CN117935341A - Automatic sign-in method based on face recognition - Google Patents
Automatic sign-in method based on face recognition Download PDFInfo
- Publication number
- CN117935341A CN117935341A CN202410324872.7A CN202410324872A CN117935341A CN 117935341 A CN117935341 A CN 117935341A CN 202410324872 A CN202410324872 A CN 202410324872A CN 117935341 A CN117935341 A CN 117935341A
- Authority
- CN
- China
- Prior art keywords
- face
- recognition
- triangular
- face recognition
- characteristic points
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000001815 facial effect Effects 0.000 claims description 5
- 230000001360 synchronised effect Effects 0.000 claims description 5
- 230000015572 biosynthetic process Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/109—Time management, e.g. calendars, reminders, meetings or time accounting
- G06Q10/1091—Recording time for administrative or management purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/165—Detection; Localisation; Normalisation using facial parts and geometric relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—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
Landscapes
- Engineering & Computer Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Collating Specific Patterns (AREA)
- Quality & Reliability (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Geometry (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Tourism & Hospitality (AREA)
Abstract
The invention relates to the technical field of electronic sign-in, in particular to a face recognition-based automatic sign-in method, which comprises the steps of inputting information of a person to be identified, storing a triangular relationship constructed by nose tip feature points, left shoulder feature points and right shoulder feature points of the person to be identified into a first identification library, and putting names and face images of the person to be identified into a second identification library; extracting the triangular relations of nose tip feature points, left shoulder feature points and right shoulder feature points of a human face in the image, comparing the triangular relations with the triangular relations in the first recognition library, if the comparison is successful, then carrying out recognition comparison on the corresponding face image, and if the comparison is successful, signing in; according to the invention, by utilizing the characteristic that the width and the distance between shoulders are large enough and easy to identify, and combining with the formation of a triangular relationship at the tip of the nose, the efficiency of calibrating the feature points is improved; and combining a triangle relationship to possibly correspond to a plurality of face information, so that all information in the whole second recognition library is not required to be compared, and the accuracy and the speed of recognition are further improved.
Description
Technical Field
The invention relates to the technical field of electronic sign-in, in particular to an automatic sign-in method based on face recognition.
Background
With the application and popularization of various electronic products, electronic check-in is widely applied in the fields of meeting check-in, staff check-in, students check-in class and the like.
Face recognition is a biological recognition technology for carrying out identity recognition based on facial feature information of people. A series of related technologies, commonly referred to as image recognition and face recognition, are used to capture images or video streams containing faces with a camera or cameras, and automatically detect and track the faces in the images, and then perform face recognition on the detected faces.
In open school such as universities and colleges, teacher can roll call to confirm the attendance condition of student in the class, and traditional roll call mode has both wasted the time of having class, and many take the student of roll call let the teacher unable correct statistics student's attendance, especially the great class of student. The classroom sign-in machine in the prior art has single structure and single function; when students concentrate, the general classroom check-in machine can not meet the use requirement, and the check-in speed is too slow, so that many students can not finish check-in before class to influence the attendance rate; the reason for the too slow sign-in speed comprises that the recognition accuracy of the camera is not high, students are required to recognize in a fixed interval, 68 characteristic points are adopted for positioning recognition in the existing face recognition, 68 points are concentrated on the face, so that the recognition is difficult, meanwhile, the requirement on the accuracy of the camera is high, if the camera is not clear, the camera is required to be re-recognized, and the recognition speed is slow; the shoulders are shot together by the existing camera, and the face recognition is carried out after the shoulders are removed, so that the recognition speed is low; a new recognition method is required to increase the recognition speed.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: a face recognition-based automatic sign-in method for positioning and identifying by re-planning feature points is provided.
In order to solve the technical problems, the invention adopts the following technical scheme: a face recognition based automatic sign-in method, comprising:
Establishing a recognition library, and inputting personnel information to be recognized, wherein the personnel information comprises a name and a portrait, the portrait comprises a face and shoulders, the face is provided with a plurality of characteristic points meeting the face recognition, the characteristic points comprise nose tip characteristic points, and the two outermost sides of the shoulders are provided with left shoulder characteristic points and right shoulder characteristic points;
the method comprises the steps of storing a triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the identified person into a first identification library, putting the name and the face image of the identified person into a second identification library, and simultaneously establishing a connection for the triangular relationship, the face image and the name; each triangle relation corresponds to a plurality of facial image information;
Face recognition, namely capturing a video frame captured in a field dynamic manner and detecting a face in the video in real time, if no face exists, not acting, if the face exists, judging whether an image of the video frame contains identifiable left shoulder characteristic points and right shoulder characteristic points, if not, directly extracting the characteristic points of the face to be compared with the face in a second recognition library for recognition, if the comparison is successful, extracting the name, otherwise, carrying out face recognition again; if the identifiable left shoulder characteristic points and right shoulder characteristic points exist, triangular relations constructed by the nose tip characteristic points, the left shoulder characteristic points and the right shoulder characteristic points of the human face in the image are extracted and compared with those in the first recognition library, if the comparison is successful, a plurality of face images in the second recognition library corresponding to the triangular relations are recognized and compared, if the comparison is successful, names are extracted, and otherwise, the human face recognition is performed again;
and automatically checking in, displaying the successful check-in of the personnel with the extracted name, and forming a check-in record.
Preferably, the triangle relation is a triangle area or an included angle formed by the nose tip characteristic point, the left shoulder characteristic point and the right shoulder characteristic point.
Preferably, when the triangular relationship is a triangular area, the step of "comparing the triangular relationship constructed by extracting the nose tip feature point, the left shoulder feature point and the right shoulder feature point of the face in the image with the triangular relationship in the first recognition library" in the face recognition further includes:
And extracting the triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the human face in the image, comparing the triangular relationship with the triangular relationship in the first recognition library in a region with the triangular area of +/-5%, and if the area value formed by the triangular relationship in the first recognition library is in the region, judging that the comparison is successful.
Preferably, the triangle relationship is an included angle formed by the nose tip feature point, the left shoulder feature point and the right shoulder feature point, and the step of extracting the triangle relationship constructed by the nose tip feature point, the left shoulder feature point and the right shoulder feature point of the face in the image and comparing the triangle relationship with the triangle relationship in the first recognition library in the face recognition further comprises:
And extracting the triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the human face in the image, comparing the triangular relationship with the triangular relationship in the first recognition library in a section with an included angle of +/-2 degrees, and if the area value formed by the triangular relationship in the first recognition library is in the section, judging that the comparison is successful.
From the above description, the triangular area + -5% interval and the included angle + -2 DEG interval are empirical values obtained by long-term test, and the accuracy of judgment is highest in both cases.
Preferably, "capturing video frames captured dynamically in the scene and detecting faces in the video in real time" further includes:
Capturing video frames captured in the field dynamically, detecting N faces of the faces in the video in real time, which are obtained from the video frames, performing face shape recognition, selecting a plurality of photos with face shapes, and performing synchronous face recognition.
Preferably, in the face recognition, if a plurality of faces appear in a video frame, the face recognition is performed according to the order of the image occupation ratio from large to small.
Preferably, in the process of sorting and performing face recognition, if there are video frames dynamically captured in the field and faces in the video are detected in real time, the faces detected in real time are subjected to face recognition preferentially.
Preferably, after extracting the name, judging whether the name is in the sign-in record, if so, ending, and if not, continuing.
Preferably, the number of times of re-face recognition is one, the second time is considered to be sign-in failure, and the images of the video frames are synchronously stored, and the staff waits for verifying the staff identity.
Preferably, the check-in record automatically counts the number of check-in successes while recording the number of check-in failures, and the total number of check-in successes and check-in failures.
The invention has the beneficial effects that: by utilizing the characteristic that the width and the distance between shoulders are large enough and easy to identify, combining with the characteristic that the nose tip is in a triangular relationship, the three characteristic points are obvious and are easy to extract in an image, the efficiency of calibrating the characteristic points can be improved, the traditional face recognition is that the characteristic points are identified one by one after all the characteristic points are carried out on the image, even the condition that the calibration characteristic points fail and then the identification fails due to insufficient resolution of a camera can occur, and the three characteristic points are not easy to be influenced by expression change due to the fact that the positions of the three characteristic points are relatively fixed; and then combining a triangle relation to possibly correspond to a plurality of face information, so that the number of the face information which is compared in face recognition is only a few, and all the information in the whole second recognition library is not required to be compared, thereby improving the accuracy and the speed of recognition and realizing the improvement of the speed of automatic sign-in; meanwhile, the captured video frame images can be directly compared with the first recognition library and the second recognition library, recognition can be performed as long as one condition can be met, the requirement that shoulders need to be additionally removed in the prior art is overcome, and the speed of recognition can be improved due to the fact that the shoulder is removed in a saving mode, and then the automatic sign-in speed is further improved.
Drawings
Fig. 1 is a basic logic diagram of a method for automatically signing in based on face recognition according to an embodiment of the present invention;
Fig. 2 is a schematic diagram of feature point distribution and recognition modes of a method for automatic sign-in based on face recognition according to a first embodiment of the present invention;
Fig. 3 is a schematic diagram of feature point distribution and recognition modes of a method for automatic sign-in based on face recognition according to a second embodiment of the present invention.
Detailed Description
In order to describe the technical contents, the achieved objects and effects of the present invention in detail, the following description will be made with reference to the embodiments in conjunction with the accompanying drawings.
It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region.
Referring to fig. 1 to 3, a method for automatically signing in based on face recognition includes:
Establishing a recognition library, and inputting personnel information to be recognized, wherein the personnel information comprises a name and a portrait, the portrait comprises a face and shoulders, the face is provided with a plurality of characteristic points meeting the face recognition, the characteristic points comprise nose tip characteristic points, and the two outermost sides of the shoulders are provided with left shoulder characteristic points and right shoulder characteristic points;
the method comprises the steps of storing a triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the identified person into a first identification library, putting the name and the face image of the identified person into a second identification library, and simultaneously establishing a connection for the triangular relationship, the face image and the name; each triangle relation corresponds to a plurality of facial image information;
Face recognition, namely capturing a video frame captured in a field dynamic manner and detecting a face in the video in real time, if no face exists, not acting, if the face exists, judging whether an image of the video frame contains identifiable left shoulder characteristic points and right shoulder characteristic points, if not, directly extracting the characteristic points of the face to be compared with the face in a second recognition library for recognition, if the comparison is successful, extracting the name, otherwise, carrying out face recognition again; if the identifiable left shoulder characteristic points and right shoulder characteristic points exist, triangular relations constructed by the nose tip characteristic points, the left shoulder characteristic points and the right shoulder characteristic points of the human face in the image are extracted and compared with those in the first recognition library, if the comparison is successful, a plurality of face images in the second recognition library corresponding to the triangular relations are recognized and compared, if the comparison is successful, names are extracted, and otherwise, the human face recognition is performed again;
and automatically checking in, displaying the successful check-in of the personnel with the extracted name, and forming a check-in record.
From the above description, by utilizing the characteristic that the width and the distance between the shoulders are large enough and easy to identify, and combining the characteristic points at the tip of the nose to form a triangular relationship, the three characteristic points are obvious and are easy to extract in the image, so that the efficiency of calibrating the characteristic points can be improved, the traditional face recognition is avoided, namely, the image is identified one by one after all the characteristic points are carried out, even the condition that the calibration characteristic points fail and then the recognition fails due to insufficient resolution of a camera can occur, and the three characteristic points are not easy to be influenced by expression change due to the fact that the positions of the three characteristic points are relatively fixed; and then combining a triangle relation to possibly correspond to a plurality of face information, so that the number of the face information which is compared in face recognition is only a few, and all the information in the whole second recognition library is not required to be compared, thereby improving the accuracy and speed of recognition; meanwhile, the captured video frame images can be directly compared with the first recognition library and the second recognition library, so that recognition can be performed only by meeting one condition, the requirement that shoulders need to be additionally removed in the prior art is overcome, and the recognition speed can be improved due to the omitted removing steps.
Further, the triangular relationship is a triangular area or an included angle formed by the nose tip characteristic point, the left shoulder characteristic point and the right shoulder characteristic point.
Further, when the triangular relationship is a triangular area, in the face recognition, the step of "extracting the triangular relationship constructed by the nose tip feature point, the left shoulder feature point and the right shoulder feature point of the face in the image and comparing the triangular relationship with the triangular relationship in the first recognition library" further includes:
And extracting the triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the human face in the image, comparing the triangular relationship with the triangular relationship in the first recognition library in a region with the triangular area of +/-5%, and if the area value formed by the triangular relationship in the first recognition library is in the region, judging that the comparison is successful.
Further, the triangle relationship is an included angle formed by the nose tip feature point, the left shoulder feature point and the right shoulder feature point, and the step of extracting the triangle relationship constructed by the nose tip feature point, the left shoulder feature point and the right shoulder feature point of the face in the image and comparing the triangle relationship with the triangle relationship in the first recognition library in the face recognition further comprises:
And extracting the triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the human face in the image, comparing the triangular relationship with the triangular relationship in the first recognition library in a section with an included angle of +/-2 degrees, and if the area value formed by the triangular relationship in the first recognition library is in the section, judging that the comparison is successful.
From the above description, the triangular area + -5% interval and the included angle + -2 DEG interval are empirical values obtained by long-term test, and the accuracy of judgment is highest in both cases.
Further, "capturing video frames captured dynamically in the scene and detecting faces in the video in real time" further includes:
Capturing video frames captured in the field dynamically, detecting N faces of the faces in the video in real time, which are obtained from the video frames, performing face shape recognition, selecting a plurality of photos with face shapes, and performing synchronous face recognition.
From the above description, the synchronous recognition is performed by a plurality of faces, so that the recognition efficiency and accuracy can be improved, and the problem of unclear recognition caused by only obtaining one photo at a time is avoided.
Further, in the face recognition process, if a plurality of faces appear in the video frame, the face recognition is performed according to the order of the image occupation ratio from large to small.
From the above description, the face recognition of the recognition personnel can be facilitated by selecting the mode of screening the faces from large to small in duty ratio.
Further, in the process of sorting and performing face recognition, if a video frame dynamically grabbed on site exists and faces in the video are detected in real time, the faces detected in real time are subjected to face recognition preferentially.
From the above description, through the queuing relationship, the interval waiting time is utilized or the face recognition is performed, so that the normal face recognition process is not affected, and the efficiency can be improved.
Further, after extracting the name, judging whether the name is in the sign-in record, if yes, stopping, and if not, continuing.
From the above description, by adding a decision, the problem of repeated check-in is avoided.
Further, the number of times of re-face recognition is one, the second time is determined as a sign-in failure, and the images of the video frames are synchronously stored, and the staff waits for verifying the personnel identity.
From the above description, the human verification is performed after the second failure by only one time of face recognition failure, so that the efficiency can be improved, and the problem that the single person always fails to be repeatedly recognized and other persons cannot be recognized is avoided.
Further, the check-in record automatically calculates the number of successful check-in, and simultaneously records the number of failed check-in and the total number of successful check-in and failed check-in.
From the above description, by calculating the total number of people, the successful number of people and the failed number of people, the staff can be convenient to see clearly, and the efficiency is improved.
Example 1
A face recognition based automatic sign-in method, comprising:
Establishing a recognition library, and inputting personnel information to be recognized, wherein the personnel information comprises a name and a portrait, the portrait comprises a face and shoulders, the face is provided with a plurality of characteristic points meeting the face recognition, the characteristic points comprise nose tip characteristic points, and the two outermost sides of the shoulders are provided with left shoulder characteristic points and right shoulder characteristic points;
the method comprises the steps of storing a triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the identified person into a first identification library, putting the name and the face image of the identified person into a second identification library, and simultaneously establishing a connection for the triangular relationship, the face image and the name; each triangle relation corresponds to a plurality of facial image information;
Face recognition, namely capturing a video frame captured in a field dynamic manner and detecting a face in the video in real time, if no face exists, not acting, if the face exists, judging whether an image of the video frame contains identifiable left shoulder characteristic points and right shoulder characteristic points, if not, directly extracting the characteristic points of the face to be compared with the face in a second recognition library for recognition, if the comparison is successful, extracting the name, otherwise, carrying out face recognition again; if the identifiable left shoulder characteristic points and right shoulder characteristic points exist, triangular relations constructed by the nose tip characteristic points, the left shoulder characteristic points and the right shoulder characteristic points of the human face in the image are extracted and compared with those in the first recognition library, if the comparison is successful, a plurality of face images in the second recognition library corresponding to the triangular relations are recognized and compared, if the comparison is successful, names are extracted, and otherwise, the human face recognition is performed again;
and automatically checking in, displaying the successful check-in of the personnel with the extracted name, and forming a check-in record.
Wherein,
The "face has a plurality of feature points satisfying face recognition" further includes:
this feature is identified for 68 feature points commonly used in the prior art, see 68 points on the face of fig. 2; meanwhile, the nose tip characteristic points are 30 points, the left shoulder characteristic points and the right shoulder characteristic points can be set to be 68 points and 69 points (the existing serial numbers start from 0), and other points except the points 30, 68 and 69 are not marked in the figure; the triangular relationship (triangular area) constructed by the nasal tip feature point, the left shoulder feature point and the right shoulder feature point is shown in fig. 2.
When the triangular relationship is a triangular area, in the face recognition, the step of extracting the triangular relationship constructed by the nose tip feature point, the left shoulder feature point and the right shoulder feature point of the face in the image and comparing the triangular relationship with the triangular relationship in the first recognition library further comprises the steps of:
And extracting the triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the human face in the image, comparing the triangular relationship with the triangular relationship in the first recognition library in a region with the triangular area of +/-5%, and if the area value formed by the triangular relationship in the first recognition library is in the region, judging that the comparison is successful.
The capturing the video frames dynamically captured on site and detecting the faces in the video in real time further comprises:
capturing video frames captured in the field dynamically, detecting N (N is more than 0) faces of the faces in the video from the video frames in real time, performing face shape recognition, selecting a plurality of photos with face shapes, and performing synchronous face recognition.
And when the face recognition is carried out, if a plurality of faces appear in the video frame, the face recognition is carried out according to the order of the image occupation ratio from large to small.
In the process of sorting and carrying out face recognition, if a video frame dynamically grabbed on site exists and faces in the video are detected in real time, the faces detected in real time are subjected to face recognition preferentially.
And after extracting the name, judging whether the name is in the sign-in record, if so, ending, and if not, continuing.
The number of the face recognition is once again, the second time is determined as the sign-in failure, the images of the video frames are synchronously stored, and the staff is waited to verify the identity of the staff.
The check-in record automatically calculates the number of successful check-ins, and simultaneously records the number of failed check-ins and the total number of successful check-ins and failed check-ins.
Example two
The method for automatically signing in based on face recognition is the same as the first embodiment and is not repeated, wherein,
The "face has a plurality of feature points satisfying face recognition" further includes:
This feature is identified for 68 feature points commonly used in the prior art, see 68 points on the face of fig. 3; meanwhile, the nose tip characteristic points are 30 points, the left shoulder characteristic points and the right shoulder characteristic points can be set to be 68 points and 69 points (the existing serial numbers start from 0), and other points except the points 30, 68 and 69 are not marked in the figure; the triangular relationship (included angle) constructed by the nasal tip feature point, the left shoulder feature point and the right shoulder feature point is shown in fig. 3.
The triangle relationship is an included angle formed by the nose tip feature point, the left shoulder feature point and the right shoulder feature point, and the step of comparing the triangle relationship constructed by the nose tip feature point, the left shoulder feature point and the right shoulder feature point of the face in the extracted image with the triangle relationship in the first recognition library in the face recognition further comprises the steps of:
And extracting the triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the human face in the image, comparing the triangular relationship with the triangular relationship in the first recognition library in a section with an included angle of +/-2 degrees, and if the area value formed by the triangular relationship in the first recognition library is in the section, judging that the comparison is successful.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent changes made by the specification and drawings of the present invention, or direct or indirect application in the relevant art, are included in the scope of the present invention.
Claims (10)
1. A method for automatically signing in based on face recognition, comprising the following steps:
Establishing a recognition library, and inputting personnel information to be recognized, wherein the personnel information comprises a name and a portrait, the portrait comprises a face and shoulders, the face is provided with a plurality of characteristic points meeting the face recognition, the characteristic points comprise nose tip characteristic points, and the two outermost sides of the shoulders are provided with left shoulder characteristic points and right shoulder characteristic points;
the method comprises the steps of storing a triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the identified person into a first identification library, putting the name and the face image of the identified person into a second identification library, and simultaneously establishing a connection for the triangular relationship, the face image and the name; each triangle relation corresponds to a plurality of facial image information;
Face recognition, namely capturing a video frame captured in a field dynamic manner and detecting a face in the video in real time, if no face exists, not acting, if the face exists, judging whether an image of the video frame contains identifiable left shoulder characteristic points and right shoulder characteristic points, if not, directly extracting the characteristic points of the face to be compared with the face in a second recognition library for recognition, if the comparison is successful, extracting the name, otherwise, carrying out face recognition again; if the identifiable left shoulder characteristic points and right shoulder characteristic points exist, triangular relations constructed by the nose tip characteristic points, the left shoulder characteristic points and the right shoulder characteristic points of the human face in the image are extracted and compared with those in the first recognition library, if the comparison is successful, a plurality of face images in the second recognition library corresponding to the triangular relations are recognized and compared, if the comparison is successful, names are extracted, and otherwise, the human face recognition is performed again;
and automatically checking in, displaying the successful check-in of the personnel with the extracted name, and forming a check-in record.
2. The automatic sign-in method based on face recognition according to claim 1, wherein the triangular relationship is a triangular area or an included angle formed by a nose tip feature point and three of a left shoulder feature point and a right shoulder feature point.
3. The automatic sign-in method based on face recognition according to claim 2, wherein when the triangular relationship is a triangular area, the step of "comparing the triangular relationship constructed by three of the nose tip feature point, the left shoulder feature point and the right shoulder feature point of the face in the extracted image" in face recognition with the triangular relationship in the first recognition library "further comprises:
And extracting the triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the human face in the image, comparing the triangular relationship with the triangular relationship in the first recognition library in a region with the triangular area of +/-5%, and if the area value formed by the triangular relationship in the first recognition library is in the region, judging that the comparison is successful.
4. The automatic sign-in method based on face recognition according to claim 2, wherein the triangle relationship is an included angle formed by the nose tip feature point, the left shoulder feature point and the right shoulder feature point, and the step of "extracting the triangle relationship constructed by the nose tip feature point, the left shoulder feature point and the right shoulder feature point of the face in the image and comparing the triangle relationship in the first recognition library" in the face recognition further comprises:
And extracting the triangular relationship constructed by the nasal tip feature points, the left shoulder feature points and the right shoulder feature points of the human face in the image, comparing the triangular relationship with the triangular relationship in the first recognition library in a section with an included angle of +/-2 degrees, and if the area value formed by the triangular relationship in the first recognition library is in the section, judging that the comparison is successful.
5. The automatic check-in method based on face recognition according to claim 1, wherein "capturing video frames captured dynamically in the field and detecting faces in the video in real time" further comprises:
Capturing video frames captured in the field dynamically, detecting faces in the video in real time, carrying out face shape recognition on N faces obtained from the video frames, selecting a plurality of photos with face shapes, and carrying out synchronous face recognition.
6. The automatic sign-in method based on face recognition according to claim 1, wherein if a plurality of faces appear in a video frame during face recognition, the face recognition is performed in a sequence from large to small according to the image ratio.
7. The automatic sign-in method based on face recognition according to claim 6, wherein in the process of sorting face recognition, if there are video frames captured dynamically in the field and faces in the video are detected in real time, the faces detected in real time are prioritized for face recognition.
8. The automatic sign-in method based on face recognition according to claim 1, wherein the name is extracted and then whether the name is in the sign-in record is judged, if yes, the method is terminated, and if not, the method is continued.
9. The automatic check-in method based on face recognition according to claim 1, wherein the number of times of re-performing face recognition is one time, the second time is determined as a check-in failure, and the images of the video frames are synchronously stored, waiting for the staff to verify the staff's identity.
10. The automatic check-in method based on face recognition according to claim 9, wherein the check-in record automatically calculates the number of successful check-ins while recording the number of failed check-ins and the total number of successful check-ins and failed check-ins.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410324872.7A CN117935341B (en) | 2024-03-21 | 2024-03-21 | Automatic sign-in method based on face recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410324872.7A CN117935341B (en) | 2024-03-21 | 2024-03-21 | Automatic sign-in method based on face recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117935341A true CN117935341A (en) | 2024-04-26 |
| CN117935341B CN117935341B (en) | 2024-06-04 |
Family
ID=90755943
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410324872.7A Active CN117935341B (en) | 2024-03-21 | 2024-03-21 | Automatic sign-in method based on face recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117935341B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106780810A (en) * | 2016-12-06 | 2017-05-31 | 浙江水马环保科技有限公司 | Work attendance method based on water purifier |
| CN109359625A (en) * | 2018-11-16 | 2019-02-19 | 南京甄视智能科技有限公司 | The method and system of customer identification is judged based on head and shoulder detection and face recognition technology |
| CN109685025A (en) * | 2018-12-27 | 2019-04-26 | 中科院合肥技术创新工程院 | Shoulder feature and sitting posture Activity recognition method |
| CN109872160A (en) * | 2019-02-01 | 2019-06-11 | 广州逗号智能科技有限公司 | Electric paying method and device |
| CN109903410A (en) * | 2019-01-23 | 2019-06-18 | 深圳壹账通智能科技有限公司 | Roll-call based on recognition of face is registered method, apparatus, equipment and storage medium |
| CN109979057A (en) * | 2019-03-26 | 2019-07-05 | 国家电网有限公司 | A kind of power communication security protection face intelligent identifying system based on cloud computing |
| CN110598647A (en) * | 2019-09-17 | 2019-12-20 | 四川爱目视光智能科技有限公司 | Head posture recognition method based on image recognition |
| CN110762411A (en) * | 2019-11-05 | 2020-02-07 | 广东麒麟精工科技有限公司 | Intelligent learning lamp based on machine identification |
| CN111563452A (en) * | 2020-05-06 | 2020-08-21 | 南京师范大学镇江创新发展研究院 | Multi-human body posture detection and state discrimination method based on example segmentation |
| JP2021007028A (en) * | 2020-09-24 | 2021-01-21 | 日本電気株式会社 | Information processing system, method for managing authentication object, and program |
| CN113537019A (en) * | 2021-07-07 | 2021-10-22 | 智洋创新科技股份有限公司 | Detection method for identifying wearing of safety helmet of transformer substation personnel based on key points |
-
2024
- 2024-03-21 CN CN202410324872.7A patent/CN117935341B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106780810A (en) * | 2016-12-06 | 2017-05-31 | 浙江水马环保科技有限公司 | Work attendance method based on water purifier |
| CN109359625A (en) * | 2018-11-16 | 2019-02-19 | 南京甄视智能科技有限公司 | The method and system of customer identification is judged based on head and shoulder detection and face recognition technology |
| CN109685025A (en) * | 2018-12-27 | 2019-04-26 | 中科院合肥技术创新工程院 | Shoulder feature and sitting posture Activity recognition method |
| CN109903410A (en) * | 2019-01-23 | 2019-06-18 | 深圳壹账通智能科技有限公司 | Roll-call based on recognition of face is registered method, apparatus, equipment and storage medium |
| CN109872160A (en) * | 2019-02-01 | 2019-06-11 | 广州逗号智能科技有限公司 | Electric paying method and device |
| CN109979057A (en) * | 2019-03-26 | 2019-07-05 | 国家电网有限公司 | A kind of power communication security protection face intelligent identifying system based on cloud computing |
| CN110598647A (en) * | 2019-09-17 | 2019-12-20 | 四川爱目视光智能科技有限公司 | Head posture recognition method based on image recognition |
| CN110762411A (en) * | 2019-11-05 | 2020-02-07 | 广东麒麟精工科技有限公司 | Intelligent learning lamp based on machine identification |
| CN111563452A (en) * | 2020-05-06 | 2020-08-21 | 南京师范大学镇江创新发展研究院 | Multi-human body posture detection and state discrimination method based on example segmentation |
| JP2021007028A (en) * | 2020-09-24 | 2021-01-21 | 日本電気株式会社 | Information processing system, method for managing authentication object, and program |
| CN113537019A (en) * | 2021-07-07 | 2021-10-22 | 智洋创新科技股份有限公司 | Detection method for identifying wearing of safety helmet of transformer substation personnel based on key points |
Non-Patent Citations (1)
| Title |
|---|
| 顾志松;沈春锋;姚文韬;董文生;: "高清人像抓拍检索***的设计与实现", 控制工程, no. 1, 20 July 2015 (2015-07-20) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117935341B (en) | 2024-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105551104B (en) | A kind of automatic attendance checking system in middle and primary schools classroom differentiated based on monitoring image seat | |
| CN110267061B (en) | News splitting method and system | |
| CN107481343A (en) | A kind of check class attendance based on face recognition technology is registered system and its method of work | |
| CN109284733A (en) | A kind of shopping guide's act of omission monitoring method based on yolo and multitask convolutional neural networks | |
| CN109740501A (en) | A kind of Work attendance method and device of recognition of face | |
| CN109858371A (en) | The method and device of recognition of face | |
| CN108154304A (en) | There is the server of Teaching Quality Assessment | |
| CN106250825A (en) | A kind of at the medical insurance adaptive face identification system of applications fields scape | |
| CN105989363B (en) | Establishing method for multi-angle face image library | |
| CN104183029A (en) | Portable quick crowd attendance method | |
| Chowdhury et al. | Development of an automatic class attendance system using cnn-based face recognition | |
| CN102194106B (en) | Human face recognition method used in door access system | |
| CN110827432A (en) | Class attendance checking method and system based on face recognition | |
| CN114760442B (en) | Monitoring system for online education management | |
| CN113269903A (en) | Face recognition class attendance system | |
| JP4374609B2 (en) | People counting device | |
| CN109345427B (en) | Classroom video frequency point arrival method combining face recognition technology and pedestrian recognition technology | |
| CN111611854A (en) | Classroom condition evaluation method based on pattern recognition | |
| CN109903410A (en) | Roll-call based on recognition of face is registered method, apparatus, equipment and storage medium | |
| CN108229124A (en) | A kind of self-service query printing testimony of a witness compares device systems and its method of work | |
| CN117935341B (en) | Automatic sign-in method based on face recognition | |
| CN117218756B (en) | Intelligent safety pick-up system and method based on face recognition | |
| CN111582195B (en) | Construction method of Chinese lip language monosyllabic recognition classifier | |
| CN115294630A (en) | Conference system and conference management method | |
| CN113052100A (en) | Traffic identification method and related device |
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 |