CN111860047A - Face rapid identification method based on deep learning - Google Patents
Face rapid identification method based on deep learning Download PDFInfo
- Publication number
- CN111860047A CN111860047A CN201910345927.1A CN201910345927A CN111860047A CN 111860047 A CN111860047 A CN 111860047A CN 201910345927 A CN201910345927 A CN 201910345927A CN 111860047 A CN111860047 A CN 111860047A
- Authority
- CN
- China
- Prior art keywords
- face
- picture
- feature point
- point data
- deep learning
- 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 32
- 238000013135 deep learning Methods 0.000 title claims abstract description 31
- 230000001815 facial effect Effects 0.000 claims description 22
- 210000000056 organ Anatomy 0.000 claims description 8
- 210000000887 face Anatomy 0.000 claims description 5
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 238000013136 deep learning model Methods 0.000 claims description 4
- 230000007547 defect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 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
- G06V40/168—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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- 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
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Library & Information Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method for quickly identifying a human face based on deep learning, which comprises the steps of collecting N pictures with human faces, processing the pictures to generate corresponding N human face sample pictures, extracting human face characteristic point data from the N human face sample pictures and storing the data into a database; deep learning is carried out on the N face pictures, and a learning model is established; comparing the input video stream with the learning model, and collecting a video face to form a face picture; extracting face feature point data; the human face feature point data is compared with the human face feature point data stored in the database for judgment, and the human face rapid identification method not only improves the human face identification efficiency, but also improves the human face identification precision.
Description
Technical Field
The invention relates to the technical field of face recognition, in particular to a face rapid recognition method based on deep learning.
Background
Face recognition, also commonly called portrait recognition, facial recognition, is a biometric technology for identifying the identity based on the facial feature information of a person, and uses a camera or a video camera to collect an image or a video stream containing a face, and automatically detects and tracks the face in the image, and further performs a series of related technologies on the face of the detected face, so as to achieve the purpose of identifying the identity of different persons.
The current face recognition technology has the following defects or defects:
(1) the human face recognition speed is low, most of the existing human face recognition systems need user cooperation, and after a user executes a corresponding instruction according to requirements, recognition and detection can be finished under the condition that the acquisition conditions are relatively ideal;
(2) the face recognition accuracy is low, which results in low recognition efficiency, such as uneven illumination, yin and yang faces, low resolution, background interference, etc.), and under various shielding conditions, the recognition rate is greatly reduced.
(3) And a deep learning algorithm is adopted for feature extraction, so that artificial interference is avoided, and the optimal features are searched through self learning of the equipment. But the method also has the defects of a mass sample library, long training period, high requirements on equipment configuration and the like.
Disclosure of Invention
The invention aims to overcome the problems in the prior art and provides a rapid face recognition method based on deep learning.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides a face rapid identification method based on deep learning, which comprises the following steps:
collecting N pictures with faces, processing the pictures to generate corresponding N face sample pictures, extracting face feature point data from the N face sample pictures and storing the data in a database;
The generation method of the face sample picture comprises the steps of firstly identifying an eye region picture, a nose region picture and a mouth region picture of a picture with a face, positioning the central positions of two eyes, a nose and a mouth, aligning the two eye positions of the eye region picture on the same horizontal line, rotating and/or translating the central positions of the two eyes, the nose and the mouth on the same vertical line, cutting the face picture through a rectangular frame to form the face sample picture, and aligning the vertical center line of the rectangular frame to the central position between the two eyes;
step two, deep learning is carried out on the N face pictures, and a learning model is established;
comparing the input video stream with the learning model, and collecting a video face to form a face picture; firstly, intercepting a picture with a human face from a video stream, identifying an eye region image, a nose region image and a mouth region image from the picture, positioning the central positions of two eyes, a nose and a mouth, aligning the two eye positions of the eye region image on the same horizontal line, rotating and/or translating the central positions of the two eyes, the nose and the mouth on the same vertical line, cutting the image through a rectangular frame to form a human face picture, and aligning the vertical central line of the rectangular frame to the central position between the two eyes;
Extracting face feature point data from the collected face picture;
and step five, comparing the face feature point data with face feature point data stored in a database, judging whether the face feature point data is similar to the face feature point data stored in the database, if so, calling face sample information corresponding to the face feature point stored in the database, and otherwise, ending.
According to the method for rapidly identifying the face based on the deep learning, the face feature point data in the face image is positioned through the deep learning model in the convolutional neural network.
Further, the face feature points include organ feature points and contour feature points, and organ position points.
In the fourth step of the method for rapidly identifying the human face based on the deep learning, before extracting the data of the human face feature points, the image resolution of the collected human face image is adjusted to be the same as the resolution of the human face sample image.
In another embodiment of the present invention, the first step of the method for rapidly identifying a face based on deep learning further includes processing N face sample pictures into M resolution face sample pictures, and respectively extracting and storing corresponding M kinds of face feature point data into M sub-databases of a database.
Judging the resolution of the acquired face picture, searching the resolution with the M resolutions closest to each other, adjusting the resolution of the face picture to the closest resolution, and extracting face feature point data; comparing the face feature point data with face feature point data stored in a sub-database with the closest resolution, judging whether the face feature point data is similar to the face feature point data stored in the database, if so, calling face sample information corresponding to the face feature point stored in the database, and otherwise, ending.
Compared with the prior art, the method for rapidly identifying the human face based on the deep learning has the following beneficial effects:
(1) compared with the existing face rapid identification method, the face sample picture is processed and then face characteristic point data is extracted and stored in the database, so that the face identification efficiency and the face identification precision are improved;
(2) the face recognition precision is continuously improved through continuous and uninterrupted deep learning;
(3) And meanwhile, the N face sample pictures are processed into M-resolution face sample pictures, corresponding M kinds of face feature point data are respectively extracted and stored in M sub databases of the databases, the resolution of the collected face images is adjusted, the face feature point data are compared with the face feature point data stored in the sub database with the closest resolution, and the recognition efficiency is further improved on the premise of not reducing the recognition accuracy.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a schematic flow chart of a method for rapidly recognizing a face based on deep learning according to embodiment 1 of the present invention;
fig. 2 is a schematic flow chart of a method for rapidly recognizing a face based on deep learning according to embodiment 2 of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantageous effects to be solved by the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Example 1
The embodiment of the invention provides a face rapid identification method based on deep learning, which comprises the following steps as shown in the attached figure 1:
the invention provides a face rapid identification method based on deep learning, which comprises the following steps:
step S11, collecting N pictures with human faces, processing and generating corresponding N human face sample pictures, extracting human face feature point data from the N human face sample pictures and storing the data in a database;
the generation method of the face sample picture comprises the steps of firstly identifying an eye region picture, a nose region picture and a mouth region picture of a picture with a face, positioning the central positions of two eyes, a nose and a mouth, aligning the two eye positions of the eye region picture on the same horizontal line, rotating and/or translating the central positions of the two eyes, the nose and the mouth on the same vertical line, cutting the face picture through a rectangular frame to form the face sample picture, and aligning the vertical center line of the rectangular frame to the central position between the two eyes;
step S12, deep learning is carried out on the N face pictures, and a learning model is established;
step S13, comparing the input video stream with the learning model, collecting the video face to form a face picture, specifically, intercepting the picture with the face from the video stream, identifying an eye region image, a nose region image and a mouth region image from the picture, positioning the central positions of two eyes, a nose and a mouth, aligning the two eye positions of the eye region image on the same horizontal line, rotating and/or translating the picture on the same vertical line at the central positions of the two eyes, the nose and the mouth, cutting the image through a rectangular frame to form the face picture, and aligning the vertical center line of the rectangular frame with the central position between the two eyes;
Step S14, extracting face feature point data from the collected face picture;
step S15, comparing the facial feature point data with the facial feature point data stored in the database, determining whether the facial feature point data is similar to the facial feature point data stored in the database, if so, retrieving facial sample information corresponding to the facial feature points stored in the database, otherwise, ending.
According to the method for rapidly identifying the face based on the deep learning, the face feature point data in the face image is positioned through the deep learning model in the convolutional neural network.
Further, the face feature points include organ feature points and contour feature points, and organ position points.
In step S14 of the above-mentioned method for fast recognizing a face based on deep learning, before extracting the face feature point data, the image resolution of the collected face image is adjusted to the same resolution as the face sample image.
Example 2
The embodiment of the invention provides another face rapid identification method based on deep learning, which comprises the following steps as shown in the attached figure 2:
step S21, collecting N pictures with human faces, processing the pictures to generate corresponding N human face sample pictures, processing the N human face sample pictures into M human face sample pictures with resolution, and respectively extracting corresponding M types of human face feature point data to store the data in M sub databases of a database;
The generation method of the face sample picture comprises the steps of firstly identifying an eye region picture, a nose region picture and a mouth region picture of a picture with a face, positioning the central positions of two eyes, a nose and a mouth, aligning the two eye positions of the eye region picture on the same horizontal line, rotating and/or translating the central positions of the two eyes, the nose and the mouth on the same vertical line, cutting the face picture through a rectangular frame to form the face sample picture, and aligning the vertical center line of the rectangular frame to the central position between the two eyes;
step S22, deep learning is carried out on the N face pictures, and a learning model is established;
step S23, comparing the input video stream with the learning model, and collecting the video face to form a face picture; firstly, intercepting a picture with a human face from a video stream, identifying an eye region image, a nose region image and a mouth region image from the picture, positioning the central positions of two eyes, a nose and a mouth, aligning the two eye positions of the eye region image on the same horizontal line, rotating and/or translating the central positions of the two eyes, the nose and the mouth on the same vertical line, cutting the image through a rectangular frame to form a human face picture, and aligning the vertical central line of the rectangular frame to the central position between the two eyes;
Step S24, judging the resolution of the collected face picture, searching the resolution with the M resolutions closest to each other, adjusting the resolution of the face picture to the closest resolution, and extracting face feature point data;
step S25, comparing the facial feature point data with the stored facial feature point data in the sub-database with the closest resolution, determining whether the facial feature point data is similar to the stored facial feature point data in the database, if so, retrieving facial sample information corresponding to the stored facial feature points in the database, otherwise, ending.
According to the method for rapidly identifying the face based on the deep learning, the face feature point data in the face image is positioned through the deep learning model in the convolutional neural network.
Further, the face feature points include organ feature points and contour feature points, and organ position points.
The above description describes preferred embodiments of the invention, but it should be understood that the invention is not limited to the above embodiments, and should not be viewed as excluding other embodiments. Modifications made by those skilled in the art in light of the teachings of this disclosure, which are well known or are within the skill and knowledge of the art, are also to be considered as within the scope of this invention.
Claims (6)
1. A rapid human face recognition method based on deep learning is characterized in that,
the method comprises the following steps:
collecting N pictures with faces, processing the pictures to generate corresponding N face sample pictures, extracting face feature point data from the N face sample pictures and storing the data in a database;
the generation method of the face sample picture comprises the steps of firstly identifying an eye region picture, a nose region picture and a mouth region picture of a picture with a face, positioning the central positions of two eyes, a nose and a mouth, aligning the two eye positions of the eye region picture on the same horizontal line, rotating and/or translating the central positions of the two eyes, the nose and the mouth on the same vertical line, cutting the face picture through a rectangular frame to form the face sample picture, and aligning the vertical center line of the rectangular frame to the central position between the two eyes;
step two, deep learning is carried out on the N face pictures, and a learning model is established;
comparing the input video stream with the learning model, and collecting a video face to form a face picture; firstly, intercepting a picture with a human face from a video stream, identifying an eye region image, a nose region image and a mouth region image from the picture, positioning the central positions of two eyes, a nose and a mouth, aligning the two eye positions of the eye region image on the same horizontal line, rotating and/or translating the central positions of the two eyes, the nose and the mouth on the same vertical line, cutting the image through a rectangular frame to form a human face picture, and aligning the vertical central line of the rectangular frame to the central position between the two eyes;
Extracting face feature point data from the collected face picture;
comparing the facial feature point data with facial feature point data stored in a database, judging whether the facial feature point data is similar to the facial feature point data stored in the database, if so, calling facial sample information corresponding to the facial feature points stored in the database, otherwise, ending;
wherein N is a positive integer.
2. The method for rapidly recognizing the human face based on the deep learning as claimed in claim 1, wherein
And positioning the data of the facial feature points in the facial image through a deep learning model in the convolutional neural network.
3. The method for rapidly recognizing the human face based on the deep learning as claimed in claim 1, wherein
The face feature points include organ feature points and contour feature points and organ position points.
4. The fast face recognition method based on deep learning of claim 1,
and in the fourth step, before extracting the face feature point data, adjusting the image resolution of the acquired face image to be the same resolution of the face sample image.
5. The fast face recognition method based on deep learning of claim 1,
The first step further comprises the steps that N face sample pictures are processed into M face sample pictures with resolution, corresponding M kinds of face feature point data are respectively extracted and stored into M sub databases of the database, and M is a positive integer.
6. The method for rapidly recognizing the human face based on the deep learning of claim 5,
step four, judging the resolution of the acquired face picture, searching the resolution with the M resolutions closest to each other, adjusting the resolution of the face picture to the closest resolution, and extracting face feature point data;
and step five, comparing the face feature point data with face feature point data stored in a sub-database with the closest resolution, judging whether the face feature point data is similar to the face feature point data stored in the database, if so, calling face sample information corresponding to the face feature point stored in the database, and otherwise, ending.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910345927.1A CN111860047A (en) | 2019-04-26 | 2019-04-26 | Face rapid identification method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910345927.1A CN111860047A (en) | 2019-04-26 | 2019-04-26 | Face rapid identification method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111860047A true CN111860047A (en) | 2020-10-30 |
Family
ID=72951784
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910345927.1A Pending CN111860047A (en) | 2019-04-26 | 2019-04-26 | Face rapid identification method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111860047A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114220142A (en) * | 2021-11-24 | 2022-03-22 | 慧之安信息技术股份有限公司 | Face feature recognition method of deep learning algorithm |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101174103B1 (en) * | 2011-12-28 | 2012-08-14 | (주)로복전자 | A face recognition method of Mathematics pattern analysis for muscloskeletal in basics |
| CN105117692A (en) * | 2015-08-05 | 2015-12-02 | 福州瑞芯微电子股份有限公司 | Real-time face identification method and system based on deep learning |
| CN107392182A (en) * | 2017-08-17 | 2017-11-24 | 宁波甬慧智能科技有限公司 | A kind of face collection and recognition method and device based on deep learning |
| CN109063696A (en) * | 2018-09-30 | 2018-12-21 | 中山市昭歌云视频安防科技有限公司 | A kind of face identification method and system |
| WO2019033571A1 (en) * | 2017-08-17 | 2019-02-21 | 平安科技（深圳）有限公司 | Facial feature point detection method, apparatus and storage medium |
| WO2019033572A1 (en) * | 2017-08-17 | 2019-02-21 | 平安科技（深圳）有限公司 | Method for detecting whether face is blocked, device and storage medium |
| CN109508700A (en) * | 2018-12-28 | 2019-03-22 | 广州粤建三和软件股份有限公司 | A kind of face identification method, system and storage medium |
-
2019
- 2019-04-26 CN CN201910345927.1A patent/CN111860047A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101174103B1 (en) * | 2011-12-28 | 2012-08-14 | (주)로복전자 | A face recognition method of Mathematics pattern analysis for muscloskeletal in basics |
| CN105117692A (en) * | 2015-08-05 | 2015-12-02 | 福州瑞芯微电子股份有限公司 | Real-time face identification method and system based on deep learning |
| CN107392182A (en) * | 2017-08-17 | 2017-11-24 | 宁波甬慧智能科技有限公司 | A kind of face collection and recognition method and device based on deep learning |
| WO2019033571A1 (en) * | 2017-08-17 | 2019-02-21 | 平安科技（深圳）有限公司 | Facial feature point detection method, apparatus and storage medium |
| WO2019033572A1 (en) * | 2017-08-17 | 2019-02-21 | 平安科技（深圳）有限公司 | Method for detecting whether face is blocked, device and storage medium |
| CN109063696A (en) * | 2018-09-30 | 2018-12-21 | 中山市昭歌云视频安防科技有限公司 | A kind of face identification method and system |
| CN109508700A (en) * | 2018-12-28 | 2019-03-22 | 广州粤建三和软件股份有限公司 | A kind of face identification method, system and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 张成成;李思成;王艳双;陈磊磊;张;: "基于深度学习的人脸识别技术在课堂签到上的应用", 时代汽车, no. 04 * |
| 陈章斌;: "基于深度学习人脸识别技术在高校课堂点名中的设计及实现", 兰州文理学院学报(自然科学版), no. 06 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114220142A (en) * | 2021-11-24 | 2022-03-22 | 慧之安信息技术股份有限公司 | Face feature recognition method of deep learning algorithm |
| CN114220142B (en) * | 2021-11-24 | 2022-08-23 | 慧之安信息技术股份有限公司 | Face feature recognition method of deep learning algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021036436A1 (en) | Facial recognition method and apparatus | |
| CN109117797A (en) | A kind of face snapshot recognition method based on face quality evaluation | |
| CN110490158B (en) | Robust face alignment method based on multistage model | |
| CN108564052A (en) | Multi-cam dynamic human face recognition system based on MTCNN and method | |
| CN110210276A (en) | A kind of motion track acquisition methods and its equipment, storage medium, terminal | |
| CN107977639B (en) | Face definition judgment method | |
| CN106529414A (en) | Method for realizing result authentication through image comparison | |
| CN104050448B (en) | A kind of human eye positioning, human eye area localization method and device | |
| CN110929679A (en) | Non-supervision self-adaptive pedestrian re-identification method based on GAN | |
| CN111666845B (en) | Small sample deep learning multi-mode sign language recognition method based on key frame sampling | |
| CN110827432B (en) | Class attendance checking method and system based on face recognition | |
| CN110991398A (en) | Gait recognition method and system based on improved gait energy map | |
| CN111488943A (en) | Face recognition method and device | |
| CN111860047A (en) | Face rapid identification method based on deep learning | |
| CN111582195B (en) | Construction method of Chinese lip language monosyllabic recognition classifier | |
| CN106980818B (en) | Personalized preprocessing method, system and terminal for face image | |
| CN113537050A (en) | Dynamic face recognition algorithm based on local image enhancement | |
| CN110633631B (en) | Pedestrian re-identification method based on component power set and multi-scale features | |
| CN112149517A (en) | Face attendance checking method and system, computer equipment and storage medium | |
| CN112200008A (en) | Face attribute recognition method in community monitoring scene | |
| CN110598569B (en) | Action recognition method based on human body posture data | |
| CN112507941A (en) | Cross-vision field pedestrian re-identification method and device for mine AI video analysis | |
| CN112733732A (en) | Face detection and recognition method based on feature analysis | |
| CN111797691A (en) | Method for improving face recognition accuracy and processing subsystem | |
| CN110830734A (en) | Abrupt change and gradual change lens switching identification method |
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 |