An intrusion detection system

Abstract

The present application provides an intrusion detection system, which includes a processing host and a plurality of intrusion detection devices arranged at preset intervals on the fence of a target area, each of which includes a visual detection component and a vibration detection component; the visual detection component identifies the object type of the intrusion object from the captured target image of the intrusion object that initially enters the early warning identification area outside the target area, and sends the captured target image and the identified object type to the vibration detection component in the same target intrusion detection device; the vibration detection component collects the vibration signal generated by the intrusion object contacting the fence, determines the behavior type of the intrusion object according to the vibration signal, and generates an intrusion detection result and sends it to the processing host; the processing host generates intrusion alarm information according to the object type and the behavior type. Through the present application, the combination of active detection and passive detection can be realized, thereby improving the detection accuracy and reducing the false alarm rate.

Description

Technical Field

[0001] The present application relates to the field of security monitoring, and in particular to an intrusion detection system. Background Art

[0002] Intrusion alarm system is a typical application of security, especially for high-level security sites, such as civil aviation airports, military sites, nuclear power plants, high-speed railways, etc. Traditional security technologies (such as infrared beams, laser beams, electronic fences, etc.) are difficult to meet the growing security needs of users.

[0003] At present, the security technical means adopted by the security system mainly include the following two types: First, passive detection technology. The advantage of passive detection is that it relies on physical protection devices, and generally has low power consumption and false alarm rate, but its disadvantage is that it lacks active detection and early warning capabilities, and cannot realize a multi-level protection system; second, active detection technology. Active detection has intelligent characteristics to a certain extent, but its disadvantage is a high false alarm rate, which is generally 1-2 orders of magnitude higher than passive detection. Summary of the invention

[0004] In view of this, the purpose of the present application is to provide an intrusion detection system, which is formed by combining a visual detection component with a vibration detection component to achieve a combination of active detection and passive detection, thereby improving the detection accuracy of intrusion objects and reducing the false alarm rate.

[0005] In a first aspect, an embodiment of the present application provides an intrusion detection system, which includes a processing host and a plurality of intrusion detection devices arranged on a fence of a target area at preset intervals, each intrusion detection device including a visual detection component and a vibration detection component, the preset interval being less than the minimum of an image capture limit of the visual detection component and a vibration detection limit of the vibration detection component; the visual detection component identifies an object type of the intrusion object from a captured target image of the intrusion object that initially enters an early warning identification area outside the target area, and sends the captured target image and the identified object type to a vibration detection component in the same target intrusion detection device; the vibration detection component collects a vibration signal generated by the intrusion object contacting the fence, determines a behavior type of the intrusion object based on the vibration signal, and generates an intrusion detection result and sends it to the processing host, the intrusion detection result including the target image, the object type, the behavior type, and a device identification of the target intrusion detection device; the processing host generates an intrusion alarm message based on the object type and the behavior type.

[0006] In an optional embodiment, the image capture ranges of the multiple visual detection components are superimposed to form a warning identification area; wherein the visual detection component is also used to perform the following processing: real-time capture of images of the warning identification area outside the target area; when an intrusion object is identified from the captured image, the image of the identified intrusion object is determined as the target image, and the object type of the intrusion object in the target image is identified; it is determined whether the object type meets the warning condition; if the object type meets the warning condition, a warning signal is generated, and the warning signal is sent to a vibration detection component in the same target intrusion detection device, and the warning signal carries a warning mark; the vibration detection component sends the warning signal directly to the processing host according to the warning mark; the processing host sends an sound and light alarm signal to a sound and light alarm device arranged near the target intrusion detection device according to the warning signal, so as to control the sound and light alarm device to perform a sound and light alarm.

[0007] In an optional embodiment, the visual detection component includes a camera module, a visual processor, a memory and a first communication module; wherein the camera module captures images of a warning identification area outside the target area in real time, and sends the captured image to the visual processor; the visual processor determines whether there is an intrusion object in the captured image, inputs the target image with the intrusion object into an object type recognition model to obtain the object type of the intrusion object, and sends the target image and the corresponding object type to the memory for storage; the visual processor also sends the target image and the corresponding object type to the first communication module; the first communication module sends the received target image and the corresponding object type to the vibration detection component in the same target intrusion detection device.

[0008] In an optional embodiment, the visual detection component also includes an illuminance sensor, an infrared lamp and a first power module; wherein the illuminance sensor detects the current light value of the external environment and sends the current light value to the visual processor; the visual processor compares the received current light value with a pre-set light threshold, and when the current light value is less than the light threshold, sends a control signal to the infrared lamp; the infrared lamp responds to the control signal for fill light; the first power module provides power for the camera module, the visual processor, the illuminance sensor and the infrared lamp.

[0009] In an optional embodiment, the vibration detection component includes a vibration sensor, a vibration processor and a second communication module; wherein the second communication module sends the target image and the object type received from the visual detection component to the vibration processor; the vibration sensor collects the vibration signal generated by the intruding object contacting the fence, and sends the vibration signal to the vibration processor; the vibration processor determines a behavior feature vector based on the received vibration signal, and inputs the behavior feature vector into a behavior type recognition model to obtain the behavior type of the intruding object, wherein the behavior feature vector includes the amplitude and frequency of the vibration signal; the vibration processor also generates an intrusion detection result based on the target image, the object type, the behavior type and the device identification of the target intrusion detection device, and sends the intrusion detection result to the second communication module; the second communication module sends the intrusion detection result to the processing host.

[0010] In an optional embodiment, the processing host is also used to perform the following processing: pre-recording the layout position of each intrusion detection device arranged on the fence; determining the target intrusion detection device that detects the intrusion object based on the device identification; determining the target layout position corresponding to the target intrusion detection device, determining the target layout position as the intrusion position, so as to add the intrusion position to the intrusion alarm information.

[0011] In an optional embodiment, the processing host is also used to perform the following processing: determining the intrusion behavior of the intrusion object based on the object type and the behavior type; matching the intrusion behavior with a type behavior comparison table, wherein the type behavior comparison table records multiple types of intrusion objects and preset behavior types implemented by each type of intrusion object; if the intrusion behavior exists in the type behavior comparison table, generating intrusion alarm information, wherein the intrusion alarm information includes the target image, the object type, the behavior type and the intrusion location.

[0012] In an optional embodiment, the type behavior comparison table includes multiple behavior entries and an alarm method corresponding to each behavior entry, and each behavior entry includes a type of intrusion object and a preset behavior type implemented by the type of intrusion object. The processing host is also used to perform the following processing: if the intrusion behavior exists in the type behavior comparison table, determine the target alarm method corresponding to the behavior entry matching the intrusion behavior; and add an alarm identifier indicating the target alarm method to the intrusion alarm information.

[0013] In a second aspect, an embodiment of the present application also provides an intrusion detection system, the system comprising: a plurality of intrusion detection devices arranged at preset intervals on a fence of a target area, each intrusion detection device comprising: a visual detection component for capturing a target image of an intruding object entering an early warning identification area outside the target area, and sending the target image to a vibration detection component; a vibration detection component for collecting a vibration signal generated by the intruding object contacting the fence, and outputting an intrusion detection result including the target image and the vibration signal; wherein the preset interval is less than the minimum value of the image capture limit of the visual detection component and the vibration detection limit of the vibration detection component.

[0014] In an optional embodiment, the intrusion detection system further includes: a plurality of brackets, each intrusion detection device being arranged on a fence of a target area through a corresponding bracket; wherein each bracket is a crenel-shaped bracket, a first extension portion of the crenel-shaped bracket being fixedly connected to the fence through a first clamping plate, a second extension portion of the crenel-shaped bracket being fixedly connected to the fence through a second clamping plate, and a middle portion of the crenel-shaped bracket being used for placing an intrusion detection device.

[0015] The present application provides an intrusion detection system, including a processing host and a plurality of intrusion detection devices arranged at preset intervals on the fence of a target area, each of which includes a visual detection component and a vibration detection component; the visual detection component identifies the object type of the intrusion object from the captured target image of the intrusion object that initially enters the early warning identification area outside the target area, and sends the captured target image and the identified object type to the vibration detection component in the same target intrusion detection device. In this way, active detection is achieved through the visual detection component, image capture and type recognition of the intrusion object entering the early warning identification area are performed, and the types of people, vehicles, small animals, etc. are accurately distinguished, so as to exclude intrusion alarms of some objects that will not cause intrusion threats in the alarm information, such as small animals and other targets, thereby preliminarily reducing the false alarm rate. The vibration detection component collects the vibration signal generated by the intrusion object contacting the fence, determines the behavior type of the intrusion object according to the vibration signal, and generates an intrusion detection result and sends it to the processing host, the intrusion detection result includes the target image, object type, behavior type and the device identification of the target intrusion detection device; the processing host generates intrusion alarm information according to the object type and behavior type. In this way, the intrusion behavior of the intruder can be identified. In addition, different levels of alarm information can be generated according to the degree of harm of the intrusion behavior, and the most appropriate processing method can be selected accordingly. Furthermore, by combining the visual detection component with the vibration detection component, the combination of active detection and passive detection is realized, thereby improving the detection accuracy and reducing the false alarm rate. In addition, multiple intrusion detection devices are arranged according to the actual situation and local conditions to improve the overall work efficiency and work accuracy of the security system.

[0016] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A schematic diagram of the structure of an intrusion detection system provided in an embodiment of the present application;

[0019] Figure 2 A schematic diagram of the layout of an intrusion detection system provided in an embodiment of the present application;

[0020] Figure 3 A schematic diagram of the arrangement of multiple visual detection components provided in an embodiment of the present application;

[0021] Figure 4 One of the structural schematic diagrams of the visual detection component provided in the embodiment of the present application;

[0022] Figure 5 A schematic diagram of the structure of a vibration detection assembly provided in an embodiment of the present application;

[0023] Figure 6 A schematic diagram of the structure of a processing host provided in an embodiment of the present application;

[0024] Figure 7 The second structural diagram of the visual detection component provided in the embodiment of the present application;

[0025] Figure 8 A topological diagram of the connection between the intrusion detection device and the processing host provided in an embodiment of the present application.

[0026] Figure numerals: 1-fence, 2-intrusion detection device, 3-visual detection component, 4-vibration detection component, 5-image capture range, 6-field of view overlap area, 7-field of view range, 8-monitoring center, 9-processing host, 21-camera module, 22-visual processor, 23-memory, 24-first communication module, 25-illuminance sensor, 26-infrared light, 27-first power module, 28-memory module, 29-image algorithm processor, 31-vibration sensor, 32-vibration processor, 33-second communication module, 51-bus communication module, 52-Ethernet communication module, 53-host processor, 54-alarm relay module. DETAILED DESCRIPTION

[0027] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application usually described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, each other embodiment obtained by those skilled in the art without making creative work belongs to the scope of protection of the present application.

[0028] First, the application scenarios to which the present application is applicable are introduced. The present application can be applied to the "security monitoring" scenario, but it should be understood that the embodiments of the present application are only introduced using the security monitoring scenario as an example, but the application scenarios of the intrusion detection system of the present application are not limited thereto.

[0029] Research has found that traditional security technologies (such as infrared beams, laser beams, electronic fences, etc.) are difficult to meet users' growing security needs. In addition to basic security needs such as zero missed reports and low false alarms, users also pay special attention to the following points:

[0030] (1) Accurately report the location of the intrusion point with an accuracy of meters; (2) Classify and identify intrusion targets, and accurately distinguish between types such as people, vehicles, and small animals; (3) Classify intrusion behaviors, and distinguish the target's action type, such as crossing the boundary, wandering, knocking, hitting, shaking, climbing, and climbing over.

[0031] For high-level security, the commonly used technologies are mainly vibration optical fiber, vibration cable, MEMS vibration, microwave radar, and intelligent video. Among them, the false alarm rate of vibration optical fiber is high, and it does not have the ability to accurately locate, classify targets and behaviors; vibration cable can accurately locate, but it also does not have the ability to classify targets and behaviors; MEMS vibration is the most widely used, and can accurately locate and classify behaviors, but it does not have the ability to classify targets. The above three are the three most widely used passive detection technologies. They must rely on security entities such as fences, and they mainly achieve intrusion detection by measuring mechanical vibration signals on physical protective devices. It can be seen that the advantage of passive detection technology is that it relies on physical protective devices, and generally has low power consumption and false alarm rate. Its disadvantage is that it lacks active detection and early warning capabilities and cannot achieve a multi-level protection system.

[0032] Compared with the above three passive detection technologies, microwave radar and video intelligent analysis belong to active detection technologies. The former uses the Doppler principle to detect moving targets, can measure the target speed, direction and reflected signal strength, can achieve accurate position measurement and non-image target recognition; the latter achieves target and behavior classification and recognition through intelligent analysis of video images, but does not have the ability to measure distance. Although active detection technology has intelligent characteristics to some extent, its disadvantage is that the false alarm rate is high, which is generally 1-2 orders of magnitude higher than passive detection.

[0033] From the above analysis, it can be seen that any single technical means cannot meet the needs of users at present. The more feasible solution is to use two or more technologies with different principles in combination, and achieve the best effect through fusion detection. The more representative product in this regard is the microwave radar and MEMS vibration fusion intrusion detection technology. Its idea is to achieve accurate measurement and behavior classification through MEMS vibration, and achieve extremely low false alarm rate through three-dimensional detection and vibration waveform recognition algorithm. At the same time, by integrating small, low-power microwave radar modules on the front-end device, active detection and early warning are realized, and the target approach, distance, wandering and other behaviors are identified. The module-level fusion perception realized by this technology on the front-end device is a typical edge computing mode, which can maximize the computing power of edge devices and has strong elasticity and scalability. Although this technology has been widely promoted, it still lacks the ability to classify intrusion targets and cannot distinguish between types such as people, vehicles and small animals. Therefore, it is impossible to output high-quality alarm signals through fusion and perception. The processing of its alarm still relies on manual review and has not reached the level of full intelligence.

[0034] In order to overcome at least one of the above-mentioned defects, an embodiment of the present application provides an intrusion detection system for improving the working efficiency of the security system and effectively reducing the false alarm rate.

[0035] See also Figure 1 , Figure 1 This is a schematic diagram of the structure of the intrusion detection system provided in the embodiment of the present application. Figure 1 As shown in , the intrusion detection system provided by the embodiment of the present application includes: a plurality of intrusion detection devices 2 installed on a fence 1 of a target area, each intrusion detection device 2 includes a visual detection component 3 and a vibration detection component 4.

[0036] Specifically, the target area refers to an object of security monitoring, and a plurality of intrusion detection devices 2 are arranged at preset intervals along the periphery of the target area.

[0037] In the first case, the target area is surrounded by rigid walls such as brick walls or metal guardrails.

[0038] At this time, a fence 1 may be set above the rigid wall, and a plurality of intrusion detection devices 2 may be arranged on the fence 1 .

[0039] In the second case, the target area is surrounded by a fence.

[0040] At this time, a fence 1 is arranged along the periphery of the target area, and a plurality of intrusion detection devices 2 are arranged on the fence 1 .

[0041] It should be understood that the height of the fence in the second case is higher than the height of the fence in the first case. Optionally, the height of the fence in the second case should be higher than the jumping height of a person or an animal to prevent a person or an animal from easily climbing over the fence. Accordingly, the total height of the rigid wall and the fence in the above-mentioned first case should also be higher than the jumping height of a person or an animal.

[0042] Exemplarily, the fence in the present application may include a fence made of a flexible material, for example, the fence may be a wire mesh. In this way, since the fence has a certain flexibility, it is not conducive for the intruder to support and climb over it. In addition, when the intruder attempts to climb over the fence to enter the target area, it is inevitable to touch the fence, thereby causing the fence to vibrate. At this time, due to the flexibility of the fence, the vibration detection component 4 can sense a relatively obvious vibration signal from the fence, so as to effectively prevent the intruder from entering the target area.

[0043] In an embodiment of the present application, the visual detection component 3 in each intrusion detection device 2 is used to capture the target image of the intrusion object entering the early warning identification zone outside the target area, and send the target image to the vibration detection component 4 in the same target intrusion detection device.

[0044] In a preferred embodiment, the visual detection component 3 can not only capture the target image of the intruding object, but also identify the object type of the intruding object.

[0045] Specifically, the visual detection component 3 identifies the object type of the intrusion object from the captured target image of the intrusion object that initially enters the early warning identification zone outside the target area, and sends the captured target image and the identified object type to the vibration detection component 4 in the same target intrusion detection device.

[0046] In a preferred embodiment of the present application, the above-mentioned intrusion detection system may further include a plurality of brackets, and each intrusion detection device 2 is arranged on the fence of the target area through a corresponding bracket. Exemplarily, each intrusion detection device 2 can be fixed on the fence through a corresponding bracket, because the fence is made of a flexible material, and the vibration generated by the fence being touched by the intrusion object will be transmitted to the intrusion detection device 2 through the bracket, and the intrusion detection device 2 can capture and process the external vibration signal through the received vibration signal.

[0047] In an optional example, the above-mentioned bracket may be a crenel-shaped bracket, and a crenel is provided in the middle part of the bracket. For example, the fence arranged along the periphery of the target area may be provided with a column at a certain interval to support the fence. When the bracket needs to be installed on the fence, the crenel of the bracket can leave space for the columns of the fence to pass through, and the bracket spans the columns of the fence to be fixed on the mesh surface of the fence on both sides of the columns of the fence.

[0048] For example, each bracket may be equipped with two clips, the first extension of the crenel-shaped bracket is fixedly connected to the fence through the first clip, and the second extension of the crenel-shaped bracket is fixedly connected to the fence through the second clip. For example, the mesh of the fence may be formed by metal wires, and the metal wires are clamped between the extension of the crenel-shaped bracket and the clips to fix the bracket on the mesh of the fence. The middle part of the crenel-shaped bracket can also be used to place the intrusion detection device 2.

[0049] For example, the inner side of the middle part of the crenel-shaped bracket protrudes outward to form a certain space for passing the upright pole of the fence, and the outer side of the middle part of the crenel-shaped bracket can be used to fix the intrusion detection device. As an example, the crenel-shaped bracket and the intrusion detection device can be fixedly connected, for example, the above-mentioned fixed connection can be connected by bolts, rivets, etc., but the present application is not limited thereto, and other fixing methods can also be used.

[0050] Figure 2 A schematic diagram of the layout of the intrusion detection system provided in the embodiment of the present application is shown in FIG. Figure 2 As shown, the shooting direction of the visual detection component 3 of each intrusion detection device 2 can be toward the periphery of the target area to capture the image of the intruding object entering the target area from the outside. The outside of the fence is the image capture range 5 of the visual detection component 3, and the image capture range 5 constitutes the early warning identification area mentioned above.

[0051] Exemplarily, a plurality of intrusion detection devices 2 are arranged at preset intervals on the fence of the target area, where the preset intervals can be determined based on the visual detection distance (i.e., image capture limit) of the visual detection component of the intrusion detection device and the vibration perception distance (vibration detection limit) of the vibration detection component. Preferably, the minimum value of the image capture limit of the visual detection component and the vibration detection limit of the vibration detection component can be selected as the preset interval for the arrangement of the intrusion detection device to ensure that the plurality of intrusion detection devices 2 can completely cover the target area and avoid areas where detection is missed.

[0052] Figure 3 A schematic diagram showing the arrangement of multiple visual detection components according to an embodiment of the present application is shown.

[0053] exist Figure 3In the example shown, each visual detection component 3 can use a wide-angle lens. To avoid visual blind spots, each visual detection component 3 can be rotated by a certain angle during arrangement.

[0054] Taking the case where the field of view 7 of each visual detection component 3 is a 120-degree sector (i.e., the viewing angle of the wide-angle lens reaches 120 degrees) as an example, multiple visual detection components 3 can be used as follows: Figure 3 The arrangement shown enables the fields of view of adjacent visual detection components 3 to overlap with each other, forming a field of view overlap area 6, thereby reducing or even eliminating the blind spot.

[0055] At this time, the above-mentioned early warning identification area can refer to the area formed by the superposition of the image capture ranges of multiple visual detection components 3. Its function is to capture the target image of the intrusion object and identify the object type of the intrusion object when the intrusion object enters the early warning identification area.

[0056] See also Figure 4 , Figure 4 This is one of the structural diagrams of the visual detection component provided in the embodiment of the present application. Figure 4 As shown, the visual detection component 3 may include a camera module 21 , a visual processor 22 , a memory 23 and a first communication module 24 .

[0057] Specifically, the camera module 21 is used to capture images of the early warning identification area outside the target area in real time. Because the fields of view of adjacent camera modules 21 overlap with each other, the image of any intrusion object entering the early warning identification area will be captured by the camera module 21 in the area.

[0058] Subsequently, the camera module 21 sends the captured image of the warning identification area to the visual processor 22 .

[0059] For example, the camera module 21 can be a miniature camera, and a wide-angle lens can be selected, with a viewing angle of not less than 120 degrees, to minimize the visual blind spot. At the same time, the camera modules 21 of each intrusion detection device can be deflected in the same direction at a certain angle, so that the fields of view of the camera modules 21 of adjacent intrusion detection devices overlap with each other, thereby eliminating the visual blind spot (such as Figure 3 As shown). As an example, the camera module can use a CMOS visual sensor with a pixel of not less than 300,000, a resolution supporting 480P, 720P and 1080P, and supporting H.264 video encoding format. For example, the camera module 21 can be connected to the visual processor 22 through the MIPI interface to send the captured image of the warning identification area to the visual processor 22.

[0060] The visual processor 22 is used to determine whether there is an intrusion object in the captured image, and input the target image with the intrusion object into the object type recognition model to obtain the object type of the intrusion object.

[0061] For example, the visual processor 22 can input the image captured by the camera module 21 into the target recognition model to determine whether there is an intrusion object in the captured image. At this time, the input of the target recognition model is the image, and the output of the target recognition model is the object identification, which is used to indicate whether there is an intrusion object in the image.

[0062] The above-mentioned object type recognition model may be a pre-trained deep learning model for identifying the type of intrusion objects, and the object type recognition model may be pre-stored in the memory 23. At this time, the input of the object type recognition model is a target image with intrusion objects, and the output of the object type recognition model is a type identifier, which is used to indicate the object type of the intrusion object in the target image.

[0063] For example, images of various types of intrusion objects, such as images of people, vehicles, small animals, etc., can be collected, and object type labels (including people, vehicles, small animals, etc.) can be set for each collected image. The type deep learning model can be trained based on the collected images and the corresponding object type labels to obtain an object type recognition model that is sufficient to accurately identify the object types of various types of intrusion objects.

[0064] The visual processor 22 sends the received target image and the corresponding object type to the memory 23 for storage. In addition, the camera module 21 can also capture the video of the warning identification area and send the video to the visual processor 22. At this time, the visual processor 22 can also send the received video to the memory 23 for storage. In other words, the memory 23 can be used to store the video and images collected by the camera module 21. Exemplarily, the memory 23 can be a removable memory card to facilitate its replacement and maintenance.

[0065] The visual processor 22 also sends the received target image and the corresponding object type to the first communication module 24, and the first communication module 24 sends the received target image and the corresponding object type to the vibration detection component 4 in the same target intrusion detection device.

[0066] Here, the first communication module 24 is used for communication between the visual detection component 3 and the vibration detection component 4. Exemplarily, the visual detection component 3 and the vibration detection component 4 can be connected via a serial port.

[0067] In this way, by capturing images and identifying the types of intrusion objects entering the early warning identification area, the object types of the intrusion objects are distinguished, thereby achieving active detection by the intrusion detection system.

[0068] In a preferred embodiment of the present application, the visual detection component 3 and the vibration detection component 4 can be used to implement hierarchical alarm control. For example, when the presence of an intruder in the early warning identification area is initially detected, a primary warning can be issued based on the recognition result of the visual detection component 3 to warn and drive away the intruder. If it is determined that the intruder is further approaching the target area, an intrusion alarm can be issued based on the recognition result of the vibration detection component 4.

[0069] For example, the visual detection component 3 can also be used to perform the following processing: capture images of the warning identification area outside the target area in real time; when an intrusion object is identified from the captured image, determine the image of the identified intrusion object as the target image, and identify the object type of the intrusion object in the target image; determine whether the identified object type meets the warning conditions; if the object type meets the warning conditions, generate a warning signal, and send the warning signal to the vibration detection component in the same target intrusion detection device.

[0070] Here, the warning condition includes at least one pre-set warning object type of the intrusion object. When the identified object type exists in the warning object type, it is determined that the identified object type meets the warning condition. When the identified object type does not exist in the warning object type, it is determined that the identified object type does not meet the warning condition.

[0071] Here, the warning signal carries a warning mark, which is used to indicate that the current warning signal is a primary alarm. After receiving the warning signal, the subsequent vibration detection component 4 and the processing host can perform the processing process for the primary alarm based on the warning mark it carries.

[0072] When the intruder moves closer to the fence and attempts to hit, shake, shear, or climb over the fence, the fence shakes due to the intrusion behavior of the intruder, generating a vibration signal. At this time, in the embodiment of the present application, the vibration detection component 4 in each intrusion detection device 2 is used to collect the vibration signal generated by the intruder contacting the fence, and determine the behavior type of the intruder according to the vibration signal. For example, the detection part of the vibration detection component 4 can be arranged on the fence to detect the vibration signal from the fence.

[0073] Here, multiple intrusion detection devices 2 are arranged on the fence at preset intervals. When an intrusion object touches a certain section of the fence, the intrusion detection device 2 arranged near the certain section will monitor the vibration signal. When there are two or more intrusion detection devices 2 that monitor vibration data, the intrusion position of the intrusion object can be determined according to the strength of the vibration signal monitored by each intrusion detection device 2. For example, the arrangement position of the intrusion detection device 2 that monitors the strongest vibration signal among the multiple intrusion detection devices 2 can be determined as the intrusion position of the intrusion object.

[0074] Preferably, each intrusion detection device 2 can be set with a unique device identifier, and after the intrusion detection device is installed, the longitude and latitude coordinates of the installation location of the intrusion detection device are recorded, and the device identifier is associated with the recorded longitude and latitude coordinates, so that security personnel can accurately locate the intrusion position of the intrusion object.

[0075] See also Figure 5 , Figure 5 This is a schematic diagram of the structure of the vibration detection component provided in the embodiment of the present application. Figure 5 As shown, the vibration detection component 4 may include a vibration sensor 31 , a vibration processor 32 and a second communication module 33 .

[0076] Specifically, the second communication module 33 sends the target image and the corresponding object type received from the visual detection component 3 to the vibration processor 32 .

[0077] The vibration sensor 31 collects a vibration signal generated by the intrusion object contacting the fence, and sends the vibration signal to the vibration processor 32 .

[0078] Exemplarily, the vibration sensor can be an acceleration sensor. Here, the intrusion detection device is fixed on the fence through a bracket. When the fence vibrates due to external force, the vibration can be transmitted to the vibration sensor 31 through the bracket, so that the vibration of the fence is captured by the vibration sensor 31, and then a vibration signal corresponding to the vibration is generated. The vibration signal here may include amplitude and frequency.

[0079] The vibration processor 32 processes the collected vibration signal.

[0080] At this time, the vibration processor 32 determines a behavior feature vector based on the received vibration signal. For example, the behavior feature vector includes the amplitude and frequency of the vibration signal. The vibration processor inputs the behavior feature vector into a behavior category recognition model pre-trained with a vibration waveform integrated in the vibration processor to obtain the behavior type of the intruder.

[0081] For example, the vibration signals collected by the vibration sensor when the fence is touched by different types of intrusion behaviors can be collected, each collected vibration signal can be matched one by one with each type of intrusion behavior, and the corresponding vibration signal and behavior type of the intrusion behavior can be used to train the behavior category recognition model.

[0082] Here, accurate measurement and behavior classification of vibration are achieved based on the amplitude and frequency of the vibration signal. That is, the behavior type of the intruder can be determined based on the amplitude and frequency of the detected vibration signal. As an example, the behavior type may include but is not limited to at least one of the following items: impact, shaking, shearing, destruction, climbing, climbing over a fence, etc.

[0083] It should be understood that the vibration processor 32 can process the vibration signal using various existing vibration recognition algorithms, that is, various existing vibration recognition algorithms for determining the location where the vibration occurs and the behavior category according to the vibration signal can be used to implement the processing function of the above-mentioned vibration processor 32. Preferably, the above-mentioned behavior category recognition model of the present application can be used to process the vibration signal.

[0084] The vibration processor 32 also generates an intrusion detection result based on the target image obtained from the visual detection component 3, the object type together with the behavior type generated by the vibration processor 32 itself and the device identification of the target intrusion detection device, and sends the intrusion detection result to the processing host.

[0085] For example, the above-mentioned intrusion detection results can be sent to the processing host via the second communication module 33. The device identifier here is the unique identification information possessed by each intrusion detection device, which can be distinguished by identification information such as the device serial number and the installation sequence number. In this case, the intrusion detection results received by the processing host include the target image, object type, behavior type and device identification of the target intrusion detection device. In an embodiment of the present application, the first communication module 24 of the visual detection component 3 is connected to the second communication module 33 of the vibration detection component 4 to send the image collected by the visual detection component 3 and the corresponding object type to the second communication module 33. At this time, the second communication module 33 can output the intrusion detection results received from the vibration processor 32 to the processing host. After receiving the intrusion detection results, the processing host processes the information therein and generates intrusion alarm information.

[0086] In the case where the visual detection component 3 outputs a warning signal, at this time, the vibration detection component directly sends the warning signal to the processing host according to the warning mark carried in the warning signal.

[0087] Here, the warning mark is used to indicate that the current warning signal is a primary alarm. The vibration detection component 4 can execute the processing process for the primary alarm after detecting the warning mark. That is, at this time, the vibration detection component 4 directly forwards the warning signal without sending the vibration signal detected by itself together with the warning signal to the processing host, thereby shortening the system's warning response time.

[0088] In this case, the processing host sends an audible and visual alarm signal to the audible and visual alarm device arranged near the target intrusion detection device according to the early warning signal, so as to control the audible and visual alarm device to make an audible and visual alarm, so as to warn and drive away the intruding object.

[0089] In the embodiment of the present application, as a preferred embodiment, the processing host is also used to perform the following processing:

[0090] The arrangement position of each intrusion detection device arranged on the fence is recorded in advance. Here, when the intrusion detection device is installed, the longitude and latitude coordinates of the installation position of each intrusion detection device can be recorded, and the device identification of the intrusion detection device can be associated with the recorded longitude and latitude coordinates, and the above information can be recorded in the processing host.

[0091] According to the device identification, the target intrusion detection device that detected the intrusion object is determined. Here, the processing host reads the device identification, determines which intrusion detection device issued the intrusion detection result, and finds the corresponding longitude and latitude coordinates based on the pre-recorded device identification and the associated longitude and latitude coordinates, thereby determining the target layout position corresponding to the target intrusion detection device, determining the target layout position as the intrusion position, and adding the intrusion position to the intrusion alarm information.

[0092] See also Figure 6 , Figure 6 This is a schematic diagram of the structure of the processing host provided in the embodiment of the present application. Figure 6 As shown, the processing host 9 may include a bus communication module 51 , an Ethernet communication module 52 , a host processor 53 and an alarm relay module 54 .

[0093] Specifically, the bus communication module 51 is connected to the second communication modules 33 of each intrusion detection device 2 , respectively, and the bus communication module 51 sends the intrusion detection results received from each second communication module 33 to the host processor 53 .

[0094] The host processor 53 separates the received intrusion detection result into object type and behavior type, and generates intrusion alarm information according to the separated object type and behavior type.

[0095] For example, the host processor 53 may determine the intrusion behavior of the intrusion object according to the object type and the behavior type.

[0096] Match the intrusion behavior with the type behavior comparison table. Here, the type behavior comparison table records multiple types of intrusion objects and the preset behavior types implemented by each type of intrusion object, and is saved in the processing host; if there is an intrusion behavior in the type behavior comparison table, determine the target alarm method corresponding to the behavior entry matching the intrusion behavior; add the alarm identifier indicating the target alarm method to the intrusion alarm information, and generate the intrusion alarm information. At this time, the intrusion alarm information includes the target image, object type, behavior type, and intrusion location.

[0097] Exemplarily, the type behavior comparison table may include multiple behavior entries and an alarm method corresponding to each behavior entry, and each behavior entry includes a type of intrusion object and a preset behavior type implemented by the type of intrusion object.

[0098] The type behavior comparison table here can be pre-set according to the actual situation. For example, the object type in a behavior entry of the type behavior comparison table is recorded as a person, and the behavior type is recorded as climbing. When the host processor 53 determines that the object type in the intrusion behavior is a person, the behavior type is climbing, and a corresponding matching result is found in the type behavior comparison table, an intrusion alarm message is generated. The type behavior comparison table can be set according to the security level. For example, security facilities with a higher security level can record people, vehicles and animals in the behavior entries of the type behavior comparison table, and record collision, shaking, shear damage, climbing and crossing in the behavior type corresponding to the behavior entry in the type behavior comparison table. As long as any combination of the above types and behaviors is found in the intrusion behavior, an intrusion alarm message is generated.

[0099] Different alarm methods can be set for different behavior items. For example, when the type of intrusion object is an animal, the alarm method is to drive the sound and light alarm device near the invasion site to drive away the intrusion object with sound and light. When the type of intrusion object is a person, the alarm method is to mobilize security personnel to the invasion site to carry out security work.

[0100] For example, the processing host can also be used to perform the following processing: if there is an intrusion behavior in the type behavior comparison table, determine the target alarm method corresponding to the behavior entry matching the intrusion behavior; add an alarm identifier indicating the target alarm method to the intrusion alarm information.

[0101] In this way, appropriate security measures can be taken according to the different threat levels of intrusion behaviors, reducing security costs and improving security work efficiency.

[0102] The host processor 53 then sends the intrusion alarm information to the Ethernet communication module 52, and the Ethernet communication module 52 sends the received intrusion alarm information to the monitoring center.

[0103] Here, the bus communication module 51 is respectively connected to the second communication module 33 of each intrusion detection device 2 through the bus, and the Ethernet communication module 52 is connected to the monitoring center through an Ethernet cable.

[0104] In the case where the vibration detection component 4 forwards the warning signal, the processing host can send an audible and visual alarm signal to the audible and visual alarm device arranged near the target intrusion detection device through the alarm relay module 54 according to the warning signal to control the audible and visual alarm device to perform an audible and visual alarm.

[0105] As an example, the alarm device may include a warning light or a siren, or a combination of the two. For example, the alarm device may be installed on the posts of the fence, and may be arranged at certain intervals according to actual needs.

[0106] See also Figure 7 , Figure 7 This is a second structural diagram of the visual detection component provided in the embodiment of the present application. Figure 7 As shown in the figure, the visual detection component 3 includes not only a camera module 21, a visual processor 22, a memory 23, and a first communication module 24, but also an illumination sensor 25, an infrared lamp 26, a first power module 27, a memory module 28, and an image algorithm processor 29.

[0107] Specifically, the illumination sensor 25 sends the detected current illumination value of the external environment to the visual processor 22 .

[0108] For example, the illumination sensor 25 may be any sensor that can be used to detect the brightness of light in the surrounding environment, and the present application does not impose any limitation on this.

[0109] The visual processor 22 sends a control signal indicating that the current illumination value is less than the illumination threshold value, which is generated by comparing the current illumination value with the illumination threshold value, to the infrared lamp 26. The infrared lamp 26 emits light in response to the received control signal, thereby providing fill light.

[0110] Here, the infrared lamp 26 is a fill light device for collecting images in a dark environment, which can emit invisible infrared light to the outside, and has the characteristics of concealment and energy saving. In the embodiment of the present application, the light threshold can be set according to actual needs, and the light threshold can be the minimum light brightness required for the camera module 21 to clearly capture videos and images.

[0111] The visual processor 22 compares the current illumination value with the illumination threshold. When it is determined that the current illumination value is less than the illumination threshold, a control signal is generated and sent to the infrared lamp 26 to control the infrared lamp 26 to emit light to supplement the environment. When it is determined that the current illumination value is not less than (i.e., greater than or equal to) the illumination threshold, no control signal is generated, and the infrared lamp 26 does not emit light.

[0112] It should be understood that, in addition to the above-mentioned method of setting the infrared lamp 26, the image recognition algorithm can also be trained based on black and white data to reduce the impact of dark light on image recognition accuracy.

[0113] The first power module 27 can provide power for each device in the visual detection component 3 , for example, it can provide power for the camera module 21 , the visual processor 22 , the illumination sensor 25 , and the infrared lamp 26 .

[0114] The memory module 28 can be connected to the visual processor 22 to temporarily store the calculation data in the visual processor 22 and the data exchanged with the external memory such as the memory 23.

[0115] The visual processor 22 is the core control element of the visual detection component 3 , and is responsible for sending the video and images collected by the camera module 21 to the image algorithm processor 29 and the memory 23 .

[0116] In one case, the image algorithm processor 29 may process the image captured by the camera module 21 .

[0117] At this time, the image recognition algorithm in the image algorithm processor 29 can be used to process the collected image to identify whether there is an intrusion object in the collected image, and further determine the category of the intrusion object. It should be understood that the image recognition algorithm running in the image algorithm processor 29 can be various existing image recognition algorithms that can be used to identify people, animals, and objects from images, and can also be various existing image recognition algorithms that can classify the identified objects, that is, the existing image recognition algorithm can be used to implement the processing function of the image algorithm processor 29, and the present application does not improve the image recognition algorithm.

[0118] In this case, the image algorithm processor 29 may send the processing result (ie, the category of the intrusion object) to the visual processor 22 , and the visual processor 22 sends the processing result to the vibration detection component 4 via the first communication module 24 .

[0119] In another case, the image algorithm processor 29 is not required to process the image captured by the camera module 21 .

[0120] At this time, the visual processor 22 can directly send the images and videos captured by the camera module 21 to the vibration detection component 4 via the first communication module 24 without processing the captured images and videos.

[0121] See also Figure 8 , Figure 8 A topological diagram of the connection between the intrusion detection device and the processing host provided in an embodiment of the present application.

[0122] exist Figure 8 In the example shown, the processing host 9 is connected to two buses, each bus being used to connect to at least one intrusion detection device 2. Exemplarily, the second communication module 33 of each intrusion detection device 2 is connected to the processing host 9 via a bus. Exemplarily, the above-mentioned bus may include an RS485 bus or a CAN bus.

[0123] In a preferred example, the intrusion detection system may further include a monitoring center 8, to which a processing host 9 is connected. The processing host 9 sends the image and vibration data received from each intrusion detection device 2 to the monitoring center 8 for display.

[0124] Optionally, after receiving the image and vibration data, the monitoring center 8 can process the image and vibration data to display the processed results. It should be understood that the image processing process of the monitoring center 8 can refer to the image processing process of the image algorithm processor 29, and the vibration data processing process of the monitoring center 8 can refer to the vibration data processing process of the vibration processor 32, and this application will not repeat this part.

[0125] In addition, in response to the above-mentioned visual processor 22 processing the collected image and / or the vibration processor 32 processing the collected vibration data, the second communication module 33 can send the received image processing results and vibration processing results to the processing host 9, and the processing host 9 sends the received image processing results and vibration processing results to the monitoring center 8 for display.

[0126] Exemplarily, the processing host 9 may be further connected to an Ethernet cable for communicating with the monitoring center 8 .

[0127] As an example, the processing host 9 can be mounted on the fence 1 or installed in a weak current box in the target area. The number of processing hosts 9 can be determined according to the size of the target area.

[0128] In summary, the workflow of the intrusion detection system provided by the embodiment of the present application is briefly described: First, the visual field coverage of the visual detection component 3 forms a video image detection area of ​​a certain width on the periphery of the fence 1. When an intruder enters the video image detection area, the visual detection component 3 collects images of the intruder and can also classify the intruder. In one example, an early warning can be issued for intruders of target types entering the video image detection area, and the target type can be set according to actual conditions. For example, people, vehicles, and aggressive animals can be determined as target types. When the above-mentioned types of intruders enter the video image detection area, the intrusion detection system sends an early warning message to the monitoring center 8 to alert the monitoring personnel.

[0129] When the intruder further approaches the fence 1 and contacts it, the vibration detection component 4 can collect vibration signals, and the intrusion position and / or intrusion behavior of the intruder can be classified based on the collected vibration signals. After confirming the position and behavior of the intruder, the intrusion detection device 2 can issue an alarm to the processing host 9 and send the intruder's intrusion alarm information to the processing host.

[0130] After receiving the intrusion alarm information, the processing host 9 matches the intrusion behavior with the type behavior comparison table, and sends out corresponding alarm information according to the corresponding information of the preset object type and behavior type. Security personnel can take appropriate security measures based on different alarm information.

[0131] The processing host 9 can also determine whether there is an intruder outside the fence 1 based on the video or image. If an intruder is found, the alarm is proven to be true and an alarm message is sent to the monitoring center 8. The monitoring personnel can determine the intrusion location of the intruder through the identification and longitude and latitude information of the intrusion detection device 2 that issued the alarm. At the same time, the processing host 9 can also drive the alarm device to issue an alarm to warn and drive the intruder away. If it is determined that there is no intruder outside the fence 1, it means that the alarm message may be a false alarm. At this time, the alarm message can be further investigated or directly filtered out.

[0132] The intrusion detection system provided in the embodiment of the present application is easy to install and use, has low cost, and structurally combines the visual detection component 3 with the vibration detection component 4 to achieve complementarity between active detection and passive detection, which can accurately locate the intrusion target and reduce the false alarm rate.

[0133] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. An intrusion detection system, characterized in that: The intrusion detection system comprises a processing host and a plurality of intrusion detection devices arranged at preset intervals on a fence of a target area, each intrusion detection device comprises a visual detection component and a vibration detection component, the preset interval is less than the minimum value of an image capture limit of the visual detection component and a vibration detection limit of the vibration detection component, the visual detection component comprises a camera module, a visual processor, a memory and a first communication module, and the vibration detection component comprises a vibration sensor, a vibration processor and a second communication module; Among them, the camera module of the visual detection component captures the image of the early warning identification area outside the target area in real time, and sends the captured image to the visual processor; the visual processor determines whether there is an intrusion object in the captured image, inputs the target image with the intrusion object into the object type recognition model to obtain the object type of the intrusion object, and sends the target image and the corresponding object type to the memory for storage; the visual processor also sends the target image and the corresponding object type to the first communication module; the first communication module sends the received target image and the corresponding object type to the vibration detection component in the same target intrusion detection device; The second communication module of the vibration detection component sends the target image and the object type received from the first communication module of the visual detection component to the vibration processor; the vibration sensor collects the vibration signal generated by the intrusion object contacting the fence, and sends the vibration signal to the vibration processor; the vibration processor determines the behavior feature vector according to the received vibration signal, and inputs the behavior feature vector into the behavior type recognition model to obtain the behavior type of the intrusion object, and the behavior feature vector includes the amplitude and frequency of the vibration signal; the vibration processor also generates an intrusion detection result according to the target image, the object type, the behavior type and the device identification of the target intrusion detection device, and sends the intrusion detection result to the second communication module; the second communication module sends the intrusion detection result to the processing host; The processing host generates intrusion alarm information according to the object type and the behavior type.

2. The intrusion detection system according to claim 1, characterized in that: The image capture ranges of the visual detection components of the multiple intrusion detection devices are superimposed to form an early warning identification area; Wherein, the visual detection component is also used to perform the following processing: Capture images of the early warning identification zone outside the target area in real time; When an intrusion object is identified from the captured image, determining the image in which the intrusion object is identified as the target image, and identifying the object type of the intrusion object in the target image; Determining whether the object type meets the warning condition; If the object type meets the warning condition, a warning signal is generated and sent to a vibration detection component in the same target intrusion detection device, wherein the warning signal carries a warning mark; The vibration detection component directly sends the warning signal to the processing host according to the warning mark; The processing host sends an audible and visual alarm signal to an audible and visual alarm device arranged near the target intrusion detection device according to the early warning signal, so as to control the audible and visual alarm device to perform an audible and visual alarm.

3. The intrusion detection system according to claim 1, characterized in that: The visual detection component also includes an illumination sensor, an infrared lamp and a first power supply module; Among them, the illumination sensor detects the current illumination value of the external environment and sends the current illumination value to the visual processor; The visual processor compares the received current light value with the preset light threshold, and sends a control signal to the infrared light when the current light value is less than the light threshold; The infrared lamp performs fill light in response to the control signal; The first power module provides power for the camera module, the visual processor, the illumination sensor, and the infrared lamp.

4. The intrusion detection system according to claim 1, characterized in that: The processing host is also used to perform the following processing: Record in advance the location of each intrusion detection device arranged on the fence; Determining, based on the device identifier, a target intrusion detection device that detects the intrusion object; The target layout position corresponding to the target intrusion detection device is determined, and the target layout position is determined as the intrusion position, so as to add the intrusion position to the intrusion alarm information.

5. The intrusion detection system according to claim 4, characterized in that: The processing host is also used to perform the following processing: Determining the intrusion behavior of the intrusion object according to the object type and the behavior type; Matching the intrusion behavior with a type behavior comparison table, wherein the type behavior comparison table records multiple types of intrusion objects and preset behavior types implemented by each type of intrusion object; If the intrusion behavior exists in the type behavior comparison table, intrusion alarm information is generated, wherein the intrusion alarm information includes the target image, the object type, the behavior type and the intrusion location.

6. The intrusion detection system according to claim 5, characterized in that: The type behavior comparison table includes a plurality of behavior items and an alarm mode corresponding to each behavior item, and each behavior item includes a type of intrusion object and a preset behavior type implemented by the type of intrusion object; The processing host is also used to perform the following processing: If the intrusion behavior exists in the type behavior comparison table, determining a target alarm mode corresponding to a behavior entry matching the intrusion behavior; An alarm identifier indicating the target alarm mode is added to the intrusion alarm information.

7. An intrusion detection system, characterized in that: The intrusion detection system comprises a plurality of intrusion detection devices arranged at preset intervals on the fence of the target area, each intrusion detection device comprising: A visual detection component, the visual detection component includes a camera module, a visual processor, a memory and a first communication module, the camera module of the visual detection component captures an image of a warning identification area outside a target area in real time, and sends the captured image to the visual processor; the visual processor determines whether there is an intrusion object in the captured image, inputs the target image with the intrusion object into an object type recognition model to obtain the object type of the intrusion object, and sends the target image and the corresponding object type to the memory for storage; the visual processor also sends the target image and the corresponding object type to the first communication module; the first communication module sends the received target image and the corresponding object type to the vibration detection component in the same target intrusion detection device; A vibration detection component, wherein the second communication module of the vibration detection component sends the target image and the object type received from the first communication module of the visual detection component to a vibration processor; the vibration sensor collects a vibration signal generated by the intrusion object contacting the fence, and sends the vibration signal to the vibration processor; the vibration processor determines a behavior feature vector according to the received vibration signal, and inputs the behavior feature vector into a behavior type recognition model to obtain the behavior type of the intrusion object, wherein the behavior feature vector includes the amplitude and frequency of the vibration signal; the vibration processor also generates an intrusion detection result according to the target image, the object type, the behavior type and the device identification of the target intrusion detection device, and sends the intrusion detection result to the second communication module; the second communication module sends the intrusion detection result to a processing host; The preset interval is smaller than a minimum value of an image capture limit of the visual detection component and a vibration detection limit of the vibration detection component.

8. The intrusion detection system according to claim 7, characterized in that: The intrusion detection system further comprises: a plurality of brackets, each intrusion detection device being arranged on a fence of a target area through a corresponding bracket; Among them, each bracket is a crenel-shaped bracket, the first extension part of the crenel-shaped bracket is fixedly connected to the fence through a first clamping plate, the second extension part of the crenel-shaped bracket is fixedly connected to the fence through a second clamping plate, and the middle part of the crenel-shaped bracket is used to place the intrusion detection device.

Images