CN113990009A - Building falling object intelligent detection and interception system based on CPU and control method thereof - Google Patents

Building falling object intelligent detection and interception system based on CPU and control method thereof Download PDF

Info

Publication number
CN113990009A
CN113990009A CN202111275440.4A CN202111275440A CN113990009A CN 113990009 A CN113990009 A CN 113990009A CN 202111275440 A CN202111275440 A CN 202111275440A CN 113990009 A CN113990009 A CN 113990009A
Authority
CN
China
Prior art keywords
building
cpu
interception
falling
infrared
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111275440.4A
Other languages
Chinese (zh)
Inventor
杨森林
王亦澜
李�浩
李荫梁
李喜龙
李林春
谢帅帅
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xian University
Original Assignee
Xian University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xian University filed Critical Xian University
Priority to CN202111275440.4A priority Critical patent/CN113990009A/en
Publication of CN113990009A publication Critical patent/CN113990009A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/16—Actuation by interference with mechanical vibrations in air or other fluid
    • G08B13/1609—Actuation by interference with mechanical vibrations in air or other fluid using active vibration detection systems
    • G08B13/1618—Actuation by interference with mechanical vibrations in air or other fluid using active vibration detection systems using ultrasonic detection means
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F41—WEAPONS
    • F41H—ARMOUR; ARMOURED TURRETS; ARMOURED OR ARMED VEHICLES; MEANS OF ATTACK OR DEFENCE, e.g. CAMOUFLAGE, IN GENERAL
    • F41H13/00—Means of attack or defence not otherwise provided for
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/181—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using active radiation detection systems
    • G08B13/183—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using active radiation detection systems by interruption of a radiation beam or barrier
    • G08B13/184—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using active radiation detection systems by interruption of a radiation beam or barrier using radiation reflectors
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/0202—Child monitoring systems using a transmitter-receiver system carried by the parent and the child
    • G08B21/0205—Specific application combined with child monitoring using a transmitter-receiver system
    • G08B21/0208—Combination with audio or video communication, e.g. combination with "baby phone" function
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/0202—Child monitoring systems using a transmitter-receiver system carried by the parent and the child
    • G08B21/028—Communication between parent and child units via remote transmission means, e.g. satellite network
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B7/00—Signalling systems according to more than one of groups G08B3/00 - G08B6/00; Personal calling systems according to more than one of groups G08B3/00 - G08B6/00
    • G08B7/06—Signalling systems according to more than one of groups G08B3/00 - G08B6/00; Personal calling systems according to more than one of groups G08B3/00 - G08B6/00 using electric transmission, e.g. involving audible and visible signalling through the use of sound and light sources

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Child & Adolescent Psychology (AREA)
  • General Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Emergency Management (AREA)
  • Business, Economics & Management (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • General Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Alarm Systems (AREA)

Abstract

The invention discloses a building falling object intelligent detection and interception system and method based on a Central Processing Unit (CPU), wherein the system comprises the CPU, the CPU is respectively connected with a monitoring camera, an infrared sensor array, an ultrasonic sensor array, an interception net, a storage, an acousto-optic alarm circuit, a clock circuit, a display unit, a wireless network module and a power supply module, and the CPU is in wireless communication with the sensor array, the interception net and an intelligent terminal through the wireless network module. The falling object is efficiently detected through the infrared sensor and the ultrasonic sensor, then rapid falling object interception, field image recording, sound-light alarm and terminal falling object intelligent detection are implemented, finally the object type is accurately detected on the mobile terminal, and the falling object source tracing is accurately realized. The intelligent detection system can provide real-time and effective intelligent detection, positioning and interception of falling objects for a high-rise system, guarantee the existence of the falling objects and accurate detection and early warning of the positions, and can effectively prevent thieves from invading and throwing objects from high altitude.

Description

Building falling object intelligent detection and interception system based on CPU and control method thereof
Technical Field
The invention relates to the technical field of wireless sensor networks, image and video information processing, artificial intelligence deep learning and intelligent terminal APP development, in particular to an intelligent implementation method for object detection and building safety monitoring.
Background
With the rapid development of urbanization in China and the increasing improvement of the living standard of people, high-rise building groups become the marks of urban buildings. However, as the number of high-rise buildings in large and medium-sized cities in China is increased, the problem of falling objects also comes along. The accident caused by the impact of falling objects is actually happened, the situation that people are injured by falling objects is not rare, however, the root of the problems is analyzed, not only due to the quality problem of citizens, but also an important factor for testing urban risk prevention. Based on the relevant experimental data, it is shown that: a30 g egg dropped on the 4 th floor can injure passerby, if the egg dropped on the 8 th floor, the scalp of passerby can be injured, if the egg dropped on the 18 th floor, the skull of passerby can be smashed, and if the egg dropped on the 25 th floor, the death of passerby can be directly caused. Based on the experimental data, the falling objects pose great threat to the public safety of the society, so that the problem of solving the falling objects is urgent.
Based on the problems, in order to avoid the potential safety hazard of falling objects in life, accidents caused by falling objects are reduced from the source, the injury caused by falling objects is reduced, and the falling objects are no longer the pain of a cage city, so that the development of an intelligent building falling object detection system is of great significance.
At present, the research on the falling object danger and the safety problem is mainly carried out from the aspects of sociology, law and management. In the specifications and standards of various domestic building designs, the related range of the number of articles such as anti-throwing and suspension prohibition is extremely limited, and the number mainly comprises the height of a railing and the distance width of a lever. The current general planning standards for residential areas require that the edges of the road surface be kept at a certain distance from the buildings and components. However, in terms of safety of high-speed falling of falling objects, the specified distance value is difficult to completely meet the requirement, and a sufficient technical basis and theoretical basis are lacked. The falling object condition is monitored by installing the camera at home, but the effect of preventing and controlling risks is limited. The domestic research aiming at the high falling object danger and safety measures is less, and sufficient action psychology basis and basis are lacked. Compared with the domestic countries, the European and American countries attach more importance to the safety research of the building environment. In the sixty-seven decades of the 20 th century, socialists, planners and architects in the countries of europe and america also begin to pay attention to the protection and safety of the quality of the construction environment, but mainly related to the falling of objects such as construction sites, exterior wall materials and the like, and the discussion and the solution of the problem of falling objects from the perspective of the construction design and planning are not reported.
Disclosure of Invention
Based on the technical problems, the invention provides a building object intelligent detection and early warning interception system based on a CPU and a control method thereof.
The purpose of the invention is realized by the following technical scheme.
In one aspect of the present invention, a CPU-based intelligent detection and interception system for falling objects in a building is provided, which comprises:
the central processing unit CPU is used for acquiring building object information transmitted by the infrared and ultrasonic sensor arrays, starting the interception net in real time to intercept quickly and sending an alarm through the acousto-optic alarm circuit; meanwhile, a monitoring camera is started to shoot a picture of a falling object or an invading foreign body of the building and is transmitted to the mobile terminal;
the infrared sensor array comprises an infrared sensor and a Central Processing Unit (CPU), wherein the infrared sensor is used for detecting whether a building falling object exists around the building and transmitting infrared information of the building falling object or the invading foreign object to the CPU;
the ultrasonic sensor array comprises ultrasonic sensors and a Central Processing Unit (CPU), wherein the ultrasonic sensors are used for detecting whether buildings fall or not around the buildings and transmitting ultrasonic information of the buildings falling or invading foreign matters to the CPU;
the wireless network module is used for wireless transmission and comprises a control instruction, a starting signal and a control signal monitoring camera which are respectively sent to the interception network, the acousto-optic alarm circuit and the monitoring camera by the central processing unit CPU;
the monitoring camera is provided with a weak light ray auxiliary illumination device, is triggered by motion detection and triggers two shooting modes by a Central Processing Unit (CPU), and is used for shooting the scene of falling objects or invading foreign bodies of the building; the shot pictures are stored in a data memory SD card;
the intercepting net is driven by a switch of the electromagnetic attraction device and is used for quickly opening a certain angle when a building falls down to intercept the building falling down;
the intelligent terminal is used for looking up whether a photo of a falling object or an invading foreign object exists or not, position information and the type of the falling object in real time through the APP;
the clock circuit is used for generating clock signals for detecting the infrared and ultrasonic sensor arrays;
the storage is composed of an SD storage and is used for storing the live photos collected from the monitoring camera, including the original video frame and all digital image information in the video frame processing process;
the acousto-optic alarm circuit is used for sending an acousto-optic alarm signal when detecting that a building falls down, reminding pedestrians downstairs of avoiding the falling object and informing property management personnel that some emergency treatment work needs to be carried out in time when the building falls down or invades foreign matters;
the power supply module is used for supplying power required by the system;
the central processing unit CPU is respectively connected with the monitoring camera, the infrared sensor array, the ultrasonic sensor array, the interception net, the storage, the acousto-optic alarm circuit, the clock circuit, the display unit, the wireless network module and the power supply module, and is in wireless communication with the sensor array, the interception net and/or the invading foreign matters through the wireless network module.
Preferably, the infrared sensor array and the ultrasonic sensor array are arranged on the outer side of each layer of window of the building, and the infrared sensors detect whether the building falls or invades foreign matters within 300cm of the periphery of the building; the ultrasonic sensor detects whether a building falling object or an invading foreign object exists within 200cm of the periphery of the building.
Preferably, the intercepting net is arranged below the infrared sensor array and the ultrasonic sensor array of each layer of window; a plurality of monitoring cameras are arranged between every two layers.
Preferably, the intercepting net comprises a frame, an electromagnetic suction device, a pull rope and a rope winding motor; the frame is the rectangle, and the lower limb of rectangle frame is fixed on the outer wall, and electromagnetic attraction device includes a plurality of opposite polarity's magnet, and positive polarity magnet is fixed on the outer wall, and negative polarity magnet is fixed on rectangle frame upper edge, and the both sides middle part of rectangle frame is connected with the cable, and the other end of cable is connected on the cable motor output shaft.
Preferably, the central processing unit CPU adopts an STM 32F407 chip.
The infrared sensor array adopts a proximity diffuse reflection type infrared induction photoelectric switch.
The ultrasonic sensor array adopts an ultrasonic sensor proximity switch.
The wireless network module adopts a Zigbee wireless network module.
The monitoring camera adopts fluorite monitoring camera C3W, passes through cloud wifi remote network connection with mobile terminal.
The external high-speed HSE of the clock circuit is externally connected with an 8MHz quartz crystal oscillator, and a 168MHz main clock is output through parameter setting of a frequency division and frequency multiplication register; the low-speed clock is 32.768 kHz.
The sound and light alarm circuit adopts an NE555 timing circuit.
In another aspect of the invention, a CPU-based intelligent detection and interception method for falling objects in a building is provided, which comprises the following steps:
initializing a system:
the monitoring camera mode is a motion trigger working mode; the detection state of the infrared sensor array and the ultrasonic sensor array is 0;
generating a sensor layout database SX according to the positions of the infrared sensor array, the ultrasonic sensor array, the interception net and the monitoring camera array;
step 1: respectively starting all infrared sensors and ultrasonic sensors to detect and sense whether there is a building falling object or reflected signals generated by the shielding of an invading foreign matter in real time, and setting X (i, j) or Y (i, j) to be '0' when there is no building falling object shielding for any ith row and jth column of infrared sensors and ultrasonic sensors; if a building falling object or an invading foreign object is detected to block X (i, j) ═ 1 'or Y (i, j) ═ 1', an infrared detection result X (i, j) or Y (i, j) and an infrared sensor number IDX (i, j) or an ultrasonic sensor number IDY (i, j) are sent to a central processing unit CPU through a wireless network module;
step 2: the central processing unit CPU processes the detection signal transmitted by the wireless network module, and if X (i, j) and Y (i, j) are '0', the detection is continued; if X (i, j) and Y (i, j) are '1', recording the position numbers (i, j) of the infrared sensor and the ultrasonic sensor; executing an algorithm for quickly searching an infrared and ultrasonic sensor array layout data class table SX to obtain an interception net LJ _ ij arranged right below the infrared and ultrasonic sensor array (i, j), and executing the step 3;
and step 3: the CPU sends an interception starting instruction to all the interception networks with the numbers LJ (i, j); the intercepting net is rapidly unfolded to immediately intercept the falling objects of the building; simultaneously starting a monitoring camera to shoot and record the falling object, and storing a recorded image P (i, j) into the SD card;
and 4, step 4: the CPU starts an acousto-optic alarm circuit in real time to send out alarm sound and a flash lamp for the longitudinal positions of the infrared and ultrasonic sensor arrays (i, j);
and 5: and transmitting the shot recorded image P (i, j) to a mobile phone APP, intelligently identifying the falling objects or the invading foreign objects of the building by adopting a deep learning target detection algorithm of the mobile phone APP, and informing property management personnel of the identification result and the occurrence position of the identification result.
Preferably, in step 2, the sensor array layout data class table SX (M, N), if the sensor position is (i, j), where i is the number of layers and j is the number of sensor columns in the ith layer; initializing an interception network identity coding array LJ _ ij to be null;
<2sa >: starting from m-i-1, if the interception net state in the infrared and ultrasonic sensor array layout data type table SX (m, j) is true, the existence of the interception net is indicated, and the mth interception net is added in LJ _ ij;
<2sb >: repeating step <4sa > if m is equal to or greater than 1, when m is equal to m-1; otherwise, executing the step <4sc >;
<2sc >: and returning, and returning the interception network identity coding array LJ _ ij to the central processing unit CPU.
Preferably, in step 3, the CPU sends an interception start instruction to all the interception nets with numbers LJ (i, j), and the interception implementation process includes:
<3sa >: under normal conditions, when an interception net control signal output by a Central Processing Unit (CPU) is at a high level, an electromagnet control device of the interception net is attracted, and the interception net is not opened;
<3sb >: when a building falling object is detected, all output control signals of each interception net corresponding to the interception net identity coding data LJ _ ij are changed into low levels, so that an electromagnet control device of the interception net is opened, and the interception net is unfolded to realize interception;
<3sc >: after the intercepted building falling object is removed, the CPU sends out an output interception net control signal to be high level, and the polarities of the magnetic poles of the electromagnet control devices of the interception net are opposite; controlling a rope winding motor to rotate forwards, so that the rope winding is tightened and shortened, and the interception net is sucked;
<3sd >: after the intercepting net is restored to the normal state, the central processing unit CPU controls the rope winding motor to rotate reversely, and the rope winding is unfolded reversely.
Preferably, in step 5, a deep learning algorithm is adopted, and the specific processing procedure of intelligently identifying the object is as follows:
<5sa >: the central processing unit CPU sends the recorded image P (i, j) to the mobile phone APP through the wireless network module;
<5sb >: the mobile phone App records an image P (i, j) by histogram equalization processing;
<5sc>: for the recorded image P1(i, j) performing segmentation processing to obtain a local image P containing a complete building falling object2(i, j), while simplifying the background;
<5sd >: carrying out interpolation processing on a local image containing a building pendant to obtain an image with 256 multiplied by 3 pixels;
<5se >: sequentially obtaining feature maps with the sizes of 256 multiplied by 64, 128 multiplied by 128, 64 multiplied by 256 and 32 multiplied by 512 by passing the images with the sizes of 256 multiplied by 3 through a pooling layer and a convolution layer; with a 4 x 4 pre-selected feature map,
<5sf >: after classifying and regressing 4096+1536+384+96+16 6128 candidate frames, performing non-maximum suppression on the detection result to obtain a final object detection result.
Preferably, in the step <5se > - <5sf >, convolution kernels with different sizes are respectively adopted for classification and feature extraction, 6128 candidate frames are generated in total, each candidate frame is subjected to non-maximum suppression operation after being classified and regressed step by step, a target detection result is obtained, and the object detection result is displayed on the mobile phone APP.
Due to the adoption of the technical scheme, the invention has the following beneficial effects:
according to the invention, an infrared sensor and an ultrasonic sensor are adopted to jointly sense a falling object of a building, and after the CPU detects that the falling object appears, an interception net is controlled to be quickly opened and intercept the falling object through a wireless sensor network; opening the double-light infrared camera to shoot a scene, and recording the picture and the appearance position of the object; then starting an object falling acousto-optic alarm system through a wireless sensor network; and finally, detecting the content of the infrared photo by an object intelligent detection algorithm to complete object detection, sending the falling object photo and the falling object content to an administrator mobile phone through the APP, informing an administrator of carrying out emergency treatment on the detected or intercepted object in real time, and completing subsequent emergency treatment work. The system adopts the sensor to sense the falling objects of the building intelligently, realizes the communication between the CPU, the interception system, the falling object identification system and the alarm system through the wireless sensor network, can provide effective and real-time intelligent detection and interception for the falling objects of the high-rise building system, ensures the existence and accurate detection of the real-time falling objects, realizes the high-efficiency interception and accurate identification of the falling objects, and early warning of the falling objects, prevents safety accidents caused by the falling objects of the building, realizes the positioning of the falling objects or the invading objects, effectively prevents thieves from invading and throwing objects at high altitude, provides legal evidence for the problem of the falling objects at high altitude, and ensures the safety of the personnel and the objects of the building inhabitants.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention:
FIG. 1 is a schematic structural diagram of a CPU-based building falling object detecting and intercepting system of the present invention;
FIGS. 2(a) and (b) are schematic structural diagrams of the reset condition of the intercepting net of the present invention;
FIGS. 3(a) and (b) are schematic structural views of the opened condition of the intercepting net of the present invention;
FIG. 4 is a schematic view of the wall sensor arrangement of the present invention;
FIG. 5 is a flow chart of the processing and control of the intelligent detecting and intercepting system for detecting falling objects in a building according to the present invention;
fig. 6 is a flowchart of the intelligent object detection based on deep learning of the intelligent mobile terminal.
In the figure: 11. a window; 12. an array of sensors; 13. a surveillance camera; 14. an intercepting net; 1. an outer wall; 2. a cable; 3. a frame; 4. an electromagnetic attraction device; 5. a rope winding motor.
Detailed Description
The present invention will now be described in detail with reference to the drawings and specific embodiments, wherein the exemplary embodiments and descriptions of the present invention are provided to explain the present invention without limiting the invention thereto.
As shown in fig. 1, the present invention provides a CPU-based building object intelligent detection and interception system, comprising:
the central processing unit CPU is used for finishing reading information of whether a building falls down or not (or invades foreign matters) from the sensor array (the infrared sensor and the ultrasonic sensor), when the building falls down or the foreign matters invade is detected, the intercepting network of a lower layer of the building is started in real time through the wireless network module to quickly intercept the falling objects of the building, the position where the building falls down or the invades the foreign matters is recorded, the acousto-optic alarm circuit is started to inform property management personnel of emergency treatment, in addition, a monitoring camera is started through the wireless network module to shoot and store field pictures when the objects appear, the intelligent identification of the object types is finished, and the field pictures are sent to the mobile phone APP of the management personnel through the wireless sensor network.
The infrared sensor array comprises an infrared sensor and a Central Processing Unit (CPU), wherein the infrared sensor is used for detecting whether a building falling object or an invading foreign object exists within 300cm of the periphery of the building, reflecting a signal to the infrared receiver when the building falling object or the invading foreign object occurs, and outputting a detection result to the CPU;
the ultrasonic sensor array comprises ultrasonic sensors and a CPU, wherein the ultrasonic sensors are used for detecting whether a building falling object or an invading foreign object exists within 200cm of the periphery of a building, when the building falling object or the invading foreign object occurs, a signal is reflected to the ultrasonic receivers, and a detection result signal is output to the CPU; the time of the echo signals detected by the infrared sensor and the ultrasonic sensor is different, the effective detection distance is different, and the two are complementary to obtain a more efficient object detection effect.
The wireless network module is used for wireless transmission of information within 100 meters, and comprises a CPU (central processing unit) which sends a control instruction to an interception network through a wireless network, a CPU which sends a starting signal to the acousto-optic early warning through the wireless network, and a CPU which sends a control signal to a monitoring camera through the wireless network;
the monitoring camera is provided with a weak light auxiliary illumination device and is used for shooting the scene of falling objects or invading foreign matters of the building in the daytime and at night, recording the scene condition, setting a starting shooting mode, and acquiring photos and transmitting the photos to the data memory SD card for storage in a motion detection triggering mode and a CPU triggering mode;
the intelligent terminal is used for looking up whether a building falling object or a foreign object invasion picture, position information and an object type identification function on the APP in real time through the APP, so that property management personnel or judicial departments can conveniently look up and use the information for subsequent event processing;
the clock circuit is used for generating a clock signal used for the detection of the sensor array;
the data memory SD card is used for storing the on-site photos collected by the monitoring camera, the original video frames and all digital image information in the video frame processing process;
the display unit is used for displaying the building falling object image and the processing result of the system;
and the power supply module is used for supplying power required by the system.
The central processing unit CPU is respectively connected with the wireless network module, the monitoring camera, the interception net, the data storage SD card, the sound and light alarm circuit, the clock circuit, the infrared and ultrasonic sensor array, the display unit and the power supply module, and carries out wireless communication with the sensor array, the interception net and the intelligent terminal by utilizing the wireless network module.
The central processing unit CPU adopts an STM 32F407 chip. The infrared sensor adopts a near diffuse reflection type infrared induction photoelectric switch, and the detection distance of the infrared sensor is adjustable and is 3m at most. Ultrasonic sensor adopts ultrasonic sensor proximity switch, and adjustable inductive switch of detection distance 2 meters provides 485 bus output. The wireless sensing network module adopts a Zigbee wireless network module. The surveillance camera head adopts fluorite surveillance camera head C3W for it is outdoor, can be waterproof, have wireless high definition night vision function, and available cell-phone cloud wifi remote network connects. The data memory adopts SD card.
The external high-speed HSE of the clock circuit is externally connected with an 8MHz quartz crystal oscillator, and is set by frequency division and frequency multiplication register parameters (PLL _ M, PLL _ N, PLL _ P and PLL _ Q) to output a 168MHz main clock; the low-speed clock is 32.768 kHz.
The intercepting net is designed as shown in figures 3(a) and (b), and comprises a frame 3, an electromagnetic suction device 4, a pull rope 2 and a rope winding motor 5; the frame 3 is the rectangle, the lower limb of rectangle frame is fixed on the outer wall 1, the electromagnetic attraction device 2 includes three pairs of magnets of controllable polarity opposite, positive polarity magnet one end is fixed on the outer wall, other end negative polarity magnet is fixed on the edge of interception net rectangle frame, the opening and reset control of interception net are accomplished to the electromagnetic attraction device, the both sides middle part of interception net rectangle frame is connected with cable 2, the other end of cable 2 is connected on the output shaft of cable coiling motor 5, realize the reset and the opening of interception net through control electromagnetic attraction device electromagnetism polarity.
As shown in fig. 2(a) and (b), when there is no falling object, the intercepting net is in a suction or reset state; as shown in fig. 3(a) and (b), when a building falling object is detected, the intercepting net is opened for intercepting; after the intercepting net is opened, the opening angle is limited by the inhaul cable 2, and the maximum opening angle is 75 degrees. The polarity of the electromagnetic device is controlled by the CPU, when the polarities of the electromagnetic devices 4a and 4b are the same, on one hand, the electromagnetic attraction device and the interception net repel each other to push the interception net to open outwards, and on the other hand, the interception net is accelerated to open outwards under the action of gravity, so that the interception net quickly falls to open, and the falling objects of the building are intercepted.
After the building falls and is cleaned, the polarities of the electromagnetic devices on the intercepting net are different under the control of the cable winding motor, the electromagnetic devices 4a and 4b are attracted, and the intercepting net is reset.
The mobile terminal adopts the smart phone to install the APP to realize early warning and identification. The sound-light alarm circuit is realized by an NE555 timing circuit.
As shown in fig. 4, an infrared sensor and ultrasonic sensor array 12 is provided outside the window 11 of each floor of the building. The intercepting net 14 is arranged below the sensor array of each window of each layer; and a plurality of monitoring cameras 13 are arranged between every two layers.
As shown in FIG. 5, the invention adopts the CPU-based building falling object detecting and intercepting method, which comprises the following steps:
initialization: initializing the working mode of the monitoring camera into a motion triggering mode; initializing an infrared sensor X (i, j) and setting an ultrasonic detection state Y (i, j) as 0; initializing an acousto-optic alarm system to be in an initial state; initializing an interception net to be in an initial state; generating a sensor layout database SX according to the positions of the infrared sensor array, the ultrasonic sensor array, the interception net and the monitoring camera array;
step 1: starting all infrared sensors to detect reflected signals generated by whether building falling object shielding exists or not in real time, and setting X (i, j) to be '0' when no foreign object shielding exists for any infrared sensor in the ith row and the jth column; and if the object shielding X (i, j) is detected to be equal to '1', simultaneously sending the infrared detection result X (i, j) and the infrared sensor number IDX (i, j) to the CPU through the wireless sensor network module.
Step 2: starting the ultrasonic sensors to sense whether a reflected signal Y is generated by shielding of a falling object of the building in real time, wherein for any i ultrasonic sensor, Y (i, j) is '0' when no foreign object is shielded; and if the object occlusion Y (i, j) is detected to be equal to '1', the ultrasonic detection result Y (i, j) and the ultrasonic sensor number IDY (i, j) are sent to the CPU through the wireless sensor network module.
And step 3: the central processing unit CPU processes the detection signal received by the wireless network module by adopting OR operation, namely X (i, j) | Y (i, j), and if the result is '0', the CPU waits in place in the step 2; if the result is '1', then: (1) recording the position numbers (i, j) of the infrared sensor and the ultrasonic sensor; (2) and (4) executing a quick search algorithm of the sensor array layout data type table SX to obtain an interception net LJ _ ij arranged under the sensor array (i, j), and executing the step 4.
In step 3, aiming at the layout diagram of the building falling object detection system shown in fig. 3, a sensor layout data class table SX (M, N) is provided, if the sensor position is (i, j), wherein i is the number of layers and j is the number of sensor columns of the ith layer; initializing an interception network identity coding array LJ _ ij to be null;
<3sa >: starting from m-i-1, if the interception net state in the infrared and ultrasonic sensor array layout data type table SX (m, j) is '1', indicating that the interception net exists, and adding the mth interception net in LJ _ ij;
<3sb >: repeating step 4sa if m is equal to or greater than 1, where m is m-1; otherwise, executing step 4 sc;
<3sc >: and returning, and returning the interception network identity coding array LJ _ ij to the CPU.
And 4, step 4: the CPU sends an interception starting instruction to all the interception nets with the serial numbers LJ (i, j); after the intercepting net receives the intercepting instruction, the intercepting net is immediately unfolded to the maximum opening position through the electromagnetic control device, and the detected falling objects or foreign matters are immediately intercepted; and simultaneously starting a camera V (i, j) to take a picture of the falling object, and storing a picture P (i, j) taken and recorded in the SD card.
In step 4, after the central processing unit CPU receives the interception net identity code array LJ _ ij right below the building falling object, the specific process of intercepting is implemented as follows:
<4sa >: under normal conditions, when an interception net control signal output by a Central Processing Unit (CPU) is at a high level, the high level enables the magnetic poles of electromagnet control devices of the interception net to have different polarities and be in an attraction state, and the interception net is not opened;
<4sb >: when a building falling object is detected, searching a corresponding intercepting net control signal for each intercepting net in the intercepting net identity coding data LJ _ ij, changing all output control signals into low levels, enabling the magnetic poles of the electromagnet control devices of the intercepting nets to be the same in polarity and in a repulsion state, rapidly unfolding the intercepting nets downwards under the action of electromagnetic thrust and self gravity, but only opening to a maximum angle (75 degrees) due to the action of a winding rope to be in an upwards inclined state, and facilitating reliable interception of falling foreign matters; meanwhile, a plurality of small suckers are arranged on the intercepting net, so that efficient intercepting is convenient to realize;
<4sc >: after intercepting the falling objects of the building, the property management personnel starts an emergency treatment mode and removes the intercepted objects in time; if the object is relatively large, manual removal or rescue is carried out in time through the nearest window; if the object is small, the object can be grabbed in time by the remote control unmanned aerial vehicle, so that the dangerous situation of falling again is avoided;
<4sd >: after the object is removed, for all the opened intercepting nets, the CPU sends out high level of an output intercepting net control signal to enable the polarities of the magnetic poles of the electromagnet control devices of the intercepting nets to be opposite; meanwhile, the central processing unit CPU controls the rope winding motor to rotate forwards through the wireless network, so that the rope winding is shortened, and the intercepting net is controlled to be tightened to the wall surface; once the electromagnet device is contacted, the electromagnet device is in an attraction state due to opposite polarity, and the interception net is recovered to the conventional state;
<4se >: after the intercepting net is restored to the conventional state, the central processing unit CPU controls the rope winding motor to rotate reversely through the wireless network, so that the rope winding motor is completely unfolded reversely to be in a natural falling state, and the intercepting net is convenient to open rapidly and stretch in a limiting mode when a falling object exists next time.
And 5: the CPU starts the sound and flash alarm devices of the inner and outer walls of the building corridor in real time, and sends out alarm sound and flash lamps for the longitudinal positions of the sensors (i, j), so that people can avoid the outer area of the building where falling objects or foreign matters invade in time.
Step 6: the shot recorded image P (i, j) and the appearance position of the shot recorded image P (i, j) are transmitted to the mobile phone APP, intelligent identification is carried out on an object through a deep learning target detection algorithm of the mobile phone APP, and a property manager is informed of an identification result and the appearance position of the identification result through the mobile phone APP.
In step 6, a deep learning algorithm is adopted, and a specific processing process of intelligently identifying the object is shown in fig. 5, specifically:
<6sa >: for the recorded image P (i, j), the CPU sends the recorded image to the mobile phone APP through a wireless network;
<6sb>: after the mobile phone App receives P (i, j), the histogram equalization processing is adopted to enhance the image, and the image P with better visual effect is obtained1(i，j)；
<6sc>: for image P1(i, j) performing segmentation processing to obtain a local image P containing the complete falling object2(i, j) while simplifying the background and facilitatingAccurate identification of the object;
<6sd >: carrying out interpolation processing on a local image containing a pendant to obtain an image with 256 multiplied by 3 pixels;
<6se >: [1] passing 256 × 256 × 3 images through a convolutional layer to obtain a feature map with 256 × 256 × 64 dimensions;
[2] obtaining a characteristic diagram with the size of 128 multiplied by 128 after passing through a pooling layer and two convolution layers;
[3] obtaining a feature map with the size of 64 multiplied by 256 after passing through a pooling layer and three convolutional layers (Conv 3);
[4] obtaining a feature map with the size of 32 × 32 × 512 after passing through a pooling layer and three convolution layers, obtaining a feature map with the size of 32 × 32 × 512, normalizing the feature map, and obtaining 32 × 32 × 4 ═ 4096 candidate frames by adopting a 4 × 4 preselected feature map for classification and regression, wherein the feature map is commonly used for detecting small objects;
<6sf >: [1] then pooling the feature map of the previous step and obtaining a 16 × 16 × 1024 feature map by using a 3 × 3 × 1024 convolutional layer;
[2] and performing convolution operation of 1 × 1 × 1024 to obtain a feature map convolution operation of 16 × 16 × 1024 to obtain a feature map of 16 × 16 × 1024, and performing classification and regression by using 6 × 6 preselected feature maps to obtain 16 × 16 × 6-1536 candidate frames.
<6sg >: then carrying out convolution and pooling operations of 1 × 1 × 1024 to obtain a 16 × 16 × 512 feature map, carrying out convolution and pooling operations of 3 × 3 × 512 to obtain an 8 × 8 × 512 feature map, and adopting a 6 × 6 preselected feature map to obtain 8 × 8 × 6-384 candidate frames for classification and regression;
<6sh >: performing convolution of 1 × 1 × 512 and 3 × 3 × 512, performing pooling operation to obtain 4 × 4 × 256 feature maps, and performing classification and regression by using 6 × 6 preselected feature maps to obtain 4 × 4 × 6-96 candidate frames;
<6si >: continuously performing convolution operations of 1 × 1 × 256 and 3 × 3 × 256, performing pooling operation to obtain 2 × 2 × 128 features, and performing classification and regression by using a 4 × 4 preselected feature map to obtain 16 candidate frames of 2 × 2 × 4;
<6sj >: after classifying and regressing 4096+1536+384+96+16 6128 candidate frames, performing non-maximum suppression on the detection result to obtain a final object detection result.
In the steps from <6se > to <6si >, convolution kernels with different sizes are respectively adopted for classification and feature extraction, 6128 candidate frames are generated in total, non-maximum suppression operation is carried out after each candidate frame is classified and regressed step by step, a target detection result is obtained, and the object detection result is displayed on the APP, so that managers can arrange emergency danger elimination treatment in time.
The method can provide effective and real-time intelligent detection and interception of falling objects for a high-rise system, ensure the real-time existence and accurate detection of the occurrence positions of the falling objects, efficiently intercept and accurately identify the falling objects, and early warn the falling objects.
The present invention is not limited to the above-mentioned embodiments, and based on the technical solutions disclosed in the present invention, those skilled in the art can make some substitutions and modifications to some technical features without creative efforts according to the disclosed technical contents, and these substitutions and modifications are all within the protection scope of the present invention.

Claims (10)

1. The utility model provides a building weighs down thing intellectual detection system and interception system based on CPU which characterized in that includes:
the central processing unit CPU is used for acquiring building object information transmitted by the infrared and ultrasonic sensor arrays, starting the interception net in real time to intercept quickly and sending an alarm through the acousto-optic alarm circuit; meanwhile, a monitoring camera is started to shoot a picture of a falling object or an invading foreign body of the building and is transmitted to the mobile terminal;
the infrared sensor array comprises an infrared sensor and a Central Processing Unit (CPU), wherein the infrared sensor is used for detecting whether a building falling object or an invading foreign object exists around the building and transmitting infrared information of the building falling object to the CPU;
the ultrasonic sensor array comprises ultrasonic sensors and a Central Processing Unit (CPU), wherein the ultrasonic sensors are used for detecting whether building falling objects or invading foreign matters exist around a building and transmitting ultrasonic information of the building falling objects to the CPU;
the wireless network module is used for wireless transmission and comprises a control instruction, a starting signal and a control signal monitoring camera which are respectively sent to the interception network, the acousto-optic alarm circuit and the monitoring camera by the central processing unit CPU;
the monitoring camera is provided with a weak light ray auxiliary illumination device, is triggered by motion detection and triggers two shooting modes by a Central Processing Unit (CPU), and is used for shooting the scene of falling objects or invading foreign bodies of the building; the shot pictures are stored in a data memory SD card;
the intercepting net is driven by a switch of the electromagnetic attraction device and is used for quickly opening a certain angle when a building falls down to intercept the building falling down;
the intelligent terminal is used for looking up whether a photo of a falling object or an invading foreign body exists or not, position information and the type of the falling object or the invading foreign body in real time through the APP;
the clock circuit is used for generating clock signals for detecting the infrared and ultrasonic sensor arrays;
the storage is composed of an SD storage and is used for storing the live photos collected from the monitoring camera, including the original video frame and all digital image information in the video frame processing process;
the acousto-optic alarm circuit is used for sending an acousto-optic alarm signal when detecting that a building falls down or invades a foreign body, reminding a pedestrian downstairs of avoiding the falling down, and informing a property management person that the building falls down and some emergency treatment work needs to be carried out in time;
the power supply module is used for supplying power required by the system;
the central processing unit CPU is respectively connected with the monitoring camera, the infrared sensor array, the ultrasonic sensor array, the intercepting net, the storage, the acousto-optic alarm circuit, the clock circuit, the display unit, the wireless network module and the power supply module, and the central processing unit CPU is in wireless communication with the sensor array, the intercepting net and the intelligent terminal through the wireless network module.
2. The intelligent detecting and intercepting system for building falling objects based on the CPU as claimed in claim 1, wherein the infrared sensor array and the ultrasonic sensor array are arranged outside each layer of windows of the building, and the infrared sensor array detects whether the building falling objects or the invading foreign objects exist within 300cm of the periphery of the building; the ultrasonic sensor detects whether a building falling object or an invading foreign object exists within 200cm of the periphery of the building.
3. The intelligent detecting and intercepting system for building falling objects based on the CPU of claim 1, wherein the intercepting net is arranged below the infrared and ultrasonic sensor arrays of each layer of the window; a plurality of monitoring cameras are arranged between every two layers.
4. The intelligent detecting and intercepting system for building falling objects based on the CPU as claimed in claim 3, wherein the intercepting net comprises a frame, an electromagnetic attraction device, a pull cable and a cable winding motor; the frame is the rectangle, and the lower limb of rectangle frame is fixed on the outer wall, and electromagnetic attraction device includes a plurality of opposite polarity's magnet, and positive polarity magnet is fixed on the outer wall, and negative polarity magnet is fixed on rectangle frame upper edge, and the both sides middle part of rectangle frame is connected with the cable, and the cable other end is connected on the cable motor output shaft of rolling up.
5. The intelligent detecting and intercepting system for building falling objects based on the CPU as claimed in claim 1, wherein the CPU employs an STM 32F407 chip;
the infrared sensor array adopts a proximity diffuse reflection type infrared induction photoelectric switch;
the ultrasonic sensor array adopts an ultrasonic sensor proximity switch;
the wireless network module adopts a Zigbee wireless network module;
the monitoring camera adopts a fluorite monitoring camera C3W and is connected with the mobile terminal through a cloud wifi remote network;
the external high-speed HSE of the clock circuit is externally connected with an 8MHz quartz crystal oscillator, and a 168MHz main clock is output through parameter setting of a frequency division and frequency multiplication register; the low-speed clock adopts 32.768 kHz;
the sound and light alarm circuit adopts an NE555 timing circuit.
6. A building falling object intelligent detection and interception method based on a CPU is characterized by comprising the following steps:
initializing a system:
the monitoring camera mode is a motion trigger working mode; the detection state of the infrared sensor and the ultrasonic sensor is 0;
generating a sensor layout database SX according to the positions of the infrared sensor array, the ultrasonic sensor array, the interception net and the monitoring camera array;
step 1: respectively starting all infrared sensors and ultrasonic sensors to detect and sense whether there is a building falling object or reflected signals generated by the shielding of an invading foreign matter in real time, and setting X (i, j) or Y (i, j) to be '0' when there is no building falling object shielding for any ith row and jth column of infrared sensors and ultrasonic sensors; if a building falling object or an invading foreign object is detected to block X (i, j) ═ 1 'or Y (i, j) ═ 1', an infrared detection result X (i, j) or Y (i, j) and an infrared sensor number IDX (i, j) or an ultrasonic sensor number IDY (i, j) are sent to a central processing unit CPU through a wireless network module;
step 2: the central processing unit CPU processes the detection signal transmitted by the wireless network module, and if X (i, j) and Y (i, j) are '0', the detection is continued; if X (i, j) and Y (i, j) are '1', recording the position numbers (i, j) of the infrared sensor and the ultrasonic sensor; executing a quick search algorithm of a sensor array layout data type table SX to obtain an interception net LJ _ ij arranged right below the infrared and ultrasonic sensor array (i, j), and executing the step 3;
and step 3: the CPU sends an interception starting instruction to all the interception networks with the numbers LJ (i, j); the intercepting net is rapidly unfolded to immediately intercept the falling objects of the building; simultaneously starting a monitoring camera to shoot and record the falling object, and storing a recorded image P (i, j) into the SD card;
and 4, step 4: the CPU starts an acousto-optic alarm circuit in real time to send out alarm sound and a flash lamp for the longitudinal positions of the infrared and ultrasonic sensor arrays (i, j);
and 5: and transmitting the shot recorded image P (i, j) to a mobile phone APP, intelligently identifying the falling objects or the invading foreign objects of the building by adopting a deep learning target detection algorithm of the mobile phone APP, and informing property management personnel of the identification result and the occurrence position of the identification result.
7. The intelligent CPU-based building falling object detecting and intercepting method according to claim 6, wherein in the step 2, the sensor array layout data class table SX (M, N) is provided, if the sensor position is (i, j), wherein i is the number of layers and j is the number of sensor columns of the ith layer; initializing an interception network identity coding array LJ _ ij to be null;
<2sa >: starting from m-i-1, if the interception net state in the infrared and ultrasonic sensor array layout data type table SX (m, j) is true, the existence of the interception net is indicated, and the mth interception net is added in LJ _ ij;
<2sb >: repeating step <4sa > if m is equal to or greater than 1, when m is equal to m-1; otherwise, executing the step <4sc >;
<2sc >: and returning, and returning the interception network identity coding array LJ _ ij to the central processing unit CPU.
8. The intelligent detection and interception method for building falling objects based on the CPU as claimed in claim 6, wherein in step 3, the CPU sends an interception starting instruction to all the interception nets with number LJ (i, j), and the interception process is implemented as follows:
<3sa >: under normal conditions, when an interception net control signal output by a Central Processing Unit (CPU) is at a high level, an electromagnet control device of the interception net is attracted, and the interception net is not opened;
<3sb >: when a building falling object is detected, all output control signals of each interception net corresponding to the interception net identity coding data LJ _ ij are changed into low levels, so that an electromagnet control device of the interception net is opened, and the interception net is unfolded to realize interception;
<3sc >: after the intercepted building falling object is removed, the CPU sends out an output interception net control signal to be high level, and the polarities of the magnetic poles of the electromagnet control devices of the interception net are opposite; controlling a rope winding motor to rotate forwards, so that the rope winding is tightened and shortened, and the interception net is sucked;
<3sd >: after the intercepting net is restored to the normal state, the central processing unit CPU controls the rope winding motor to rotate reversely, and the rope winding is unfolded reversely.
9. The intelligent detection and interception method for building falling objects based on the CPU as claimed in claim 6, wherein in the step 5, a deep learning algorithm is adopted, and the specific processing procedure for intelligently identifying objects is as follows:
<5sa >: the central processing unit CPU sends the recorded image P (i, j) to the mobile phone APP through the wireless network module;
<5sb >: the mobile phone App records an image P (i, j) by histogram equalization processing;
<5sc>: for the recorded image P1(i, j) performing segmentation processing to obtain a local image P containing a complete building falling object2(i, j), while simplifying the background;
<5sd >: carrying out interpolation processing on a local image containing a building pendant to obtain an image with 256 multiplied by 3 pixels;
<5se >: sequentially obtaining feature maps with the sizes of 256 multiplied by 64, 128 multiplied by 128, 64 multiplied by 256 and 32 multiplied by 512 by passing the images with the sizes of 256 multiplied by 3 through a pooling layer and a convolution layer; with a 4 x 4 pre-selected feature map,
<5sf >: after classifying and regressing 4096+1536+384+96+16 6128 candidate frames, performing non-maximum suppression on the detection result to obtain a final object detection result.
10. The intelligent detection and interception method for building falling objects based on the CPU as claimed in claim 6, characterized in that in the step <5se > - <5se >, convolution kernels with different sizes are respectively adopted for classification and feature extraction, 6128 candidate frames are generated in total, after classification and regression for each candidate frame step by step, non-maximum suppression operation is performed, a target detection result is obtained, and the object detection result is displayed on the mobile phone APP.
CN202111275440.4A 2021-10-29 2021-10-29 Building falling object intelligent detection and interception system based on CPU and control method thereof Pending CN113990009A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111275440.4A CN113990009A (en) 2021-10-29 2021-10-29 Building falling object intelligent detection and interception system based on CPU and control method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111275440.4A CN113990009A (en) 2021-10-29 2021-10-29 Building falling object intelligent detection and interception system based on CPU and control method thereof

Publications (1)

Publication Number Publication Date
CN113990009A true CN113990009A (en) 2022-01-28

Family

ID=79744700

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111275440.4A Pending CN113990009A (en) 2021-10-29 2021-10-29 Building falling object intelligent detection and interception system based on CPU and control method thereof

Country Status (1)

Country Link
CN (1) CN113990009A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115424401A (en) * 2022-11-07 2022-12-02 南京合识科技有限公司 Big data security monitoring management method and system with early warning processing function
WO2023204741A1 (en) * 2022-04-22 2023-10-26 Telefonaktiebolaget Lm Ericsson (Publ) Method and apparatus for detecting falling objects via a wireless communication network

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023204741A1 (en) * 2022-04-22 2023-10-26 Telefonaktiebolaget Lm Ericsson (Publ) Method and apparatus for detecting falling objects via a wireless communication network
CN115424401A (en) * 2022-11-07 2022-12-02 南京合识科技有限公司 Big data security monitoring management method and system with early warning processing function
CN115424401B (en) * 2022-11-07 2023-09-19 盛世数字科技(天津)有限公司 Big data security monitoring management method and system with early warning processing function

Similar Documents

Publication Publication Date Title
CN109686109B (en) Parking lot safety monitoring management system and method based on artificial intelligence
CN109068099B (en) Virtual electronic fence monitoring method and system based on video monitoring
US9449398B2 (en) Directional object detection
CN113990009A (en) Building falling object intelligent detection and interception system based on CPU and control method thereof
Lin et al. A vision-based parking lot management system
CN108275114B (en) Oil tank anti-theft monitoring system
CN107336669A (en) Vehicle safety protection system and method thereof
US20120188370A1 (en) Surveillance systems and methods to monitor, recognize, track objects and unusual activities in real time within user defined boundaries in an area
CN106926794B (en) Vehicle monitoring system and method thereof
CN113593250A (en) Illegal parking detection system based on visual identification
CN109886129B (en) Prompt message generation method and device, storage medium and electronic device
CN210110004U (en) Oil field behavior monitoring system based on artificial intelligence
CN104917889A (en) Mobile monitoring terminal and monitoring method thereof
KR20080044812A (en) The automatic guard system to prevent the crime and accident using computer video image analysis technology
CN106060471A (en) Intelligent monitoring method and system based on augment reality technology
CN111429681A (en) Comprehensive perimeter security system with infrared thermal image verification and method
CN110703760A (en) Newly-increased suspicious object detection method for security inspection robot
KR102353724B1 (en) Apparatus and method for monitoring city condition
TW201801520A (en) Image device and method for detecting a chief
JP2014119827A (en) Imaging system
CN216486694U (en) Building falling object intelligent detection and interception system based on CPU
CN207020820U (en) Red-running Monitor System
CN114092885A (en) High-speed service area vehicle protection method and system based on video analysis
CN212278365U (en) Multifunctional pole for patrol inspection alarm
JP6967868B2 (en) Surveillance systems, surveillance programs, and storage media

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination