CN110390229B - Face picture screening method and device, electronic equipment and storage medium - Google Patents
Face picture screening method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110390229B CN110390229B CN201810360907.7A CN201810360907A CN110390229B CN 110390229 B CN110390229 B CN 110390229B CN 201810360907 A CN201810360907 A CN 201810360907A CN 110390229 B CN110390229 B CN 110390229B
- Authority
- CN
- China
- Prior art keywords
- face
- picture
- detected
- preset
- dimensional reference
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
- G06V20/653—Three-dimensional objects by matching three-dimensional models, e.g. conformal mapping of Riemann surfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- 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)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention provides a method and a device for screening face pictures, electronic equipment and a storage medium, and is characterized in that the method comprises the following steps: acquiring a picture to be detected containing a human face target; comparing a face target in a picture to be detected with a three-dimensional reference face model in a preset three-dimensional reference database to determine a characteristic value of the face target; and comparing the numerical value between the characteristic value and a preset threshold value, and classifying the pictures to be detected into different categories according to the comparison result, wherein the categories comprise high-quality pictures and common-quality pictures. By comparing the value of the characteristic value with the preset threshold value, whether the picture to be detected meets the requirement of subsequent operation can be measured more accurately, and the picture to be detected is screened more accurately.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to a method and an apparatus for screening a face image, an electronic device, and a storage medium.
Background
With the continuous development of internet and intelligent hardware device technologies, face recognition technology has been applied to more and more fields, such as security, finance, entertainment, and the like. Provides more and more convenience and fun for the life of people. Through the face recognition technology, the real face or the face in the picture can be recognized, so that the corresponding identity information and the like can be determined, and further subsequent responses such as operation or flow of information registration, payment and the like can be executed. Especially in the security protection field, the face recognition technology can rapidly identify the identity of people, thereby being beneficial to the development of security protection work.
When the identity information is determined by the face recognition technology, pictures containing faces can be obtained through shooting equipment such as a video camera, a camera and a camera, and in practical application, a plurality of pictures containing faces can be generally obtained, then the plurality of face pictures can be screened or classified according to certain rules, so that the face pictures meeting the requirements are selected, and then subsequent operations or processes are performed on the selected face pictures.
However, in the prior art, when a face picture is screened, the screening is generally performed only according to the definition of the face picture, so that in many cases, the face picture which can meet the requirement of the subsequent operation cannot be accurately selected, and the subsequent operation or process cannot be accurately completed. For example, when a face in a face picture is tilted downward or tilted upward, although the picture definition is high, due to an excessively large angle of the tilted downward or tilted upward, a part of a face region is blocked, and identity information corresponding to the face cannot be accurately identified through the face picture.
Disclosure of Invention
The embodiment of the invention aims to provide a method and a device for screening face pictures, electronic equipment and a storage medium, which can more accurately screen the face pictures. The specific technical scheme is as follows:
the embodiment of the invention provides a method for screening face pictures, which comprises the following steps:
acquiring a picture to be detected containing a human face target;
comparing the face target in the picture to be detected with the three-dimensional reference face model in the preset three-dimensional reference database to determine a characteristic value of the face target;
and comparing the numerical value between the characteristic value and a preset threshold value, and classifying the pictures to be detected into different categories according to the comparison result, wherein the categories comprise high-quality pictures and common-quality pictures.
Optionally, the comparing the face target in the picture to be detected with the three-dimensional reference face model in the preset three-dimensional reference database to determine the characteristic value of the face target includes:
respectively comparing the face target in the picture to be detected with a plurality of three-dimensional reference face models in the preset three-dimensional reference database, wherein the three-dimensional reference face models are three-dimensional reference face models with different characteristic values;
determining a three-dimensional reference face model matched with the face target;
and taking the characteristic value of the three-dimensional reference human face model as the characteristic value of the human face target.
Optionally, the characteristic values include a binocular pupil distance, a pitch angle, and left and right side face angles;
the preset threshold comprises a binocular pupil distance threshold, a pitch angle threshold and left and right side face angle thresholds.
Optionally, the comparing the value between the characteristic value and a preset threshold value, and according to the comparison result, classifying the pictures to be detected into different categories, including:
when the binocular pupil distance is larger than the binocular pupil distance threshold value, the pitch angle is smaller than the pitch angle threshold value, and the left and right side face angles are smaller than the left and right side face angle threshold values, the picture to be detected is a high-quality picture;
otherwise, the picture to be detected is a common quality picture.
Optionally, the binocular pupil distance threshold includes a first binocular pupil distance threshold and a second binocular pupil distance threshold, the pitch angle threshold includes a first pitch angle threshold and a second pitch angle threshold, the left and right side face angle thresholds include a first left and right side face angle threshold and a second left and right side face angle threshold, and the category further includes medium quality pictures;
the comparison of the numerical values between the characteristic values and the preset threshold values and the classification of the pictures to be detected into different categories according to the comparison results comprise:
when the binocular pupil distance is larger than the first binocular pupil distance threshold value, the pitch angle is smaller than the first pitch angle threshold value, and the left and right side face angles are smaller than the first left and right side face angle threshold value, the picture to be detected is a high-quality picture;
when the binocular interpupillary distance is less than the first binocular interpupillary distance threshold and greater than the second binocular interpupillary distance threshold,
and the pitch angle is greater than the first pitch angle threshold and less than the second pitch angle threshold,
when the left and right side face angles are larger than the first left and right side face angle threshold value and smaller than the second left and right side face angle threshold value, the picture to be detected is a medium-quality picture;
otherwise, the picture to be detected is a common quality picture.
Optionally, the comparing the face target in the picture to be detected with the three-dimensional reference face model in the preset three-dimensional reference database to determine the characteristic value of the face target includes:
identifying the human face target in the picture to be detected;
and comparing the face target with a three-dimensional reference face model in the preset three-dimensional reference database to determine the characteristic value of the face target.
Optionally, the method further includes:
establishing a face model corresponding to the face target in the picture to be detected;
when the picture to be detected is a high-quality picture, inputting the face model into a preset database, and comparing the face model with each preset face model stored in the preset database;
when a preset face model matched with the face model exists in the preset database, outputting alarm information;
or when the preset face model matched with the face model does not exist in the preset database, alarm information is output.
The embodiment of the invention also provides a face picture screening device, which comprises:
the acquisition module is used for acquiring a picture to be detected containing a human face target;
the comparison module is used for comparing the face target in the picture to be detected with the three-dimensional reference face model in the preset three-dimensional reference database to determine the characteristic value of the face target;
and the screening module is used for comparing the numerical value between the characteristic value and a preset threshold value, and classifying the pictures to be detected into different categories according to the comparison result, wherein the categories comprise high-quality pictures and common-quality pictures.
Optionally, the comparison module is specifically configured to:
respectively comparing the face target in the picture to be detected with a plurality of three-dimensional reference face models in the preset three-dimensional reference database, wherein the three-dimensional reference face models are three-dimensional reference face models with different characteristic values; determining a three-dimensional reference face model matched with the face target; and taking the characteristic value of the three-dimensional reference human face model as the characteristic value of the human face target.
Optionally, the characteristic values in the device include a binocular pupil distance, a pitch angle, and left and right side face angles;
the preset threshold comprises a binocular pupil distance threshold, a pitch angle threshold and left and right side face angle thresholds.
Optionally, the screening module is specifically configured to:
when the binocular pupil distance is larger than the binocular pupil distance threshold value, the pitch angle is smaller than the pitch angle threshold value, and the left and right side face angles are smaller than the left and right side face angle threshold values, the picture to be detected is a high-quality picture; otherwise, the picture to be detected is a common quality picture.
Optionally, the binocular pupil distance threshold in the comparison module includes a first binocular pupil distance threshold and a second binocular pupil distance threshold, the pitch angle threshold includes a first pitch angle threshold and a second pitch angle threshold, the left and right side face angle thresholds include a first left and right side face angle threshold and a second left and right side face angle threshold, and the category in the screening module further includes medium quality pictures;
the screening module is specifically configured to:
when the binocular pupil distance is larger than the first binocular pupil distance threshold value, the pitch angle is smaller than the first pitch angle threshold value, and the left and right side face angles are smaller than the first left and right side face angle threshold value, the picture to be detected is a high-quality picture;
when the binocular pupil distance is smaller than the first binocular pupil distance threshold and larger than the second binocular pupil distance threshold, the pitch angle is larger than the first pitch angle threshold and smaller than the second pitch angle threshold, and the left and right side face angles are larger than the first left and right side face angle threshold and smaller than the second left and right side face angle threshold, the picture to be detected is a medium quality picture;
otherwise, the picture to be detected is a common quality picture.
Optionally, the comparison module is specifically configured to:
identifying the human face target in the picture to be detected;
and comparing the face target with a three-dimensional reference face model in the preset three-dimensional reference database to determine the characteristic value of the face target.
Optionally, the apparatus further comprises:
the modeling module is used for establishing a face model corresponding to the face target in the picture to be detected;
the analysis module is used for inputting the face model into a preset database and comparing the face model with each preset face model stored in the preset database when the picture to be detected is a high-quality picture;
the alarm module is used for outputting alarm information when a preset face model matched with the face model exists in the preset database; or when the preset face model matched with the face model does not exist in the preset database, alarm information is output.
The embodiment of the invention also provides electronic equipment which comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete mutual communication through the communication bus;
a memory for storing a computer program;
and the processor is used for realizing any one of the face picture screening methods when executing the program stored in the memory.
In another aspect of the present invention, there is also provided a computer-readable storage medium, where instructions are stored, and when the instructions are executed on a computer, the computer is enabled to execute any one of the above-mentioned face picture screening methods.
In another aspect of the present invention, an embodiment of the present invention further provides a computer program product including instructions, which when run on a computer, causes the computer to execute any one of the above-mentioned face picture screening methods.
According to the face picture screening method, the face picture screening device, the electronic equipment and the storage medium, the face target in the obtained picture to be detected is compared with the three-dimensional reference face model in the preset three-dimensional reference database to obtain the characteristic value of the face target, and then the characteristic value is compared with the preset threshold value in numerical value, so that the picture to be detected is determined to be a high-quality picture or a common-quality picture. In the embodiment of the invention, when the picture to be detected is screened, the picture to be detected can be screened through various characteristic values instead of only considering one factor of definition, so that whether the picture to be detected meets the requirement of subsequent operation can be more accurately measured, and the picture to be detected is more accurately screened. In addition, the preset threshold value can be configured in advance, so that the screening standard of the picture to be detected can be adjusted more flexibly to adapt to application under different actual conditions or scenes. Of course, it is not necessary for any product or method of practicing the invention to achieve all of the above-described advantages at the same time.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart of a method for screening a face picture according to an embodiment of the present invention;
fig. 2 is another flowchart of a face image screening method according to an embodiment of the present invention;
fig. 3 is a structural diagram of a face image screening apparatus according to an embodiment of the present invention;
fig. 4 is a structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a flowchart of a face image screening method according to an embodiment of the present invention, including:
The embodiment of the invention can be used for various electronic equipment with data processing capability, and the electronic equipment can be used for processing image data, for example, the electronic equipment can be a mobile phone, a snapshot machine, a camera, a server, a monitoring camera, a monitoring system connected with the monitoring camera, and the like.
The picture to be detected can be an image or a picture containing one or more face targets, and the face targets are images capable of reflecting face features of people. For example, the picture to be detected can be a picture taken at any time by a capturing machine installed at an entrance and an exit of a specific place, and the picture contains one or more human face targets; or, the picture to be detected may also be a monitoring picture shot by a monitoring camera, wherein the picture may contain a human face target in a monitored scene.
The electronic device may obtain the picture to be detected in various ways, for example, manually input the picture to be detected, or the electronic device may directly obtain the picture to be detected, for example, a picture containing a human face target shot by a camera or a snapshot machine may be directly used as the picture to be detected.
The feature value is a parameter value capable of reflecting the features of the face target, and the feature value may be of various types, for example, the feature value may be a position angle of the face, an area size of the face target, a position relationship of feature points in the face, a proportion and a definition between five sense organs in the face target, and the like.
After the picture to be detected is obtained, the picture to be detected can be input into a preset three-dimensional reference database. And a large number of three-dimensional reference face models are stored in a preset three-dimensional reference database. The three-dimensional reference human face model is established by acquiring a front image, side images at different angles, images at different head raising angles, images at different head lowering angles and images shot at different shooting distances and then in an orthogonal picture mode. Each reference face model may have different feature values such as area size, position angle, or position pose.
Therefore, specifically, in the embodiment of the present invention, in step 102, comparing the face target in the picture to be detected with the three-dimensional reference face model in the preset three-dimensional reference database, and determining the feature value of the face target may include:
102a, respectively comparing the face target in the picture to be detected with a plurality of three-dimensional reference face models in a preset three-dimensional reference database, wherein the plurality of three-dimensional reference face models are three-dimensional reference face models with different characteristic values.
And 102b, determining a three-dimensional reference face model matched with the face target.
The human face target is compared with the three-dimensional reference human face model one by one, so that the three-dimensional reference human face model which is the closest to the size, position angle or position posture of the human face target can be determined, and the three-dimensional reference human face model is used as the three-dimensional reference human face model matched with the human face target.
In addition, in practical application, in order to more rapidly match the face target with the three-dimensional reference face model, a face model corresponding to the face target can be established for the face target, and then the face model is compared with a plurality of three-dimensional reference face models, so that the three-dimensional reference face model matched with the face target can be determined more rapidly.
And 102c, taking the characteristic value of the three-dimensional reference human face model as the characteristic value of the human face target.
When the three-dimensional reference face model matched with the face target is determined, the characteristic value of the matched three-dimensional reference face model can be used as the characteristic value of the face target, so that the characteristic value of the face target is determined.
And 103, comparing the numerical value between the characteristic value and a preset threshold value, and classifying the pictures to be detected into different categories according to the comparison result, wherein the categories comprise high-quality pictures and common-quality pictures.
After the picture to be detected is input into the preset three-dimensional reference database, the characteristic value of the human face target in the picture to be detected can be obtained through the preset three-dimensional reference database. When the characteristic value is obtained, the characteristic value can be used for comparison with a preset threshold value.
The preset threshold is a preset threshold, and whether the characteristic value meets the required standard or not or whether the characteristic value is in a proper range can be measured by comparing the characteristic value with the preset threshold. When the characteristic value meets the standard or is in a proper range, the human face target corresponding to the characteristic value meets the condition or requirement for executing subsequent processes or operations, so that the picture to be detected where the human face target is located can be determined to be a high-quality picture, and otherwise, the picture to be detected is a high-quality picture with common quality.
Specifically, there may be different preset thresholds corresponding to different types of feature values. When the types of the characteristic values are different, the preset threshold values are correspondingly different, and the specific comparison mode is also different.
For example, when the feature value is an angle of a head-up of a human face, the preset threshold may be an angle value, and when the feature value exceeds the angle, the requirement is not met, and the identity information corresponding to the human face target cannot be accurately identified.
Or, when the feature value is the area of the human face target, the preset threshold may be an area value, and when the feature value is smaller than the area, the requirement is not satisfied, so that when the feature value is smaller than the preset threshold, the picture to be detected is a normal quality picture.
Therefore, when the feature value is compared with the preset threshold, it is determined that the picture to be detected is a high-quality picture or a normal-quality picture under the condition of various different comparison results according to different feature value types.
In the embodiment of the invention, the human face target in the acquired picture to be detected is compared with the three-dimensional reference human face model in the preset three-dimensional reference database to obtain the characteristic value of the human face target, and then the numerical value of the characteristic value is compared with the preset threshold value, so that the picture to be detected is determined to be a high-quality picture or a common-quality picture. In the embodiment of the invention, when the picture to be detected is screened, the picture to be detected can be screened through various characteristic values instead of only considering one factor of definition, so that whether the picture to be detected meets the requirement of subsequent operation can be more accurately measured, and the picture to be detected is more accurately screened. In addition, the preset threshold value can be configured in advance, so that the screening standard of the picture to be detected can be adjusted more flexibly to adapt to application under different actual conditions or scenes.
When the picture to be detected is screened out to be a high-quality picture or a common-quality picture, corresponding subsequent operations or processes can be carried out according to different categories of the picture to be detected. Therefore, in order to perform the subsequent steps more conveniently, referring to fig. 2, the method for screening a face picture according to the embodiment of the present invention further includes:
and 104, establishing a face model corresponding to the face target in the picture to be detected.
The face model is a digital model constructed by using face features in a face target, such as relative positions and proportional relations among facial features and the like. Through the digital model, various operations or processes can be realized on the face target, for example, identification information confirmation is carried out through the digital model.
The step of establishing the corresponding face model for the face target in the picture to be detected can be performed at any stage after the picture to be detected is obtained, for example, the corresponding face model can be established for the face target in the picture to be detected immediately after the picture to be detected is obtained, so that each picture to be detected can obtain one corresponding face model; or, the method can be carried out after screening whether the picture to be detected is a high-quality picture, so that a corresponding face model can be established only for the screened picture to be detected with the high-quality picture, the data processing amount is reduced, and the operation efficiency is improved. Of course, if the face model corresponding to the face target is already established when the face target is compared with the three-dimensional reference face model in the preset three-dimensional reference data, the face model can also be directly applied without repeatedly establishing the face model corresponding to the face target.
And if the category of the picture to be detected is a high-quality picture, indicating that the human face target in the picture to be detected meets the requirement of performing subsequent operation or flow. Subsequent operations or procedures can be performed. Specifically, in the embodiment of the invention, the high-quality picture to be detected can be used for identifying the identity of personnel, and an alarm is given according to various identified identities.
Therefore, the method for screening a face picture provided by the embodiment of the invention further comprises the following steps:
and 105, when the picture to be detected is a high-quality picture, inputting the face model into a preset database, and comparing the face model with each preset face model stored in the preset database.
The preset database is a database which stores a large amount of personnel information and contains a large amount of preset face models, and each preset face model corresponds to one piece of identity information. The preset face model is a model established by collecting face information of people in an earlier stage, for example, for an application scene of a school, the face information of people in the school can be collected in advance, respective face models are generated and stored in a preset database, and the preset database can be used as a face model database of the people in the school.
After the face model corresponding to the face target in the picture to be detected is obtained, the face model can be input into a preset database and then compared with each preset face model in the preset database, so that whether the preset face model matched with the face model exists in the preset database or not can be determined.
Specifically, a deep learning algorithm mode can be adopted for comparison, the similarity between the face model and each preset face model in the preset database can be obtained through the deep learning algorithm, and when the similarity between the face model and one preset face model is greater than a preset threshold value, the face model can be considered to be matched with the preset face model. Of course, in the embodiment of the present invention, the comparison between the face models may be performed in other manners in the prior art, and may be selected according to the needs and applied to the embodiment of the present invention.
When the preset database is a blacklist database, and a preset face model matched with the face model exists in the preset database, it indicates that a face target corresponding to the face model is matched with the identity information in the blacklist database, so that alarm information can be sent for the obtained face target in the picture to be detected, for example, an alarm is sent in a monitoring system, or an alarm mail or a short message is sent to related personnel.
Alternatively, when the preset database is a white list database, for example, in the above-mentioned example, the preset database is a face model database of persons in the school. If the preset face model matched with the face model does not exist in the preset database, the fact that the face target in the acquired picture to be detected is a person in the non-school is shown, and therefore strangers can be triggered to give an alarm. And informing the relevant personnel that the non-campus internal personnel are present in the monitoring range in time.
In the embodiment of the invention, the picture to be detected is screened firstly, the picture to be detected with higher picture quality is selected, and comparison and identity recognition are carried out in the preset database. Therefore, the accuracy of comparison and identity recognition is improved, and the problem of low operation efficiency caused by comparison and recognition of all pictures to be detected in the prior art is solved. In addition, the problem of inaccurate comparison and identity recognition caused by selecting the pictures to be compared only through definition is solved, and the false alarm rate of alarming is reduced.
Meanwhile, in the embodiment of the invention, the screening scale or the screening standard of the picture to be detected can be adjusted by configuring different preset threshold values, so that the picture to be detected can be flexibly screened according to different application scenes. For example, when the missing report rate needs to be reduced, the preset threshold value can be set more loosely, so that more pictures to be detected can be screened, and subsequent comparison or identification steps can be performed, thereby reducing the missing report rate; when the false alarm rate needs to be reduced, the preset threshold value can be set strictly, so that the quality of the screened to-be-detected picture is further improved, subsequent comparison or identification is carried out more accurately, and the false alarm rate is reduced.
In practical application, in order to more accurately judge the category of the picture to be detected through the characteristic values, the used characteristic values can be multiple, that is, the multiple characteristic values are compared with multiple corresponding preset threshold values, so that the picture to be detected can be more accurately measured or evaluated, and the picture to be detected can be more accurately screened and determined to be a high-quality picture or a common-quality picture.
Therefore, in an embodiment of the present invention, the characteristic values may include the binocular pupil distance, the pitch angle, and the left and right side face angles. Accordingly, the preset threshold includes: a binocular pupil distance threshold, a pitch angle threshold and left and right side face angle thresholds.
The binocular pupil distance is the distance between two eyes in the human face target, changes according to the distance between the electronic equipment and the shot human face target, and is smaller when the distance between the electronic equipment and the human face target is farther. Therefore, in order to make the human face target in the picture to be detected clearer, the binocular pupil distance is larger in a certain range, and the binocular pupil distance is better. For example, the binocular pupillary distance may generally range from 18 to 571 pixel units, and a larger binocular pupillary distance in this range indicates a clearer human face object.
The pitch angle refers to the angle of the head of a shot human face target rising or falling, and in order to enable the human face target to be clearer and easier to identify, the shielding degree is smaller, and the smaller the pitch angle is, the better the pitch angle is. When the pitch angle is 0, it indicates that the face target is not lowered or raised. The pitch angle can generally range from-90 degrees to 90 degrees, where positive and negative denote head up or head down, respectively.
The left and right side face angles refer to angles that a face object to be photographed turns around to the left or right, and similarly, the smaller the left and right side face angles, the better. When the left side face angle and the right side face angle are smaller, the face target is closer to the front, so that the face features of the face target are displayed more clearly and are shielded to a smaller extent. The left and right side face angles may generally range from-90 to 90 degrees, where positive and negative respectively denote left or right rotation.
In the embodiment of the invention, the position angle and the shielding condition of the face target in the picture to be detected can be reflected more accurately through the three characteristic values. Therefore, the high-quality pictures meeting the requirements can be screened out more accurately through the three characteristic values and the corresponding preset threshold values.
Specifically, when the characteristic value may include a binocular pupil distance, a pitch angle, and left and right side face angles, in the method for screening a face picture provided in the embodiment of the present invention, step 103, compares the value between the characteristic value and a preset threshold, and according to a comparison result, the pictures to be detected are classified into different categories, which may include:
and 103a, when the binocular pupil distance is larger than the binocular pupil distance threshold value, the pitch angle is smaller than the pitch angle threshold value, and the left and right side face angles are smaller than the left and right side face angle threshold values, the picture to be detected is a high-quality picture.
And 103b, otherwise, the picture to be detected is a common quality picture.
In order to obtain a picture with higher quality, in an implementation manner of the embodiment of the present invention, when the three feature values uniformly meet the corresponding preset threshold, the picture to be detected where the face target is located may be regarded as a picture with high quality.
For example, the face target in the picture to be detected is compared with a three-dimensional reference face model in a preset three-dimensional reference database, and the characteristic values of the face target are determined as follows: the pupil distance of the eyes is 50, the pitch angle is 20, and the left and right angles are 30.
Meanwhile, the preset thresholds are respectively as follows: the binocular pupillary distance threshold is 18, the pitch angle threshold is 45 and the left and right side face angle thresholds 45.
Through comparison, the binocular pupil distance of the human face target in the picture to be detected is larger than the binocular pupil distance threshold value, the pitch angle is smaller than the pitch angle threshold value, and the left side face angle and the right side face angle are smaller than the left side face angle threshold value and the right side face angle threshold value, so that the picture to be detected is a high-quality picture.
In another implementation manner of the embodiment of the present invention, when any two feature values of the three feature values satisfy corresponding preset threshold values, the picture to be detected may also be divided into categories of high-quality pictures. Therefore, the method avoids the excessively low passing rate of screening, and is more suitable for screening the face pictures under the condition of low requirement on the quality of the pictures.
By combining the above embodiments, in practical application, in order to more flexibly adapt to various different application scenarios, when the picture to be detected is screened, the picture to be detected can be screened into multiple grades, so as to meet different service requirements.
Therefore, in the embodiment of the present invention, when the pictures to be detected are screened, besides two categories, i.e., a high-quality picture and a normal-quality picture, a category of a medium-quality picture may also be included. And, the binocular pupil distance threshold of the preset thresholds may include a first binocular pupil distance threshold and a second binocular pupil distance threshold, the pitch angle threshold may include a first pitch angle threshold and a second pitch angle threshold, and the left and right side face angle thresholds may include a first left and right side face angle threshold and a second left and right side face angle threshold.
Wherein the first binocular interpupillary distance threshold is greater than the second binocular interpupillary distance threshold; the first pitch angle threshold is less than the second pitch angle threshold; the first left-right side face angle threshold is less than the second left-right side face angle threshold.
Correspondingly, in the method for screening a face picture provided in the embodiment of the present invention, step 103, comparing the values between the characteristic values and the preset threshold values, and classifying the pictures to be detected into different categories according to the comparison result, which may include:
and 103c, when the binocular pupil distance is larger than a first binocular pupil distance threshold value, the pitch angle is smaller than a first pitch angle threshold value, and the left and right side face angles are smaller than a first left and right side face angle threshold value, the picture to be detected is a high-quality picture.
And 103d, when the binocular pupil distance is smaller than a first binocular pupil distance threshold value and is larger than a second binocular pupil distance threshold value, the pitch angle is larger than a first pitch angle threshold value and is smaller than a second pitch angle threshold value, and the left and right side face angles are larger than a first left and right side face angle threshold value and are smaller than a second left and right side face angle threshold value, the picture to be detected is a medium-quality picture.
And step 103e, if not, the picture to be detected is a common quality picture.
Since the first binocular interpupillary distance threshold is greater than the second binocular interpupillary distance threshold; the first pitch angle threshold is less than the second pitch angle threshold; the first left-right side face angle threshold is less than the second left-right side face angle threshold.
Therefore, when each feature value corresponding to the face target respectively meets the measurement standard established by the first binocular pupil distance threshold, the first pitch angle threshold and the first left and right side face angle thresholds, it indicates that the face target is closer to the motor device for shooting, and the deflection of the position angle and the like of the face is smaller, and the face belongs to a high-quality picture.
When each feature value corresponding to the human face target does not accord with the measurement standard constructed by the first binocular pupil distance threshold, the first pitch angle threshold and the first left and right side face angle threshold, but accords with the measurement standard constructed by the second binocular pupil distance threshold, the second pitch angle threshold and the second left and right side face angle threshold, it indicates that the human face target is not high in quality, but can be applied to a scene with a low picture quality requirement, so that the picture to be detected can be a picture with medium quality.
If the characteristic values corresponding to the face target do not meet the measurement standard established by the second binocular pupil distance threshold, the second pitch angle threshold and the second left and right side face angle threshold, the image quality of the image to be detected where the face target is located is low, the image is of ordinary quality, subsequent operation or flow can not be carried out on the image, and the problem of false alarm caused by inaccurate recognition can be avoided.
Certainly, in practical application, the preset threshold values of a plurality of gears can be set according to needs, so that more detailed screening of the pictures to be detected is realized, the screening of the pictures to be detected is more accurate, and the needs of different application scenes or different services are met.
By combining the above embodiments, in practical application, when the face target in the picture to be detected is compared with the three-dimensional reference face model in the preset three-dimensional reference database, the comparison efficiency is further improved, so that the comparison process is more accurate. In the method for screening a face picture provided in the embodiment of the present invention, step 102, comparing a face target in a picture to be detected with a three-dimensional reference face model in a preset three-dimensional reference database to determine a feature value of the face target, may further include:
firstly, recognizing a human face target in a picture to be detected.
Before comparing with the three-dimensional reference face model in the preset three-dimensional reference database, the face target can be determined from the picture to be detected by various face detection methods, such as a neural network algorithm, a principal component analysis algorithm, an independent component analysis algorithm, a singular value feature-based method and the like. Specifically, the determining of the face target may be determining an image area where the face target is located, and identifying the face target by a frame selection method.
Or after the region where the face target is located is determined, the face target can be extracted from the picture to be detected through modes such as matting and the like.
And secondly, comparing the face target with a three-dimensional reference face model in a preset three-dimensional reference database to determine the characteristic value of the face target.
After the face target is determined, the face target can be input into a preset three-dimensional reference database and compared with a three-dimensional reference face model in the preset three-dimensional reference database, and therefore the characteristic value of the face target can be determined more quickly and conveniently.
Referring to fig. 3, fig. 3 is a structural diagram of a face image screening apparatus according to an embodiment of the present invention, including:
an obtaining module 301, configured to obtain a to-be-detected picture including a face target;
a comparison module 302, configured to compare the face target in the picture to be detected with the three-dimensional reference face model in the preset three-dimensional reference database, and determine a feature value of the face target;
and the screening module 303 is configured to compare the value between the characteristic value and a preset threshold, and according to a comparison result, classify the pictures to be detected into different categories, where the categories include high-quality pictures and common-quality pictures.
In the embodiment of the invention, the human face target in the acquired picture to be detected is compared with the three-dimensional reference human face model in the preset three-dimensional reference database to obtain the characteristic value of the human face target, and then the numerical value of the characteristic value is compared with the preset threshold value, so that the picture to be detected is determined to be a high-quality picture or a common-quality picture. In the embodiment of the invention, when the picture to be detected is screened, the picture to be detected can be screened through various characteristic values instead of only considering one factor of definition, so that whether the picture to be detected meets the requirement of subsequent operation can be more accurately measured, and the picture to be detected is more accurately screened. In addition, the preset threshold value can be configured in advance, so that the screening standard of the picture to be detected can be adjusted more flexibly to adapt to application under different actual conditions or scenes.
Optionally, in the face image screening apparatus provided in the embodiment of the present invention, the comparison module 302 is specifically configured to:
respectively comparing the face target in the picture to be detected with a plurality of three-dimensional reference face models in the preset three-dimensional reference database, wherein the three-dimensional reference face models are three-dimensional reference face models with different characteristic values; determining a three-dimensional reference face model matched with the face target; and taking the characteristic value of the three-dimensional reference human face model as the characteristic value of the human face target.
Optionally, in the face image screening apparatus provided in the embodiment of the present invention, the characteristic values in the apparatus include a binocular pupil distance, a pitch angle, and left and right side face angles;
the preset threshold comprises a binocular pupil distance threshold, a pitch angle threshold and left and right side face angle thresholds.
Optionally, in the face image screening apparatus provided in the embodiment of the present invention, the screening module 303 is specifically configured to:
when the binocular pupil distance is larger than the binocular pupil distance threshold value, the pitch angle is smaller than the pitch angle threshold value, and the left and right side face angles are smaller than the left and right side face angle threshold values, the picture to be detected is a high-quality picture; otherwise, the picture to be detected is a common quality picture.
Optionally, in the face image screening apparatus provided in the embodiment of the present invention, the binocular pupil distance threshold in the comparison module 302 includes a first binocular pupil distance threshold and a second binocular pupil distance threshold, the pitch angle threshold includes a first pitch angle threshold and a second pitch angle threshold, the left and right side face angle thresholds include a first left and right side face angle threshold and a second left and right side face angle threshold, and the category in the screening module 303 further includes a medium quality image;
the screening module 303 is specifically configured to:
when the binocular pupil distance is larger than the first binocular pupil distance threshold value, the pitch angle is smaller than the first pitch angle threshold value, and the left and right side face angles are smaller than the first left and right side face angle threshold value, the picture to be detected is a high-quality picture;
when the binocular pupil distance is smaller than the first binocular pupil distance threshold and larger than the second binocular pupil distance threshold, the pitch angle is larger than the first pitch angle threshold and smaller than the second pitch angle threshold, and the left and right side face angles are larger than the first left and right side face angle threshold and smaller than the second left and right side face angle threshold, the picture to be detected is a medium quality picture;
otherwise, the picture to be detected is a common quality picture.
Optionally, in the face image screening apparatus provided in the embodiment of the present invention, the comparison module 302 is specifically configured to:
identifying the human face target in the picture to be detected;
and comparing the face target with a three-dimensional reference face model in the preset three-dimensional reference database to determine the characteristic value of the face target.
Optionally, in the apparatus for screening a face picture provided in the embodiment of the present invention, the apparatus further includes:
the modeling module is used for establishing a face model corresponding to the face target in the picture to be detected;
the analysis module is used for inputting the face model into a preset database and comparing the face model with each preset face model stored in the preset database when the picture to be detected is a high-quality picture;
the alarm module is used for outputting alarm information when a preset face model matched with the face model exists in the preset database; or when the preset face model matched with the face model does not exist in the preset database, alarm information is output.
An embodiment of the present invention further provides an electronic device, as shown in fig. 4, including a processor 401, a communication interface 402, a memory 403, and a communication bus 404, where the processor 401, the communication interface 402, and the memory 403 complete mutual communication through the communication bus 404,
a memory 403 for storing a computer program;
the processor 401, when executing the program stored in the memory 403, implements the following steps:
acquiring a picture to be detected containing a human face target;
comparing the face target in the picture to be detected with the three-dimensional reference face model in the preset three-dimensional reference database to determine a characteristic value of the face target;
and comparing the numerical value between the characteristic value and a preset threshold value, and classifying the pictures to be detected into different categories according to the comparison result, wherein the categories comprise high-quality pictures and common-quality pictures.
The communication bus mentioned in the electronic device may be a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the electronic equipment and other equipment.
The Memory may include a RAM (Random Access Memory) or an NVM (Non-Volatile Memory), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component.
In another embodiment of the present invention, a computer-readable storage medium is further provided, where instructions are stored in the computer-readable storage medium, and when the instructions are run on a computer, the computer is caused to execute the method for screening a face picture as described in any of the above embodiments.
In another embodiment of the present invention, a computer program product containing instructions is further provided, which when run on a computer, causes the computer to execute the method for screening a face picture as described in any of the above embodiments.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for embodiments such as the apparatus, since they are substantially similar to the method embodiments, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiments.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (8)
1. A method for screening face pictures is characterized by comprising the following steps:
acquiring a picture to be detected containing a human face target;
comparing the face target in the picture to be detected with a three-dimensional reference face model in a preset three-dimensional reference database to determine a characteristic value of the face target;
comparing the numerical value between the characteristic value and a preset threshold value, and classifying the pictures to be detected into different categories according to the comparison result, wherein the categories comprise high-quality pictures and common-quality pictures;
the step of comparing the face target in the picture to be detected with the three-dimensional reference face model in the preset three-dimensional reference database to determine the characteristic value of the face target includes:
establishing a face model corresponding to the face target in the picture to be detected, and respectively comparing the established face model with a plurality of three-dimensional reference face models in the preset three-dimensional reference database, wherein the three-dimensional reference face models are three-dimensional reference face models with different characteristic values; determining a three-dimensional reference face model matched with the face target; taking the characteristic value of the three-dimensional reference human face model as the characteristic value of the human face target;
the characteristic values comprise binocular pupil distance, a pitching angle and left and right side face angles; the preset threshold comprises a binocular pupil distance threshold, a pitch angle threshold and left and right side face angle thresholds;
the comparison of the numerical values between the characteristic values and the preset threshold values and the classification of the pictures to be detected into different categories according to the comparison results comprise: when the binocular pupil distance is larger than the binocular pupil distance threshold value, the pitch angle is smaller than the pitch angle threshold value, and the left and right side face angles are smaller than the left and right side face angle threshold values, the picture to be detected is a high-quality picture; otherwise, the picture to be detected is a common quality picture;
alternatively, the first and second electrodes may be,
the binocular pupil distance threshold comprises a first binocular pupil distance threshold and a second binocular pupil distance threshold, the pitch angle threshold comprises a first pitch angle threshold and a second pitch angle threshold, the left and right side face angle thresholds comprise a first left and right side face angle threshold and a second left and right side face angle threshold, and the category further comprises medium-quality pictures; the comparison of the numerical values between the characteristic values and the preset threshold values and the classification of the pictures to be detected into different categories according to the comparison results comprise: when the binocular pupil distance is larger than the first binocular pupil distance threshold value, the pitch angle is smaller than the first pitch angle threshold value, and the left and right side face angles are smaller than the first left and right side face angle threshold value, the picture to be detected is a high-quality picture; when the binocular pupil distance is smaller than the first binocular pupil distance threshold and larger than the second binocular pupil distance threshold, the pitch angle is larger than the first pitch angle threshold and smaller than the second pitch angle threshold, and the left and right side face angles are larger than the first left and right side face angle threshold and smaller than the second left and right side face angle threshold, the picture to be detected is a medium quality picture; otherwise, the picture to be detected is a common quality picture.
2. The method according to claim 1, wherein the comparing the face target in the picture to be detected with the three-dimensional reference face model in the preset three-dimensional reference database to determine the feature value of the face target comprises:
identifying the human face target in the picture to be detected;
and comparing the face target with a three-dimensional reference face model in the preset three-dimensional reference database to determine the characteristic value of the face target.
3. The method according to any one of claims 1 to 2, further comprising:
establishing a face model corresponding to the face target in the picture to be detected;
when the picture to be detected is a high-quality picture, inputting the face model into a preset database, and comparing the face model with each preset face model stored in the preset database;
when a preset face model matched with the face model exists in the preset database, outputting alarm information;
or when the preset face model matched with the face model does not exist in the preset database, alarm information is output.
4. The utility model provides a people's face picture sieving mechanism which characterized in that includes:
the acquisition module is used for acquiring a picture to be detected containing a human face target;
the comparison module is used for comparing the face target in the picture to be detected with a three-dimensional reference face model in a preset three-dimensional reference database to determine a characteristic value of the face target;
the screening module is used for comparing the numerical value between the characteristic value and a preset threshold value, and classifying the pictures to be detected into different categories according to the comparison result, wherein the categories comprise high-quality pictures and common-quality pictures;
wherein, the comparison module is specifically configured to: establishing a face model corresponding to the face target in the picture to be detected, and respectively comparing the established face model with a plurality of three-dimensional reference face models in the preset three-dimensional reference database, wherein the three-dimensional reference face models are three-dimensional reference face models with different characteristic values; determining a three-dimensional reference face model matched with the face target; taking the characteristic value of the three-dimensional reference human face model as the characteristic value of the human face target;
wherein the characteristic values in the device include a binocular pupillary distance, a pitch angle, and left and right side face angles; the preset threshold comprises a binocular pupil distance threshold, a pitch angle threshold and left and right side face angle thresholds;
the screening module is specifically configured to: when the binocular pupil distance is larger than the binocular pupil distance threshold value, the pitch angle is smaller than the pitch angle threshold value, and the left and right side face angles are smaller than the left and right side face angle threshold values, the picture to be detected is a high-quality picture; otherwise, the picture to be detected is a common quality picture;
alternatively, the first and second electrodes may be,
the binocular pupil distance threshold in the comparison module comprises a first binocular pupil distance threshold and a second binocular pupil distance threshold, the pitch angle threshold comprises a first pitch angle threshold and a second pitch angle threshold, the left and right side face angle thresholds comprise a first left and right side face angle threshold and a second left and right side face angle threshold, and the category in the screening module further comprises medium-quality pictures; the screening module is specifically configured to: when the binocular pupil distance is larger than the first binocular pupil distance threshold value, the pitch angle is smaller than the first pitch angle threshold value, and the left and right side face angles are smaller than the first left and right side face angle threshold value, the picture to be detected is a high-quality picture; when the binocular pupil distance is smaller than the first binocular pupil distance threshold and larger than the second binocular pupil distance threshold, the pitch angle is larger than the first pitch angle threshold and smaller than the second pitch angle threshold, and the left and right side face angles are larger than the first left and right side face angle threshold and smaller than the second left and right side face angle threshold, the picture to be detected is a medium quality picture; otherwise, the picture to be detected is a common quality picture.
5. The apparatus of claim 4, wherein the comparison module is specifically configured to:
identifying the human face target in the picture to be detected;
and comparing the face target with a three-dimensional reference face model in the preset three-dimensional reference database to determine the characteristic value of the face target.
6. The apparatus of any of claims 4 to 5, further comprising:
the modeling module is used for establishing a face model corresponding to the face target in the picture to be detected;
the analysis module is used for inputting the face model into a preset database and comparing the face model with each preset face model stored in the preset database when the picture to be detected is a high-quality picture;
the alarm module is used for outputting alarm information when a preset face model matched with the face model exists in the preset database; or when the preset face model matched with the face model does not exist in the preset database, alarm information is output.
7. An electronic device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing mutual communication by the memory through the communication bus;
a memory for storing a computer program;
a processor for implementing the method steps of any of claims 1 to 3 when executing a program stored in the memory.
8. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of the claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810360907.7A CN110390229B (en) | 2018-04-20 | 2018-04-20 | Face picture screening method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810360907.7A CN110390229B (en) | 2018-04-20 | 2018-04-20 | Face picture screening method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110390229A CN110390229A (en) | 2019-10-29 |
| CN110390229B true CN110390229B (en) | 2022-03-04 |
Family
ID=68283691
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810360907.7A Active CN110390229B (en) | 2018-04-20 | 2018-04-20 | Face picture screening method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110390229B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111178162B (en) * | 2019-12-12 | 2023-11-07 | 北京迈格威科技有限公司 | Image recognition method, device, computer equipment and storage medium |
| CN111046806A (en) * | 2019-12-12 | 2020-04-21 | 天地伟业技术有限公司 | Heterogeneous image face recognition target library generation method |
| CN113095110B (en) * | 2019-12-23 | 2024-03-08 | 浙江宇视科技有限公司 | Method, device, medium and electronic equipment for dynamically warehousing face data |
| CN111159445A (en) * | 2019-12-30 | 2020-05-15 | 深圳云天励飞技术有限公司 | Picture filtering method and device, electronic equipment and storage medium |
| EP4124029A4 (en) * | 2020-03-17 | 2023-04-05 | NEC Corporation | Gate system, gate device, image processing method therefor, program, and arrangement method for gate device |
| CN111640207A (en) * | 2020-06-04 | 2020-09-08 | 福建票付通信息科技有限公司 | Non-inductive face collection ticket checking method |
| CN112220450B (en) * | 2020-08-21 | 2023-08-15 | 上海交通大学医学院附属第九人民医院 | Orbital disease screening method, system and terminal based on three-dimensional model |
| CN112580434B (en) * | 2020-11-25 | 2024-03-15 | 奥比中光科技集团股份有限公司 | Face false detection optimization method and system based on depth camera and face detection equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799877A (en) * | 2012-09-11 | 2012-11-28 | 上海中原电子技术工程有限公司 | Method and system for screening face images |
| CN103927515A (en) * | 2014-04-08 | 2014-07-16 | 章建国 | Three-dimension dynamic facial comparison method |
| CN104657817A (en) * | 2015-01-28 | 2015-05-27 | 四川君逸易视科技有限公司 | Face snapshotting, comparing, identifying, retrieving, and inquiring method for bank counter |
| CN105243357A (en) * | 2015-09-15 | 2016-01-13 | 深圳市环阳通信息技术有限公司 | Identity document-based face recognition method and face recognition device |
| CN105631439A (en) * | 2016-02-18 | 2016-06-01 | 北京旷视科技有限公司 | Human face image collection method and device |
| CN106557723A (en) * | 2015-09-25 | 2017-04-05 | 北京市商汤科技开发有限公司 | A kind of system for face identity authentication with interactive In vivo detection and its method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| PT106430B (en) * | 2012-07-03 | 2018-08-07 | Cesar Augusto Dos Santos Silva | INTERPUPILARY DISTANCE MEASUREMENT SYSTEM USING A SCREEN AND CAMERA DEVICE |
| CN103745208A (en) * | 2014-01-27 | 2014-04-23 | 中国科学院深圳先进技术研究院 | Face recognition method and device |
| CN104463117B (en) * | 2014-12-02 | 2018-07-03 | 苏州科达科技股份有限公司 | A kind of recognition of face sample collection method and system based on video mode |
| CN107122054A (en) * | 2017-04-27 | 2017-09-01 | 青岛海信医疗设备股份有限公司 | A kind of detection method and device of face deflection angle and luffing angle |
| CN107277265A (en) * | 2017-07-13 | 2017-10-20 | 广东欧珀移动通信有限公司 | Solve lock control method and Related product |
-
2018
- 2018-04-20 CN CN201810360907.7A patent/CN110390229B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799877A (en) * | 2012-09-11 | 2012-11-28 | 上海中原电子技术工程有限公司 | Method and system for screening face images |
| CN103927515A (en) * | 2014-04-08 | 2014-07-16 | 章建国 | Three-dimension dynamic facial comparison method |
| CN104657817A (en) * | 2015-01-28 | 2015-05-27 | 四川君逸易视科技有限公司 | Face snapshotting, comparing, identifying, retrieving, and inquiring method for bank counter |
| CN105243357A (en) * | 2015-09-15 | 2016-01-13 | 深圳市环阳通信息技术有限公司 | Identity document-based face recognition method and face recognition device |
| CN106557723A (en) * | 2015-09-25 | 2017-04-05 | 北京市商汤科技开发有限公司 | A kind of system for face identity authentication with interactive In vivo detection and its method |
| CN105631439A (en) * | 2016-02-18 | 2016-06-01 | 北京旷视科技有限公司 | Human face image collection method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110390229A (en) | 2019-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110390229B (en) | Face picture screening method and device, electronic equipment and storage medium | |
| US9875392B2 (en) | System and method for face capture and matching | |
| CN109670441A (en) | A kind of realization safety cap wearing knows method for distinguishing, system, terminal and computer readable storage medium | |
| CN110659397B (en) | Behavior detection method and device, electronic equipment and storage medium | |
| CN105956572A (en) | In vivo face detection method based on convolutional neural network | |
| CN112381075B (en) | Method and system for carrying out face recognition under specific scene of machine room | |
| CN105631439A (en) | Human face image collection method and device | |
| CN111210399B (en) | Imaging quality evaluation method, device and equipment | |
| US20150262068A1 (en) | Event detection apparatus and event detection method | |
| US20200242336A1 (en) | Method for selecting images in video of faces in the wild | |
| CN111414858B (en) | Face recognition method, target image determining device and electronic system | |
| CN111079694A (en) | Counter assistant job function monitoring device and method | |
| CN111738120A (en) | Person identification method, person identification device, electronic equipment and storage medium | |
| CN111444555B (en) | Temperature measurement information display method and device and terminal equipment | |
| CN108171135A (en) | Method for detecting human face, device and computer readable storage medium | |
| CN111415339A (en) | Image defect detection method for complex texture industrial product | |
| CN113850183A (en) | Method for judging behaviors in video based on artificial intelligence technology | |
| CN116863297A (en) | Monitoring method, device, system, equipment and medium based on electronic fence | |
| CN111325181A (en) | State monitoring method and device, electronic equipment and storage medium | |
| CN113869115A (en) | Method and system for processing face image | |
| JP6893812B2 (en) | Object detector | |
| CN115719428A (en) | Face image clustering method, device, equipment and medium based on classification model | |
| CN114758384A (en) | Face detection method, device, equipment and storage medium | |
| JP2019066909A (en) | Object distribution estimation apparatus | |
| KR20210031444A (en) | Method and Apparatus for Creating Labeling Model with Data Programming |
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 |