CN108388889B (en) * | 2018-03-23 | 2022-02-18 | 百度在线网络技术(北京)有限公司 | Method and device for analyzing face image |
CN108764052B (en) | 2018-04-28 | 2020-09-11 | Oppo广东移动通信有限公司 | Image processing method, image processing device, computer-readable storage medium and electronic equipment |
CN108737728B (en) * | 2018-05-03 | 2021-06-11 | Oppo广东移动通信有限公司 | Image shooting method, terminal and computer storage medium |
CN108694389A (en) * | 2018-05-15 | 2018-10-23 | Oppo广东移动通信有限公司 | Safe verification method based on preposition dual camera and electronic equipment |
CN108830229A (en) * | 2018-06-20 | 2018-11-16 | 哈尔滨理工大学 | The vivo identification method of Face datection is combined under a kind of frame based on caffe |
US20200082160A1 (en) * | 2018-09-12 | 2020-03-12 | Kneron (Taiwan) Co., Ltd. | Face recognition module with artificial intelligence models |
CN110008878B (en) * | 2019-03-27 | 2021-07-30 | 熵基科技股份有限公司 | Anti-fake method for face detection and face recognition device with anti-fake function |
CN110096861A (en) * | 2019-04-12 | 2019-08-06 | 檀鹏程 | A kind of bi-directional distributed formula authentication system based on biological characteristic |
CN110555931A (en) * | 2019-08-31 | 2019-12-10 | 华南理工大学 | Face detection and gate inhibition system device based on deep learning recognition |
CN111325139B (en) * | 2020-02-18 | 2023-08-04 | 浙江大华技术股份有限公司 | Lip language identification method and device |
TWI734454B (en) * | 2020-04-28 | 2021-07-21 | 鴻海精密工業股份有限公司 | Identity recognition device and identity recognition method |
CN111582118A (en) * | 2020-04-29 | 2020-08-25 | 福州瑞芯微电子股份有限公司 | Face recognition method and device |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102221937A (en) * | 2010-04-15 | 2011-10-19 | 上海天派无线科技有限公司 | Real-time video image coordinate recognition system and method |
CN104147778A (en) * | 2014-08-20 | 2014-11-19 | 赵东方 | Desktop projection interactive game machine host |
CN104318237A (en) * | 2014-10-28 | 2015-01-28 | 厦门大学 | Fatigue driving warning method based on face identification |
CN105023005B (en) * | 2015-08-05 | 2018-12-07 | 王丽婷 | Face identification device and its recognition methods |
-
2017
- 2017-02-15 CN CN201710081460.5A patent/CN106874871B/en not_active Expired - Fee Related
Also Published As
Publication number | Publication date |
---|---|
CN106874871A (en) | 2017-06-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106874871B (en) | Living body face double-camera identification method and identification device | |
CN108229362B (en) | Binocular face recognition living body detection method based on access control system | |
CN107862299B (en) | Living body face detection method based on near-infrared and visible light binocular cameras | |
CN109359634B (en) | Face living body detection method based on binocular camera | |
CN102622588B (en) | Dual-certification face anti-counterfeit method and device | |
CN105243357A (en) | Identity document-based face recognition method and face recognition device | |
CN104361326A (en) | Method for distinguishing living human face | |
CN104933344A (en) | Mobile terminal user identity authentication device and method based on multiple biological feature modals | |
CN104794458A (en) | Fuzzy video person identifying method | |
CN105975938A (en) | Smart community manager service system with dynamic face identification function | |
CN110458041B (en) | Face recognition method and system based on RGB-D camera | |
CN109902604B (en) | High-safety face comparison system and method based on Feiteng platform | |
CN103902958A (en) | Method for face recognition | |
CN112396011B (en) | Face recognition system based on video image heart rate detection and living body detection | |
CN104408780A (en) | Face recognition attendance system | |
CN104484652A (en) | Method for fingerprint recognition | |
CN107909063B (en) | Biological identification video playback attack detection method based on gray level change | |
CN105335691A (en) | Smiling face identification and encouragement system | |
CN112818722A (en) | Modular dynamically configurable living body face recognition system | |
CN104484803A (en) | Mobile phone payment method employing three-dimensional human face recognition based on neural network | |
CN109308436B (en) | Living body face recognition method based on active infrared video | |
CN113313057A (en) | Face living body detection and recognition system | |
CN106388441A (en) | Method and device for displaying, and intelligent mirror | |
CN113411507A (en) | Skin measurement image acquisition method, device, equipment and storage medium | |
CN110321782B (en) | System for detecting human body characteristic signals |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20200703 Address after: 230000 west side of Xianghe North Road, Feidong Economic Development Zone, Feidong County, Hefei City, Anhui Province Patentee after: ANHUI GUANGZHEN PHOTOELECTRIC TECHNOLOGY Co.,Ltd. Address before: 523000 Guangdong Province, Dongguan City Industrial Zone Qingxi Town Silver Star Patentee before: GUANGDONG LITE ARRAY Co.,Ltd. |
|
TR01 | Transfer of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200605 |
|
CF01 | Termination of patent right due to non-payment of annual fee |