CN110717357A - Early warning method and device, electronic equipment and storage medium - Google Patents
Early warning method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110717357A CN110717357A CN201810762447.0A CN201810762447A CN110717357A CN 110717357 A CN110717357 A CN 110717357A CN 201810762447 A CN201810762447 A CN 201810762447A CN 110717357 A CN110717357 A CN 110717357A
- Authority
- CN
- China
- Prior art keywords
- face image
- recognized
- warned
- early
- matched
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the invention provides an early warning method, an early warning device, electronic equipment and a storage medium, which are applied to the technical field of video monitoring, wherein if a white list face image matched with a face image to be recognized does not exist, a person corresponding to the face image to be recognized does not belong to a person in a white list, and the face image to be recognized is compared with the face image to be early warned; when the occurrence frequency of the to-be-early-warned face image matched with the to-be-recognized face image is larger than a preset frequency threshold value, it is indicated that a person corresponding to the to-be-recognized face image appears for multiple times in the current monitoring scene, and a pedestrian corresponding to the to-be-recognized face image may be a potential customer or a suspicious person, so that early warning for the to-be-early-warned face image matched with the to-be-recognized face image is generated. By the early warning method, automatic early warning for personnel can be realized.
Description
Technical Field
The invention relates to the technical field of video monitoring, in particular to an early warning method, an early warning device, electronic equipment and a storage medium.
Background
For scenes such as large shopping malls, shopping centers, chain stores, airports, stations, museums, exhibition halls and the like, the flow of people is large, and the components of people are complex. For regulatory purposes, it is desirable to be able to identify suspicious people at the location or to identify potential customers at the location.
Therefore, it is desirable to be able to warn the personnel in the aforementioned locations.
Disclosure of Invention
The embodiment of the invention aims to provide an early warning method, an early warning device, electronic equipment and a storage medium so as to realize automatic early warning of personnel. The specific technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides an early warning method, where the method includes:
acquiring a face image to be recognized in a monitoring scene;
comparing the face image to be recognized with a white list face image in a preset white list, and judging whether a white list face image matched with the face image to be recognized exists or not;
if the white list face image matched with the face image to be recognized does not exist, comparing the face image to be recognized with the face image to be early-warned in an early-warning list, and judging whether the face image to be early-warned matched with the face image to be recognized exists or not;
if a to-be-early-warned face image matched with the to-be-warned face image exists, acquiring the occurrence time interval of the to-be-early-warned face image matched with the to-be-warned face image, wherein the occurrence time interval of the to-be-early-warned face image represents a pedestrian corresponding to the to-be-early-warned face image and repeatedly appears in the time interval of the monitoring scene aiming at any to-be-early-warned face image;
judging whether the occurrence time interval is smaller than a preset interval threshold value or not;
if the appearance time interval is not smaller than the preset interval threshold, increasing the appearance times of the face image to be early-warned matched with the face image to be recognized according to a preset time increasing rule;
and when the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is greater than a preset frequency threshold value, generating early warning aiming at the face image to be early-warned matched with the face image to be recognized.
Optionally, the acquiring a to-be-recognized face image in a monitoring scene includes:
acquiring face images of each pedestrian newly appearing in a monitoring scene;
judging whether each face image accords with a preset recognizable rule or not;
and taking the face image which accords with the recognizable rule as a face image to be recognized.
Optionally, after the determining whether the to-be-early-warned face image matched with the to-be-recognized face image exists, the method further includes:
and if the face image to be pre-warned which is matched with the face image to be recognized does not exist, taking the face image to be recognized as a new face image to be pre-warned, and adding the new face image to be pre-warned into the pre-warning list.
Optionally, after the determining whether the occurrence time interval is smaller than a preset interval threshold, the method further includes:
and if the occurrence time interval is smaller than the preset interval threshold, storing the face image to be recognized into a snapshot library.
In a second aspect, an embodiment of the present invention provides an early warning apparatus, where the apparatus includes:
the image acquisition module is used for acquiring a face image to be recognized in a monitoring scene;
the face comparison module is used for comparing the face image to be recognized with a white list face image in a preset white list and judging whether a white list face image matched with the face image to be recognized exists or not;
the first judgment module is used for comparing the face image to be recognized with the face image to be early-warned in an early warning list if the white list face image matched with the face image to be recognized does not exist, and judging whether the face image to be early-warned matched with the face image to be recognized exists or not;
the time acquisition module is used for acquiring the appearance time interval of the to-be-early-warned face image matched with the to-be-warned face image if the to-be-early-warned face image matched with the to-be-warned face image exists, wherein the appearance time interval of the to-be-early-warned face image represents the pedestrian corresponding to the to-be-early-warned face image aiming at any to-be-early-warned face image, and the time interval appears in the monitoring scene repeatedly;
the second judgment module is used for judging whether the occurrence time interval is smaller than a preset interval threshold value or not;
the frequency increasing module is used for increasing the frequency of the face image to be pre-warned matched with the face image to be recognized according to a preset frequency increasing rule if the appearance time interval is not smaller than the preset interval threshold;
and the early warning generation module is used for generating early warning aiming at the face image to be early warned matched with the face image to be recognized when the occurrence frequency of the face image to be early warned matched with the face image to be recognized is greater than a preset frequency threshold value.
Optionally, the image obtaining module includes:
the face image acquisition sub-module is used for acquiring face images of all people newly appearing in the monitoring scene;
the recognizable judging submodule is used for judging whether each face image accords with a preset recognizable rule or not;
and the facial image to be recognized determining submodule is used for taking the facial image which accords with the recognizable rule as the facial image to be recognized.
Optionally, the early warning apparatus according to the embodiment of the present invention further includes:
and the early warning list updating module is used for taking the face image to be recognized as a new face image to be early warned and adding the new face image to be early warned into the early warning list if the face image to be early warned matched with the face image to be recognized does not exist.
Optionally, the early warning apparatus according to the embodiment of the present invention further includes:
and the snapshot library storage module is used for storing the face image to be recognized into the snapshot library if the occurrence time interval is smaller than the preset interval threshold.
In a third aspect, an embodiment of the present invention provides an electronic device, including a processor and a memory;
the memory is used for storing a computer program;
the processor is configured to implement the warning method according to any one of the first aspect when executing the program stored in the memory.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements the warning method according to any one of the first aspects.
The early warning method, the early warning device, the electronic equipment and the storage medium provided by the embodiment of the invention are used for acquiring a face image to be recognized in a monitoring scene; comparing the face image to be recognized with a white list face image in a preset white list, and judging whether a white list face image matched with the face image to be recognized exists or not; if the white list face image matched with the face image to be recognized does not exist, comparing the face image to be recognized with the face image to be early-warned in the early-warning list, and judging whether the face image to be early-warned matched with the face image to be recognized exists or not; if a to-be-early-warned face image matched with the to-be-warned face image exists, acquiring the occurrence time interval of the to-be-early-warned face image matched with the to-be-warned face image, wherein the occurrence time interval of the to-be-early-warned face image represents a pedestrian corresponding to the to-be-early-warned face image and repeatedly appears in a monitoring scene; judging whether the occurrence time interval is smaller than a preset interval threshold value or not; if the appearance time interval is not smaller than the preset interval threshold, increasing the appearance times of the face image to be early-warned matched with the face image to be recognized according to a preset time increasing rule; and when the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is greater than a preset frequency threshold value, generating early warning aiming at the face image to be early-warned matched with the face image to be recognized. The method comprises the steps that a white list face image matched with a face image to be recognized does not exist, the fact that a person corresponding to the face image to be recognized does not belong to a person in the white list is shown, when the occurrence frequency of a face image to be early-warned matched with the face image to be recognized is larger than a preset frequency threshold value, the fact that the person corresponding to the face image to be recognized appears for multiple times in a current monitoring scene is shown, a pedestrian corresponding to the face image to be recognized may be a potential client or a suspicious person, early warning for the face image to be early-warned matched with the face image to be recognized is generated, and automatic early warning for the person can be. Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of an early warning method according to an embodiment of the present invention;
fig. 2 is another schematic flow chart of an early warning method according to an embodiment of the present invention;
fig. 3 is a schematic diagram of an early warning device according to an embodiment of the present invention;
fig. 4 is a schematic diagram of an electronic device according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In large shopping malls, shopping centers, chain stores, airports, stations, museums, exhibition halls and other scenes, the flow of people is large, the components of personnel are complex, and thieves, potential customers or returning passengers and the like may exist, so that people in the scenes are expected to be warned.
In view of this, an embodiment of the present invention provides an early warning method, and referring to fig. 1, the method includes:
s101, obtaining a face image to be recognized in a monitored scene.
The early warning method in the embodiment of the invention can be realized by an early warning system, and the early warning system is any system capable of realizing the early warning method in the embodiment of the invention. For example:
the early warning system may be an electronic device comprising: a processor, a memory, a communication interface, and a bus; the processor, the memory and the communication interface are connected through a bus and complete mutual communication; the memory stores executable program code; the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the early warning method of the embodiment of the invention.
The early warning system can also be an application program used for executing the early warning method of the embodiment of the invention when running.
The early warning system may also be a storage medium for storing executable code for performing the early warning method of the embodiments of the present invention.
The early warning system acquires a face image to be recognized in the current monitoring scene through image acquisition equipment, such as a camera, in the current monitoring scene.
And S102, comparing the face image to be recognized with a white list face image in a preset white list, and judging whether a white list face image matched with the face image to be recognized exists or not.
The face images of the people in the white list, namely the face images in the white list, are stored in the preset white list. The white list personnel can be staff in a monitoring scene, and the early warning system cannot give early warning to the white list personnel. The early warning system compares the face image to be recognized with each white list face image through a preset face recognition algorithm, judges whether the white list face image matched with the face image to be recognized exists or not, and namely judges whether the pedestrian corresponding to the face image to be recognized is a white list person or not. For example, the face image to be recognized and the white list face image are input into a pre-trained neural network for face recognition, and whether the face image to be recognized and the white list face image are matched is judged. Or based on the comparison of the face models, calculating the similarity between the face image to be recognized and the white list face image, and when the similarity is greater than a preset similarity threshold, judging that the face image to be recognized is matched with the white list face image, wherein the preset similarity threshold can be set according to actual conditions, for example, the preset similarity threshold is 70%, 80% or 90%.
And S103, if no white list face image matched with the face image to be recognized exists, comparing the face image to be recognized with the face image to be early-warned in an early-warning list, and judging whether a face image to be early-warned matched with the face image to be recognized exists or not.
The early warning list stores the face image of the person to be early warned, namely the face image to be early warned. The early warning system carries out model comparison on the face image to be recognized and each face image to be early warned through a preset face recognition algorithm, judges whether the face image to be early warned matched with the face image to be recognized exists or not, and namely judges whether a pedestrian corresponding to the face image to be recognized is a person to be early warned or not.
For example, a convolutional neural network which is subjected to face recognition training in advance is used for carrying out model comparison on a face image to be recognized and each face image to be pre-warned, and judging whether a face image to be pre-warned matched with the face image to be recognized exists or not; or the target detection method based on the image segmentation technology is used for comparing the face image to be recognized with each face image to be early-warned in a model manner and judging whether the face image to be early-warned matched with the face image to be recognized exists or not; or the target detection method based on the characteristic matching technology is used for comparing the facial image to be recognized with each facial image to be pre-warned, judging whether the facial image to be pre-warned matched with the facial image to be recognized exists or not, and the like.
And S104, if a to-be-early-warned face image matched with the to-be-warned face image exists, acquiring the appearance time interval of the to-be-early-warned face image matched with the to-be-warned face image, wherein the appearance time interval of the to-be-early-warned face image represents a pedestrian corresponding to the to-be-early-warned face image and repeatedly appears in the time interval of the monitoring scene aiming at any to-be-early-warned face image.
Representing the appearance time interval of the face image to be pre-warned, and repeatedly showing the pedestrian in the monitoring scene by the face image to be pre-warned. For example, for the monitoring scene a, the time that the pedestrian a appears in the monitoring scene a at the nth time is 08:30, after the pedestrian a leaves the monitoring scene a, the time that the pedestrian a appears in the monitoring scene a at the (N + 1) th time is 09:30, when the face image to be recognized of the pedestrian a is obtained at the (N + 1) th time, the appearance time interval of the early warning face image corresponding to the pedestrian a is 1 hour, wherein N is a positive integer.
Specifically, the occurrence time interval of the face image to be pre-warned may be: the difference value of the current time and the time for updating the occurrence frequency of the face image to be early-warned for the last time; or the occurrence time interval of the face image to be pre-warned can be as follows: the time for acquiring the face image to be pre-warned corresponding to the face image to be pre-warned at this time, the difference value between the time for acquiring the face image to be pre-warned corresponding to the face image to be pre-warned at the last time, and the like.
And S105, judging whether the appearance time interval is smaller than a preset interval threshold value.
The preset interval threshold may be set according to actual conditions, for example, the preset interval threshold may be set to 30 seconds, 1 minute, 3 minutes, or the like. The early warning system judges whether the occurrence time interval is smaller than a preset interval threshold value.
And S106, if the appearance time interval is not smaller than the preset interval threshold, increasing the appearance times of the face image to be early-warned matched with the face image to be recognized according to a preset time increasing rule.
The preset number increasing rule is a rule for arbitrarily increasing the occurrence number of the face image to be pre-warned, for example, if the time interval is not less than a preset interval threshold, the occurrence number of the face image to be pre-warned, which is matched with the face image to be recognized, is increased by 1.
And S107, when the occurrence frequency of the face image to be pre-warned matched with the face image to be recognized is greater than a preset frequency threshold value, generating pre-warning aiming at the face image to be pre-warned matched with the face image to be recognized.
The preset time threshold may be set according to actual conditions, for example, the preset time threshold may be set to 3, 5, or 10, etc. And if the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is greater than a preset frequency threshold value, the early warning system generates early warning aiming at the face image to be early-warned matched with the face image to be recognized. According to the early warning, a security guard or a salesperson and the like pay attention to the personnel corresponding to the face image to be early warned, so that the occurrence of theft events is reduced, or sales performance is increased for potential customers.
In the embodiment of the invention, a white list face image matched with the face image to be recognized does not exist, which indicates that the person corresponding to the face image to be recognized does not belong to the person in the white list, and when the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is greater than a preset frequency threshold value, which indicates that the person corresponding to the face image to be recognized appears for multiple times in the current monitoring scene, the pedestrian corresponding to the face image to be recognized may be a potential customer or a suspicious person, and generates early warning aiming at the face image to be early-warned matched with the face image to be recognized, so that automatic early warning aiming at the person can be realized.
Optionally, the obtaining of the facial image to be recognized in the monitoring scene includes:
acquiring a face image of each pedestrian newly appearing in a monitoring scene;
the early warning system acquires the face images of each pedestrian newly appearing in the monitoring scene through image acquisition equipment such as a camera in the monitoring scene. For example, the early warning system detects newly appearing pedestrians in the monitoring image of the current monitoring scene through a preset recognition algorithm, and extracts face images of the newly appearing pedestrians. The newly appeared pedestrians refer to pedestrians newly entering the monitoring scene, including the pedestrian entering the current monitoring scene for the first time, and also including the pedestrian entering the current monitoring scene again after leaving the current monitoring scene. Optionally, the early warning system may detect a newly-appearing pedestrian only for an image at an entrance or an exit of the monitored scene, so as to save computational resources.
Step two, judging whether each face image accords with a preset recognizable rule;
the preset recognizable rule can be set according to actual conditions, for example, the resolution of the face image is greater than a preset resolution threshold, or whether the score of the face image is greater than a preset score threshold or not is judged. The early warning system carries out face modeling on the face image, scores the modeled face image, and the scoring rule is that the face image is compared with a three-dimensional reference database of the face one by one to obtain parameters such as pupil distance, pitch angle, left and right angles and the like, and the quality score of the face image is obtained according to the parameters. And judging the face image with the score higher than a preset score threshold value, and conforming to a preset recognizable rule. And judging the face image with the score not higher than the preset score threshold value, wherein the face image does not conform to the preset recognizable rule.
And step three, taking the face image which accords with the recognizable rule as a face image to be recognized.
The early warning system takes the face image which accords with the recognizable rule as the face image to be recognized. And discarding or storing the face image which does not accord with the recognizable rule to a specified position. Optionally, the face images which do not accord with the preset recognizable rules are added into a snapshot library to facilitate subsequent query and analysis.
In the embodiment of the invention, the face image which accords with the recognizable rule is taken as the face image to be recognized, so that the success rate of comparing the subsequent face images to be recognized can be improved.
Optionally, after determining whether the to-be-early-warned face image matched with the to-be-recognized face image exists, the method further includes:
and if the face image to be pre-warned which is matched with the face image to be recognized does not exist, taking the face image to be recognized as a new face image to be pre-warned, and adding the new face image to be pre-warned into the pre-warning list.
And if the face image to be pre-warned does not exist, the person corresponding to the face image to be pre-warned appears in the current monitoring scene for the first time, and the pre-warning system takes the face image to be pre-warned as a new face image to be pre-warned and adds the new face image to be pre-warned into the pre-warning list. Optionally, each person to be pre-warned corresponding to the face image to be pre-warned in the pre-warning list corresponds to a unique ID, so as to conveniently count the occurrence frequency of the person to be pre-warned.
In the embodiment of the invention, if the face image to be pre-warned does not exist, which is matched with the face image to be recognized, the face image to be recognized is taken as a new face image to be pre-warned and is added into the pre-warning list, so that the follow-up comparison of the personnel newly added into the monitoring scene can be supported.
Optionally, after the determining whether the occurrence time interval is smaller than a preset interval threshold, the method further includes:
and if the appearance time interval is smaller than the preset interval threshold, storing the face image to be recognized into a snapshot library.
The occurrence time interval is smaller than a preset interval threshold, which indicates that a person corresponding to the face image to be recognized may not be acquired in a short time due to the fact that the person is shielded by an object or pacing is conducted at the edge of the current monitored scene, so that the occurrence frequency of the face image to be pre-warned matched with the face image to be recognized is not increased when the time interval is smaller than the preset interval threshold, and the face image to be recognized is directly stored in the snapshot library.
In the embodiment of the invention, if the occurrence time interval is smaller than the preset interval threshold, the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is not increased, but the face image to be recognized is stored in the snapshot library, so that the face image to be recognized is conveniently read and analyzed from the snapshot library subsequently, the invalid increase of the occurrence frequency can be reduced, and the condition of false early warning is reduced.
The statistics of personnel information and early warning have significance for a plurality of scenes, the method for judging whether the same person appears for a plurality of times in the early warning method implemented by the invention is based on the comparison of face models, and if the similarity of the comparison of the two models is greater than a set similarity threshold, the person is judged to be the same person. The function is mainly used for mining potential abnormal personnel when a user cannot provide photos of concerned people in advance, such as potential inertial theft and high-risk personnel, and providing auxiliary decision for the user to determine suspicious personnel, and realizing early warning and control of the people of the concerned people.
Another flow diagram of the early warning method implemented by the present invention is shown in fig. 2. The early warning method implemented by the invention adopts a face detection algorithm, a face grading algorithm, a face modeling and a face model comparison algorithm to firstly identify faces of pictures grabbed by a face grabber in real time, then sends face grabbed pictures with face coordinates to a face modeling module to carry out face modeling, sends the face grabbed pictures with the models to the face grading module to be compared with a three-dimensional reference database of the faces one by one to obtain parameters such as pupil distance, pitch angle, left and right angles and the like, and obtains whether the quality of the face grabbed pictures is high or low according to the parameters. And (4) for the low-score face snapshot picture, directly storing the snapshot picture into a face snapshot library of the system without a function of personnel frequency statistics, and ending the process. If the quality of the snapshot image is high at the moment, if the pitch angle and the left and right angles are moderate, the frequency statistics of the personnel is carried out, the face scoring is adopted in combination to improve the accuracy of the frequency statistics, and the face with low scoring quality of a certain low head or side face is not counted.
And comparing the modeling data of the face snapshot picture with the set white list library one by one, if the snapshot face indicates that the person is an internal person in the white list library, directly storing the snapshot picture into the face snapshot library of the system without counting, and ending the process. If the name is not in the white list library, the name is described as a stranger, and the name is the group of people needing important attention in the scheme.
In the early warning method implemented by the invention, a database of personnel frequency information needs to be maintained, and the unique ID, storage position, frequency, time of each occurrence and face model data corresponding to the snapshot are maintained in the database. And when the snapshot is identified as a stranger through comparison of the white list library, performing model one-to-one comparison with each face ID maintained in the personnel frequency database, judging that the snapshot is the same person if the similarity of the two faces is greater than the specified similarity, judging the interval between the real-time snapshot and the latest snapshot time corresponding to the successfully-compared face ID in the personnel frequency database, if the interval is within the set snapshot interval, not updating the times, only updating the latest snapshot time in the personnel frequency database, storing the snapshot in the snapshot library, and ending the process. If the interval between the face ID and the face ID is larger than the designated snapshot interval, updating the frequency of the face ID in the personnel frequency database, adding 1 to the frequency, updating the time of the latest snapshot image, judging whether the frequency corresponding to the face ID is larger than a set frequency threshold or not at the moment, and if the frequency in the designated time is larger than the set frequency threshold, performing early warning processing; if the frequency does not reach the set frequency threshold, no processing is performed, the process of the snapshot image is finished, and the next snapshot image of the face snapshot machine is waited.
In the process, for each face snapshot, one ID in the scheme corresponding to one face snapshot corresponds to the frequency of occurrence, and the frequency is used for judging the early warning processing, so that the real-time statistics and early warning processing of the frequency information of the personnel are realized.
In the early warning method implemented by the invention, parameters can be configured according to actual requirements, such as: the alarm frequency threshold value of the personnel appears, the snapshot interval for counting the occurrence frequency is set, the white list library personnel are set, whether the similarity threshold value belongs to the same person is judged, the time and the number of days for monitoring are designated, and even which monitoring points are subjected to personnel counting and early warning can be selected.
The early warning method provided by the embodiment of the invention can be used for counting the occurrence frequency information of the personnel in the specified time period, carrying out early warning treatment on the personnel repeatedly occurring for a certain number of times, and can be used for specifying the white list, namely, the white list personnel which can not trigger early warning, such as internal personnel and the like. The number of times of occurrence is accumulated, a snapshot interval concept is provided, if the number of times of occurrence is 1, the number of times of occurrence is considered to be 1, the setting of the snapshot interval enables the counted personnel to be suitable for different scenes according to the length change of the snapshot interval, and the number of times of occurrence of the counted personnel is more averaged. And carrying out early warning treatment on the personnel reaching a certain number of times, and informing corresponding security guards or managers to pay key attention to the appointed personnel.
An embodiment of the present invention further provides an early warning apparatus, referring to fig. 3, the apparatus includes:
the image acquisition module 301 is configured to acquire a face image to be recognized in a monitored scene;
a face comparison module 302, configured to compare the to-be-recognized face image with a white list face image in a preset white list, and determine whether a white list face image matching the to-be-recognized face image exists;
a first judging module 303, configured to, if there is no white list face image matching the face image to be recognized, compare the face image to be recognized with a face image to be early-warned in an early-warning list, and judge whether there is a face image to be early-warned matching the face image to be recognized;
a time obtaining module 304, configured to obtain an occurrence time interval of a to-be-early-warned face image matched with the to-be-warned face image if the to-be-early-warned face image matched with the to-be-warned face image exists, where, for any to-be-early-warned face image, the occurrence time interval of the to-be-early-warned face image represents a pedestrian corresponding to the to-be-early-warned face image, and repeatedly appears in the time interval of the monitoring scene;
a second determining module 305, configured to determine whether the occurrence time interval is smaller than a preset interval threshold;
a frequency increasing module 306, configured to increase, according to a preset frequency increasing rule, the frequency of occurrence of the to-be-early-warned face image matching the to-be-recognized face image if the occurrence time interval is not smaller than the preset interval threshold;
and an early warning generation module 307, configured to generate an early warning for the to-be-early-warned face image matched with the to-be-recognized face image when the occurrence frequency of the to-be-early-warned face image matched with the to-be-recognized face image is greater than a preset frequency threshold.
In the embodiment of the invention, a white list face image matched with the face image to be recognized does not exist, which indicates that the person corresponding to the face image to be recognized does not belong to the person in the white list, and when the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is greater than a preset frequency threshold value, which indicates that the person corresponding to the face image to be recognized appears for multiple times in the current monitoring scene, the pedestrian corresponding to the face image to be recognized may be a potential customer or a suspicious person, and generates early warning aiming at the face image to be early-warned matched with the face image to be recognized, so that automatic early warning aiming at the person can be realized.
Optionally, the image obtaining module 301 includes:
the face image acquisition sub-module is used for acquiring face images of all people newly appearing in the monitoring scene;
the recognizable judging submodule is used for judging whether each face image accords with a preset recognizable rule or not;
and the facial image to be recognized determining submodule is used for taking the facial image which accords with the recognizable rule as the facial image to be recognized. .
In the embodiment of the invention, the face image which accords with the recognizable rule is taken as the face image to be recognized, so that the success rate of comparing the subsequent face images to be recognized can be improved.
Optionally, the early warning apparatus according to the embodiment of the present invention further includes:
and the early warning list updating module is used for taking the face image to be recognized as a new face image to be early warned and adding the new face image to be early warned into the early warning list if the face image to be early warned matched with the face image to be recognized does not exist.
In the embodiment of the invention, if the face image to be pre-warned does not exist, which is matched with the face image to be recognized, the face image to be recognized is taken as a new face image to be pre-warned and is added into the pre-warning list, so that the follow-up comparison of the personnel newly added into the monitoring scene can be supported.
Optionally, the early warning apparatus according to the embodiment of the present invention further includes:
and the snapshot library storage module is used for storing the face image to be recognized into the snapshot library if the appearance time interval is smaller than the preset interval threshold.
In the embodiment of the invention, if the occurrence time interval is smaller than the preset interval threshold, the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is not increased, but the face image to be recognized is stored in the snapshot library, so that the face image to be recognized is conveniently read and analyzed from the snapshot library subsequently, the invalid increase of the occurrence frequency can be reduced, and the condition of false early warning is reduced.
An embodiment of the present invention further provides an electronic device, as shown in fig. 4, including a processor 401 and a memory 402;
a memory 402 for storing a computer program;
the processor 401, when executing the program stored in the memory 402, implements the following steps:
acquiring a face image to be recognized in a monitoring scene;
comparing the face image to be recognized with a white list face image in a preset white list, and judging whether a white list face image matched with the face image to be recognized exists or not;
if the white list face image matched with the face image to be recognized does not exist, comparing the face image to be recognized with the face image to be early-warned in an early-warning list, and judging whether the face image to be early-warned matched with the face image to be recognized exists or not;
if a to-be-early-warned face image matched with the to-be-warned face image exists, acquiring the occurrence time interval of the to-be-early-warned face image matched with the to-be-warned face image, wherein the occurrence time interval of the to-be-early-warned face image represents a pedestrian corresponding to the to-be-early-warned face image and repeatedly appears in the time interval of the monitoring scene aiming at any to-be-early-warned face image;
judging whether the occurrence time interval is smaller than a preset interval threshold value or not;
if the appearance time interval is not smaller than the preset interval threshold, increasing the appearance times of the face images to be early-warned matched with the face images to be recognized according to a preset time increasing rule;
and when the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is greater than a preset frequency threshold value, generating early warning aiming at the face image to be early-warned matched with the face image to be recognized.
In the embodiment of the invention, a white list face image matched with the face image to be recognized does not exist, which indicates that the person corresponding to the face image to be recognized does not belong to the person in the white list, and when the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is greater than a preset frequency threshold value, which indicates that the person corresponding to the face image to be recognized appears for multiple times in the current monitoring scene, the pedestrian corresponding to the face image to be recognized may be a potential customer or a suspicious person, and generates early warning aiming at the face image to be early-warned matched with the face image to be recognized, so that automatic early warning aiming at the person can be realized.
Optionally, the processor 401, when configured to execute the program stored in the memory 402, is further configured to implement any of the above-described warning methods.
Optionally, the electronic device implemented by the present invention further includes: a communication interface and a communication bus, through which the processor 401, the communication interface and the memory 402 communicate with each other.
The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the electronic equipment and other equipment.
The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component.
A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when executed by a processor, the computer program implements the steps of:
acquiring a face image to be recognized in a monitoring scene;
comparing the face image to be recognized with a white list face image in a preset white list, and judging whether a white list face image matched with the face image to be recognized exists or not;
if the white list face image matched with the face image to be recognized does not exist, comparing the face image to be recognized with the face image to be early-warned in an early-warning list, and judging whether the face image to be early-warned matched with the face image to be recognized exists or not;
if a to-be-early-warned face image matched with the to-be-warned face image exists, acquiring the occurrence time interval of the to-be-early-warned face image matched with the to-be-warned face image, wherein the occurrence time interval of the to-be-early-warned face image represents a pedestrian corresponding to the to-be-early-warned face image and repeatedly appears in the time interval of the monitoring scene aiming at any to-be-early-warned face image;
judging whether the occurrence time interval is smaller than a preset interval threshold value or not;
if the appearance time interval is not smaller than the preset interval threshold, increasing the appearance times of the face images to be early-warned matched with the face images to be recognized according to a preset time increasing rule;
and when the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is greater than a preset frequency threshold value, generating early warning aiming at the face image to be early-warned matched with the face image to be recognized.
In the embodiment of the invention, a white list face image matched with the face image to be recognized does not exist, which indicates that the person corresponding to the face image to be recognized does not belong to the person in the white list, and when the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is greater than a preset frequency threshold value, which indicates that the person corresponding to the face image to be recognized appears for multiple times in the current monitoring scene, the pedestrian corresponding to the face image to be recognized may be a potential customer or a suspicious person, and generates early warning aiming at the face image to be early-warned matched with the face image to be recognized, so that automatic early warning aiming at the person can be realized.
Optionally, the computer program, when executed by the processor, is further capable of implementing any of the above-mentioned warning methods.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the embodiments of the apparatus, the electronic device, and the storage medium, since they are substantially similar to the method embodiments, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiments.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. An early warning method, characterized in that the method comprises:
acquiring a face image to be recognized in a monitoring scene;
comparing the face image to be recognized with a white list face image in a preset white list, and judging whether a white list face image matched with the face image to be recognized exists or not;
if the white list face image matched with the face image to be recognized does not exist, comparing the face image to be recognized with the face image to be early-warned in an early-warning list, and judging whether the face image to be early-warned matched with the face image to be recognized exists or not;
if a to-be-early-warned face image matched with the to-be-warned face image exists, acquiring the occurrence time interval of the to-be-early-warned face image matched with the to-be-warned face image, wherein the occurrence time interval of the to-be-early-warned face image represents a pedestrian corresponding to the to-be-early-warned face image and repeatedly appears in the time interval of the monitoring scene aiming at any to-be-early-warned face image;
judging whether the occurrence time interval is smaller than a preset interval threshold value or not;
if the appearance time interval is not smaller than the preset interval threshold, increasing the appearance times of the face image to be early-warned matched with the face image to be recognized according to a preset time increasing rule;
and when the occurrence frequency of the face image to be early-warned matched with the face image to be recognized is greater than a preset frequency threshold value, generating early warning aiming at the face image to be early-warned matched with the face image to be recognized.
2. The method according to claim 1, wherein the acquiring the face image to be recognized in the monitored scene comprises:
acquiring face images of each pedestrian newly appearing in a monitoring scene;
judging whether each face image accords with a preset recognizable rule or not;
and taking the face image which accords with the recognizable rule as a face image to be recognized.
3. The method according to claim 1, wherein after the determining whether there is a to-be-pre-warned face image matching the to-be-recognized face image, the method further comprises:
and if the face image to be pre-warned which is matched with the face image to be recognized does not exist, taking the face image to be recognized as a new face image to be pre-warned, and adding the new face image to be pre-warned into the pre-warning list.
4. The method of claim 1, wherein after said determining whether said time interval of occurrence is less than a preset interval threshold, said method further comprises:
and if the occurrence time interval is smaller than the preset interval threshold, storing the face image to be recognized into a snapshot library.
5. An early warning device, the device comprising:
the image acquisition module is used for acquiring a face image to be recognized in a monitoring scene;
the face comparison module is used for comparing the face image to be recognized with a white list face image in a preset white list and judging whether a white list face image matched with the face image to be recognized exists or not;
the first judgment module is used for comparing the face image to be recognized with the face image to be early-warned in an early warning list if the white list face image matched with the face image to be recognized does not exist, and judging whether the face image to be early-warned matched with the face image to be recognized exists or not;
the time acquisition module is used for acquiring the appearance time interval of the to-be-early-warned face image matched with the to-be-warned face image if the to-be-early-warned face image matched with the to-be-warned face image exists, wherein the appearance time interval of the to-be-early-warned face image represents the pedestrian corresponding to the to-be-early-warned face image aiming at any to-be-early-warned face image, and the time interval appears in the monitoring scene repeatedly;
the second judgment module is used for judging whether the occurrence time interval is smaller than a preset interval threshold value or not;
the frequency increasing module is used for increasing the frequency of the face image to be pre-warned matched with the face image to be recognized according to a preset frequency increasing rule if the appearance time interval is not smaller than the preset interval threshold;
and the early warning generation module is used for generating early warning aiming at the face image to be early warned matched with the face image to be recognized when the occurrence frequency of the face image to be early warned matched with the face image to be recognized is greater than a preset frequency threshold value.
6. The apparatus of claim 5, wherein the image acquisition module comprises:
the face image acquisition sub-module is used for acquiring face images of all people newly appearing in the monitoring scene;
the recognizable judging submodule is used for judging whether each face image accords with a preset recognizable rule or not;
and the facial image to be recognized determining submodule is used for taking the facial image which accords with the recognizable rule as the facial image to be recognized.
7. The apparatus of claim 5, further comprising:
and the early warning list updating module is used for taking the face image to be recognized as a new face image to be early warned and adding the new face image to be early warned into the early warning list if the face image to be early warned matched with the face image to be recognized does not exist.
8. The apparatus of claim 5, further comprising:
and the snapshot library storage module is used for storing the face image to be recognized into the snapshot library if the occurrence time interval is smaller than the preset interval threshold.
9. An electronic device comprising a processor and a memory;
the memory is used for storing a computer program;
the processor, when executing the program stored in the memory, implementing the method steps of any of claims 1-4.
10. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810762447.0A CN110717357B (en) | 2018-07-12 | 2018-07-12 | Early warning method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810762447.0A CN110717357B (en) | 2018-07-12 | 2018-07-12 | Early warning method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110717357A true CN110717357A (en) | 2020-01-21 |
| CN110717357B CN110717357B (en) | 2022-12-06 |
Family
ID=69209113
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810762447.0A Active CN110717357B (en) | 2018-07-12 | 2018-07-12 | Early warning method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110717357B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111814627A (en) * | 2020-06-29 | 2020-10-23 | 深圳市商汤科技有限公司 | Person detection method and device, electronic device and storage medium |
| CN112750274A (en) * | 2020-12-17 | 2021-05-04 | 青岛以萨数据技术有限公司 | Facial feature recognition-based aggregation early warning system, method and equipment |
| CN113343743A (en) * | 2020-03-03 | 2021-09-03 | 杭州海康威视***技术有限公司 | Target person identification method and device, electronic equipment and storage medium |
| CN113468948A (en) * | 2021-04-26 | 2021-10-01 | 深圳市安软科技股份有限公司 | View data based security and protection control method, module, equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679155A (en) * | 2013-12-30 | 2014-03-26 | 深圳锐取信息技术股份有限公司 | Multi-target automatic tracking system and method |
| CN106295469A (en) * | 2015-05-21 | 2017-01-04 | 北京文安智能技术股份有限公司 | A kind of guest's property analysis method based on face, Apparatus and system |
| CN106372572A (en) * | 2016-08-19 | 2017-02-01 | 北京旷视科技有限公司 | Monitoring method and apparatus |
| CN106529373A (en) * | 2015-09-10 | 2017-03-22 | 上海银晨智能识别科技有限公司 | Human face recognition monitoring alarm method and system |
| CN107066921A (en) * | 2016-12-23 | 2017-08-18 | 深圳市大唐盛世智能科技有限公司 | The pedestrian detection method and device of a kind of intelligent video monitoring |
| CN107480246A (en) * | 2017-08-10 | 2017-12-15 | 北京中航安通科技有限公司 | A kind of recognition methods of associate people and device |
| CN107644190A (en) * | 2016-07-20 | 2018-01-30 | 北京旷视科技有限公司 | Pedestrian's monitoring method and device |
-
2018
- 2018-07-12 CN CN201810762447.0A patent/CN110717357B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679155A (en) * | 2013-12-30 | 2014-03-26 | 深圳锐取信息技术股份有限公司 | Multi-target automatic tracking system and method |
| CN106295469A (en) * | 2015-05-21 | 2017-01-04 | 北京文安智能技术股份有限公司 | A kind of guest's property analysis method based on face, Apparatus and system |
| CN106529373A (en) * | 2015-09-10 | 2017-03-22 | 上海银晨智能识别科技有限公司 | Human face recognition monitoring alarm method and system |
| CN107644190A (en) * | 2016-07-20 | 2018-01-30 | 北京旷视科技有限公司 | Pedestrian's monitoring method and device |
| CN106372572A (en) * | 2016-08-19 | 2017-02-01 | 北京旷视科技有限公司 | Monitoring method and apparatus |
| CN107066921A (en) * | 2016-12-23 | 2017-08-18 | 深圳市大唐盛世智能科技有限公司 | The pedestrian detection method and device of a kind of intelligent video monitoring |
| CN107480246A (en) * | 2017-08-10 | 2017-12-15 | 北京中航安通科技有限公司 | A kind of recognition methods of associate people and device |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113343743A (en) * | 2020-03-03 | 2021-09-03 | 杭州海康威视***技术有限公司 | Target person identification method and device, electronic equipment and storage medium |
| CN113343743B (en) * | 2020-03-03 | 2024-02-13 | 杭州海康威视***技术有限公司 | Target person identification method, device, electronic equipment and storage medium |
| CN111814627A (en) * | 2020-06-29 | 2020-10-23 | 深圳市商汤科技有限公司 | Person detection method and device, electronic device and storage medium |
| CN111814627B (en) * | 2020-06-29 | 2021-11-26 | 深圳市商汤科技有限公司 | Person detection method and device, electronic device and storage medium |
| CN112750274A (en) * | 2020-12-17 | 2021-05-04 | 青岛以萨数据技术有限公司 | Facial feature recognition-based aggregation early warning system, method and equipment |
| CN113468948A (en) * | 2021-04-26 | 2021-10-01 | 深圳市安软科技股份有限公司 | View data based security and protection control method, module, equipment and storage medium |
| CN113468948B (en) * | 2021-04-26 | 2023-11-10 | 深圳市安软科技股份有限公司 | View data-based security control method, module, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110717357B (en) | 2022-12-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110717357B (en) | Early warning method and device, electronic equipment and storage medium | |
| CN106203458B (en) | Crowd video analysis method and system | |
| EP3654285B1 (en) | Object tracking using object attributes | |
| CN108629791B (en) | Pedestrian tracking method and device and cross-camera pedestrian tracking method and device | |
| JP6905850B2 (en) | Image processing system, imaging device, learning model creation method, information processing device | |
| CN110717358B (en) | Visitor number counting method and device, electronic equipment and storage medium | |
| US8744125B2 (en) | Clustering-based object classification | |
| CN112507813A (en) | Event detection method and device, electronic equipment and storage medium | |
| WO2018180588A1 (en) | Facial image matching system and facial image search system | |
| WO2015166612A1 (en) | Image analysis system, image analysis method, and image analysis program | |
| CN109376639B (en) | Accompanying personnel early warning system and method based on portrait recognition | |
| CN110263633B (en) | Method, system and storage medium for detecting and early warning of toxic person based on space-time correlation | |
| CN110390229B (en) | Face picture screening method and device, electronic equipment and storage medium | |
| CN108694399B (en) | License plate recognition method, device and system | |
| CN106355154B (en) | Method for detecting frequent passing of people in surveillance video | |
| CN111353338B (en) | Energy efficiency improvement method based on business hall video monitoring | |
| CN111814510A (en) | Detection method and device for remnant body | |
| CN110544312A (en) | Video display method and device in virtual scene, electronic equipment and storage device | |
| Chandran et al. | Identifying social groups in pedestrian crowd videos | |
| Iazzi et al. | Fall detection based on posture analysis and support vector machine | |
| US11537639B2 (en) | Re-identification of physical objects in an image background via creation and storage of temporary data objects that link an object to a background | |
| US20200012866A1 (en) | System and method of video content filtering | |
| KR101407394B1 (en) | System for abandoned and stolen object detection | |
| CN113837066A (en) | Behavior recognition method and device, electronic equipment and computer storage medium | |
| CN114067248A (en) | Behavior detection method and device, electronic equipment and 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |