Practical population monitoring method and device, electronic equipment and storage medium

Abstract

In the embodiment of the invention, an elevator taking image sequence of a target person is acquired, an elevator access behavior of the target person and a behavior attribute of the target person are determined based on the elevator taking image sequence of the target person, and whether the target person is a real population or not is determined based on the elevator access behavior and the behavior attribute. The behavior attribute and the access behavior of the target person are determined by processing the elevator taking image sequence, the correlation of the behavior attribute and the access behavior of the target person in the elevator to whether the target person is the actual population or not is fully considered, and the accuracy is high; and the real population monitoring method can be deployed to any server or server cluster, automatic monitoring can be realized, the objectivity of real population monitoring is further realized, and the efficiency is relatively high.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring, and in particular to a real population monitoring method, device, electronic equipment and storage medium. Background Art

[0002] The monitoring of the actual population plays a positive role in urban management, construction and planning. At present, conventional population statistics methods require household registration or inquiries, which are highly subjective, and the data may differ from the actual situation. At the same time, the time and labor costs of household registration or inquiries are high, and the efficiency and accuracy are low. Therefore, how to provide a method that can monitor the local actual population in real time and improve the efficiency and accuracy of actual population monitoring and ensure the objectivity of monitoring has become an urgent problem to be solved. Summary of the invention

[0003] The embodiment of the present invention provides a real population monitoring method, which aims to solve the problems of strong subjectivity, low efficiency and accuracy in existing real population monitoring. By processing the elevator riding image sequence to determine the behavioral attributes and entry and exit behaviors of the target person, the correlation between the behavioral attributes and entry and exit behaviors of the target person in the elevator and whether the target person is a real population is fully considered, and the accuracy is high. The above-mentioned real population monitoring method can be deployed to any server or server cluster, and can realize automatic monitoring, thereby realizing the objectivity of real population monitoring, and the efficiency is high.

[0004] In a first aspect, an embodiment of the present invention provides a method for monitoring a real population, the method comprising:

[0005] Obtain an image sequence of the target person riding in an elevator;

[0006] Based on the elevator riding image sequence of the target person, determining the elevator entry and exit behavior of the target person and the behavior attribute of the target person;

[0007] Based on the elevator entry and exit behavior and the behavior attributes, it is determined whether the target person is an actual population.

[0008] Optionally, obtaining a sequence of images of the target person riding in an elevator includes:

[0009] When a target person is detected entering an elevator car, image tracking processing is performed on the target person in the elevator car to obtain an elevator riding image sequence of the target person.

[0010] Optionally, the determining the elevator entry and exit behavior of the target person and the behavior attribute of the target person based on the elevator riding image sequence of the target person includes:

[0011] Determining the elevator entry and exit behavior of the target person based on the head frame image and the tail frame image of the elevator riding image sequence;

[0012] Based on the middle frame images of the elevator riding image sequence, the behavior attributes of the target person are determined, the image acquisition time of the head frame image is earlier than the image acquisition time of the middle frame image, and the image acquisition time of the middle frame image is earlier than the tail frame image.

[0013] Optionally, determining the elevator entry and exit behavior of the target person based on the head frame image and the tail frame image of the elevator riding image sequence includes:

[0014] Performing digital recognition processing on the head frame image to obtain the first floor number when the target person enters the elevator car;

[0015] and performing digital recognition processing on the tail frame to obtain the second floor number when the target person leaves the elevator car;

[0016] Based on the first floor data number and the second floor number, the elevator entry and exit behavior of the target person is determined.

[0017] Optionally, determining the behavior attribute of the target person based on the intermediate frame image of the elevator riding image sequence includes:

[0018] Performing behavior recognition processing on the intermediate frame image to obtain a behavior recognition result of the target person;

[0019] Based on the behavior recognition result, the behavior attribute of the target person is determined.

[0020] Optionally, the determining whether the target person is an actual population based on the elevator entry and exit behavior and the behavior attribute includes:

[0021] Based on the elevator entry and exit behavior, determining the entry and exit pattern of the target person;

[0022] Based on the behavior attributes, determining the behavior intention of the target person;

[0023] Based on the entry and exit patterns and the behavioral intentions, determine whether the target person is an actual population.

[0024] Optionally, determining whether the target person is an actual population based on the entry and exit pattern and the behavioral intention includes:

[0025] Determine the entry and exit patterns of the actual population and the behavioral intentions of the actual population;

[0026] When the entry and exit pattern conforms to the entry and exit pattern of the actual population, and the behavioral intention conforms to the behavioral intention of the actual population, the target person is determined to be the actual population.

[0027] In a second aspect, an embodiment of the present invention further provides a real population monitoring device, the real population monitoring device comprising:

[0028] An acquisition module, used to acquire an image sequence of a target person riding in an elevator;

[0029] A first determination module is used to determine the elevator entry and exit behavior of the target person and the behavior attribute of the target person based on the elevator riding image sequence of the target person;

[0030] The second determination module is used to determine whether the target person is an actual population based on the elevator entry and exit behavior and the behavior attributes.

[0031] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the actual population monitoring method provided in the embodiment of the present invention when executing the computer program.

[0032] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the actual population monitoring method provided in the embodiment of the invention are implemented.

[0033] In the embodiment of the present invention, an elevator riding image sequence of a target person is obtained, and based on the elevator riding image sequence of the target person, the elevator entry and exit behavior of the target person and the behavioral attributes of the target person are determined, and based on the elevator entry and exit behavior and the behavioral attributes, it is determined whether the target person is a real population. By processing the elevator riding image sequence to determine the behavioral attributes and entry and exit behavior of the target person, the correlation between the behavioral attributes and entry and exit behavior of the target person in the elevator and whether the target person is a real population is fully considered, and the accuracy is high. The above-mentioned real population monitoring method can be deployed to any server or server cluster, and can realize automatic monitoring, thereby realizing the objectivity of real population monitoring, and the efficiency is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 is a flow chart of a real population monitoring method provided by an embodiment of the present invention;

[0036] Figure 2 is a schematic structural diagram of a real population monitoring device provided in an embodiment of the present invention;

[0037] Figure 3 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] like Figure 1 As shown, Figure 1 : is a method flow chart of a real population monitoring method provided by an embodiment of the present invention. The real population monitoring method comprises the steps of:

[0040] 101. Obtain an elevator riding image sequence of the target person.

[0041] In an embodiment of the present invention, the above-mentioned actual population monitoring method can be deployed in a population management platform. The above-mentioned population management platform can be a server or server cluster with functions such as image processing, image recognition, data transmission, and data storage. The above-mentioned population management platform can store historical elevator riding image sequences. The above-mentioned historical elevator riding image sequences can include historical elevator riding times and historical elevator riding locations. Each historical elevator riding image sequence corresponds to one or more passengers. The above-mentioned population management platform can also store architectural drawings of buildings corresponding to the above-mentioned elevator riding image sequences. The above-mentioned architectural drawings can be used to express the design ideas and achievements of the above-mentioned buildings. The above-mentioned architectural drawings can describe in detail the information on the plane, facade, details and structure of the above-mentioned buildings in the form of graphics and text.

[0042] The target person may be any person who enters the elevator. An image acquisition device may be deployed in the elevator. The image acquisition device may be an electronic device with functions such as image recognition, image acquisition, data storage and data transmission. The image acquisition device may be deployed in any building equipped with an elevator. The image acquisition device acquires an elevator riding image sequence of the target person through the image acquisition function. The elevator riding image sequence may include one or more passengers. If there is only one passenger in the elevator riding image sequence, the passenger is the target person. If there are multiple passengers in the elevator riding image sequence, there are multiple target persons, and each passenger corresponds to one target person.

[0043] In a possible embodiment, when the image acquisition device recognizes that a person has entered the elevator through the image recognition function, the image acquisition function acquires an elevator riding image sequence of the person until the person leaves the acquisition range of the image acquisition device. The image acquisition device then sends the acquired elevator riding image sequence to the population management platform.

[0044] 102. Based on the elevator riding image sequence of the target person, determine the elevator entry and exit behavior of the target person and the behavior attributes of the target person.

[0045] In an embodiment of the present invention, the above-mentioned entry and exit behaviors may include elevator entry behaviors and elevator departure behaviors. The above-mentioned elevator entry behaviors are determined through the head frame of the above-mentioned elevator riding image sequence, the above-mentioned elevator departure behaviors are determined through the tail frame of the above-mentioned elevator riding image sequence, and the behavior attributes of the above-mentioned target person are determined through the middle frames of the above-mentioned elevator riding image sequence.

[0046] Specifically, recognition processing can be performed based on the head frame of the above-mentioned elevator riding image sequence to determine the entry behavior of the above-mentioned target person when he just enters the elevator. For example, the above-mentioned entry behavior may include the behavior of the above-mentioned target person pressing the floor button when he just enters the elevator, the above-mentioned entry behavior may include the behavior of the above-mentioned target person putting down the items in his hands when he just enters the elevator, and the above-mentioned entry behavior may also include the behavior of the above-mentioned target person holding a child or a small animal when entering the elevator. The entry behavior of the above-mentioned target person is determined based on the above-mentioned behaviors. After the above-mentioned entry behavior is determined, recognition processing is performed based on the tail frame of the above-mentioned elevator riding image sequence to determine the departure behavior of the above-mentioned target person when leaving the elevator.

[0047] The above-mentioned leaving behavior may include the behavior of the above-mentioned target person pressing the door opening button when preparing to leave the elevator, the above-mentioned leaving behavior may include the behavior of the above-mentioned target person picking up items on the ground when preparing to leave the elevator, and the above-mentioned behavior may also include the behavior of the above-mentioned target person picking up a child or small animal on the ground when preparing to leave the elevator. Recognition processing is performed based on the intermediate frames of the above-mentioned elevator riding image sequence to determine the behavioral attributes of the above-mentioned target person. Specifically, the behavioral attributes of the above-mentioned target person can be determined through features such as the target person's clothing, handheld items, facial expressions, etc.

[0048] The above-mentioned recognition processing can be performed on the above-mentioned elevator riding image sequence through a model matching method using perspective projection hashing (PnP-Hash), visual SLAM (Simultaneous Localization and Mapping) and other algorithms to perform image recognition on the above-mentioned elevator riding image sequence to determine the entry and exit behavior and behavior attributes of the above-mentioned target personnel, or it can also be performed on the above-mentioned elevator riding image sequence through a deep learning method using yolov7, generative adversarial networks (Generative Adversarial Networks, GAN), spatiotemporal convolutional neural networks (3D Convolutional Neural Networks, 3D CNN) and other algorithms to perform image recognition on the above-mentioned elevator riding image sequence to determine the entry and exit behavior and behavior attributes of the above-mentioned target personnel.

[0049] In a possible embodiment, when the image capture device captures the elevator riding image sequence and sends the elevator riding image sequence to the population management platform, the population management platform recognizes and processes the elevator riding image sequence through the model matching method or the deep learning method to determine the elevator entry and exit behavior of the target person and the behavioral attributes of the target person.

[0050] 103. Based on the elevator entry and exit behavior and behavior attributes, determine whether the target person is an actual population.

[0051] In an embodiment of the present invention, the actual population may be the actual population living in the building corresponding to the elevator. The floors frequently entered and exited by the target person and the entry and exit times corresponding to the floors are determined based on the entry and exit behavior of the elevator. The entry and exit patterns of the target person are determined based on the floors frequently entered and exited by the target person and the entry and exit times corresponding to the floors. The clothing, hand-held items, and facial expressions of the target person are determined based on the behavioral attributes of the target person. The behavioral intentions of the target person are determined based on the clothing, hand-held items, and facial expressions of the target person. Based on the behavioral intentions of the target person and the entry and exit patterns of the target person, it is determined whether the target person is the actual population.

[0052] In a possible embodiment, feature extraction may be performed on the target person according to the elevator riding image sequence to obtain feature data corresponding to the target person, extract a historical elevator riding image sequence corresponding to the feature data in the population management platform, determine the entry and exit pattern of the target person and the behavior intention of the target person based on the historical elevator riding image sequence and the elevator riding image sequence, and determine whether the target person is the actual resident of the building corresponding to the elevator riding image sequence based on the entry and exit pattern of the target person and the behavior intention of the target person. The historical elevator riding image sequence is collected earlier than the elevator riding image sequence.

[0053] Specifically, features of the target person's face can be extracted based on the elevator riding image sequence to obtain feature data corresponding to the target person, and the feature data can be compared with the feature data of historical passengers in the historical elevator riding image sequence stored in the population management platform. Image recognition can be performed based on the successfully compared historical elevator riding image sequence to determine the target person's historical elevator entry and exit behavior and the corresponding historical behavior attributes. Based on the historical elevator entry and exit behavior and the corresponding historical behavior attributes, the target person's historical entry and exit pattern and the corresponding historical behavior intention can be determined. Based on the historical entry and exit pattern, historical behavior intention, entry and exit pattern, and behavior intention, it can be determined whether the target person is the actual population living in the building corresponding to the elevator riding image sequence.

[0054] The elevators corresponding to the historical elevator riding image sequence and the elevator riding image sequence are elevators in the same building.

[0055] For example, if the above-mentioned feature data fails to be compared with the feature data of historical passengers in the historical elevator riding image sequence stored in the above-mentioned population management platform, it means that the above-mentioned target person appears in the building corresponding to the above-mentioned elevator riding image sequence for the first time. If the elevator entry behavior of the above-mentioned target person is the first floor and the elevator departure behavior is the ninth floor, according to the architectural drawing corresponding to the previous elevator riding image sequence, the ninth floor is determined to be the fire protection floor, according to the above-mentioned behavioral attributes, the clothes of the above-mentioned target person are determined to be work clothes, according to the above-mentioned behavioral attributes, the object held by the above-mentioned target person is determined to be a toolbox, according to the above-mentioned behavioral attributes, the facial expression of the above-mentioned target person is determined to be a gentle mouth corner, then it can be determined that the above-mentioned target person is not the actual population living in the building corresponding to the above-mentioned elevator riding image sequence.

[0056] If the above-mentioned feature data is successfully matched with the feature data of historical passengers in the historical elevator riding image sequence stored in the above-mentioned population management platform, it means that the above-mentioned target person is not the first time to appear in the building corresponding to the above-mentioned elevator riding image sequence. Image recognition is performed based on the successfully matched historical elevator riding image sequence to determine the historical elevator entry and exit behavior of the above-mentioned target person and the corresponding historical behavior attributes.

[0057] According to the elevator riding image sequence of the above-mentioned target person, it is determined that the elevator entry behavior of the above-mentioned target person is the first floor, and the elevator departure behavior is the eighth floor. According to the architectural drawing corresponding to the above-mentioned elevator riding image sequence, it is determined that the above-mentioned eighth floor is a residential floor. According to the above-mentioned behavior history elevator riding image sequence, it is determined that the historical elevator entry behavior of the above-mentioned target person is all the first floor or the eighth floor, and the historical elevator departure behavior is all the first floor or the eighth floor. The entry time corresponding to the historical elevator entry behavior or the departure time corresponding to the historical elevator departure behavior is 8 am or 7 pm. According to the above-mentioned elevator entry behavior, elevator departure behavior, historical elevator entry behavior, historical elevator departure behavior, and the departure time and entry time corresponding to the above behaviors, it can be determined that the entry and exit pattern of the above-mentioned target person is entering and exiting the above-mentioned elevator during commute times.

[0058] According to the above-mentioned historical elevator riding image sequence, the historical behavior attributes corresponding to the above-mentioned target person are determined; based on the historical behavior attributes corresponding to the above-mentioned target person, it is determined that the above-mentioned target person was dressed in a suit or casual clothes when riding the elevator in the past; the items held by the above-mentioned target person when riding the elevator in the past were briefcases or small animals; based on the above-mentioned clothing and the items held by the above-mentioned target person when riding the elevator, it is determined that the behavior intention of the above-mentioned target person is commuting to get off work or walking the dog; based on the above-mentioned entry and exit patterns and the above-mentioned behavior intention, it is determined that the above-mentioned target person is the actual resident of the building corresponding to the above-mentioned elevator riding image sequence.

[0059] In the embodiment of the present invention, an elevator riding image sequence of a target person is obtained, and based on the elevator riding image sequence of the target person, the elevator entry and exit behavior of the target person and the behavioral attributes of the target person are determined, and based on the elevator entry and exit behavior and the behavioral attributes, it is determined whether the target person is a real population. By processing the elevator riding image sequence to determine the behavioral attributes and entry and exit behavior of the target person, the correlation between the behavioral attributes and entry and exit behavior of the target person in the elevator and whether the target person is a real population is fully considered, and the accuracy is high. The above-mentioned real population monitoring method can be deployed to any server or server cluster, and can realize automatic monitoring, thereby realizing the objectivity of real population monitoring, and the efficiency is high.

[0060] Optionally, in the step of acquiring the target person's elevator riding image sequence, it is also possible to detect when the target person enters the elevator car, perform image tracking processing on the target person in the elevator car, and obtain the target person's elevator riding image sequence.

[0061] In the embodiment of the present invention, the image acquisition device deployed in the elevator car can be used to detect in real time whether a person enters the elevator car. When the image acquisition device detects that a person enters the elevator car, if the person is one person, the person is determined as a target person, and the image acquisition device performs image tracking processing on the target person in the elevator car to obtain an elevator riding image sequence of the target person. If there are multiple people, all of the multiple people are determined as target people, and image tracking processing is performed on each of them to obtain elevator riding image sequences corresponding to the multiple people.

[0062] In a possible embodiment, when the image acquisition device detects that a person enters the elevator car, the person is determined as a target person, and the image acquisition device performs image tracking processing on the target person to obtain an elevator riding image sequence of the target person, and the elevator riding image sequence of the target person is transmitted to the population management platform through the data transmission function.

[0063] Optionally, in the step of determining the target person's elevator entry and exit behavior and the target person's behavioral attributes based on the target person's elevator riding image sequence, the target person's elevator entry and exit behavior can also be determined based on the head frame image and the tail frame image of the elevator riding image sequence; and the target person's behavioral attributes can be determined based on the middle frame image of the elevator riding image sequence.

[0064] In an embodiment of the present invention, the number of frames of the elevator riding image sequence can be determined based on the elevator riding image sequence. According to the image acquisition time, the first 20% of the image frames of the total number of frames of the elevator riding image sequence are used as the head frame images, the last 20% of the image frames of the total number of frames of the elevator riding image sequence are used as the tail frame images, and the remaining image frames are used as intermediate frame images. Alternatively, the age of the target person can be determined according to the elevator riding image sequence, and the number of frames of the head frame image and the number of frames of the tail frame image can be determined according to the age of the target person.

[0065] The older the target person is, the more frames the head frame image and the tail frame image have. If the target person is younger, the fewer frames the head frame image and the tail frame image have. The image acquisition time of the head frame image is earlier than the image acquisition time of the middle frame image, and the image acquisition time of the middle frame image is earlier than the tail frame image.

[0066] In a possible embodiment, after the population management platform obtains the elevator riding image sequence of the target person, the elevator entry behavior of the target person is determined based on the head frame image corresponding to the elevator riding image sequence, the elevator departure behavior of the target person is determined based on the tail frame image corresponding to the elevator riding image sequence, the elevator entry and exit behavior of the target person is determined based on the elevator entry behavior and the elevator departure behavior, and the behavior attributes of the target person are determined based on the middle frame image of the elevator riding image sequence.

[0067] Optionally, in the step of determining the target person's elevator entry and exit behavior based on the head frame image and the tail frame image of the elevator riding image sequence, the head frame image can also be digitally recognized to obtain the first floor number when the target person enters the elevator car; and the tail frame can be digitally recognized to obtain the second floor number when the target person leaves the elevator car; based on the first floor data number and the second floor number, the target person's elevator entry and exit behavior is determined.

[0068] In an embodiment of the present invention, the digital recognition process may be a recognition method based on template matching or a recognition method based on digital features. Specifically, the recognition method based on template matching may be to construct a digital template and match the numbers in the head frame image and the tail frame image with the digital template to recognize the numbers in the head frame image and the tail frame image. The recognition method based on digital features may be to analyze and count the geometric features such as the shape and structure of the numbers in the head frame image and the tail frame image, extract the features of the numbers in the head frame image and the tail frame image for recognition. The first floor number and the second floor number may be the numbers on the floor display device in the elevator.

[0069] In a possible embodiment, after the population management platform obtains the elevator riding image sequence, digital recognition processing is performed on the numbers on the floor display device in the elevator corresponding to the elevator riding image sequence according to the template matching-based recognition method or the digital feature-based recognition method to obtain the first floor number and the second floor number, and the elevator entry and exit behavior of the target person is determined based on the first floor number and the second floor number.

[0070] In another possible embodiment, after the population management platform obtains the elevator riding image sequence, it can also determine the floor number of the elevator at the same time corresponding to the acquisition time of the head frame image based on the acquisition time of the head frame image, and determine the first floor number based on the floor number of the elevator at the same time corresponding to the acquisition time of the head frame image. Determine the floor number of the elevator at the same time corresponding to the acquisition time of the tail frame image based on the acquisition time of the tail frame image, determine the second floor number based on the floor number of the elevator at the same time corresponding to the acquisition time of the tail frame image, and determine the elevator entry and exit behavior of the target person based on the first floor number and the second floor number.

[0071] Optionally, in the step of determining the behavioral attributes of the target person based on the intermediate frame images of the elevator riding image sequence, the intermediate frame images can also be subjected to behavior recognition processing to obtain the behavior recognition results of the target person; and the behavior attributes of the target person are determined based on the behavior recognition results.

[0072] In an embodiment of the present invention, the intermediate frame image may be subjected to behavior recognition processing by a behavior recognition algorithm to obtain a behavior recognition result of the target person. The behavior recognition algorithm may be a 3D-Convolutional Neural Networks (3D-CNN) algorithm or a Two-Stream Convolutional Networks algorithm. Based on the behavior recognition result of the target person, the behavior attributes of the target person are determined.

[0073] Specifically, if the above-mentioned intermediate frame image is subjected to behavior recognition processing according to the above-mentioned behavior recognition algorithm, and it is determined that the behavior recognition result of the above-mentioned target person includes that the clothes worn are work clothes and the handheld item is a toolbox, then the behavior attribute of the above-mentioned target person can be determined as a construction worker based on the above-mentioned behavior recognition result; if the behavior recognition result of the above-mentioned target person includes that the clothes worn are pajamas and the handheld item is a bag of garbage, then the behavior attribute of the above-mentioned target person can be determined as a nearby resident based on the above-mentioned behavior recognition result.

[0074] In a possible embodiment, after the above-mentioned population management platform divides the above-mentioned elevator riding image sequence and obtains the intermediate frames corresponding to the above-mentioned elevator riding image sequence, it performs behavior recognition processing on the above-mentioned intermediate frames based on the above-mentioned behavior recognition algorithm to obtain the behavior recognition results of the above-mentioned target personnel, and determines the behavior attributes of the above-mentioned target personnel based on the above-mentioned behavior recognition results.

[0075] Optionally, in the step of determining whether the target person is an actual population based on the elevator entry and exit behavior and behavior attributes, the entry and exit pattern of the target person can also be determined based on the elevator entry and exit behavior; the behavior intention of the target person can be determined based on the behavior attributes; and whether the target person is an actual population can be determined based on the entry and exit pattern and behavior intention.

[0076] In an embodiment of the present invention, the historical elevator entry and exit behavior of the target person can be determined based on the elevator entry and exit behavior, the entry and exit pattern of the target person can be determined based on the elevator entry and exit behavior and the historical elevator entry and exit behavior, and the behavioral intention of the target person can be determined based on the behavioral attributes.

[0077] In a possible embodiment, after the population management platform determines the entry and exit behavior and behavior attributes of the target person according to the elevator riding image sequence, the characteristic data of the target person in the elevator riding image sequence can be determined according to the elevator entry and exit behavior, and a query and comparison is performed in the population management platform based on the characteristic data of the target person to determine the historical elevator riding image sequence of the target person. The historical elevator entry and exit behavior of the target person is determined based on the historical elevator riding image sequence of the target person, and the entry and exit floors and historical entry and exit floors corresponding to the target person are determined based on the entry and exit behavior and historical entry and exit behavior of the target person. The entry and exit pattern of the target person is determined according to the entry and exit floors and historical entry and exit floors corresponding to the target person. The behavioral intention of the target person is determined according to the behavioral attributes of the target person. The acquisition time of the historical elevator riding image sequence is earlier than the elevator riding image sequence.

[0078] Optionally, in the step of determining whether the target person is the actual population based on the entry and exit patterns and behavioral intentions, the entry and exit patterns of the actual population and the behavioral intentions of the actual population can also be determined; when the entry and exit patterns are consistent with the entry and exit patterns of the actual population, and the behavioral intentions are consistent with the behavioral intentions of the actual population, the target person is determined to be the actual population.

[0079] In an embodiment of the present invention, the actual population entry and exit pattern can be determined based on the local people's living customs, and the actual population behavior intention can be determined based on the local people's living customs and behavior habits. The population entry and exit pattern and population behavior intention can also be determined based on the building usage scene corresponding to the elevator riding image sequence.

[0080] In a possible embodiment, when the population management platform determines the entry and exit pattern of the target person based on the elevator entry and exit behavior and the historical elevator entry and exit behavior, and determines the behavioral intention of the target person based on the behavioral attributes, the actual population entry and exit pattern is determined according to the local residents' living customs, and the actual population behavioral intention is determined according to the local residents' living customs and behavioral habits. When the entry and exit pattern of the target person is consistent with the entry and exit pattern of the actual population, and the behavioral intention is consistent with the behavioral intention of the actual population, the target person is determined to be the actual population.

[0081] like Figure 2 As shown, an embodiment of the present invention provides a real population monitoring device, the real population monitoring device comprising:

[0082] An acquisition module 201 is used to acquire an image sequence of a target person riding in an elevator;

[0083] A first determination module 202 is used to determine the elevator entry and exit behavior of the target person and the behavior attribute of the target person based on the elevator riding image sequence of the target person;

[0084] The second determination module 203 is used to determine whether the target person is an actual population based on the elevator entry and exit behavior and the behavior attribute.

[0085] Optionally, the acquisition module 201 includes:

[0086] The image tracking submodule is used to detect that a target person enters the elevator car, and then perform image tracking processing on the target person in the elevator car to obtain an elevator riding image sequence of the target person.

[0087] Optionally, the first determining module 202 includes:

[0088] A first determination submodule is used to determine the elevator entry and exit behavior of the target person based on the head frame image and the tail frame image of the elevator riding image sequence;

[0089] The second determination submodule is used to determine the behavior attributes of the target person based on the middle frame image of the elevator riding image sequence, the image acquisition time of the head frame image is earlier than the image acquisition time of the middle frame image, and the image acquisition time of the middle frame image is earlier than the tail frame image.

[0090] Optionally, the first determining submodule includes:

[0091] A first recognition unit, configured to perform digital recognition processing on the head frame image to obtain a first floor number when the target person enters the elevator car;

[0092] A second recognition unit is used to perform digital recognition processing on the tail frame to obtain the second floor number when the target person leaves the elevator car;

[0093] The first determining unit is used to determine the elevator entry and exit behavior of the target person based on the first floor data number and the second floor number.

[0094] Optionally, the second determining submodule includes:

[0095] A third recognition unit is used to perform behavior recognition processing on the intermediate frame image to obtain a behavior recognition result of the target person;

[0096] The second determining unit is used to determine the behavior attribute of the target person based on the behavior recognition result.

[0097] Optionally, the second determining module 203 includes:

[0098] A third determination submodule is used to determine the entry and exit pattern of the target person based on the elevator entry and exit behavior;

[0099] A fourth determination submodule, used to determine the behavior intention of the target person based on the behavior attribute;

[0100] The fifth determination submodule is used to determine whether the target person is an actual population based on the entry and exit pattern and the behavioral intention.

[0101] Optionally, the fifth determining submodule includes:

[0102] The third determination unit is used to determine the entry and exit patterns of the actual population and the behavioral intentions of the actual population;

[0103] The fourth determination unit is used to determine that the target person is the actual population when the entry and exit pattern conforms to the entry and exit pattern of the actual population and the behavior intention conforms to the behavior intention of the actual population.

[0104] The actual population monitoring device provided in the embodiment of the present invention can implement each process implemented by the actual population monitoring method in the above method embodiment, and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0105] See also Figure 3 , Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, it includes: a memory 302, a processor 301, and a computer program of a population monitoring method stored in the memory 302 and executable on the processor 301, wherein:

[0106] The processor 301 is used to call the computer program stored in the memory 302 and execute the following steps:

[0107] Obtain an image sequence of the target person riding in an elevator;

[0108] Based on the elevator riding image sequence of the target person, determining the elevator entry and exit behavior of the target person and the behavior attribute of the target person;

[0109] Based on the elevator entry and exit behavior and the behavior attributes, it is determined whether the target person is an actual population.

[0110] Optionally, the acquiring of the elevator riding image sequence of the target person executed by the processor 301 includes:

[0111] When a target person is detected entering an elevator car, image tracking processing is performed on the target person in the elevator car to obtain an elevator riding image sequence of the target person.

[0112] Optionally, the processor 301 determines the elevator entry and exit behavior of the target person and the behavior attribute of the target person based on the elevator riding image sequence of the target person, including:

[0113] Determining the elevator entry and exit behavior of the target person based on the head frame image and the tail frame image of the elevator riding image sequence;

[0114] Based on the middle frame images of the elevator riding image sequence, the behavior attributes of the target person are determined, the image acquisition time of the head frame image is earlier than the image acquisition time of the middle frame image, and the image acquisition time of the middle frame image is earlier than the tail frame image.

[0115] Optionally, the processor 301 determines the elevator entry and exit behavior of the target person based on the head frame image and the tail frame image of the elevator riding image sequence, including:

[0116] Performing digital recognition processing on the head frame image to obtain the first floor number when the target person enters the elevator car;

[0117] and performing digital recognition processing on the tail frame to obtain the second floor number when the target person leaves the elevator car;

[0118] Based on the first floor data number and the second floor number, the elevator entry and exit behavior of the target person is determined.

[0119] Optionally, the determining of the behavior attribute of the target person based on the intermediate frame image of the elevator riding image sequence performed by the processor 301 includes:

[0120] Performing behavior recognition processing on the intermediate frame image to obtain a behavior recognition result of the target person;

[0121] Based on the behavior recognition result, the behavior attribute of the target person is determined.

[0122] Optionally, the processor 301 determines whether the target person is an actual population based on the elevator entry and exit behavior and the behavior attribute, including:

[0123] Based on the elevator entry and exit behavior, determining the entry and exit pattern of the target person;

[0124] Based on the behavior attributes, determining the behavior intention of the target person;

[0125] Based on the entry and exit patterns and the behavioral intentions, determine whether the target person is an actual population.

[0126] Optionally, the processor 301 determines whether the target person is an actual population based on the entry and exit pattern and the behavioral intention, including:

[0127] Determine the entry and exit patterns of the actual population and the behavioral intentions of the actual population;

[0128] When the entry and exit pattern conforms to the entry and exit pattern of the actual population, and the behavioral intention conforms to the behavioral intention of the actual population, the target person is determined to be the actual population.

[0129] The electronic device provided in the embodiment of the present invention can implement each process implemented by the actual population monitoring method in the above method embodiment, and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0130] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the computer program implements the various processes of the actual population monitoring method or the application-side actual population monitoring method provided in the embodiment of the present invention, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0131] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0132] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for monitoring a real population, characterized in that: include: Obtain an image sequence of the target person riding in an elevator; Based on the elevator riding image sequence of the target person, determining the elevator entry and exit behavior of the target person and the behavior attribute of the target person; Based on the elevator entry and exit behavior and the behavior attributes, it is determined whether the target person is an actual population.

2. The method according to claim 1, characterized in that The step of obtaining a sequence of images of a target person riding in an elevator comprises: When a target person is detected entering an elevator car, image tracking processing is performed on the target person in the elevator car to obtain an elevator riding image sequence of the target person.

3. The method according to claim 1 or 2, characterized in that The step of determining the elevator entry and exit behavior of the target person and the behavior attribute of the target person based on the elevator riding image sequence of the target person includes: Determining the elevator entry and exit behavior of the target person based on the head frame image and the tail frame image of the elevator riding image sequence; Based on the middle frame images of the elevator riding image sequence, the behavior attributes of the target person are determined, the image acquisition time of the head frame image is earlier than the image acquisition time of the middle frame image, and the image acquisition time of the middle frame image is earlier than the tail frame image.

4. The method according to claim 3, characterized in that The step of determining the elevator entry and exit behavior of the target person based on the head frame image and the tail frame image of the elevator riding image sequence includes: Performing digital recognition processing on the head frame image to obtain the first floor number when the target person enters the elevator car; and performing digital recognition processing on the tail frame to obtain the second floor number when the target person leaves the elevator car; Based on the first floor data number and the second floor number, the elevator entry and exit behavior of the target person is determined.

5. The method according to claim 3 or 4, characterized in that The determining the behavior attribute of the target person based on the intermediate frame image of the elevator riding image sequence includes: Performing behavior recognition processing on the intermediate frame image to obtain a behavior recognition result of the target person; Based on the behavior recognition result, the behavior attribute of the target person is determined.

6. The method according to any one of claims 1 to 5, characterized in that: The determining whether the target person is an actual population based on the elevator entry and exit behavior and the behavior attribute includes: Based on the elevator entry and exit behavior, determining the entry and exit pattern of the target person; Based on the behavior attributes, determining the behavior intention of the target person; Based on the entry and exit patterns and the behavioral intentions, determine whether the target person is an actual population.

7. The method according to claim 6, characterized in that The determining whether the target person is an actual population based on the entry and exit pattern and the behavioral intention includes: Determine the entry and exit patterns of the actual population and the behavioral intentions of the actual population; When the entry and exit pattern conforms to the entry and exit pattern of the actual population, and the behavioral intention conforms to the behavioral intention of the actual population, the target person is determined to be the actual population.

8. A real population monitoring device, characterized in that: The actual population monitoring device comprises: An acquisition module, used to acquire an image sequence of a target person riding in an elevator; A first determination module is used to determine the elevator entry and exit behavior of the target person and the behavior attribute of the target person based on the elevator riding image sequence of the target person; The second determination module is used to determine whether the target person is an actual population based on the elevator entry and exit behavior and the behavior attributes.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the actual population monitoring method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the real population monitoring method as described in any one of claims 1 to 7 are implemented.

Images