Security monitoring system, security monitoring method, and security monitoring program

Abstract

To achieve personalized security monitoring tailored to the unique environment and circumstances of each monitored area. [Solution] The security monitoring system 100 comprises a shooting unit 1, an image analysis unit 31, a monitoring necessity information storage unit 32, a monitoring information notification unit 33, and a monitoring necessity information update unit 35. The image analysis unit 31 analyzes the monitoring image using the function of artificial intelligence to detect objects that require confirmation, and further determines the identity of the objects that require confirmation with the objects that require monitoring included in the monitoring necessity information. The monitoring information notification unit 33 outputs caution information about the objects that require confirmation if it is determined that they are not the same as the objects that require monitoring. The monitoring necessity information update unit 35 dynamically updates the content of the monitoring necessity information to match the judgment result when the user 5 inputs the judgment result regarding whether or not to continue monitoring the objects that require confirmation output as caution information.

Description

Technical Field

[0001] The present invention relates to a security monitoring system, a security monitoring method, and a program for security monitoring. More specifically, the present invention relates to a security monitoring technology utilizing artificial intelligence (AI). When security monitoring adapted to the unique environment and situation of each individual monitored area is required, such as for the surrounding area of a specific private residence or a specific cultural property building, the present invention relates to a security monitoring system, a security monitoring method, and a program for security monitoring that can perform security monitoring in a mode optimized for each monitored area.

Background Art

[0002] For example, even in the conventional security monitoring system disclosed in Patent Document 1, when an image captured by a camera is analyzed to detect an object matching specific conditions such as abnormal behavior or a suspicious person, and an object determined to be a suspicious person or a suspicious object is found, a notification to that effect is sent to the "user" or "system administrator" of the security monitoring system. In this specification, a person who enjoys the operational effects of the security monitoring system, or an agent of such person, is referred to as a "user". In the case of a security monitoring system for a private residence, the resident, owner of the residence, or a person entrusted with actual monitoring work by such persons corresponds to the "user". Also, in this specification, an information processing engineer in charge of maintenance work for continuing to operate the system normally or in a more favorable mode during system operation is referred to as a "system administrator".

[0003] However, the security monitoring system disclosed in Patent Document 1 is basically a system that performs routine monitoring and warnings based on a predetermined algorithm set by default. For this reason, it was difficult for this security monitoring system to flexibly respond to the unique environment and circumstances of each "individual monitored area" which are each under different unique conditions. Specifically, it was difficult to ensure that certain targets that should only be detected under the unique environment and circumstances of each monitored area (for example, an ordinary vehicle that is not generally considered highly suspicious, but which is unrelated to the area and repeatedly appears in the area) are not always omitted from the detection targets.

[0004] To address this challenge, Patent Document 1 discloses that information regarding the detected object can be recorded in a storage device and used for re-analysis or as reference material as needed. However, in order to continue making highly accurate judgments of suspiciousness in accordance with the unique environment and circumstances of each monitored area, manual operations by technicians, such as relearning artificial intelligence (AI) or rewriting programs, by the "system administrator" responsible for system maintenance and management are required frequently and at irregular intervals to correct erroneous judgments of the security monitoring system whenever a false positive occurs. This excessive workload on the "system administrator" has been a practical obstacle to realizing security monitoring tailored to the unique environment and circumstances of each monitored area in the field of AI-based security monitoring. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2008-294921 [Overview of the project] [Problems that the invention aims to solve]

[0006] The present invention aims to provide a security monitoring system that uses artificial intelligence (AI) image analysis capabilities to detect targets for monitoring, and to enable detailed security monitoring tailored to the unique environment and circumstances of each target area, such as a specific private residence or the surrounding area of ​​a specific cultural heritage building, without relying on an excessive burden on the "system administrator." [Means for solving the problem]

[0007] The inventors of this invention have conceived of a security monitoring system that utilizes the advanced image analysis capabilities of artificial intelligence (AI). By incorporating the user's decision-making process for selecting monitoring targets into the system in a unique cyclical process, the system can continuously provide personalized monitoring capabilities for each user without relying on an excessive burden on the system administrator. This led to the completion of the present invention. Specifically, the present invention provides the following security monitoring system, etc.

[0008] (1) A security monitoring system for monitoring a target area within a specific range, comprising: an imaging unit for acquiring monitoring images of the target area; an image analysis unit for performing image analysis of the monitoring images using an artificial intelligence image analysis function; a monitoring necessity information storage unit for storing monitoring necessity information including a list of targets to be monitored specific to the target area; a monitoring information notification unit for notifying the user of the security monitoring system of monitoring information; and a monitoring necessity information update unit, wherein the image analysis unit includes means for detecting targets to be checked, which analyzes the monitoring images and determines that the targets to be checked have a certain level of suspicion or higher; and means for matching the targets to be checked with the targets to be monitored included in the monitoring necessity information. A security monitoring system comprising: a means for determining whether monitoring is necessary to determine identity; the monitoring information notification unit has a monitoring information selection output means that outputs warning information about the object to be confirmed if it is determined that the object to be confirmed is the same as the object to be monitored, and outputs caution information about the object to be confirmed if it is determined that the object to be confirmed is not the same as the object to be monitored; and the monitoring necessity information update unit has a monitoring necessity information update means that dynamically updates the content of the monitoring necessity information to match the determination result when the determination result regarding whether the user of the security monitoring system should continue monitoring the object to be confirmed, which has been output as caution information, is input.

[0009] According to the security monitoring system described in (1), in the field of security monitoring utilizing artificial intelligence (AI), when security monitoring tailored to the unique environment and circumstances of each monitoring target area is required, such as the area surrounding a specific private residence or a specific cultural heritage building, it is possible to achieve detailed security monitoring in a manner that is optimally personalized to that environment and circumstances, without relying on an excessive burden on the "system administrator".

[0010] (2) The security monitoring system as described in (1), wherein the monitoring requirement information also includes a list of targets that do not need to be monitored, which is specific to the area being monitored, and the monitoring information notification unit does not output information about the target that needs to be checked if it determines that the target that needs to be checked is the same as the target that does not need to be monitored.

[0011] According to the security monitoring system in (2), even if an object is detected in the security monitoring system in (1) by artificial intelligence (AI) as being temporarily deemed suspicious, if the "user" determines that the detected object does not require monitoring based on their own criteria, the "user" can quickly remove it from the list of objects requiring monitoring through an easy operation performed by the "user" themselves, without relying on an excessive burden on the "system administrator". This allows for more efficient and thorough optimization of the system through dynamic updates of monitoring requirements information.

[0012] (3) The security monitoring system according to (1) or (2), further comprising: a recording unit that records the monitoring images as stored images; and an image search unit that searches the stored images to extract a specific monitoring target, wherein the image analysis unit has a means for changing the suspiciousness criteria for the monitoring target, which automatically changes the suspiciousness criteria for the monitoring target according to the results of analyzing the stored images of the monitoring target extracted by the image search unit.

[0013] According to the security monitoring system in (3), in the security monitoring system in (1) or (2), specific images that can contribute to real-time suspiciousness determination can be extracted from past monitoring images (stored images) accumulated in the recording unit, and the analysis results can also be automatically contributed to the determination of suspiciousness in real-time detection of targets requiring attention. This makes it possible to make more advanced and complex judgments that incorporate not only the environment and situation in the monitored area, but also the history of changes in past events and events up to the present in chronological order, enabling more detailed and personalized security monitoring that is tailored to the actual situation of the monitored area.

[0014] (4) A security monitoring system for monitoring a target area within a specific range, comprising: an imaging unit for acquiring monitoring images of the target area; a recording unit for recording the monitoring images as stored images; an image search unit for searching the stored images to extract specific targets; an image analysis unit for performing image analysis of the monitoring images using artificial intelligence; and a monitoring information notification unit for notifying the user of the security monitoring system of monitoring information, wherein the image analysis unit includes a means for detecting targets requiring confirmation that are determined to have a certain level of suspiciousness or higher after analyzing the monitoring images; and a means for changing the criteria for determining the suspiciousness of a target, according to the results of analyzing the stored images of the target extracted by the image search unit.

[0015] According to the security monitoring system in (4), when security monitoring tailored to the unique environment and circumstances of each monitored area is required, such as the area surrounding a specific private residence or a specific cultural heritage building, specific images that can contribute to real-time suspiciousness assessment can be extracted from past monitoring images stored in the recording unit, and the analysis results can be automatically contributed to the assessment of suspiciousness in real-time detection of targets requiring attention. This enables sophisticated and complex judgments that incorporate the history of past events and the chronological changes in events up to the present, and allows for detailed security monitoring in a manner that is optimally personalized to the environment and circumstances, without relying on an excessive burden on the "system administrator".

[0016] (5) A security monitoring method for monitoring a target area within a specific range, comprising: a shooting step in which a shooting unit photographs the target area and acquires a monitoring image; a detection step in which an image analysis unit detects a target area that is determined to have a certain level of suspiciousness or higher by performing an image analysis of the monitoring image using artificial intelligence; a monitoring necessity determination step in which the image analysis unit determines whether the target area to be checked is the same as a target area to be checked stored in the security monitoring system; a monitoring information notification step in which a monitoring information notification unit outputs warning information about the target area to be checked if it is determined to be the same as a target area to be checked, and outputs caution information about the target area to be checked if it is determined to be different from a target area to be checked; and a monitoring necessity information update step in which a monitoring necessity information update unit dynamically updates the content of the monitoring necessity information to match the determination result when the determination result regarding whether the user of the security monitoring system should continue monitoring the target area to be checked, which has been output as caution information, is input.

[0017] According to the security monitoring method in (5), in the field of security monitoring utilizing artificial intelligence (AI), when security monitoring tailored to the unique environment and circumstances of each monitoring target area is required, such as the area surrounding a specific private residence or a specific cultural heritage building, detailed security monitoring in a manner optimally personalized to that environment and circumstances can be realized without relying on an excessive burden on the "system administrator".

[0018] (6) The security monitoring method according to (5), wherein the monitoring requirement information also includes a list of targets that do not need to be monitored, which is specific to the area to be monitored, and the monitoring information output step unit does not output information about the target that needs to be checked if it is determined that the target that does not need to be monitored is the same as the target that does not need to be monitored.

[0019] According to the security monitoring method in (6), even if an artificial intelligence (AI) has detected a person or object that it has initially judged to be suspicious in the security monitoring method described in (5), if the "user" has judged that the detected object does not require monitoring based on its own criteria, the "user" can quickly remove it from the list of objects requiring monitoring through an operation that is easy for the "user" to perform, without relying on an excessive burden on the "system administrator". This makes it possible to optimize the system more efficiently and without omissions through dynamic updates of the monitoring necessity information.

[0020] (7) The security monitoring method according to (5) or (6), further comprising: a recording step in which a recording unit records the monitoring image as an stored image; and a stored image search step in which an image search unit searches the recorded monitoring image to extract a stored image of a specific monitoring target, wherein in the object requiring verification detection step, the criteria for determining the degree of suspicion of the monitoring target are automatically updated according to the results of analyzing the stored image extracted in the stored image search step, and the object requiring verification is detected based on the updated criteria.

[0021] According to the security monitoring method of (7), in the security monitoring method of (5) or (6), specific images that can contribute to real-time suspiciousness judgment can be extracted from past monitoring images stored in the recording unit, and the analysis results can also be automatically contributed to the judgment of suspiciousness in real-time detection of targets requiring attention. This makes it possible to make more advanced and complex judgments that incorporate not only the environment and situation in the monitored area, but also the history of changes in past events and events up to the present in chronological order, thereby realizing security monitoring in a manner that is more detailed and personalized to the actual situation of the monitored area.

[0022] (8) A security monitoring method for monitoring a target area within a specific range, comprising: a shooting step in which a shooting unit photographs the target area and acquires a monitoring image; a recording step in which a recording unit records the monitoring image as an stored image; a stored image search step st3 in which an image search unit searches the recorded monitoring image and extracts the stored image of a specific target; and a detection step for targets requiring confirmation in which an image analysis unit performs image analysis of the monitoring image using artificial intelligence and detects targets that are determined to have a certain level of suspicion or higher as targets requiring confirmation, wherein in the detection step for targets requiring confirmation, the criteria for determining the level of suspicion of the target are automatically updated according to the results of the analysis of the stored image extracted in the stored image search step st3, and the targets requiring confirmation are detected based on the updated criteria.

[0023] According to the security monitoring method in (8), when security monitoring tailored to the unique environment and circumstances of each monitoring target area is required, such as the area surrounding a specific private residence or a specific cultural heritage building, specific images that can contribute to real-time suspiciousness assessment can be extracted from past monitoring images stored in the recording unit, and the analysis results can be automatically contributed to the assessment of suspiciousness in real-time detection of targets requiring investigation. This enables sophisticated and complex judgments that incorporate the history of past events and the chronological changes in events up to the present, and allows for detailed security monitoring in a manner that is optimally personalized to the environment and circumstances, without relying on an excessive burden on the "system administrator".

[0024] (9) A security monitoring program for monitoring a monitoring target area within a specific range, comprising: a confirmation-required target detecting step of detecting, as a confirmation-required target, a monitoring target determined to have a suspicion level equal to or higher than a predetermined level by performing image analysis of a monitoring image using artificial intelligence; a monitoring necessity determining step of determining the identity between the confirmation-required target and a monitoring-required target stored in a security monitoring system; a monitoring information notifying step of outputting alert information for the confirmation-required target when it is determined that the confirmation-required target is identical to the monitoring-required target, and outputting caution information for the confirmation-required target when it is determined that the confirmation-required target is not identical to the monitoring-required target; and a monitoring necessity information updating step of dynamically updating the content of monitoring necessity information to match the determination result when the determination result regarding whether the user of the security monitoring system needs to continue monitoring the confirmation-required target output as the caution information is input. A security monitoring program that causes a computer to execute the security monitoring method in which the above steps are performed.

[0025] According to the security monitoring program of (9), when security monitoring adapted to the unique environment and situation of each individual monitoring target area is required, such as in the surrounding area of a specific private residence or a specific cultural property building, detailed security monitoring in an optimally personalized manner according to the environment and situation can be realized through operations that are easy for the "user" to perform by themselves, without depending on an excessive burden on a "system administrator".

[0026] (10) A security monitoring program for monitoring a monitoring target area within a specific range, comprising: a stored image searching step of searching recorded monitoring images to extract stored images of a specific monitoring target; and a confirmation-required target detecting step of detecting, as a confirmation-required target, a monitoring target determined to have a suspicion level equal to or higher than a predetermined level by performing image analysis of a monitoring image using artificial intelligence. A security monitoring program that causes a computer to execute the security monitoring method in which the determination criterion for the suspicion level is automatically changed according to a result of analyzing the stored image in the confirmation-required target detecting step.

[0027] According to the security monitoring program of (10), when security monitoring adapted to the unique environment and situation of each individual monitoring area is required, such as the surrounding area of a specific private residence or a specific cultural property building, specific images that can contribute to real-time suspicion degree determination can be extracted from past monitoring images stored in a recording unit, and the analysis results thereof can automatically contribute to the determination of the suspicion degree in the detection of objects requiring confirmation in real time. This enables advanced and complex determination that incorporates the historical context of past events and time-series changes of events up to the present, and realizes detailed security monitoring in an optimally personalized manner according to the environment and situation without depending on an excessive burden on the "system administrator". [Effect of the Invention]

[0028] According to the present invention, when security monitoring adapted to the unique environment and situation of each individual monitoring area is required, such as the surrounding area of a specific private residence or a specific cultural property building, detailed security monitoring in an optimally personalized manner adapted to said environment and situation can be realized without depending on an excessive burden on the "system administrator". Brief Description of the Drawings

[0029] [Figure 1] It is a block diagram showing the configuration of the security monitoring system of the present invention. [Figure 2] It is a flowchart showing the operation flow of the security monitoring method of the present invention. [Figure 3] It is a conceptual diagram schematically showing the basic design concept, configuration and operation aspects of an image analysis unit provided in the security monitoring system of the present invention. [Figure 4] It is a diagram schematically showing an example of a specific overall configuration arrangement of the security monitoring system of the present invention. Mode for Carrying Out the Invention

[0030] <Security Monitoring System> The best mode for carrying out the present invention will be described below with reference to the drawings as appropriate. The security monitoring system of the present invention (hereinafter also simply referred to as the "security monitoring system") is a monitoring system for maintaining security within a specific range of a monitored area.

[0031] The "security monitoring system" first automatically detects "targets requiring attention" using artificial intelligence (AI) image analysis and notifies the system's "users," who are the beneficiaries of the monitoring function, of the "attention information" or "warning information" requesting them to determine whether monitoring is necessary. Furthermore, this "security monitoring system" is configured so that users can easily determine whether the notified "targets requiring attention (attention information)" require continued monitoring and, if necessary, register the "targets requiring attention" to a list of "targets requiring monitoring" or exclude them from future monitoring ("addition to the list of items not requiring monitoring") by operating their own smartphone or other terminal. As a result, users of this "monitoring system" can continuously enjoy monitoring effects that are more finely tailored to each user's security requirements compared to conventional security monitoring systems, without relying on the support of the "system administrator." The specific configuration of this "security monitoring system," which is configured to realize such unique operations, will be explained in detail below.

[0032] [Overall structure] The basic configuration of the security monitoring system 100, which is an example of a preferred embodiment of the "security monitoring system," is shown in the block diagram of Figure 1. The security monitoring system 100 is composed of an imaging unit 1 and a processing unit 3. The processing unit 3 is composed of at least an image analysis unit 31, a monitoring necessity information storage unit 32, a monitoring information notification unit 33, and a monitoring necessity information update unit 35. Furthermore, it is more preferable that the security monitoring system 100 further includes a recording unit 2 in addition to the above components, and that the processing unit 3 further includes an image search unit 34.

[0033] Furthermore, regarding the user terminal 4 used by user 5 (see Figure 4), who is a "user" of the security monitoring system 100, the present invention can be implemented by user 5 using any general-purpose information processing terminal (such as a smartphone owned before starting to use the security monitoring system) that user 5 already possesses before acquiring the system, without any particular restrictions, as appropriate as needed. Therefore, the following explanation will assume that this user terminal 4 is not included in the minimum essential constituent elements of the "security monitoring system" of the present invention.

[0034] The specific arrangement of each component of the security monitoring system 100 can be, for example, as shown in Figure 4, such that a camera unit 1 (for example, a digital IP camera for video recording) installed in the area to be monitored (for example, the area where the monitored target 6 (user 5's home) is located), a processing unit 3 located on the cloud, and a user terminal 4 that user 5 can carry are all connected via a network. In this case, the communication means that constitute the network connecting the above components is not particularly limited and can be various wired or wireless telecommunication lines. This communication means can be a wired connection using a dedicated communication cable, or it can be a communication means that goes through the internet or various telephone lines. Furthermore, it is preferable that these communication means are configured to enable communication using a communication protocol compliant with the ONVIF (Open Network Video Interface Forum) standard, but are not limited to this, and may be configured to enable communication using other communication protocols, such as a proprietary communication protocol.

[0035] [Photography Department] The imaging unit 1 is a device that captures images of the area to be monitored and acquires surveillance images. The camera constituting the imaging unit 1 can output the captured surveillance images as digital image data on the network constituting the security monitoring system 100 so that they can be processed by the processing unit 3. Specifically, various existing digital IP cameras for video recording that have video recording capabilities can preferably be used as the camera constituting the imaging unit 1.

[0036] Furthermore, it is preferable to use a camera equipped with a wide-angle lens and zoom function as the camera constituting the imaging unit 1, so that a wide monitoring area can be covered with a single camera. Alternatively, it is also preferable to use a PTZ camera that can be remotely controlled for pan, tilt, and zoom. With these cameras, a wider area can be monitored with a single camera. In other words, by using these cameras, the entire area to be monitored can be covered with fewer cameras.

[0037] Furthermore, especially when the area to be monitored is outdoors or is expected to be dark even indoors, it is preferable to use a camera equipped with a high-sensitivity sensor capable of shooting in low-light environments as the camera constituting the shooting unit 1. Outdoor-compatible security cameras equipped with high-sensitivity sensors that can capture subjects in color even with a faint light source of as little as 0.009 lux are available on the market, and by adopting a camera equipped with such a high-sensitivity sensor as the camera constituting the shooting unit 1, the security monitoring system 100 can be made sufficiently effective even for nighttime or dark-area monitoring.

[0038] Furthermore, among the cameras comprising the shooting unit 1, it is particularly effective to equip the camera that monitors the entrances and exits of the monitored area (for example, the front door of a private residence) with a facial recognition function capable of authenticating the faces of people entering and exiting the aforementioned entrances and exits. The facial recognition function may be built into the shooting unit 1 or incorporated into the processing unit 3. By configuring the security monitoring system 100 in this way, the security monitoring system 100 can distinguish between people who normally reside in the monitored area and other people, and can take special measures for individuals on blacklists and whitelists. This makes it possible to achieve a more highly adaptable and detailed monitoring effect tailored to the specific circumstances of each monitored area.

[0039] Furthermore, a general monocular camera capable of continuously capturing the three-dimensional space to be captured as a two-dimensional image can be used as the camera constituting the shooting unit 1. Alternatively, a 3D camera capable of directly acquiring three-dimensional positional information within the space that can be captured can also be used as the camera constituting the shooting unit 1. When the above monocular camera is used as the camera constituting the shooting unit 1, it is more preferable that the security monitoring system 100 further includes a coordinate setting unit (not shown in Figure 1) that sets coordinates that can identify the position in the "monitoring image" in relation to the actual position within the three-dimensional space that is the captureable area. The coordinate setting unit may be built into the shooting unit 1 or incorporated into the calculation processing unit 3. By providing a coordinate setting unit capable of generating three-dimensional coordinates from two-dimensional image information, it is possible to detect human movements and object movements with high accuracy even from images that only have two-dimensional information obtainable by an inexpensive monocular camera, without introducing an expensive 3D camera.

[0040] [Recording Section] The recording unit 2 is a device that records past images of the monitoring area and stores them as stored images. The recording unit 2 should have the function of saving the monitoring images (past recorded images) captured by the shooting unit 1 as digital image data so that they can be processed by the processing unit 3 after a certain period of time has elapsed, and to retrieve the necessary parts as digital data as needed. Specifically, existing digital recorders with various video recording functions and storage for data storage can be used as the recorders that make up the recording unit 2. Alternatively, the recording unit 2 may be a recording device in which a digital recording device and storage for data storage are integrated, or it may be a recording system in which devices having the above functions are distributed on a network.

[0041] [Processing Unit] The processing unit 3 analyzes surveillance images using artificial intelligence (AI) image analysis technology, detects highly suspicious objects, and notifies the user terminal 4 held by the user 5 of security-related information. To perform this function, the processing unit 3 is equipped with an image analysis unit 31, a monitoring necessity information storage unit 32, a monitoring information notification unit 33, and a monitoring necessity information update unit 35 as essential components.

[0042] Furthermore, if the security monitoring system 100 is configured to include a recording unit 2, it is preferable that the arithmetic processing unit 3 also be configured to include an image search unit 34. This allows the security monitoring system 100 to automatically contribute the analysis results of past monitoring images (stored images) to real-time artificial intelligence (AI)-based judgment of suspiciousness.

[0043] The arithmetic processing unit 3 is a known information processing device equipped with hardware such as a CPU, memory, and communication unit, and can be configured as a computer system that appropriately combines various information processing equipment and electronic devices capable of performing the functions required for each of the above-mentioned units (image analysis unit 31, monitoring necessity information storage unit 32, monitoring information notification unit 33, image search unit 34, and monitoring necessity information update unit 35). Furthermore, the arithmetic processing unit 3 only needs to be connected to the shooting unit 1, recording unit 2, and user terminal 4 so that data can be sent and received from each other via the various communication networks described above.

[0044] Furthermore, in the security monitoring system 100, the computer system constituting the arithmetic processing unit 3 can be placed on the cloud as a server, and this server can be shared by multiple clients (individual cameras constituting the imaging unit 1, multiple smartphones used as user terminals 4, etc.). In addition, the arithmetic processing unit 3 can automatically execute each operation of the security monitoring system 100 according to the instructions of the computer program (security monitoring program) according to the present invention.

[0045] [Image Analysis Department] The image analysis unit 31 analyzes surveillance images using the object recognition and behavioral analysis functions of artificial intelligence (AI), and includes a "detection means for objects requiring verification" that detects surveillance objects (people, objects, vehicles, etc.) that the artificial intelligence (AI) has determined to have a certain level of suspicion or higher as "objects requiring verification," and a "monitoring necessity determination means" that determines the identity of the detected "objects requiring verification" with the "objects requiring monitoring" included in the monitoring necessity information stored in the monitoring necessity information storage unit 32.

[0046] Furthermore, as described above, the security monitoring system 100 can be configured to include a recording unit 2 and an image search unit 34. In this case, the image analysis unit 31 can be equipped with a "suspiciousness judgment criterion changing means" that automatically changes the suspiciousness judgment criteria for the monitored target according to the results of analyzing the stored images of the monitored target extracted by the image search unit 34. This enables advanced and complex judgments by artificial intelligence (AI) that incorporate the history of past events and the chronological changes in events up to the present, thereby realizing detailed security monitoring in a manner that is optimally personalized to the environment and situation.

[0047] (Methods for changing the criteria for determining the degree of suspicion) The means for changing the suspiciousness judgment criteria can be comprised of various computational processing means that have the function of automatically changing the judgment criteria (for example, the threshold for the suspiciousness score) used by artificial intelligence (AI) to determine the suspiciousness of the monitored target, in accordance with the results of analyzing the stored images of the monitored target extracted by the image search unit 34. Specific details of this judgment criterion changing process will be described later as an embodiment of the "security monitoring method" of the present invention.

[0048] (Means for detecting objects requiring verification) The object detection means can be comprised of various known image analysis methods capable of extracting the form, size, category, and specific movements and actions of "monitoring targets" through image analysis. Specific examples of such image analysis methods include various machine learning-type image analysis devices with neural networks (so-called "image recognition devices using deep learning technology"). By mounting a machine learning model trained to detect monitoring targets (people, objects, vehicles, etc.) onto the image analysis device, the object detection means constituting the image analysis unit 31 can be configured. Specific examples of image recognition technology using deep learning are also publicly available below. "Deep Learning and Image Recognition, Operations Research" (http: / / www.orsj.o.jp / archive2 / or60-4 / or60_4_198.pdf)

[0049] Furthermore, as an example of a technical means for mechanically recognizing the details of the actions and movements of a monitored object in a surveillance image, a known image analysis technique called "OpenPose" can be cited. As disclosed by the present inventor in Japanese Patent Publication No. 6534499, by using "OpenPose," it is possible to extract skeletal information of a person in the image and recognize the actions of a person in the image by analyzing the position and velocity of each feature point constituting the skeleton. In addition, by using image analysis techniques such as "OpenPose," it is also possible to extract the hand movements of the monitored person and the movements of objects in the surveillance image as independent information from the "surveillance image," so it is possible to comprehensively judge the degree of suspiciousness of a person's actions toward an object, such as "a person grabbing an object and taking it away."

[0050] (Method for determining whether monitoring is necessary) The means for determining whether monitoring is necessary can be comprised of various known image analysis means (for example, the "image recognition device using deep learning technology" mentioned above) that can automatically determine the identity (or non-identity) of an image by comparing it with an image of an object that requires verification, or an image of an object that requires verification with an image of an object that does not require monitoring.

[0051] [Monitoring Requirement Information Storage Unit] The monitoring necessity information storage unit 32 has the function of storing "monitoring necessity information" which serves as a criterion for determining whether monitoring is necessary in a monitoring target area within a specific range. The monitoring necessity information includes at least a list of "targets requiring monitoring" which are targets that require monitoring in the monitoring target area where monitoring is actually performed. Preferably, it also includes a list of "targets not requiring monitoring" which are targets that do not require monitoring in the same monitoring target area. The monitoring necessity information storage unit 32 can be configured with various information storage means (storage devices) capable of registering database information of each of these pieces of information.

[0052] [Monitoring information notification department] The monitoring information notification unit 33 has the function of selectively notifying the user 5 of multiple types of monitoring information with different required alert levels, namely "alert information" and "warning information". If the monitoring information notification unit 33 determines that the "object requiring confirmation" detected by the image analysis unit 31 is the same as any of the "objects requiring monitoring" stored in the monitoring necessity information storage unit 32, it selectively outputs "alert information" for that "object requiring confirmation". On the other hand, if the above "object requiring confirmation" is determined to be different from the above "objects requiring monitoring", it selectively outputs "warning information" for that "object requiring confirmation". The monitoring information notification unit 33 outputs the monitoring information ("alert information" or "warning information") that has been selectively output in this way to the user terminal 4 held by the user 5.

[0053] In the security monitoring system 100, the distinction between "warning information" and "caution information" is clearly defined, and information with different required warning levels depending on the monitored object is selectively output and notified to the user 5. In particular, by selecting "caution information" and notifying the user 5 to decide whether or not to continue monitoring, and by cyclically executing a process in which the result of this decision is reliably reflected in the subsequent suspicion level judgments made by artificial intelligence (AI) through an operation that is easy for the user 5 to perform, the system proceeds while always optimally integrating the advanced computational processing capabilities of artificial intelligence (AI) with the user's judgment that is in line with the actual situation of the monitored area. This is a technical feature that differs from conventional security monitoring systems equipped with artificial intelligence (AI).

[0054] Furthermore, if the monitoring necessity information stored in the monitoring necessity information storage unit 32 includes a list of non-monitoring targets specific to the monitored area, the monitoring information notification unit 33 may choose not to output either warning information or caution information for a target requiring confirmation detected by the image analysis unit 31 if it determines that the target requiring confirmation is the same as a non-monitoring target.

[0055] [Image Search Section] The image search unit 34 has the function of searching the recorded surveillance images (stored images) recorded and saved by the recording unit 2 to extract a specific surveillance target. The image analysis unit 31 automatically changes the suspiciousness judgment criteria for the surveillance target using the suspiciousness judgment criteria changing means described above, according to the results of analyzing the stored images of the extracted surveillance target.

[0056] As the image search unit 34, various known information processing devices that implement algorithms for rapidly searching for specific features (e.g., color, shape, or human movement) from image data can be used. A specific example of such an information processing device is the information processing device disclosed by the present inventor in Japanese Patent No. 6593949. This information processing device has a category designation unit, an object (monitoring target) extraction unit, an object (monitoring target) analysis unit, and an analysis unit, and has the function of extracting analysis target objects (monitoring targets) belonging to the analysis target category designated by the category designation unit from recorded images (stored images).

[0057] In the information processing device according to the above-mentioned patent invention, the extraction of objects (monitoring images) by the object (monitoring target) extraction unit is performed at high speed by a machine learning type image recognition means having a neural network, a so-called deep learning type image recognition means.

[0058] The algorithm for the image recognition processing means that extracts objects (monitoring targets) from recorded images (stored images) is not particularly limited, but "You Only Look Once (YOLO)" can be preferably used. For example, by using "You Only Look Once (YOLO)" as the image recognition means for extracting and identifying objects (monitoring targets) in the object (monitoring target) extraction unit, it is possible to simultaneously and individually extract approximately 1000 types of analysis target objects (monitoring targets). By utilizing this function, useful objects that the security monitoring system 100 needs to perform monitoring in a more preferable manner can be extracted quickly and accurately from stored images accumulated within a certain period of time in the past.

[0059] Furthermore, it is also effective to provide the image search unit 34 with a function to tag people and objects that are being monitored. This allows for even faster searching and detection of tagged monitored objects, dramatically improving the detection accuracy and speed of the AI-based object detection means. Additionally, tagging allows for the transmission of image data for some communications to be handled solely by the exchange of tag information, significantly reducing the volume of data transmitted over the network in the security monitoring system 100 and the processing burden on the server side.

[0060] Furthermore, in the information processing device according to the aforementioned patent invention, it is also effective to equip the object (monitoring target) extraction unit with a function that allows for the individual recognition of a person's gender and age using facial recognition technology. This makes it possible to specify the analysis target category when searching stored images as an attribute such as "woman in her 30s."

[0061] [Monitoring necessity information update section] The monitoring necessity information update unit 35 has monitoring necessity information update means that, when the user 5 inputs a decision result regarding whether or not monitoring should be continued for an item requiring confirmation output as warning information by the monitoring information notification unit 33, dynamically updates the content of the monitoring necessity information stored in the monitoring necessity information storage unit 32 to match the user 5's decision result.

[0062] (Monitoring necessity information update means) The monitoring necessity information update means constituting the monitoring necessity information update unit 35 can be composed of various calculation processing means that have the function of dynamically updating the content of the monitoring necessity information to match the user 5's judgment result. Specific details of this update process will be described later as part of the description of an embodiment of the "security monitoring method" of the present invention.

[0063] [User terminal] In the security monitoring system 100, user 5 receives monitoring information ("warning information" or "caution information") output by the monitoring information notification unit 33 via user terminal 4. Therefore, firstly, user terminal 4 must have the function to allow user 5 of the security monitoring system 100 to receive monitoring information ("warning information" or "caution information") output from the monitoring information notification unit 33.

[0064] Furthermore, the user 5 transmits the result of their decision regarding whether further monitoring is necessary for the received monitoring information to the monitoring necessity information update unit 35 via the user terminal 4. Therefore, secondly, the user terminal 4 must have a function to input the user's own decision regarding whether further monitoring is necessary for the received monitoring information and transmit the said decision result to the monitoring necessity information update unit 35. In detail, it is preferable that the user 5 be able to input the decision result, such as "monitoring is necessary" or "monitoring is not necessary," with a single click on the monitoring information notification screen.

[0065] As long as the device has the first and second functions described above, the user 5 of the security monitoring system 100 can use various information processing terminals as user terminals 4, such as small portable information processing terminals like smartphones that can be easily carried, or personal computers with monitors, which can appropriately display the above-mentioned monitoring information in recognizable video, text, audio, etc., by installing a dedicated application to enable the first and second functions described above on such devices. In other words, the user 5 can use the security monitoring system 100 and enjoy its crime prevention effects by using various information processing terminals that they already possess or have been lent to them as user terminals 4.

[0066] <Security Monitoring Method (Operation of the Security Monitoring System)> Figure 2 is a flowchart showing the flow of the security monitoring method of the present invention (hereinafter also simply referred to as the "security monitoring method"). As shown in Figure 2, the "security monitoring method" is a process that cyclically performs the following steps: shooting step st1 (acquisition of monitoring images), recording step st2 (recording and saving of monitoring images), stored image search step st3 (adjustment of suspiciousness judgment criteria based on stored image data), detection of target requiring confirmation step st4 (automatic detection of monitoring targets (targets requiring confirmation) by artificial intelligence (AI)), determination of monitoring necessity step st5 (determination of identity between targets requiring confirmation and targets requiring monitoring), notification of monitoring information step st6 (notification of monitoring information (warning information or alert information)), update of monitoring necessity information step st7 (manual update of monitoring necessity information by the user of the security monitoring system), and completion of monitoring step st8.

[0067] The "security monitoring method" according to the above embodiment (hereinafter referred to as the "first embodiment") can be implemented using the security monitoring device 100 of the present invention, which comprises a shooting unit 1, a recording unit 2, an image analysis unit 31, a monitoring necessity information storage unit 32, a monitoring information notification unit 33, an image search unit 34, and a monitoring necessity information update unit 35. The "security monitoring method" according to this "first embodiment" will now be described in detail.

[0068] [Shooting Steps] In the shooting step st1, the shooting unit 1 captures images of the area to be monitored, and the monitoring images are acquired in real time or periodically. The monitoring images captured in the shooting step st1 are transmitted as digital image data to each unit that performs subsequent processing.

[0069] [Recording Steps] In recording step st2, the surveillance image acquired in shooting step st1 is transmitted in real time to recording unit 2, where it is recorded and stored as digital data. In this specification, the surveillance image stored in recording unit 2 in this manner is referred to as "stored image". It is preferable to use a compression algorithm (H.264, H.265) to reduce the data size of the stored image during recording in recording step st2. Furthermore, it is preferable to store the image data of the stored image in various storage devices on a cloud server or various standalone storage devices connected to a network, and to manage it in a data structure that allows for quick access in subsequent steps.

[0070] Furthermore, in recording step st2, it is preferable to save the "stored images" in the recording unit 2 with various metadata attached (for example, "shooting date and time (timestamp)", "shooting location (camera ID and coordinate information)", "subject characteristics (attribute information of detected people or objects)", etc.). By automatically storing video footage from a certain period in the past in this data format in recording step st2, it becomes possible to quickly search and extract the necessary image data from the stored images based on a specific date, time, or characteristics, such as specifying "the person who entered the garden last night".

[0071] Furthermore, while it is preferable to continuously record surveillance images 24 hours a day, 365 days a year in recording step st2, depending on the purpose of monitoring, it is also possible to start recording only when an abnormality or a precursor to an abnormality is detected in the monitored area by image analysis of the surveillance images using artificial intelligence (AI).

[0072] [Accumulated Image Search Step] In the stored image search step st3, the image search unit 34 searches the stored images for images of specific monitored objects that can contribute to determining the degree of suspicion in extracting objects requiring verification, based on specific search conditions (e.g., time, location, characteristics of the object). Through such a search, image data of the monitored object can be extracted as information including metadata about its behavioral history and movement routes within the monitored area. Specifically, as an example, image data that satisfies search conditions such as "person wearing a red hat" and "vehicle license plate number XYZ123" can be extracted and transmitted to the units that perform subsequent processing.

[0073] [Step to detect items requiring verification] In step st4, the object requiring verification detection unit 31 performs image analysis of the surveillance images using artificial intelligence (AI). As a result, surveillance objects that the AI ​​determines to have a level of suspicion above a certain threshold (above a predetermined threshold) are detected as objects requiring verification.

[0074] The detection of objects requiring verification is performed by using artificial intelligence (AI) image analysis functions to extract feature quantities of objects (people, vehicles, etc.) from the surveillance images, recognize people and objects, and then analyze the movement and actions of the recognized objects (people, vehicles, etc.) over time to calculate a suspiciousness score. Objects that are judged to have a suspiciousness score above a certain level are then labeled as "objects requiring verification" and detected, and transmitted to the respective units for subsequent processing. Furthermore, in the object requiring verification detection step st4, the image analysis unit 31 works in cooperation with multiple cameras that make up the shooting unit 1 to comprehensively analyze information from the entire surveillance area, thereby analyzing the relationships and interrelationships between multiple surveillance objects and generating more accurate warning and caution information.

[0075] Furthermore, in the object detection step st4, the criteria for determining the degree of suspicion of the monitored object are automatically adjusted as appropriate based on the results of analyzing the accumulated images. This adjustment of the criteria for determining the degree of suspicion is carried out by collecting data on past detection frequency, detection location, and characteristic movement patterns for each monitored object detected by the object detection means through analysis of accumulated images, and adjusting the process to appropriately change the criteria for determining the degree of suspicion for each monitored object based on the newly obtained data.

[0076] As a concrete example of an embodiment of adjusting the suspiciousness judgment criteria, the process shown in Table 1 below can be illustrated. For example, as shown in Table 1, if "for a certain vehicle in the surveillance image, the same vehicle is also found in the stored images, and the analysis results indicate that the vehicle appears in the same location with a certain frequency or higher," then "the weighting of the artificial intelligence (AI) suspiciousness score calculation model should be changed so that a higher suspiciousness score (a numerical value representing the level of suspiciousness) is calculated for the vehicle in question." Alternatively, this adjustment may be made by lowering the threshold for determining something as suspicious so that a suspicious judgment result is more likely to be output. In any case, by making such dynamic adjustments using past surveillance data (stored images) performed automatically, the suspiciousness judgment criteria can be continuously adapted to the specific environmental changes and behavioral patterns of the target of the actual monitored area without imposing additional work burdens on system administrators or users, such as manually searching for past data.

[0077] [Table 1]

[0078] [Monitoring necessity determination step] In the monitoring necessity determination step st5, the monitoring necessity determination means implemented in the image analysis unit 31 compares the image data of the "object requiring confirmation" detected in the object requiring confirmation detection step st4 with the image data of the "object requiring monitoring" stored in the monitoring necessity information storage unit 32, or the image data of the "object requiring confirmation" with the image data of the "object not requiring monitoring," and automatically determines whether they are identical (or not identical).

[0079] [Monitoring Information Notification Step] In the monitoring information notification step st6, based on the determination result in the monitoring necessity determination step st5, the monitoring information notification unit 33 outputs "warning information" or "caution information" and transmits it to the user 5 in real time. The "warning information" and "caution information" include detailed information such as the image of the monitored object, the degree of suspicion and the reason for the degree of suspicion determination, the detection location, and the time.

[0080] Furthermore, "warning information" is applied to targets of high urgency and is notified to the user 5 via the user terminal 4 through voice alarms and highlighting. On the other hand, "caution information" is applied to targets of relatively low risk, and detailed information about the detected target is notified to the user 5 so that the user 5 can quickly determine whether monitoring is necessary, and the system prompts the user 5 to confirm whether monitoring is necessary.

[0081] Specifically, as shown in Table 2 below, if the object requiring verification detected in the monitoring necessity determination step st5 is determined to be the same as the "object requiring monitoring," a "warning information" for that object is output. If the object requiring verification is determined to be different from the "object requiring monitoring," a "caution information" for that object is output.

[0082] Furthermore, as shown in Table 2, if the object requiring verification is determined to be the same as an object that does not require monitoring, neither piece of information can be output. This allows users to quickly remove objects that they have initially identified as suspicious by artificial intelligence (AI) from the list of objects requiring monitoring, without placing an excessive burden on the system administrator. This is achieved through an easy operation performed by the user themselves. As a result, the system can be optimized more efficiently and without omissions through dynamic updates of monitoring requirements information.

[0083] [Table 2]

[0084] (Monitoring requirement information update step) In the monitoring necessity information update step st7, the monitoring necessity information update unit 35 dynamically updates the contents of the "targets requiring monitoring" stored in the monitoring necessity information storage unit 32 based on the user 5's judgment. For example, if the monitoring information notification unit 33 outputs a target requiring verification (e.g., a suspicious person) to the user 5 as a warning, and the user 5 determines that "this person is an acquaintance of mine and no further monitoring is necessary," and sends an instruction to that effect from the user terminal 5 to the arithmetic processing unit 3 which constitutes the security monitoring system 100 via a telecommunications line, the monitoring necessity information update unit 35, upon receiving the instruction, executes a process to register the suspicious person in the list of targets that do not require monitoring in the monitoring necessity information. Similarly, if, in the same manner as above, user 5 determines that a vehicle requiring further investigation (e.g., a suspicious vehicle) has been output as a warning, and sends an instruction to that effect from user terminal 5 to the processing unit 3 constituting the security monitoring system 100 via a telecommunications line, the monitoring necessity information update unit 35, upon receiving the instruction, executes a process to register the suspicious vehicle in the list of items requiring monitoring in the monitoring necessity information. This enables continuous and detailed determination of the degree of suspiciousness that is tailored to the specific circumstances of the monitored area.

[0085] The "security monitoring method" of the present invention can also be a process (second embodiment) in which only the following steps are performed cyclically: the shooting step st1, the step of detecting an object requiring confirmation st4, the step of determining whether monitoring is necessary st5, the step of notifying monitoring information st6, the step of updating monitoring necessity information st7, and the step of ending monitoring st8. This "security monitoring method" according to the second embodiment can also be implemented using the "security monitoring device" of the present invention. In this case, the recording unit 2 and the image search unit 34 are not essential components of the above-described configuration, and the "security monitoring method" of the present invention can be implemented using the "security monitoring device" of the embodiment that has the other components described above.

[0086] Alternatively, the "security monitoring method" of the present invention can also be a process (third embodiment) in which only the following steps are performed cyclically: the shooting step st1, the recording step st2, the stored image search step st3, the step st4 for detecting objects requiring confirmation, the monitoring information notification step st6, and the monitoring termination step st8. This "security monitoring method" according to the third embodiment can also be implemented using the "security monitoring device" of the present invention. In this case, the monitoring necessity information storage unit 32 and the monitoring necessity information update unit 35 are not essential components of the above-described configuration, and the "security monitoring method" of the present invention can be implemented using the "security monitoring device" of the embodiment having the other components described above. [Examples]

[0087] The following are examples of more specific "implements" of the "security monitoring system" of the present invention.

[0088] [Example 1] <Example of "security monitoring system" operation in a high-end residential area> [System Overview] (Installation location) A private residence in a high-end residential area. It boasts a large plot of land and features multiple entrances, gardens, and a garage. (Purpose of using the system) Ensuring the safety of residents, detecting suspicious individuals and vehicles, and monitoring mail and deliveries. [Monitoring Flow] (Camera installation) High-resolution IP cameras will be installed in the garden and in front of the garage as part of the "Shooting Unit 1." The cameras will operate 24 hours a day to capture surveillance images in real time (shooting step st1). (Recorded) The surveillance images are saved as stored images in the recording unit 2, and metadata (shooting time, camera ID, target characteristics) is automatically added (recording step st2). (Anomaly detection trigger) A vehicle was parked in front of the garage, and camera unit 1 (high-resolution IP camera) detected that it had been there for more than 3 minutes. (Setting search criteria) The system searches the archived images for the history of a vehicle, using criteria such as vehicle color, partial license plate information, and parking location. (Search results) The video footage from the past week was searched, and it was determined that the vehicle in question had been parked at the same location several times (accumulated image search step st3). (Analysis process) The retrieved stored images were re-analyzed using AI to detect a history of the vehicle appearing multiple times during nighttime parking restrictions. This confirmed a behavioral pattern of the vehicle loitering near the garage. (Detection of items requiring verification and updating of suspiciousness assessment criteria) Based on past behavioral history, the suspiciousness score increases. The AI ​​determines that the individual has a certain level of suspiciousness and is a "subject to monitoring" (subject to investigation). (Subject to Investigation Detection Step st4) (Determination of necessity of monitoring) For items requiring monitoring, the need to notify the user (resident of the managed property) is determined by cross-referencing with monitoring necessity information (monitoring necessity determination step st5). (Notification to users) The user's smartphone receives a notification of the monitoring information (monitoring information notification step st6). An example of the notification content is, "There is a vehicle parked in front of the garage for an extended period of time. There is a history of similar activity in the past week. Please be vigilant as it is a suspicious vehicle." A thumbnail image or a link to play the video is attached to the notification, allowing the user 5 to immediately check the situation. (Update of monitoring requirements information) User 5 reviews the notification and selects either "Add to monitoring list" or "No monitoring required" with a single click on their smartphone. This allows the system to dynamically update the monitoring list and reflect the changes in future decisions (monitoring requirement information update step st7). For example, if the detected person requiring monitoring is a family member or acquaintance, they can be registered as a person authorized to visit, eliminating the need for further notifications.

[0089] In the embodiment of Example 1, it is possible to enjoy the effect of crime prevention by improving the safety of private residences and detecting suspicious persons early. Furthermore, by utilizing past data and continuously optimizing the criteria for suspicious behavior, automatic analysis and notification by AI are performed cyclically, which reduces false alarms and alleviates the burden on users (residents) in responding, while strengthening the security system.

[0090] [Example 2] <Example of using a "security monitoring system" in a parking lot> [System Overview] (Installation location) Parking lot of a large commercial facility. Surveillance cameras are positioned to cover parking spaces, vehicle entrances and exits, and pedestrian walkways. Entrance gate camera: Records vehicle license plates and vehicle types. Parking space camera: Monitors vehicle parking status and records dwell time and movement. Pedestrian walkway cameras: Monitor pedestrian movement and detect suspicious behavior. (Purpose of using the system) Ensuring safety within the parking lot, detecting suspicious vehicles and behavior, and monitoring for abandoned items and parking violations. [Monitoring Flow] (Recorded) The surveillance camera records images of vehicles and pedestrians (shooting step st1) and stores them in the recording unit 2. Characteristic information such as the vehicle's license plate, color, shape, and parking location is recorded as metadata. The surveillance images are saved as stored images in the recording unit 2, and metadata (shooting time, camera ID, target characteristics) is automatically added (recording step st2). (Anomaly detection trigger) If a vehicle stops outside a parking space or a pedestrian makes suspicious movements, it will be detected as suspicious behavior. (Setting search criteria) Enter search criteria related to vehicles or people (e.g., specific license plate number, specific time period). (Search results) The system performs a high-speed search of footage from the past week and lists the relevant videos. The search results display timestamps and vehicle location information (stored image search step st3). (Analysis process) The system analyzes the retrieved stored images to evaluate past behavioral patterns and any anomalies. For example, vehicles that have been illegally parked in the past will have their suspiciousness score increased. (Detection of items requiring verification and updating of suspiciousness assessment criteria) Based on past behavioral history, the suspiciousness score increases. The AI ​​determines that the individual is a "subject to surveillance" (subject to investigation) with a suspiciousness level above a certain threshold. (Subject to Investigation Detection Step st4). (Determination of necessity of monitoring) For items requiring monitoring, the need to notify the user (parking lot manager) is determined by comparing them with the monitoring necessity information (monitoring necessity determination step st5). (Notification to users) Monitoring information is sent to the user's smartphone (monitoring information notification step st6). A thumbnail image and a link to play the video are attached to the notification, allowing user 5 to immediately check the situation. (Update of monitoring requirements information) User 5 reviews the notification and selects either "Add to monitoring target" or "No monitoring required" with a single click on their smartphone. This allows the system to dynamically update the monitoring target list and reflect the changes in future decisions (monitoring requirement information update step st7).

[0091] In the embodiment of Example 2, the system can improve safety within the parking lot and detect illegal parking, thereby preventing accidents and crimes. Furthermore, by utilizing past data and continuously optimizing the criteria for suspicious behavior, the system can cyclically perform automated analysis and notifications using AI, significantly reducing the workload of the user (monitoring staff). [Explanation of symbols]

[0092] 1. Photography Department 2. Recording section 3. Processing Unit 31 Image Analysis Department 32. Monitoring Requirement Information Storage Unit 33 Monitoring information notification section 34 Image Search Section 35 Monitoring necessity information update section 4. User terminal 5. Users (Users of the security monitoring system) 6. Target of monitoring (user's home) 7. Suspicious Person 8. Suspicious vehicle 100 Security Monitoring Systems st1 Shooting Steps st2 recording step st3 Stored Image Search Step st4 Step to detect items requiring confirmation st5 Monitoring necessity determination step st6 Monitoring Information Notification Step st7 Monitoring necessity information update step st8 Monitoring termination step

Claims

1. A security monitoring system that monitors a target area within a specific range, A camera unit that acquires a surveillance image of the area to be monitored, An image analysis unit that performs image analysis of the aforementioned surveillance images using the image analysis function of artificial intelligence, A monitoring necessity information storage unit stores monitoring necessity information, which includes a list of items that need to be monitored specific to the monitored area. A monitoring information notification unit that notifies the user of the security monitoring system of monitoring information, It includes a monitoring necessity information update unit, The image analysis unit includes a means for detecting objects requiring confirmation that analyze the surveillance image and determine that the surveillance object has a certain level of suspicion or higher, and a means for determining whether an object requires confirmation is the same as the object requiring surveillance included in the surveillance necessity information. The monitoring information notification unit has monitoring information selection output means that outputs warning information for the target requiring confirmation if it is determined that the target requiring confirmation is the same as the target requiring monitoring, and outputs caution information for the target requiring confirmation if it is determined that the target requiring confirmation is not the same as the target requiring monitoring. The security monitoring system includes a monitoring necessity information updating unit which, when the user's decision result regarding whether or not to continue monitoring the object requiring confirmation, which has been output as warning information, is input, dynamically updates the content of the monitoring necessity information to match the decision result.

2. The monitoring requirement information also includes a list of items that do not require monitoring, which is specific to the area being monitored. If the monitoring information notification unit determines that the item requiring confirmation is the same as the item not requiring monitoring, it will not output information about the item requiring confirmation. The security monitoring system according to claim 1.

3. A recording unit that records the aforementioned monitoring image as an accumulated image, The system further includes an image search unit that searches the stored images to extract specific monitoring targets, The image analysis unit has a means for changing the suspiciousness criteria for the monitored target, which automatically changes the suspiciousness criteria for the monitored target in accordance with the results of analyzing the stored images of the monitored target extracted by the image search unit. The security monitoring system according to claim 1 or 2.

4. A security monitoring system that monitors a target area within a specific range, A camera unit that acquires surveillance images of the area to be monitored, A recording unit that records the aforementioned monitoring image as an accumulated image, An image search unit that searches the stored images and extracts a specific target for monitoring, An image analysis unit that performs image analysis of the aforementioned surveillance images using artificial intelligence, The system includes a monitoring information notification unit that notifies the user of the security monitoring system of monitoring information, The image analysis unit includes a means for detecting objects requiring confirmation that analyze the surveillance images and determine that they have a certain level of suspicion or higher, and a means for changing the criteria for determining the suspiciousness of the surveillance object according to the results of analyzing the accumulated images of the surveillance object extracted by the image search unit. Security monitoring system.

5. A security monitoring method that monitors a target area within a specific range, The camera unit performs a shooting step to capture the area under surveillance and acquire a surveillance image, The image analysis unit performs image analysis of the surveillance images using artificial intelligence, and detects surveillance targets that are determined to have a certain level of suspicion or higher as targets requiring further investigation in the process of detecting targets requiring further investigation. The image analysis unit performs a monitoring necessity determination step in which it determines whether the object requiring verification is the same as the object requiring monitoring stored in the security monitoring system, A monitoring information notification step in which the monitoring information notification unit outputs warning information for the target requiring confirmation if it determines that the target requiring confirmation is the same as the target requiring monitoring, and outputs caution information for the target requiring confirmation if it determines that the target requiring confirmation is not the same as the target requiring monitoring, The monitoring necessity information update unit dynamically updates the content of the monitoring necessity information to match the judgment result when it receives the judgment result from the user of the security monitoring system regarding the item requiring confirmation, which has been output as warning information, so as to be consistent with the judgment result. A security monitoring method that includes the following features.

6. The aforementioned monitoring requirement information also includes a list of items that do not require monitoring, which is specific to the area being monitored. If the monitoring information output step unit determines that the object requiring verification is the same as the object not requiring monitoring, it will not output information about the object requiring verification. The security monitoring method according to claim 5.

7. A recording step in which the recording unit records the aforementioned monitoring image as an accumulated image, The system further includes a stored image search step in which an image search unit searches the recorded surveillance images and extracts stored images of a specific surveillance target, In the aforementioned step of detecting an object requiring verification, the criteria for determining the degree of suspicion of the monitored object are automatically updated according to the results of analyzing the stored images extracted in the stored image search step, and the object requiring verification is detected based on the updated criteria. The security monitoring method according to claim 5 or 6.

8. A security monitoring method that monitors a target area within a specific range, The camera unit performs a shooting step to capture the area under surveillance and acquire a surveillance image, The recording unit records the aforementioned monitoring image as an accumulated image in a recording step, The image search unit performs a stored image search step st3, which searches for the recorded surveillance images and extracts the stored images of a specific surveillance target. The system includes a step of detecting objects requiring confirmation, in which the image analysis unit performs image analysis of the surveillance images using artificial intelligence and detects surveillance objects that are determined to have a certain level of suspicion or higher as objects requiring confirmation. In the aforementioned step of detecting an object requiring verification, the criteria for determining the degree of suspicion of the monitored object are automatically updated according to the results of analyzing the stored images extracted in the stored image search step, and the object requiring verification is detected based on the updated criteria. Security monitoring methods.

9. A security monitoring program that monitors a target area within a specific range, A step to detect objects requiring verification, in which artificial intelligence is used to analyze surveillance images and detects surveillance targets that are judged to have a certain level of suspicion or higher as objects requiring verification, A monitoring necessity determination step that determines whether the aforementioned object requiring verification is the same as the object requiring monitoring stored in the security monitoring system, A monitoring information notification step in which, if it is determined that the object requiring confirmation is the same as the object requiring monitoring, warning information is output for the object requiring confirmation, and if it is determined that the object requiring confirmation is not the same as the object requiring monitoring, caution information is output for the object requiring confirmation. A security monitoring method is performed by causing a computer to execute a monitoring necessity information update step, which dynamically updates the content of the monitoring necessity information to match the judgment result when the user of the security monitoring system inputs the result of whether or not to continue monitoring the object requiring confirmation, which has been output as the aforementioned caution information. A program for security monitoring.

10. A security monitoring program that monitors a target area within a specific range, A stored image search step that searches for recorded surveillance images and extracts stored images of a specific surveillance target, A step is performed in which surveillance targets deemed to have a certain level of suspiciousness are detected as targets requiring further investigation by performing image analysis of surveillance images using artificial intelligence. In the aforementioned step of detecting objects requiring verification, the computer is instructed to execute a security monitoring method in which the criteria for determining the degree of suspicion are automatically changed according to the results of analyzing the accumulated images. A program for security monitoring.

Images