Search

CN-121982864-A - Intelligent monitoring and early warning method and system for building outer facade

CN121982864ACN 121982864 ACN121982864 ACN 121982864ACN-121982864-A

Abstract

The invention relates to the technical field of intelligent security and building monitoring and provides a building facade intelligent monitoring and early warning method and system, wherein the method comprises the following steps of S1, generating a reference map and dividing areas, S2, monitoring and filtering in real time, S3, identifying abnormality and generating a signal, S4, performing early warning response operation according to the abnormal event signal, namely, avoiding shooting indoor scenes by deploying a camera array with a specific pitch angle, and strictly protecting privacy; the multi-view image stitching and the space calibration model based on the real physical size are utilized to generate a high-precision floor-level digital reference image, so that pixels in the image can be abnormally and accurately mapped to specific floors and even resident units, and the problem of fuzzy traditional monitoring and positioning is solved.

Inventors

  • CANG XIAODAN

Assignees

  • 无锡驰众保安服务有限公司

Dates

Publication Date
20260505
Application Date
20260205

Claims (8)

  1. 1. The intelligent monitoring and early warning method for the building outer facade is characterized by comprising the following steps of: S1, reference picture generation and region division, namely collecting building outer elevation images, generating digital reference pictures corresponding to all floors, and dividing independent monitoring regions on the digital reference pictures based on building structure information; s2, monitoring and filtering in real time, namely continuously acquiring a real-time video stream, filtering environmental interference on the image, and outputting a filtered real-time image; S3, abnormality identification and signal generation, namely comparing the filtered real-time image with the digital reference image of the corresponding monitoring area, and generating an abnormal event signal containing abnormality type and position information when the abnormality characteristic of a preset type is identified; S4, early warning response, namely executing the following early warning response operation according to the abnormal event signal: S41, sending a first control instruction to the remote monitoring terminal to trigger an audible and visual alarm and start timing of a preset time limit; s42, if a confirmation signal from the remote monitoring terminal is received before the timing is stopped, the alarm is released; s43, if the confirmation signal is not received when the timing is stopped, a second control instruction is sent to a local early warning indicator lamp of the floor corresponding to the abnormal position, so that the indicator lamp is activated to carry out on-site warning.
  2. 2. The monitoring and early warning method according to claim 1, wherein in the step S1, generating the digital reference map and dividing the monitoring area includes: the acquired multiple inclined view angle images are fused through image stitching and geometric correction to generate a building outer elevation panoramic image, wherein the geometric correction adopts a perspective transformation model, and the calculation formula is as follows: , In the middle of (a) , ) To correct the pixel coordinates of the front oblique image # , ) To correct the pixel coordinates of the rear panoramic image, For homogeneous co-ordinate coefficients, in matrix ~ Obtaining perspective transformation parameters through at least 4 pairs of image characteristic points by matching and solving; According to the physical size of the building, a mapping relation from the pixel coordinates of the panoramic image to the positions of the actual floors and the resident units is established, and the mapping formula is as follows: , In the formula, For the actual building height to be practical, For a practical horizontal distance, Is a vertical direction pixel-to-physical size conversion factor, Is a horizontal pixel-physical size conversion coefficient , ) For the pixel coordinates corresponding to the origin of the panorama, 、 The actual physical coordinates corresponding to the origin are obtained; And generating the digital reference pictures aligned according to floors, and completing the division and identification of each independent monitoring area.
  3. 3. The monitoring and early warning method according to claim 1, wherein the filtering of the environmental interference in the step S2 includes: and carrying out illumination normalization processing on the interference filtered image based on real-time environment illumination data acquired from an environment sensing module or a meteorological data interface, and outputting the filtered real-time image.
  4. 4. The monitoring and early warning method according to claim 3, wherein the background modeling and foreground detection are realized by an adaptive Gaussian mixture model, and a probability density function and a parameter updating formula of each pixel in the model are as follows: , In the formula, The gray value of the pixel at time t, In order to be a number of gaussian models, The weight of the ith gaussian component at time t, Is the mean value of the ith gaussian component at time t, For the variance of the ith gaussian component at time t, Is a Gaussian probability density function; Parameter updating rules: , In the formula, In order to achieve a weight-learning rate, Learning rate for mean and variance, updating weight of unmatched Gaussian components to The mean and variance remain unchanged; The illumination normalization processing is realized through an algorithm based on the Retinex theory, and the multi-scale Retinex calculation formula is as follows: , In the process, the For the normalized gray value of the image, In order to be a number of dimensions, For the weight of the s-th scale, The image gray values are filtered out for the incoming disturbances, In the case of a convolution operation, Is a Gaussian kernel function, and the expression is as follows: , In the formula, Is the standard deviation of the Gaussian kernel of the s-th scale.
  5. 5. The monitoring and early warning method according to claim 1, wherein in the step S3, identifying the abnormality feature of the preset type includes: calculating a difference map between the filtered real-time image and the digital reference map, and calculating a difference value by adopting a gray level difference square sum method, wherein the formula is as follows: , In the formula, Is that The difference value of the positions is used to determine, Is a window The number of pixels in the pixel array is, In order to filter real-time image The gray value of the location is used, Gray values of the digital reference map at the (i, j) position when When the pixel is larger than a preset threshold value T, marking the pixel as a difference pixel, and further extracting a significant change region in the pixel; and analyzing the motion trail, appearance and time sequence change characteristics of the significant change region, and classifying and identifying the significant change region as at least one abnormal type of high altitude parabolic, fire smoke, building structure falling or personnel dangerous exposure.
  6. 6. The intelligent monitoring and early warning system for the outer facade of the building is used for realizing the method as claimed in any one of claims 1 to 5, and is characterized by comprising an image acquisition unit, a data processing and control center and an early warning execution unit which are interconnected and cooperatively operated through a communication network: The image acquisition unit consists of a pitching camera array which is arranged at the periphery of the building and is arranged at a preset inclination angle, and is used for acquiring image data of the outer facade of the building and transmitting the image data to the data processing and control center through the communication network; the data processing and control center is used for receiving and processing the image data from the image acquisition unit, generating and sending a control instruction to the early warning execution unit; The early warning execution unit comprises a remote monitoring terminal and local early warning indication lamps arranged on all floors, wherein the remote monitoring terminal is used for receiving and responding to the alarm and control instructions sent by the data processing and control center, and the local early warning indication lamps are used for receiving and executing the control instructions from the data processing and control center.
  7. 7. The system of claim 6, wherein the data processing and control center further comprises a data interface for accessing an environmental awareness module or a meteorological data interface to obtain real-time environmental data for assisting in the filtering of environmental interference.
  8. 8. The system of claim 6, wherein the communication network comprises a data transmission network for connecting the image acquisition unit and the data processing and control center, and a building control network for connecting the data processing and control center and the local warning indicator lights for each floor in the warning execution unit.

Description

Intelligent monitoring and early warning method and system for building outer facade Technical Field The invention relates to the technical field of intelligent security and building monitoring, in particular to an intelligent monitoring and early warning method and system for an outer facade of a building. Background With the acceleration of the urban process, high-rise and super-high-rise buildings are increasingly dense, and public safety problems of the outer facade are obvious, wherein main risks include ① high-altitude parabolic articles falling from high positions, high speed and large impact force, serious threat to pedestrian safety, ② fire smoke, smoke generated by indoor fire overflows from windows and is a key signal of early fire, ③ outer wall components fall off, wall coverings, ceramic tiles or decorations fall loose due to aging, ④ personnel are exposed in danger, such as climbing windowsills by children, safety measures of outdoor operators are insufficient, and the like. These events are characterized by burstiness, concealment, and hazard. At present, prevention and control of the risks mainly depend on the following modes, but have obvious limitations that manual inspection and ground monitoring are low in manual inspection efficiency and long in interval, and all-weather coverage cannot be realized. The common monitoring camera installed on the ground is limited by the visual angle, is difficult to completely cover building facades with the depth of hundreds of meters, has a large number of blind areas, and has long shooting distance and insufficient image details. Conventional video analysis systems some systems attempt to detect movement of a fixed camera frame. However, the outer facade of the high-rise building has complex and changeable scenes, strong day and night illumination changes, season and weather influences (such as rain and snow, haze and light reflection), and non-target movements of flying birds, insects, drifting sundries and the like, can generate a large amount of interference signals, so that the false alarm rate of the system is extremely high, and the practicability is poor. In order to realize effective monitoring, the view angle of a camera is often required to be aimed at areas such as a window and a balcony, but the privacy of households is invaded, legal and ethical problems are caused, and a plurality of monitoring schemes are difficult to be practically deployed. The early warning mechanism is single, and the prior art usually finishes after the background generates an alarm and completely depends on real-time response of monitoring personnel. If the personnel is missed, delayed or cannot judge, the early warning information cannot be effectively transmitted to the risk site, and a response delay and a blind area of 'last kilometer of early warning' exist. Based on the problems, we propose a building facade intelligent monitoring and early warning method and system. Disclosure of Invention Aiming at the defects of the prior art, the invention realizes the automatic, accurate and reliable identification of various potential safety hazards of the outer facade of the high-rise building on the premise of protecting the privacy of the householder, and establishes a set of intelligent hierarchical response mechanism so as to promote the instantaneity and the effectiveness of the safety pre-warning. In order to achieve the above purpose, the invention is realized by the following technical scheme: an intelligent monitoring and early warning method for an outer facade of a building comprises the following steps: S1, reference picture generation and region division, namely collecting building outer elevation images, generating digital reference pictures corresponding to all floors, and dividing independent monitoring regions on the digital reference pictures based on building structure information; s2, monitoring and filtering in real time, namely continuously acquiring a real-time video stream, filtering environmental interference on the image, and outputting a filtered real-time image; S3, abnormality identification and signal generation, namely comparing the filtered real-time image with the digital reference image of the corresponding monitoring area, and generating an abnormal event signal containing abnormality type and position information when the abnormality characteristic of a preset type is identified; S4, early warning response, namely executing the following early warning response operation according to the abnormal event signal: S41, sending a first control instruction to the remote monitoring terminal to trigger an audible and visual alarm and start timing of a preset time limit; s42, if a confirmation signal from the remote monitoring terminal is received before the timing is stopped, the alarm is released; s43, if the confirmation signal is not received when the timing is stopped, a second control instruction is sent to a local early warning indicat