CN110569809A - coal mine dynamic face recognition attendance checking method and system based on deep learning - Google Patents
coal mine dynamic face recognition attendance checking method and system based on deep learning Download PDFInfo
- Publication number
- CN110569809A CN110569809A CN201910859933.9A CN201910859933A CN110569809A CN 110569809 A CN110569809 A CN 110569809A CN 201910859933 A CN201910859933 A CN 201910859933A CN 110569809 A CN110569809 A CN 110569809A
- Authority
- CN
- China
- Prior art keywords
- face
- picture
- deep learning
- coal mine
- network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
-
- 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
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/10—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people together with the recording, indicating or registering of other data, e.g. of signs of identity
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Software Systems (AREA)
- Evolutionary Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
the invention discloses a coal mine dynamic face recognition attendance checking method and system based on deep learning, wherein the method comprises the following steps: step 1: acquiring a monitoring video at a coal mine well entrance, and extracting a picture of each frame of the monitoring video; step 2: carrying out ambiguity detection on the picture, and when the detected ambiguity is lower than a set threshold, the picture is not processed and is higher than the set threshold, and executing the step 3; and step 3: detecting a face region in the preprocessed picture by using a face detection algorithm based on deep learning, and outputting coordinates of the face in the picture if the face exists; and 4, step 4: positioning face key points in a face area; and 5: and performing face recognition by adopting a deep learning-based method, and calculating an image closest to a face database by utilizing similarity comparison, namely finishing the face recognition. By adopting the method, the efficiency of the coal mine downhole staff attendance is improved, and the actions of card taking, card leakage, staff identity authentication and attendance fake making existing in the coal mine attendance are avoided.
Description
Technical Field
The invention belongs to the technical field of face recognition, and particularly relates to a coal mine dynamic face recognition attendance checking method and system based on deep learning.
Background
At present, the management and attendance of the personnel in the pit commonly used in the coal mine mainly have 3 types: 1. punching and swiping the traditional attendance card; 2. fingerprint identification; 3. and (5) iris recognition. However, in the conventional attendance card punching, a card reader is installed at a wellhead, attendance is completed by detecting identification card information, and the phenomena of card reading by personnel instead of workers and work replacement can occur. The attendance checking is completed by verifying the fingerprint information of the personnel going into the well at the well entrance through fingerprint identification, but the fingerprints of workers in the first line of the coal mine are easy to damage, the identification rate is caused by covering dust, coal ash and the like with the fingerprints, and the effect is not ideal in the coal mine industry. The iris recognition utilizes the characteristics of the invariance and the difference of the iris for the whole life to recognize the identity, has stable performance and high recognition precision, cannot be counterfeited, is suitable for the special industry of coal mines without contact, and more mines begin to use the iris recognition technology as the identity recognition means of mine safety attendance at present, and are used in coal mines of the provinces of the northeast, the three provinces of Shaanxi, the Shanxi, the Shandong and the like. However, iris attendance still has some defects: the condition that the queue at the well entrance is crowded due to the fact that the personnel going into the well need to do cooperative activities and the attendance checking efficiency is poor is shown in figure 1; some personnel are not put into the well after being authenticated on the iris attendance machine; and when some personnel go into the well, the attendance is not checked, and the attendance is checked after the personnel go out of the well and return to the well and then enter the well.
disclosure of Invention
The invention aims to solve the technical problem that in order to overcome the defects of the prior art, the invention provides the coal mine dynamic face recognition attendance method and the system based on deep learning, so that the coal mine downhole staff attendance efficiency is improved, and the coal mine attendance behaviors such as card taking, card leakage, personal identity authentication, attendance fake creation and the like are avoided.
In order to solve the technical problems, the invention adopts the technical scheme that a coal mine dynamic face recognition attendance method based on deep learning comprises the following steps:
step 1: acquiring a monitoring video at a coal mine well entrance, converting the monitoring video into a frame of picture, and extracting each frame of picture;
step 2: preprocessing the picture in the step 1: carrying out ambiguity detection on the picture, when the detected ambiguity is lower than a set threshold value, the picture is not processed, and when the detected ambiguity is higher than the set threshold value, executing the step 3; the threshold was set to 0.7;
and step 3: detecting a face area in a picture: detecting a face region in the preprocessed picture by using a face detection algorithm based on deep learning, and outputting coordinates of the face in the picture if the face exists;
And 4, step 4: positioning face key points in a face region, and aligning the face region by adopting a deep learning-based method to obtain an aligned face picture;
And 5: and (3) adopting a deep learning-based method to carry out face recognition, outputting the feature vector of each face by the aligned face image (slice) through a neural network, and calculating the image closest to the face database by utilizing similarity comparison, namely finishing the face recognition. When the image with the nearest distance exists, the duty is indicated, and if the image with the nearest distance does not exist, the absence of duty is indicated.
Further, the specific process of step 3 is as follows: inputting the preprocessed picture into a feature extraction network to generate a feature mapping graph, and generating a candidate face region in the picture by utilizing a candidate region on the output feature mapping graph; then, classifying the face region by using a face detection network, and outputting the final face target position and probability; classifying the face area into a face target and a background; the feature extraction network employs a 50-layer residual network Res 50.
The candidate area generation network adopts anchor point windows to determine candidate face areas in the pictures, the anchor point windows in the candidate area generation network adopt 8 scales and 3 proportions, and the 8 scales are [1, 2, 3, 4, 6, 8, 16, 32 and 64]]and 3 ratios are [ 2: 1,1: 1,1: 2](ii) a The face detection network comprises a 1 x 1 convolution layer and a 2k convolution layer which are connected in sequence2A convolutional layer of dimensions and a pooling layer; wherein: k is not less than 3, and the maximum value of k is determined by the required detection speed.
Further, the process of classifying the face by the face detection network is as follows: the face detection network firstly uses a 1 × 1 convolution layer to reduce dimension on the generated feature mapping graph, and then generates a position sensitive score graph through a specific convolution layer; and the position sensitive score map and the candidate face region generated by the candidate region generation network are jointly sent to a position sensitive pooling layer to generate a category score map and a range frame prediction map, an average pooling operation is respectively utilized on the category score map and the range frame prediction map to aggregate the feature mapping maps into a vector, and finally the probability and the coordinate position of each candidate region including the face are output.
further, the specific process of step 4 is as follows: the coordinates of the preprocessed pictures and the positions of the face regions are sent into a first convolution neural network, an average face shape is put into a face region coordinate range frame, face key point coordinates are initialized, the offset of the face key point coordinates is output through the first neural network, face key points are obtained, the face regions are geometrically normalized to a standard posture through similarity transformation, face alignment is achieved, and aligned face images are obtained.
Further, the specific process of step 5 is as follows: adopting a frame of a convolutional neural network and a loss function, and training the whole convolutional neural network by using the loss function in a supervision way; the convolutional neural network adopts a 20-layer residual error network;
The formula of the loss function is as follows:
wherein: l isAdenotes the A-softmax loss function, LCrepresenting the central loss function, λ is used to balance the weight of the central loss function with the A-softmax loss function, N represents the number of training set samples, xirepresenting the characteristics of the ith training set sample, m is a hyper-parameter used to quantify the decision boundary, j represents the class,represents the y thiThe center of the features of each of the categories,is a weight vectorand xiangle between them, thetaj,iRepresents WjAnd a feature vector xithe included angle between the two parts is included,And Wjrespectively representing that the network full connection layer corresponds to the category yiAnd the weight of j.
the invention also discloses a coal mine dynamic face recognition attendance system based on deep learning, which comprises an image acquisition module, a database, a video preprocessing module, a deep learning server and a detection recognition management module; an image acquisition module: the face characteristic information acquisition system is arranged at a coal mine well entrance, is used for acquiring face information of coal mine well descending operators, generates corresponding face characteristic information and stores the face characteristic information into a database;
A database: the face recognition system is used for receiving and storing face feature information and identity information;
The video preprocessing module: acquiring a real-time monitoring video from a monitoring camera, converting the video into a frame of picture, detecting the fuzziness of each frame of picture, and sending the picture to a deep learning server when the fuzziness is higher than a set threshold value and the picture is not processed and is lower than the set threshold value;
the deep learning server: the system comprises a face detection module, a face alignment module and a face recognition module which are sequentially connected;
the method comprises the steps that a preprocessed picture is sent to a face detection module, a face detection algorithm based on deep learning is used for detecting a face area in the preprocessed picture, and if a face exists, coordinates of the face in the picture are output; then, the whole picture (preprocessed picture) and the detected face region coordinates are sent to a face alignment module together, the face key point feature information is detected, and the face is aligned; finally, the aligned face picture is sent to a face recognition module to extract the features of the face, and the features of the face are compared with the information in the face feature database to recognize the identity information of each face to be detected;
the detection, identification and management module: the system is used for managing the identity information of the detected human face and realizing the inquiry and recording of the attendance information of the personnel.
The invention has the following advantages: compared with the traditional attendance checking method, the method has the advantages that the identification speed is high, the whole identification process does not need to make cooperative behaviors, and the whole process is noninductive. The method is used for realizing face recognition, the model can be rapidly deployed in a production environment without writing extra codes, the trained model can be rapidly deployed in a deep learning server, and a picture is sent to the deep learning server, so that information such as coordinates of a detected face, key coordinates of the face, characteristics of the face and the like can be returned.
drawings
Fig. 1 is an architecture diagram of a coal mine dynamic face recognition attendance method based on deep learning.
Detailed Description
the invention relates to a coal mine dynamic face recognition attendance method based on deep learning, which comprises the following steps as shown in figure 1:
Step 1: acquiring a monitoring video at a coal mine well entrance, converting the monitoring video into a picture of one frame, and extracting each frame of picture of the monitoring video; specifically, 4 ~ 6 high definition cameras are installed at the different angles of income well head, catch the picture of personnel's face when going into the well head in the pit, through the real time monitoring of a plurality of cameras, can 100% guarantee.
step 2: preprocessing the picture: carrying out ambiguity detection on the picture, when the detected ambiguity is higher than a set threshold value, the picture is not processed, and when the detected ambiguity is higher than the set threshold value, executing the step 3; the threshold was set to 0.7;
and step 3: detecting a face area in a picture: detecting a face region in the preprocessed picture by using a face detection algorithm based on deep learning, and outputting coordinates of the face in the picture if the face exists;
The specific process is as follows: inputting the whole preprocessed picture into a feature extraction network to generate a feature mapping map, and generating a candidate face region in the picture by utilizing a candidate region on the output feature mapping map on the output feature map; then, classifying the face region by using a face detection network, and outputting the final face target position and probability; classifying the face area into a face target and a background; the feature extraction network employs a 50-layer residual network Res 50.
the candidate area generation network adopts anchor point windows to determine candidate face areas in the pictures, the anchor point windows in the candidate area generation network adopt 8 scales and 3 proportions, and the 8 scales are [1, 2, 3, 4, 6, 8, 16, 32 and 64]]and 3 ratios are [ 2: 1,1: 1,1: 2](ii) a The face detection network comprises a 1 x 1 convolution layer and a 2k convolution layer which are connected in sequence2A convolutional layer of dimensions and a pooling layer; wherein: k is not less than 3, and the maximum value of k is determined by the required detection speed. If k is larger, the space division is finer and the positioning is more accurate, but since k is increased, the space position grid to be processed is increased, the memory occupation is increased, and the processing speed is reduced. Typically k is chosen to be 3.
The process of classifying the face by the face detection network is as follows: the face detection network firstly uses a 1 × 1 convolution layer to reduce dimension on the generated feature mapping graph, and then generates a position sensitive score graph through a specific convolution layer; and the position sensitive score map and the candidate face region generated by the candidate region generation network are jointly sent to a position sensitive pooling layer to generate a category score map and a range frame prediction map, an average pooling operation is respectively utilized on the category score map and the range frame prediction map to aggregate the feature mapping maps into a vector, and finally the probability and the coordinate position of each candidate region including the face are output.
The position sensitive score map is mainly used for improving the performance of face detection, because a feature mapping map output by a convolutional neural network has translation irreplaceability, and a face detection task needs to position the face and requires the network to have good translation variability. Firstly, using a 1 × 1 convolution layer to reduce dimension on the feature diagram generated by the feature extraction network, and making the feature diagram after dimension reduction pass through a 2k2Convolution operation of dimensions (conv) output dimension of 2k2The k × k position sensitivity score maps are generated for each face. Wherein: 2 denotes the category: face + background, so there are 2 categories in total; k × k denotes k × k relative spatial positions; thus each class generates k2Position sensitive score map, therefore 2k in total2a map of individual location sensitivity; k is a radical of2The position sensitive score map is encodedK describing spatial position information2And (4) a position. Taking k as an example of 3, 9 spatial position information { upper left corner, top center, upper right.
and 4, step 4: positioning face key points in the face area, and aligning the face area to obtain an aligned face picture; the method for aligning the human face is specifically realized as follows: the states of the human face such as the posture, the direction, the size and the like are different in the image, the different facial postures can cause difficulty in recognition, and the accuracy of the human face recognition can be improved by aligning the human face in the image to the uniform template, so that the human face is aligned in the human face recognition process.
the method comprises the steps of sending an original picture and coordinates of a face region detected by a face detection algorithm into a convolutional neural network, initializing face key point coordinates by putting an average face shape into a face region coordinate range frame returned by the face detection algorithm, outputting the offset of the face key point coordinates through the neural network to obtain face key points, and geometrically normalizing the face region to a standard posture by using the detected face key points through similarity transformation, so that the face alignment is realized. In the invention, the whole face image is used as input to align the face image with large face posture range change, thereby further improving the face recognition performance.
And 5: and (3) performing face recognition by adopting a deep learning-based method, outputting the feature vector of each face by the aligned face image through a neural network, and calculating the image closest to the face database by utilizing similarity comparison, namely completing the face recognition. When the image with the nearest distance exists, the duty is indicated, and if the image with the nearest distance does not exist, the absence of duty is indicated.
the specific process of the step 5 is as follows: adopting a frame of a convolutional neural network and a loss function, and training the whole convolutional neural network by using the loss function in a supervision way; the convolutional neural network adopts a 20-layer residual error network; a frame of a convolutional neural network and a loss function is adopted, and the loss function is used for training the whole convolutional neural network in a supervision mode, so that the network learns the face features with strong discriminability. In consideration of recognition speed, the convolutional neural network adopts 20 layers of residual error networks, the loss function provides an improved loss function, and the convolutional neural network is trained through joint supervision of a central loss function and an A-softmax loss function so as to obtain the human face features with more compact intra-class distance, and finally the performance of human face recognition is improved. The central loss function can only compress the features in the class, and a supervised training convolutional neural network is required to be shared with the softmax function to obtain the face features with strong resolution; the A-softmax loss function defines a learning method with large angle intervals, can learn the human face features with large inter-class distance and compact intra-class distance, and can further compress the intra-class distance of the human face through the combination of the two loss functions.
The formula for this loss function is as follows:
Wherein: l isADenotes the A-softmax loss function, LCRepresenting the central loss function, λ is used to balance the weight of the central loss function with the A-softmax loss function, N represents the number of training set samples, xirepresenting the characteristics of the ith training set sample, m is a hyper-parameter used to quantify the decision boundary, j represents the class,Represents the y thithe center of the features of each of the categories,Is a weight vectorAnd xiAngle between them, thetaj,irepresents WjAnd a feature vector xiThe included angle between the two parts is included,And WjRespectively representing that the network full connection layer corresponds to the category yiAnd the weight of j.
The invention also discloses a coal mine dynamic face recognition attendance system based on deep learning, which comprises an image acquisition module, a database, a video preprocessing module, a deep learning server and a detection recognition management module; an image acquisition module: the face characteristic information acquisition system is arranged at a coal mine well entrance, is used for acquiring face information of coal mine well descending operators, generates corresponding face characteristic information and stores the face characteristic information into a database;
A database: the face recognition system is used for receiving and storing face feature information and identity information;
The video preprocessing module: acquiring a real-time monitoring video from a monitoring camera, converting the video into a frame of picture, detecting the fuzziness of each frame of picture, and sending the picture to a deep learning server when the fuzziness is higher than a set threshold value and the picture is not processed and is lower than the set threshold value;
The deep learning server: the system comprises a face detection module, a face alignment module and a face recognition module which are sequentially connected;
the method comprises the steps that a preprocessed picture is sent to a face detection module, a face detection algorithm based on deep learning is used for detecting a face area in the preprocessed picture, and if a face exists, coordinates of the face in the picture are output; then, the whole picture (preprocessed picture) and the detected face region coordinates are sent to a face alignment module together, the face key point feature information is detected, and the face is aligned; finally, the aligned face picture is sent to a face recognition module to extract the features of the face, and the features of the face are compared with the information in the face feature database to recognize the identity information of each face to be detected;
the detection, identification and management module: the system is used for managing the identity information of the detected human face and realizing the inquiry and recording of the attendance information of the personnel.
In order to realize dynamic face recognition attendance, namely, coal mine downhole personnel do not need to make cooperative behaviors, and face recognition and attendance are realized in the process of walking through a coal mine and entering a wellhead. 4 ~ 6 high definition digtal cameras are installed to different angles of income well head, and the purpose is in order to catch the picture of personnel of going into the well head face when the well head is gone into to the process, through the real time monitoring of a plurality of cameras, can 100% guarantee the catch of people's face, no matter what wrench movement appears in the personnel of going into the well in the walking process, can guarantee the catch of people's face.
Claims (6)
1. A coal mine dynamic face recognition attendance method based on deep learning is characterized by comprising the following steps:
Step 1: acquiring a monitoring video at a coal mine well entrance, converting the monitoring video into a frame of picture, and extracting each frame of picture;
Step 2: preprocessing the picture: carrying out ambiguity detection on the picture, when the detected ambiguity is lower than a set threshold value, the picture is not processed, and when the detected ambiguity is higher than the set threshold value, executing the step 3; the threshold was set to 0.7;
And step 3: detecting a face area in a picture: detecting a face region in the preprocessed picture, and outputting coordinates of the face in the picture if the face exists;
And 4, step 4: positioning face key points in the face area, and aligning the face area to obtain an aligned face picture;
And 5: adopting a deep learning-based method to carry out face recognition, outputting the feature vector of each face by the aligned face image through a neural network, and calculating the image closest to the face database by utilizing similarity comparison, namely finishing the face recognition; when the image with the nearest distance exists, the duty is indicated, and if the image with the nearest distance does not exist, the absence of duty is indicated.
2. The coal mine dynamic face recognition attendance method based on deep learning of claim 1, wherein the specific process of step 3 is as follows:
The method comprises the following specific steps: inputting the preprocessed picture into a feature extraction network to generate a feature mapping graph, and generating a candidate face region in the picture by utilizing a candidate region on the output feature mapping graph; then, classifying the face region by using a face detection network, and outputting the final face target position and probability; classifying the face area into a face target and a background; the feature extraction network adopts a residual error network Res50 with 50 layers;
The candidate area generation network determines the candidate face area in the picture by adopting an anchor point window, the anchor point window in the candidate area generation network adopts 8 scales and 3 proportions, the 8 scales are [1, 2, 3, 4, 6, 8, 16, 32, 64], and the 3 proportions are [ 2: 1,1: 1,1: 2 ];
The face detection network comprises a 1 x 1 convolution layer and a 2k convolution layer which are connected in sequence2a convolutional layer of dimensions and a pooling layer; wherein: k is equal to or more than 3, and the maximum value of k is determined by the required detection speed.
3. The coal mine dynamic face recognition attendance method based on deep learning of claim 2, wherein the face detection network classifies the faces as follows: the face detection network firstly uses a 1 × 1 convolution layer to reduce dimension on the generated feature mapping graph, and then generates a position sensitive score graph through a specific convolution layer; and the position sensitive score map and the candidate face region generated by the candidate region generation network are jointly sent to a position sensitive pooling layer to generate a category score map and a range frame prediction map, an average pooling operation is respectively utilized on the category score map and the range frame prediction map to aggregate the feature mapping maps into a vector, and finally the probability and the coordinate position of each candidate region including the face are output.
4. the coal mine dynamic face recognition attendance method based on deep learning of claim 1, 2 or 3, characterized in that the specific process of step 4 is as follows: the coordinates of the preprocessed pictures and the positions of the face regions are sent into a first convolution neural network, an average face shape is put into a face region coordinate range frame, face key point coordinates are initialized, the offset of the face key point coordinates is output through the first neural network, face key points are obtained, the face regions are geometrically normalized to a standard posture through similarity transformation, face alignment is achieved, and aligned face images are obtained.
5. The coal mine dynamic face recognition attendance method based on deep learning of claim 4, wherein the specific process of the step 5 is as follows: adopting a frame of a convolutional neural network and a loss function, and training the whole convolutional neural network by using the loss function in a supervision way; the convolutional neural network adopts a 20-layer residual error network;
The formula of the loss function is as follows:
Wherein: l isADenotes the A-softmax loss function, LCrepresenting the central loss function, λ is used to balance the weight of the central loss function with the A-softmax loss function, N represents the number of training set samples, xiRepresenting the characteristics of the ith training set sample, m is a hyper-parameter used to quantify the decision boundary, j represents the class,represents the y thithe center of the features of each of the categories,Is a weight vectorAnd xiAngle between them, thetaj,irepresents Wjand a feature vector xiThe included angle between the two parts is included,And Wjrespectively representing that the network full connection layer corresponds to the category yiAnd the weight of j.
6. A coal mine dynamic face recognition attendance system based on deep learning is characterized by comprising an image acquisition module, a database, a video preprocessing module, a deep learning server and a detection recognition management module;
The image acquisition module: the face characteristic information acquisition system is arranged at a coal mine well entrance, is used for acquiring face information of coal mine well descending operators, generates corresponding face characteristic information and stores the face characteristic information into a database;
A database: the face recognition system is used for receiving and storing face feature information and identity information;
The video preprocessing module: acquiring a real-time monitoring video from a monitoring camera, converting the video into a frame of picture, detecting the fuzziness of each frame of picture, and sending the picture to a deep learning server when the fuzziness is higher than a set threshold value and the picture is not processed and is lower than the set threshold value;
The deep learning server: the system comprises a face detection module, a face alignment module and a face recognition module which are sequentially connected;
The method comprises the steps that a preprocessed picture is sent to a face detection module, a face detection algorithm based on deep learning is used for detecting a face area in the preprocessed picture, and if a face exists, coordinates of the face in the picture are output; then, the whole picture (preprocessed picture) and the detected face region coordinates are sent to a face alignment module together, the face key point feature information is detected, and the face is aligned; finally, the aligned face picture is sent to a face recognition module to extract the features of the face, and the features of the face are compared with the information in the face feature database to recognize the identity information of each face to be detected;
The detection, identification and management module: the system is used for managing the identity information of the detected human face and realizing the inquiry and recording of the attendance information of the personnel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910859933.9A CN110569809A (en) | 2019-09-11 | 2019-09-11 | coal mine dynamic face recognition attendance checking method and system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910859933.9A CN110569809A (en) | 2019-09-11 | 2019-09-11 | coal mine dynamic face recognition attendance checking method and system based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110569809A true CN110569809A (en) | 2019-12-13 |
Family
ID=68779321
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910859933.9A Pending CN110569809A (en) | 2019-09-11 | 2019-09-11 | coal mine dynamic face recognition attendance checking method and system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110569809A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111209860A (en) * | 2020-01-06 | 2020-05-29 | 上海海事大学 | Video attendance system and method based on deep learning and reinforcement learning |
| CN111339869A (en) * | 2020-02-18 | 2020-06-26 | 北京拙河科技有限公司 | Face recognition method, face recognition device, computer readable storage medium and equipment |
| CN111881876A (en) * | 2020-08-06 | 2020-11-03 | 桂林电子科技大学 | Attendance checking method based on single-order anchor-free detection network |
| CN111931671A (en) * | 2020-08-17 | 2020-11-13 | 青岛北斗天地科技有限公司 | Face recognition method for illumination compensation in underground coal mine adverse light environment |
| CN113111847A (en) * | 2021-04-29 | 2021-07-13 | 四川隧唐科技股份有限公司 | Automatic monitoring method, device and system for process circulation |
| WO2021184894A1 (en) * | 2020-03-20 | 2021-09-23 | 深圳市优必选科技股份有限公司 | Deblurred face recognition method and system and inspection robot |
| CN114821844A (en) * | 2021-01-28 | 2022-07-29 | 深圳云天励飞技术股份有限公司 | Attendance checking method and device based on face recognition, electronic equipment and storage medium |
| CN115050068A (en) * | 2022-05-30 | 2022-09-13 | 盛视科技股份有限公司 | Coal mine worker face recognition method |
| CN115529475A (en) * | 2021-12-29 | 2022-12-27 | 北京智美互联科技有限公司 | Method and system for detecting video flow content and controlling wind |
| CN116110100A (en) * | 2023-01-14 | 2023-05-12 | 深圳市大数据研究院 | Face recognition method, device, computer equipment and storage medium |
| CN116452878A (en) * | 2023-04-20 | 2023-07-18 | 广东工业大学 | Attendance checking method and system based on deep learning algorithm and binocular vision |
| CN117894091A (en) * | 2024-02-23 | 2024-04-16 | 江西远格科技有限公司 | Face attendance checking method for railway industry |
| CN118429862A (en) * | 2024-05-09 | 2024-08-02 | 南京广播电视集团有限责任公司 | Video auditing system and method for broadcasting and television industry based on face recognition |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104110271A (en) * | 2013-04-22 | 2014-10-22 | 开封市测控技术有限公司 | Under-mine equipment monitoring alarm system based on face recognition technology |
| CN105354565A (en) * | 2015-12-23 | 2016-02-24 | 北京市商汤科技开发有限公司 | Full convolution network based facial feature positioning and distinguishing method and system |
| CN105678248A (en) * | 2015-12-31 | 2016-06-15 | 上海科技大学 | Face key point alignment algorithm based on deep learning |
| CN107480640A (en) * | 2017-08-16 | 2017-12-15 | 上海荷福人工智能科技（集团）有限公司 | A kind of face alignment method based on two-value convolutional neural networks |
| CN108022318A (en) * | 2017-12-28 | 2018-05-11 | 上海享服信息技术有限公司 | More people's recognition of face attendance checking systems and its Work attendance method |
| CN108256450A (en) * | 2018-01-04 | 2018-07-06 | 天津大学 | A kind of supervised learning method of recognition of face and face verification based on deep learning |
| CN108446689A (en) * | 2018-05-30 | 2018-08-24 | 南京开为网络科技有限公司 | A kind of face identification method |
| CN108960136A (en) * | 2018-06-29 | 2018-12-07 | 杭州西纬软件科技有限公司 | The determination method and apparatus of Initial Face shape in face alignment algorithm |
| CN109033938A (en) * | 2018-06-01 | 2018-12-18 | 上海阅面网络科技有限公司 | A kind of face identification method based on ga s safety degree Fusion Features |
| CN109508690A (en) * | 2018-11-29 | 2019-03-22 | 浙江工业大学 | A kind of non-active cooperation attendance checking system based on recognition of face |
| CN109740501A (en) * | 2018-12-28 | 2019-05-10 | 广东亿迅科技有限公司 | A kind of Work attendance method and device of recognition of face |
| CN109754478A (en) * | 2017-11-06 | 2019-05-14 | 北京航天长峰科技工业集团有限公司 | A kind of face intelligent Checking on Work Attendance method of low user's fitness |
| CN109858466A (en) * | 2019-03-01 | 2019-06-07 | 北京视甄智能科技有限公司 | A kind of face critical point detection method and device based on convolutional neural networks |
-
2019
- 2019-09-11 CN CN201910859933.9A patent/CN110569809A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104110271A (en) * | 2013-04-22 | 2014-10-22 | 开封市测控技术有限公司 | Under-mine equipment monitoring alarm system based on face recognition technology |
| CN105354565A (en) * | 2015-12-23 | 2016-02-24 | 北京市商汤科技开发有限公司 | Full convolution network based facial feature positioning and distinguishing method and system |
| CN105678248A (en) * | 2015-12-31 | 2016-06-15 | 上海科技大学 | Face key point alignment algorithm based on deep learning |
| CN107480640A (en) * | 2017-08-16 | 2017-12-15 | 上海荷福人工智能科技（集团）有限公司 | A kind of face alignment method based on two-value convolutional neural networks |
| CN109754478A (en) * | 2017-11-06 | 2019-05-14 | 北京航天长峰科技工业集团有限公司 | A kind of face intelligent Checking on Work Attendance method of low user's fitness |
| CN108022318A (en) * | 2017-12-28 | 2018-05-11 | 上海享服信息技术有限公司 | More people's recognition of face attendance checking systems and its Work attendance method |
| CN108256450A (en) * | 2018-01-04 | 2018-07-06 | 天津大学 | A kind of supervised learning method of recognition of face and face verification based on deep learning |
| CN108446689A (en) * | 2018-05-30 | 2018-08-24 | 南京开为网络科技有限公司 | A kind of face identification method |
| CN109033938A (en) * | 2018-06-01 | 2018-12-18 | 上海阅面网络科技有限公司 | A kind of face identification method based on ga s safety degree Fusion Features |
| CN108960136A (en) * | 2018-06-29 | 2018-12-07 | 杭州西纬软件科技有限公司 | The determination method and apparatus of Initial Face shape in face alignment algorithm |
| CN109508690A (en) * | 2018-11-29 | 2019-03-22 | 浙江工业大学 | A kind of non-active cooperation attendance checking system based on recognition of face |
| CN109740501A (en) * | 2018-12-28 | 2019-05-10 | 广东亿迅科技有限公司 | A kind of Work attendance method and device of recognition of face |
| CN109858466A (en) * | 2019-03-01 | 2019-06-07 | 北京视甄智能科技有限公司 | A kind of face critical point detection method and device based on convolutional neural networks |
Non-Patent Citations (4)
| Title |
|---|
| JIFENG DAI 等: ""R-FCN: Object Detection via Region-based Fully Convolutional Networks"", 《30TH CONFERENCE ON NEURAL INFORMATION PROCESSING SYSTEMS (NIPS 2016)》 * |
| SLUMBERS: ""人脸识别损失函数综述（附开源实现）"", 《HTTPS://ZHUANLAN.ZHIHU.COM/P/51324547》 * |
| YITONG WANG 等: "Detecting Faces Using Region-based Fully Convolutional Networks", 《ARXIV.ORG》 * |
| 戴海能 等: "一种改进的基于Ｒ- FCN 模型的人脸检测算法", 《计算机与现代化》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111209860A (en) * | 2020-01-06 | 2020-05-29 | 上海海事大学 | Video attendance system and method based on deep learning and reinforcement learning |
| CN111209860B (en) * | 2020-01-06 | 2023-04-07 | 上海海事大学 | Video attendance system and method based on deep learning and reinforcement learning |
| CN111339869A (en) * | 2020-02-18 | 2020-06-26 | 北京拙河科技有限公司 | Face recognition method, face recognition device, computer readable storage medium and equipment |
| WO2021184894A1 (en) * | 2020-03-20 | 2021-09-23 | 深圳市优必选科技股份有限公司 | Deblurred face recognition method and system and inspection robot |
| CN111881876A (en) * | 2020-08-06 | 2020-11-03 | 桂林电子科技大学 | Attendance checking method based on single-order anchor-free detection network |
| CN111881876B (en) * | 2020-08-06 | 2022-04-08 | 桂林电子科技大学 | Attendance checking method based on single-order anchor-free detection network |
| CN111931671A (en) * | 2020-08-17 | 2020-11-13 | 青岛北斗天地科技有限公司 | Face recognition method for illumination compensation in underground coal mine adverse light environment |
| CN114821844B (en) * | 2021-01-28 | 2024-05-07 | 深圳云天励飞技术股份有限公司 | Attendance checking method and device based on face recognition, electronic equipment and storage medium |
| CN114821844A (en) * | 2021-01-28 | 2022-07-29 | 深圳云天励飞技术股份有限公司 | Attendance checking method and device based on face recognition, electronic equipment and storage medium |
| CN113111847A (en) * | 2021-04-29 | 2021-07-13 | 四川隧唐科技股份有限公司 | Automatic monitoring method, device and system for process circulation |
| CN115529475A (en) * | 2021-12-29 | 2022-12-27 | 北京智美互联科技有限公司 | Method and system for detecting video flow content and controlling wind |
| CN115529475B (en) * | 2021-12-29 | 2024-07-16 | 北京国瑞数智技术有限公司 | Method and system for detecting and wind controlling video flow content |
| CN115050068A (en) * | 2022-05-30 | 2022-09-13 | 盛视科技股份有限公司 | Coal mine worker face recognition method |
| CN116110100B (en) * | 2023-01-14 | 2023-11-14 | 深圳市大数据研究院 | Face recognition method, device, computer equipment and storage medium |
| CN116110100A (en) * | 2023-01-14 | 2023-05-12 | 深圳市大数据研究院 | Face recognition method, device, computer equipment and storage medium |
| CN116452878B (en) * | 2023-04-20 | 2024-02-02 | 广东工业大学 | Attendance checking method and system based on deep learning algorithm and binocular vision |
| CN116452878A (en) * | 2023-04-20 | 2023-07-18 | 广东工业大学 | Attendance checking method and system based on deep learning algorithm and binocular vision |
| CN117894091A (en) * | 2024-02-23 | 2024-04-16 | 江西远格科技有限公司 | Face attendance checking method for railway industry |
| CN118429862A (en) * | 2024-05-09 | 2024-08-02 | 南京广播电视集团有限责任公司 | Video auditing system and method for broadcasting and television industry based on face recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110569809A (en) | coal mine dynamic face recognition attendance checking method and system based on deep learning | |
| Lee et al. | Partial fingerprint matching using minutiae and ridge shape features for small fingerprint scanners | |
| Anagnostopoulos et al. | License plate recognition from still images and video sequences: A survey | |
| Galar et al. | A survey of fingerprint classification Part I: Taxonomies on feature extraction methods and learning models | |
| Lavi et al. | Survey on deep learning techniques for person re-identification task | |
| CN106529468A (en) | Finger vein identification method and system based on convolutional neural network | |
| CN102629320B (en) | Ordinal measurement statistical description face recognition method based on feature level | |
| CN103902970A (en) | Automatic fingerprint gesture estimation method and system | |
| CN102938065A (en) | Facial feature extraction method and face recognition method based on large-scale image data | |
| Saleem et al. | Face recognition using facial features | |
| CN111126240A (en) | Three-channel feature fusion face recognition method | |
| CN112132157B (en) | Gait face fusion recognition method based on raspberry pie | |
| CN111639580A (en) | Gait recognition method combining feature separation model and visual angle conversion model | |
| CN113505717A (en) | Online passing system based on face and facial feature recognition technology | |
| CN110188646B (en) | Human ear identification method based on fusion of gradient direction histogram and local binary pattern | |
| CN106886771B (en) | Image main information extraction method and face recognition method based on modular PCA | |
| Daramola et al. | Algorithm for fingerprint verification system | |
| CN111652048A (en) | A deep learning based 1: n face comparison method | |
| Pilania et al. | Implementation of image-based attendance system | |
| Farhi et al. | Smart identity management system by face detection using multitasking convolution network | |
| CN107506694B (en) | Robust face recognition method based on local median representation | |
| Sai et al. | Student Attendance Monitoring System Using Face Recognition | |
| CN116229511A (en) | Identification re-recognition method based on golden monkey trunk feature extraction | |
| Ghofrani et al. | Farsi license plate detection and recognition based on characters features | |
| Huang et al. | Skew correction of handwritten Chinese character based on ResNet |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191213 |
|
| RJ01 | Rejection of invention patent application after publication |