CN114999003A - Method and device for identifying target living body and computer readable storage medium - Google Patents
Method and device for identifying target living body and computer readable storage medium Download PDFInfo
- Publication number
- CN114999003A CN114999003A CN202210552507.2A CN202210552507A CN114999003A CN 114999003 A CN114999003 A CN 114999003A CN 202210552507 A CN202210552507 A CN 202210552507A CN 114999003 A CN114999003 A CN 114999003A
- Authority
- CN
- China
- Prior art keywords
- living body
- vector
- score
- target living
- target
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/40—Spoof detection, e.g. liveness detection
-
- 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
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Computer Interaction (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a target living body identification method and device and a computer readable storage medium. Wherein, the method comprises the following steps: acquiring position information of a target living body by using a first neural network; acquiring living body key point information of the target living body by utilizing a second neural network based on the position information; and acquiring a score vector of the target living body based on the living body key point information, processing the historical score vector and the score vector in the database to acquire a similarity, and acquiring an identification result of the target living body based on the similarity, wherein the score vector is generated by describing the living body key point information of the target living body, and the identification result is used for representing whether the target living body is successfully identified. The invention solves the technical problem of inaccurate identification caused by incomplete feature extraction when a living body is identified by vision in the related technology.
Description
Technical Field
The invention relates to the field of image recognition, in particular to a method and a device for recognizing a target living body and a computer-readable storage medium.
Background
At present, the video monitoring technology is rapidly developed, and a large amount of data can be generated in the monitoring process. The living body weight identification technology can detect and identify a specific living body from an image or video sequence, and has important significance for security monitoring and urban public security. However, since the living body image may have the problems of too bright and too dark light, blurring, serious occlusion, various postures and the like, the existing method directly extracts the features of the detected living body, and the extracted features are not necessarily representative, so that the retrieval accuracy is not high, and the retrieval fails.
In view of the above problems, no effective solution has been proposed.
Disclosure of Invention
The embodiment of the invention provides a method and a device for identifying a target living body and a computer readable storage medium, which are used for at least solving the technical problem of inaccurate identification caused by incomplete feature extraction when a living body is identified by vision in the related technology.
According to an aspect of an embodiment of the present invention, there is provided a method for identifying a target living body, including: acquiring position information of a living target body by using a first neural network, wherein the first neural network is obtained by machine learning training by using a plurality of groups of first training data, and each group of the plurality of groups of first training data comprises: historical location information of the target living body; acquiring living body key point information of a target living body by utilizing a second neural network based on the position information, wherein the second neural network is obtained by using a plurality of groups of second training data through machine learning training, and each group of the second training data in the plurality of groups of second training data comprises: historical living body key point information of the target living body; and acquiring a score vector of a target living body based on the living body key point information, processing a historical score vector in a database with the score vector to acquire a similarity, and acquiring an identification result of the target living body based on the similarity, wherein the score vector is generated by describing the living body key point information of the living body, and the identification result is used for representing whether the target living body is successfully identified.
Optionally, the living body key point information includes at least: position information of human five sense organs and position information of human joint points.
Optionally, the score vector comprises: the method comprises the steps of obtaining a brightness quality score, a definition score, an occlusion degree score and a posture score, wherein the brightness quality score is a vector obtained based on the brightness quality of a region where the target living body is located, the definition score is a vector obtained based on the image definition of an acquired image of the target living body, the occlusion degree score is a vector obtained based on the occlusion degree of the target living body occluded by an interfering object, and the posture score is a vector obtained based on the current posture of the target living body.
Optionally, before processing the historical score vector in the database with the score vector, the method further includes: acquiring a first feature vector by using a third neural network; carrying out standardization processing on the score vector to obtain a first score vector; and processing the historical feature vector to obtain the historical score vector corresponding to the historical feature vector.
Optionally, the obtaining the first feature vector by using a third neural network includes: acquiring original characteristic information of the target living body through the third neural network based on the position information; and carrying out standardization processing on the original characteristic information to obtain the first characteristic vector.
Optionally, processing a historical score vector in a database with the score vector to obtain a similarity, including: obtaining the similarity by using a second formula based on the first feature vector, the first score vector, the historical feature vector and the historical score vector, wherein the second formula is as follows:
s is the similarity, F1 is the first score vector, F2 is the historical score vector, W1 is the first feature vector, and W2 is the historical feature vector.
Optionally, acquiring a recognition result of the target living body based on the similarity includes: comparing the similarity with a preset threshold value to obtain a comparison result of which the similarity is greater than the preset threshold value; and selecting the comparison result with the highest similarity value in the comparison results with the similarity larger than a preset threshold value as the identification result of the target living body.
According to another aspect of the embodiments of the present invention, there is also provided an identification apparatus of a target living body, including: a first obtaining module, configured to obtain location information of a living target using a first neural network, where the first neural network is obtained through machine learning training using multiple sets of first training data, and each set of the multiple sets of first training data includes: historical location information of the target living body; a second obtaining module, configured to obtain living body key point information of a target living body based on the location information and by using a second neural network, where the second neural network is obtained through machine learning training using multiple sets of second training data, and each set of the second training data in the multiple sets of second training data includes: historical living body key point information of the target living body; the first processing module is used for acquiring a score vector of a target living body based on the living body key point information, processing a historical score vector in a database with the score vector to acquire similarity, and acquiring an identification result of the target living body based on the similarity, wherein the score vector is generated by describing the living body key point information of the living body, and the identification result is used for representing whether the target living body is successfully identified.
Optionally, the living body keypoint information comprises at least: position information of human five sense organs and position information of human joint points.
Optionally, the score vector comprises: the mobile terminal comprises a brightness quality score, a definition score, an occlusion degree score and a posture score, wherein the brightness quality score is a vector acquired based on the brightness quality of a region where the target living body is located, the definition score is a vector acquired based on the image definition of an acquired image of the target living body, the occlusion degree score is a vector acquired based on the occlusion degree of the target living body occluded by an interfering object, and the posture score is a vector acquired based on the current posture of the target living body.
Optionally, the apparatus further comprises: the third acquisition module is used for acquiring a first feature vector by using a third neural network before processing the historical score vector in the database and the score vector; the standardization processing module is used for carrying out standardization processing on the score vector to obtain a first score vector; and the second processing module is used for processing the historical characteristic vector and acquiring the historical score vector corresponding to the historical characteristic vector.
Optionally, the third obtaining module includes: a first acquisition unit configured to acquire original feature information of the living target body through the third neural network based on the position information; and the normalization processing unit is used for performing normalization processing on the original characteristic information to obtain the first characteristic vector.
Optionally, the first processing module includes: a second obtaining unit, configured to obtain the similarity using a second formula based on the first feature vector, the first score vector, the historical feature vector, and the historical score vector, where the second formula is:
s is the similarity, F1 is the first score vector, F2 is the historical score vector, W1 is the first feature vector, and W2 is the historical feature vector.
Optionally, the first processing module includes: the third obtaining unit is used for comparing the similarity with a preset threshold value and obtaining a comparison result of which the similarity is greater than the preset threshold value; and the selecting unit is used for selecting the comparison result with the highest similarity value in the comparison results with the similarity greater than the preset threshold value as the identification result of the target living body.
According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is executed by a processor, the apparatus in which the computer-readable storage medium is located is controlled to execute any one of the above methods for identifying a target living body.
According to another aspect of the embodiments of the present invention, there is also provided a processor, configured to execute a computer program, where the computer program executes to execute any one of the above methods for identifying a target living body.
In an embodiment of the present invention, position information of a target living body is acquired by using a first neural network, where the first neural network is obtained by machine learning training using a plurality of sets of first training data, and each of the plurality of sets of first training data includes: historical location information of the target living body; obtaining living body key point information of a target living body by utilizing a second neural network based on the position information, wherein the second neural network is obtained by using a plurality of groups of second training data through machine learning training, and each group of the second training data in the plurality of groups of second training data comprises: historical living body key point information of the target living body; and acquiring a score vector of the target living body based on the living body key point information, processing the historical score vector and the score vector in the database to acquire similarity, and acquiring an identification result of the target living body based on the similarity, wherein the score vector is generated by describing the living body key point information of the living body, and the identification result is used for representing whether the target living body is successfully identified. By the method for identifying the target living body, the aim of acquiring the feature vectors of multiple indexes based on the acquired key point information of the living body and then searching the vectors in the database to judge whether the detected target living body is accurate is fulfilled, so that the technical effect of improving the accuracy of visual identification of the target living body is achieved, and the technical problem of insufficient identification caused by incomplete feature extraction in the process of visual identification of the living body in the related technology is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
fig. 1 is a flowchart of a target living body identification method according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a target living body identification apparatus according to an embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
In accordance with an embodiment of the present invention, there is provided a method embodiment of a method for identifying a target living body, it should be noted that the steps illustrated in the flowchart of the accompanying drawings may be executed in a computer system such as a set of computer executable instructions, and that while a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be executed in an order different from that herein.
Fig. 1 is a flowchart of a method for identifying a target living body according to an embodiment of the present invention, as shown in fig. 1, the method including the steps of:
step S102, obtaining position information of a target living body by using a first neural network, where the first neural network is obtained by machine learning training using a plurality of sets of first training data, and each set of the first training data in the plurality of sets of first training data includes: historical location information of the target living body;
optionally, the living body position (i.e. the position information of the target living body) is detected first by using a target detection neural network, wherein the target detection neural network used includes, but is not limited to: YOLO networks, R-CNN series networks, SSDs, etc.
In addition, living organisms refer to living animals and plants, including but not limited to: living human, living animal.
Step S104, acquiring living body key point information of the target living body by utilizing a second neural network based on the position information, wherein the second neural network is obtained by using a plurality of groups of second training data through machine learning training, and each group of second training data in the plurality of groups of second training data comprises: historical living body key point information of the target living body;
it should be noted that, the living body detection model may use a neural network model, such as SSD, RetinaNet, YOLO series network, etc.; artificial design features may also be used in conjunction with machine learning models, such as HOG features in conjunction with SVM models.
And step S106, acquiring a score vector of the target living body based on the living body key point information, processing the historical score vector and the score vector in the database to acquire similarity, and acquiring an identification result of the target living body based on the similarity, wherein the score vector is generated by describing the living body key point information of the target living body, and the identification result is used for representing whether the target living body is successfully identified.
In the living body database, a plurality of living body characteristics and corresponding score vectors need to be stored in one living body. In the database searching process, a single feature W2 and a corresponding score vector F2 in the database are set. And calculating the similarity between the living characteristics to be inquired and the living characteristics of the database, and calculating by combining the score vector and the characteristic vector.
As can be seen from the above, in the embodiment of the present invention, first, the position information of the target living body may be acquired by using a first neural network, where the first neural network is obtained by machine learning training using a plurality of sets of first training data, and each of the plurality of sets of first training data includes: historical location information of the target living body; then, living body key point information of the target living body can be acquired by using a second neural network based on the position information, wherein the second neural network is obtained by machine learning training using a plurality of sets of second training data, and each set of the plurality of sets of second training data comprises: historical living body key point information of the target living body; and finally, acquiring a score vector of the target living body based on the living body key point information, processing the historical score vector and the score vector in the database to acquire a similarity, and acquiring an identification result of the target living body based on the similarity, wherein the score vector is generated by describing the living body key point information of the living body, and the identification result is used for representing whether the target living body is successfully identified. By the method for identifying the target living body, the aim of acquiring the key point information of the living body, acquiring the feature vectors of multiple aspect indexes based on the key point information, and searching the vectors in the database to judge whether the detected target living body is accurate is fulfilled, so that the technical effect of improving the accuracy of visual identification of the target living body is achieved, and the technical problem that identification is not accurate due to incomplete feature extraction in visual identification of the living body in the related technology is solved.
As an alternative embodiment, the living body key point information includes at least: position information of the five sense organs of the living body and position information of the joint points of the living body.
In the above optional embodiments, the human key points are extracted using a human key point extraction network, which includes but is not limited to: OpenPose, CPN, HRNet, etc.
The extracted living body key points include, but are not limited to, 17 key points, which are, respectively, a nose, left and right eyes, left and right ears, left and right shoulders, left and right elbows, left and right wrists, left and right hips, left and right knees, left and right ankles, and the like.
As an alternative embodiment, the score vector comprises: the mobile terminal comprises a brightness quality score, a definition score, a shielding degree score and a posture score, wherein the brightness quality score is a vector acquired based on the brightness quality of a region where a target living body is located, the definition score is a vector acquired based on the image definition of an acquired image of the target living body, the shielding degree score is a vector acquired based on the shielding degree of the target living body shielded by an interference object, and the posture score is a vector acquired based on the current posture of the target living body.
In the above alternative embodiment, a score vector may be calculated, where the score vector includes scores such as brightness, sharpness, occlusion degree, and pose.
(1) The luminance quality score may be based on specifying an appropriate luminance range interval, the closer the living body is within the specified interval to the interval center, the higher the score. One luminance quality score calculation formula is as follows:
where f1 is a brightness score, v is an average gray scale value of the living body image, and T1 and T2 are set thresholds, which are selected from T1-128 and T2-128.
It should be noted that the luminance quality score is calculated according to the luminance suitable interval, and other calculation methods such as a gaussian function may be used.
(2) The sharpness score calculation method can use a gradient algorithm, such as a Brenner gradient, a Laplacian gradient and the like, and the higher the gradient is, the clearer the image is, and the sharper the edges and corners are. One sharpness score calculation formula is as follows:
where f2 is the sharpness score, w, h are the width and height of the live image, x is the [0, w ] range, y is the [0, h ] range, and p (x, y) represents the pixel value of the [0,1] range.
(3) The occlusion degree score (i.e., the occlusion score value) can be used for classifying whether the identified living body is occluded by using a neural network classifier, and the last layer of the neural network can output an occlusion probability value as a softmax layer. The occlusion degree score is denoted as f 3.
(4) The pose score calculation is calculated from the live keypoint results. Specifically, a template method may be used, in which a standard front live body template is established, the positions of the key points of the live body template are set, the detected key points are scaled to the size of the template as a standard, for each key point, the distance value between the key point and the corresponding key point of the standard template is calculated, and the maximum value is truncated. For example, the detected key points of a certain joint are (x1, y1), the key points of a certain joint in the template are (x2, y2), and the calculation formula is as follows:
(5) the scores are combined to form a score vector and normalized by L2, denoted as F1 (i.e., the first score vector).
It should be noted that, when the recognition object is a human, the score vector may include brightness, sharpness, shielding degree, and pose score, and may also include a human-shaped clothing attribute type (including jacket, trousers, shoes, hat, mask, etc.), human-shaped skin color, human face features, and the like.
As an alternative embodiment, before processing the historical score vector and the score vector in the database, the method further includes: acquiring a first feature vector by using a third neural network; carrying out standardization processing on the score vector to obtain a first score vector; and processing the historical feature vector to obtain a historical score vector corresponding to the historical feature vector.
In the above alternative embodiment, the pedestrian feature is extracted using a feature extraction network (i.e., a third neural network), and ResNet, MobileNet, or the like may be used, wherein the extracted living body feature vector is denoted as W1 (i.e., a first feature vector).
As an alternative embodiment, the obtaining the first feature vector by using the third neural network includes: acquiring original characteristic information of the target living body through a third neural network based on the position information; and carrying out standardization processing on the original characteristic information to obtain a first characteristic vector.
As an alternative embodiment, processing the historical score vector and the score vector in the database to obtain the similarity includes: and obtaining the similarity by using a second formula based on the first feature vector, the first score vector, the historical feature vector and the historical score vector, wherein the second formula is as follows:
s is similarity, F1 is a first score vector, F2 is a history score vector, W1 is a first feature vector, and W2 is a history feature vector.
It should be noted that the module values of F1 (i.e., the first score vector), F2 (i.e., the historical score vector), W1 (i.e., the first feature vector), and W2 (i.e., the historical feature vector) are set to 1, the calculation result S is a similarity, the range is [0,1], and the larger S indicates that the living body to be queried has a higher similarity to the database. And finally, selecting the living body with the similarity greater than the set threshold and the highest similarity as the retrieval result, wherein the similarity threshold can be set to be 0.75.
As an alternative embodiment, acquiring the recognition result of the target living body based on the similarity includes: comparing the similarity with a preset threshold value to obtain a comparison result of which the similarity is greater than the preset threshold value; and selecting the comparison result with the highest similarity value in the comparison results with the similarity greater than the preset threshold value as the identification result of the target living body.
In the above optional embodiment, the obtained similarity is compared with a preset similarity threshold, a result that the similarity is greater than the preset similarity threshold is obtained, and the comparison result with the highest similarity data is selected from the comparison results as the identification result of the target living body, where the result indicates whether the identification is successful.
From the above, according to the identification method of the target living body, the detection neural network is used for detecting the human shape, the living body key point network is combined to extract the living body key point, and then the characteristic extraction network is used for extracting the human shape characteristic; then, calculating values such as brightness, definition, shielding degree, symmetry and the like; and finally, comparing the calculated score vector and the calculated feature vector with a database to obtain a retrieval result, and performing living body matching by combining the score vector and the feature vector of the target living body, so that the retrieval precision of the target living body is improved.
Example 2
According to another aspect of the embodiment of the present invention, there is also provided an identification apparatus of a living target, and fig. 2 is a schematic diagram of the identification apparatus of a living target according to the embodiment of the present invention, as shown in fig. 2, including: a first obtaining module 21, a second obtaining module 23 and a first processing module 25. The following describes the apparatus for identifying the target living body.
A first obtaining module 21, configured to obtain location information of a target living body by using a first neural network, where the first neural network is obtained through machine learning training using multiple sets of first training data, and each set of the multiple sets of first training data includes: historical location information of the target living body;
a second obtaining module 23, configured to obtain living body key point information of the target living body based on the location information and by using a second neural network, where the second neural network is obtained through machine learning training using multiple sets of second training data, and each set of the multiple sets of second training data includes: historical living body key point information of the target living body;
and the first processing module 25 is configured to obtain a score vector of the target living body based on the living body key point information, process the historical score vector and the score vector in the database to obtain a similarity, and obtain an identification result of the target living body based on the similarity, where the score vector is a vector generated by describing the living body key point information of the living body, and the identification result is used for representing whether the target living body has been successfully identified.
It should be noted that the first acquiring module 21, the second acquiring module 23 and the first processing module 25 correspond to steps S102 to S106 in embodiment 1, and the modules are the same as the corresponding steps in implementation example and application scenarios, but are not limited to the disclosure in embodiment 1. It should be noted that the modules described above as part of the apparatus may be implemented in a computer system such as a set of computer executable instructions.
As can be seen from the above, in the embodiment of the present invention, first, the first obtaining module 21 may obtain the position information of the target living body by using a first neural network, where the first neural network is obtained by machine learning training using a plurality of sets of first training data, and each of the plurality of sets of first training data includes: historical location information of the target living body; then, the second obtaining module 23 may be used to obtain the living body key point information of the target living body based on the position information and by using a second neural network, where the second neural network is obtained by machine learning training using a plurality of sets of second training data, and each set of the plurality of sets of second training data includes: historical living body key point information of the target living body; finally, the first processing module 25 may be utilized to obtain a score vector of the target living body based on the living body key point information, process the historical score vector and the score vector in the database to obtain a similarity, and obtain an identification result of the target living body based on the similarity, where the score vector is a vector generated by describing the living body key point information of the living body, and the identification result is used to represent whether the target living body has been successfully identified. By the method for identifying the target living body, the aim of acquiring the feature vectors of multiple indexes based on the acquired key point information of the living body and then searching the vectors in the database to judge whether the detected target living body is accurate is fulfilled, so that the technical effect of improving the accuracy of visual identification of the target living body is achieved, and the technical problem of insufficient identification caused by incomplete feature extraction in the process of visual identification of the living body in the related technology is solved.
Optionally, the living body keypoint information comprises at least: position information of human five sense organs and position information of human joint points.
Optionally, the score vector comprises: the mobile terminal comprises a brightness quality score, a definition score, a shielding degree score and a posture score, wherein the brightness quality score is a vector acquired based on the brightness quality of a region where a target living body is located, the definition score is a vector acquired based on the image definition of an acquired image of the target living body, the shielding degree score is a vector acquired based on the shielding degree of the target living body shielded by an interference object, and the posture score is a vector acquired based on the current posture of the target living body.
Optionally, the apparatus further comprises: the third acquisition module is used for acquiring the first feature vector by using a third neural network before processing the historical score vector and the score vector in the database; the standardization processing module is used for standardizing the score vector to obtain a first score vector; and the second processing module is used for processing the historical characteristic vector and acquiring a historical score vector corresponding to the historical characteristic vector.
Optionally, the third obtaining module includes: a first acquisition unit for acquiring original characteristic information of the target living body through a third neural network based on the position information; and the normalization processing unit is used for normalizing the original characteristic information to acquire a first characteristic vector.
Optionally, the first processing module includes: a second obtaining unit, configured to obtain a similarity using a second formula based on the first feature vector, the first score vector, the historical feature vector, and the historical score vector, where the second formula is:
s is similarity, F1 is a first score vector, F2 is a history score vector, W1 is a first feature vector, and W2 is a history feature vector.
Optionally, the first processing module includes: the third obtaining unit is used for comparing the similarity with a preset threshold value and obtaining a comparison result of which the similarity is greater than the preset threshold value; and the selecting unit is used for selecting the comparison result with the highest similarity value in the comparison results with the similarity larger than the preset threshold value as the identification result of the target living body.
Example 3
According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium including a stored computer program, wherein when the computer program is executed by a processor, the apparatus on which the computer-readable storage medium is located is controlled to execute any one of the above-mentioned methods for identifying a target living body.
Example 4
According to another aspect of the embodiments of the present invention, there is also provided a processor for executing a computer program, wherein the computer program executes to execute the method for identifying a target living body according to any one of the above methods.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described apparatus embodiments are merely illustrative, and for example, the division of the units may be a logical division, and in actual implementation, there may be another division, for example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or may not be executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (10)
1. A method for identifying a target living body, comprising:
acquiring position information of a target living body by using a first neural network, wherein the first neural network is obtained by machine learning training by using a plurality of groups of first training data, and each group of the plurality of groups of first training data comprises: historical location information of the target living body;
acquiring living body key point information of a target living body by utilizing a second neural network based on the position information, wherein the second neural network is obtained by using a plurality of groups of second training data through machine learning training, and each group of the second training data in the plurality of groups of second training data comprises: historical living body key point information of the target living body;
and acquiring a score vector of a target living body based on the living body key point information, processing a historical score vector in a database with the score vector to acquire a similarity, and acquiring an identification result of the target living body based on the similarity, wherein the score vector is generated by describing the living body key point information of the target living body, and the identification result is used for representing whether the target living body is successfully identified.
2. The method according to claim 1, wherein the living body keypoint information comprises at least: position information of the five sense organs of the living body and position information of the joint points of the living body.
3. The method of claim 1, wherein the score vector comprises: the mobile terminal comprises a brightness quality score, a definition score, an occlusion degree score and a posture score, wherein the brightness quality score is a vector acquired based on the brightness quality of a region where the target living body is located, the definition score is a vector acquired based on the image definition of an acquired image of the target living body, the occlusion degree score is a vector acquired based on the occlusion degree of the target living body occluded by an interfering object, and the posture score is a vector acquired based on the current posture of the target living body.
4. The method of claim 1, wherein prior to processing the historical score vector with the score vector in the database, the method further comprises:
acquiring a first feature vector by using a third neural network;
carrying out standardization processing on the score vector to obtain a first score vector;
and processing the historical characteristic vector to obtain the historical score vector corresponding to the historical characteristic vector.
5. The method of claim 4, wherein obtaining the first feature vector using a third neural network comprises:
acquiring original characteristic information of the target living body through the third neural network based on the position information;
and carrying out standardization processing on the original characteristic information to obtain the first characteristic vector.
6. The method of claim 4, wherein processing historical score vectors in a database with the score vectors to obtain similarities comprises:
obtaining the similarity by using a second formula based on the first feature vector, the first score vector, the historical feature vector and the historical score vector, wherein the second formula is as follows:
s is the similarity, F1 is the first score vector, F2 is the historical score vector, W1 is the first feature vector, and W2 is the historical feature vector.
7. The method according to claim 1, wherein acquiring the recognition result of the target living body based on the similarity comprises:
comparing the similarity with a preset threshold value to obtain a comparison result of which the similarity is greater than the preset threshold value;
and selecting the comparison result with the highest similarity value in the comparison results with the similarity greater than the preset threshold value as the identification result of the target living body.
8. An apparatus for identifying a target living body, comprising:
a first obtaining module, configured to obtain location information of a living target using a first neural network, where the first neural network is obtained through machine learning training using multiple sets of first training data, and each set of the multiple sets of first training data includes: historical location information of the target living body;
a second obtaining module, configured to obtain living body key point information of a target living body based on the location information and by using a second neural network, where the second neural network is obtained through machine learning training using multiple sets of second training data, and each set of the second training data in the multiple sets of second training data includes: historical living body key point information of the target living body;
the first processing module is used for acquiring a score vector of a target living body based on the living body key point information, processing a historical score vector in a database and the score vector to acquire a similarity, and acquiring an identification result of the target living body based on the similarity, wherein the score vector is generated by describing the living body key point information of the target living body, and the identification result is used for representing whether the target living body is successfully identified.
9. A computer-readable storage medium, comprising a stored computer program, wherein when the computer program is executed by a processor, the computer-readable storage medium controls an apparatus to execute the method for identifying a target living body according to any one of claims 1 to 7.
10. A processor for executing a computer program, wherein the computer program executes to perform the method for identifying a living target body according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210552507.2A CN114999003A (en) | 2022-05-20 | 2022-05-20 | Method and device for identifying target living body and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210552507.2A CN114999003A (en) | 2022-05-20 | 2022-05-20 | Method and device for identifying target living body and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114999003A true CN114999003A (en) | 2022-09-02 |
Family
ID=83026456
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210552507.2A Pending CN114999003A (en) | 2022-05-20 | 2022-05-20 | Method and device for identifying target living body and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114999003A (en) |
-
2022
- 2022-05-20 CN CN202210552507.2A patent/CN114999003A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111898547B (en) | Training method, device, equipment and storage medium of face recognition model | |
| Cheng et al. | Custom pictorial structures for re-identification. | |
| US10049262B2 (en) | Method and system for extracting characteristic of three-dimensional face image | |
| CN107346409B (en) | pedestrian re-identification method and device | |
| Zhang et al. | Fast and robust occluded face detection in ATM surveillance | |
| CN109389074B (en) | Facial feature point extraction-based expression recognition method | |
| CN109376604B (en) | Age identification method and device based on human body posture | |
| Dagnes et al. | Occlusion detection and restoration techniques for 3D face recognition: a literature review | |
| CN106897675A (en) | The human face in-vivo detection method that binocular vision depth characteristic is combined with appearance features | |
| CN110909618B (en) | Method and device for identifying identity of pet | |
| CN108171138B (en) | Biological characteristic information acquisition method and device | |
| CN113449704B (en) | Face recognition model training method and device, electronic equipment and storage medium | |
| WO2021174941A1 (en) | Physical attribute recognition method, system, computer device, and storage medium | |
| WO2019153175A1 (en) | Machine learning-based occluded face recognition system and method, and storage medium | |
| Almaadeed et al. | Partial shoeprint retrieval using multiple point-of-interest detectors and SIFT descriptors | |
| CN107844742A (en) | Facial image glasses minimizing technology, device and storage medium | |
| Xie et al. | A novel transient wrinkle detection algorithm and its application for expression synthesis | |
| Bąk et al. | Person re-identification using deformable patch metric learning | |
| Arif et al. | Human pose estimation and object interaction for sports behaviour | |
| CN114821786A (en) | Gait recognition method based on human body contour and key point feature fusion | |
| CN116664585B (en) | Scalp health condition detection method and related device based on deep learning | |
| CN111723688B (en) | Human body action recognition result evaluation method and device and electronic equipment | |
| CN116129473B (en) | Identity-guide-based combined learning clothing changing pedestrian re-identification method and system | |
| Das | Human’s facial parts extraction to recognize facial expression | |
| CN114999003A (en) | Method and device for identifying target living body and computer readable storage medium |
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 |