Video monitoring and fire alarm linkage system and method

Abstract

The present invention relates to a video surveillance and fire alarm linkage system and method, and relates to the general field of image data processing or generation. The system includes: a fire analysis mechanism, which is used to input the floor area of ​​a fire monitoring area, multiple foreground information of a panoramic image, and various content data of the panoramic image into an intelligent fire analysis model to obtain the interval time from the image shooting time to the fire occurrence time; a wireless transmission mechanism, which is used to wirelessly transmit the interval time to a fire management service network element. The present invention also relates to a video surveillance and fire alarm linkage method. Through the present invention, in the face of the technical problem that the time of fire occurrence cannot be predicted based on current visualization data, the specific time of subsequent fire occurrence is intelligently judged by using an intelligent fire analysis model based on the floor area of ​​the fire monitoring area and multiple targeted screened visualization data of the panoramic picture of the fire monitoring area at the current moment.

Description

Technical Field

[0001] The present invention relates to the general field of image data processing or generation, and in particular to a video monitoring and fire alarm linkage system and method. Background Art

[0002] Video surveillance systems involve general image data processing or generation. Since its development, in addition to the continuous innovation of monitoring technology, monitoring products have also begun to form a powerful security monitoring system with other products. For example, video surveillance systems can use electronic fences to enclose the monitored area into a closed area. Electronic fences have blocking and alarm systems. They are an active intrusion prevention fence that counterattacks intrusion attempts, repel intruders, delay intrusion time, and do not threaten human life. Once someone invades, the system activates the alarm, and the monitoring system also monitors at the same time, and sends the intrusion signal to the security department's monitoring equipment to ensure that management personnel can understand the situation in the alarm area in a timely manner and deal with it quickly. The more common application of video surveillance systems is to collect and analyze visual data in key fire monitoring areas, and perform fire alarm linkage operations based on the analysis results.

[0003] For example, a Chinese invention patent with publication number CN106385558A proposes a method and system for video surveillance and fire alarm linkage, including collecting alarm signals, and sending a first information containing the fire alarm signal to a linkage server; collecting video images that meet the early warning conditions, and sending a second information that meets the video alarm conditions to the linkage server; the linkage server makes a judgment, if the fire alarm signal and the video alarm signal are for the same alarm area, and the alarm conditions are met at the same time, then the corresponding linkage action is performed according to the preset linkage rules. This application uses the fire alarm and video surveillance intelligent analysis alarm joint alarm, and has the links of mutual verification and review of the alarm situation by fire and video surveillance, first complementing and confirming each other, and then sending the linkage system, making corresponding linkage actions, and automatically linking to reduce the false alarm rate, improve the accuracy of the alarm, improve the degree of system automation, and improve the efficiency of alarm handling, so as to save costs for customers and ensure the safety of the monitored targets.

[0004] For example, a video surveillance and fire alarm linkage system and method thereof proposed in the Chinese invention patent with publication number CN111669488A includes a fire detection module for detecting the occurrence of a fire and transmitting fire warning data to a linkage control module; a video surveillance module for monitoring the fire scene and transmitting fire video data to the linkage control module, and accurately determining the location of the fire; a fire alarm for sounding a fire alarm when a fire occurs; a linkage control module for analyzing and processing the fire warning data, comparing the fire warning data with a first preset value, and if the fire warning data is greater than or equal to the preset value, controlling the video surveillance module and the fire extinguishing module to operate, and sending the fire warning data and the fire video data to a monitoring center; and a monitoring center for receiving the fire warning data and the fire video data transmitted by the linkage control module to generate an alarm signal, and sending the alarm signal to the fire alarm. The dual purposes of video surveillance and fire alarm are achieved.

[0005] Obviously, the above-mentioned prior art is limited to judging whether a fire has occurred in the fire monitoring area based on the analysis results of the picture content collected in real time by the video monitoring system, and then issuing a fire alarm when a fire is determined to have occurred. This operation mode has a certain lag, and both the transmission of fire information and the response of the remote fire department need to consume a certain amount of time, which is likely to cause immeasurable losses to the scene. It can be seen that the prior art lacks a predictive fire information judgment mechanism that can predict the time of fire based on the current picture content of the video monitoring system, thereby helping the fire department to extract and respond, and effectively ensure the safety of equipment and personnel at the monitoring site. Summary of the invention

[0006] In order to solve the technical problems in the prior art, the present invention provides a video surveillance and fire alarm linkage system and method, which can use an intelligent fire analysis model to intelligently judge the specific time of a subsequent fire based on the area of ​​the fire monitoring area and a plurality of targeted screened visual data of the panoramic picture of the fire monitoring area at the current moment, and wirelessly transmit the specific time of the intelligent analysis as the fire alarm linkage information of the fire monitoring area to a remote fire management service network element, thereby realizing distributed and intelligent fire alarm linkage operation of various fire monitoring areas.

[0007] According to one aspect of the present invention, a video surveillance and fire alarm linkage system is provided, the system comprising:

[0008] A first capturing mechanism is used to obtain a panoramic image of a fire monitoring area, and output a Y component value, a U component value, and a V component value of each foreground pixel point of the panoramic image, as well as a depth value and a coordinate value of each foreground pixel point of the panoramic image as multiple foreground information of the panoramic image;

[0009] The second capturing mechanism is used to obtain a panoramic image of the fire monitoring area, and output the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image;

[0010] A step-by-step training mechanism, used for performing a set number of trainings on the BP neural network to obtain the BP neural network after each training, and outputting the BP neural network after each training as an intelligent fire analysis model, wherein the value of the set number is positively correlated with the area occupied by the fire monitoring area and is positively correlated with the number of pixels of the panoramic image;

[0011] a fire analysis mechanism, connected to the first capture mechanism, the second capture mechanism and the step-by-step training mechanism respectively, for inputting the area of ​​the fire monitoring area, the multiple foreground information of the panoramic image and the various content data of the panoramic image in parallel into the intelligent fire analysis model, running the intelligent fire analysis model to obtain the interval time from the shooting time corresponding to the panoramic image to the occurrence time of the fire in the fire monitoring area output by the intelligent fire analysis model, and outputting it as the fire warning time;

[0012] A wireless transmission mechanism, connected to the fire analysis mechanism, for wirelessly transmitting the fire warning duration as the fire alarm linkage information of the fire monitoring area to a remote fire management service network element;

[0013] Among them, the overall grayscale value, the overall depth of field value and the number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image are output as various content data of the panoramic image, including: determining whether the pixel point is an edge pixel point based on the pixel value gradient of each pixel point of the panoramic image, and fitting each edge pixel point in the panoramic image to obtain each target imaging area occupied by each target in the panoramic image.

[0014] According to another aspect of the present invention, a method for video surveillance and fire alarm linkage is provided, the method comprising:

[0015] Acquire a panoramic image of the fire monitoring area, and output the Y component value, U component value and V component value of each foreground pixel point of the panoramic image and the depth of field value and coordinate value of each foreground pixel point of the panoramic image as multiple foreground information of the panoramic image;

[0016] Obtain a panoramic image of the fire monitoring area, and output the overall grayscale value, overall depth of field value, and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image;

[0017] Performing a set number of trainings on the BP neural network to obtain the BP neural network after each training, and outputting the BP neural network after each training as an intelligent fire analysis model, wherein the value of the set number is positively correlated with the area of ​​the fire monitoring area and the number of pixels of the panoramic image;

[0018] Inputting the area of ​​the fire monitoring area, multiple foreground information of the panoramic image and various content data of the panoramic image into the intelligent fire analysis model in parallel, running the intelligent fire analysis model to obtain the interval time from the shooting time corresponding to the panoramic image to the occurrence time of the fire in the fire monitoring area output by the intelligent fire analysis model, and outputting it as the fire warning time;

[0019] Wirelessly transmit the fire warning duration as the fire alarm linkage information of the fire monitoring area to a remote fire management service network element;

[0020] Among them, the overall grayscale value, the overall depth of field value and the number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image are output as various content data of the panoramic image, including: determining whether the pixel point is an edge pixel point based on the pixel value gradient of each pixel point of the panoramic image, and fitting each edge pixel point in the panoramic image to obtain each target imaging area occupied by each target in the panoramic image.

[0021] It can be seen that the present invention has at least the following significant technical advances:

[0022] Firstly: the intelligent fire analysis model is used to intelligently analyze the specific time of subsequent fire occurrence based on the panoramic image collection content of the set fire monitoring area at the current moment, and the specific time of the intelligent analysis is used as the fire alarm linkage information of the set fire monitoring area and wirelessly transmitted to the remote fire management service network element, thereby realizing the distributed and intelligent fire alarm linkage operation of various fire monitoring areas, and improving the safety of various fire monitoring areas;

[0023] Secondly: the BP neural network after each training is used as the intelligent fire analysis model to perform intelligent analysis of the later fire occurrence time based on the visual collection content. The number of BP neural network training times is positively correlated with the area of ​​the set fire monitoring area and the number of pixels in the panoramic image, thereby completing the structural customization of the intelligent fire analysis model;

[0024] Again: in each training of the BP neural network, the area of ​​a certain fire monitoring area, multiple foreground information of a panoramic image corresponding to a certain fire monitoring area, and various content data of a panoramic image corresponding to a certain fire monitoring area are used as various input contents of the BP neural network, and the interval time from the shooting time corresponding to the panoramic image corresponding to a certain fire monitoring area to the time when a fire occurs in a certain fire monitoring area is used as a single output content of the BP neural network to complete this training action. The targeted design of the above training actions ensures the training effect of each training;

[0025] Finally: the various content data of the panoramic image used for intelligent analysis at the moment of fire include the Y component value, U component value and V component value of each foreground pixel point of the panoramic image, the depth of field value and coordinate value of each foreground pixel point of the panoramic image, the overall grayscale value, the overall depth of field value and the number of occupied pixels of each target with an area ratio exceeding the limit in the panoramic image, thereby providing comprehensive and sufficient basic information for intelligent analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein:

[0027] Figure 1 It is a technical flow chart of the video surveillance and fire alarm linkage system and method according to the present invention.

[0028] Figure 2 The internal structure diagram of the video surveillance and fire alarm linkage system according to the first embodiment of the present invention is shown.

[0029] Figure 3 The internal structure diagram of the video surveillance and fire alarm linkage system according to the second embodiment of the present invention is shown.

[0030] Figure 4 1 is an internal structure diagram of a video surveillance and fire alarm linkage system according to a third embodiment of the present invention.

[0031] Figure 5 1 is an internal structure diagram of a video surveillance and fire alarm linkage system according to a fourth embodiment of the present invention.

[0032] Figure 6 1 is an internal structure diagram of a video surveillance and fire alarm linkage system according to a fifth embodiment of the present invention.

[0033] Figure 7 The present invention is a flowchart showing the steps of a method for linking video surveillance and fire alarm according to a sixth embodiment of the present invention. DETAILED DESCRIPTION

[0034] like Figure 1 As shown, a technical flow chart of the video surveillance and fire alarm linkage system and method according to the present invention is given.

[0035] like Figure 1 As shown, the specific technical process of the present invention is as follows:

[0036] Step 1: Targeted screening of multiple basic data for intelligent analysis of the specific moment of subsequent fire. The comprehensive and sufficient screening of the above multiple basic data ensures the stability and reliability of the intelligent analysis results;

[0037] Specifically, the above-mentioned multiple basic data include the area occupied by the fire monitoring area, multiple foreground information of the panoramic image taken at the current moment of the fire monitoring area, and various content data of the panoramic image;

[0038] More specifically, the multiple foreground information of the panoramic image includes a Y component value, a U component value and a V component value of each foreground pixel point of the panoramic image, and a depth value and a coordinate value of each foreground pixel point of the panoramic image, and the various content data of the panoramic image include an overall grayscale value, an overall depth value and a number of occupied pixels of each target whose occupied area ratio exceeds a limit in the panoramic image;

[0039] Step 2: Design an intelligent fire analysis model for performing intelligent analysis of the specific moment of subsequent fire occurrence;

[0040] Specifically, the intelligent fire analysis model is a BP neural network after each training. The number of training times of the BP neural network is positively correlated with the area occupied by the fire monitoring area and the number of pixels of the panoramic image, thereby completing the structural customization of the intelligent fire analysis model.

[0041] And specifically, in each training performed on the BP neural network, the area of ​​a certain fire monitoring area, multiple foreground information of a panoramic image corresponding to a certain fire monitoring area, and various content data of a panoramic image corresponding to a certain fire monitoring area are used as various input contents of the BP neural network, and the interval time from the shooting time corresponding to the panoramic image corresponding to a certain fire monitoring area to the occurrence time of a fire in a certain fire monitoring area is used as a single output content of the BP neural network to complete this training action. The targeted design of the above training actions ensures the training effect of each training;

[0042] Step 3: Using the intelligent fire analysis model designed in the second step to intelligently analyze the interval time from the shooting time corresponding to the panoramic image to the time when the fire occurred in the fire monitoring area based on the multiple basic data selected in the first step, and output it as the fire warning time;

[0043] Step 4: wirelessly transmit the fire warning duration obtained by the intelligent analysis in the third step as the fire alarm linkage information of the fire monitoring area to a remote fire management service network element;

[0044] In this way, the fire management service network element can obtain the fire alarm linkage information sent by various fire monitoring areas, and only make necessary responses to the fire monitoring areas with short fire warning duration, thereby ensuring the full utilization of limited fire alarm linkage resources.

[0045] The key points of the present invention are: distributed management of fire alarm linkage in various fire monitoring areas, targeted screening of multiple basic data for intelligent analysis of the specific moment of subsequent fire occurrence, and customized structural design of an artificial intelligence model for intelligent analysis of the specific moment of subsequent fire occurrence.

[0046] The video surveillance and fire alarm linkage system and method of the present invention will be specifically described below by way of embodiments.

[0047] First embodiment

[0048] Figure 2 The internal structure diagram of the video surveillance and fire alarm linkage system according to the first embodiment of the present invention is shown.

[0049] like Figure 2 As shown, the video surveillance and fire alarm linkage system includes the following components:

[0050] A first capturing mechanism is used to obtain a panoramic image of a fire monitoring area, and output a Y component value, a U component value, and a V component value of each foreground pixel point of the panoramic image, as well as a depth value and a coordinate value of each foreground pixel point of the panoramic image as multiple foreground information of the panoramic image;

[0051] For example, the resolution of the acquired panoramic image of the fire monitoring area is fixed, that is, the number of pixel rows and the number of pixel columns of the acquired panoramic image of the fire monitoring area are both fixed values;

[0052] The second capturing mechanism is used to obtain a panoramic image of the fire monitoring area, and output the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image;

[0053] Specifically, each target in the panoramic image refers to each object corresponding to each block image area distribution except the background area in the picture, such as an electrical meter box, an instrument or a pipeline;

[0054] A step-by-step training mechanism, used for performing a set number of trainings on the BP neural network to obtain the BP neural network after each training, and outputting the BP neural network after each training as an intelligent fire analysis model, wherein the value of the set number is positively correlated with the area occupied by the fire monitoring area and is positively correlated with the number of pixels of the panoramic image;

[0055] For example, a set number of trainings are performed on the BP neural network to obtain the BP neural network after each training, and the BP neural network after each training is output as an intelligent fire analysis model, and the value of the set number is positively correlated with the floor area of ​​the fire monitoring area and positively correlated with the number of pixels of the panoramic image, including: when the floor area of ​​the fire monitoring area remains unchanged, the more the number of pixels of the panoramic image is, the larger the value of the set number is; and when the number of pixels of the panoramic image remains unchanged, the larger the floor area of ​​the fire monitoring area is, the larger the value of the set number is;

[0056] In this way, corresponding intelligent fire analysis models can be customized for different fire monitoring areas and panoramic images with different resolutions;

[0057] a fire analysis mechanism, connected to the first capture mechanism, the second capture mechanism and the step-by-step training mechanism respectively, for inputting the area of ​​the fire monitoring area, the multiple foreground information of the panoramic image and the various content data of the panoramic image in parallel into the intelligent fire analysis model, running the intelligent fire analysis model to obtain the interval time from the shooting time corresponding to the panoramic image to the occurrence time of the fire in the fire monitoring area output by the intelligent fire analysis model, and outputting it as the fire warning time;

[0058] A wireless transmission mechanism, connected to the fire analysis mechanism, for wirelessly transmitting the fire warning duration as the fire alarm linkage information of the fire monitoring area to a remote fire management service network element;

[0059] Specifically, wirelessly transmitting the fire warning duration as the fire alarm linkage information of the fire monitoring area to the remote fire management service network element includes: wirelessly transmitting the fire warning duration as the fire alarm linkage information of the fire monitoring area to the remote fire management service network element through a time division duplex communication link or a frequency division duplex communication link;

[0060] Wherein, outputting the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image comprises: determining whether the pixel point is an edge pixel point based on the pixel value gradient of each pixel point of the panoramic image, performing fitting processing on each edge pixel point in the panoramic image to obtain each target imaging area respectively occupied by each target in the panoramic image;

[0061] The BP neural network is trained for a set number of times to obtain the BP neural network after each training, and the BP neural network after each training is output as an intelligent fire analysis model, and the value of the set number is positively correlated with the area of ​​the fire monitoring area and the number of pixels of the panoramic image, including: in each training performed on the BP neural network, the area of ​​a certain fire monitoring area, multiple foreground information of the panoramic image corresponding to the certain fire monitoring area, and various content data of the panoramic image corresponding to the certain fire monitoring area are used as various input contents of the BP neural network, and the interval time from the shooting time corresponding to the panoramic image corresponding to the certain fire monitoring area to the occurrence time of the fire in the certain fire monitoring area is used as the single output content of the BP neural network to complete this training action;

[0062] The BP neural network is trained for a set number of times to obtain the BP neural network after each training, and the BP neural network after each training is output as an intelligent fire analysis model, wherein the value of the set number is positively correlated with the area of ​​the fire monitoring area and the number of pixels of the panoramic image, and further comprises: using a double-input single-output function to represent the numerical mapping relationship that the value of the set number is positively correlated with the area of ​​the fire monitoring area and the number of pixels of the panoramic image;

[0063] The use of a double-input single-output function to represent a numerical mapping relationship in which the value of the set number is positively correlated with the area occupied by the fire monitoring area and is positively correlated with the number of pixels in the panoramic image includes: in the double-input single-output function, the value of the set number is a single output information, and the area occupied by the fire monitoring area and the number of pixels in the panoramic image are two input information;

[0064] Wherein, outputting the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image further comprises: for each target in the panoramic image, taking the arithmetic mean of each grayscale value corresponding to each pixel point of the target imaging area occupied by the target as its overall grayscale value;

[0065] Wherein, outputting the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image further comprises: for each target in the panoramic image, taking the arithmetic mean of each depth of field value corresponding to each pixel point of the target imaging area occupied by the target as its overall depth of field value;

[0066] And wherein, the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio in the panoramic image exceeds the limit are output as various content data of the panoramic image, which also includes: for each target in the panoramic image, the total number of each pixel point of the target imaging area occupied by it is taken as its number of occupied pixels.

[0067] Second embodiment

[0068] Figure 3 The internal structure diagram of the video surveillance and fire alarm linkage system according to the second embodiment of the present invention is shown.

[0069] like Figure 3 As shown, compared with Figure 2 , the video surveillance and fire alarm linkage system also includes:

[0070] A panoramic imaging mechanism, connected to the first capturing mechanism and the second capturing mechanism respectively, and used to send the captured panoramic image of the fire monitoring area to the first capturing mechanism and the second capturing mechanism respectively;

[0071] By way of example, the panoramic imaging mechanism is respectively connected to the first capturing mechanism and the second capturing mechanism, and is used to send the captured panoramic images of the fire monitoring area to the first capturing mechanism and the second capturing mechanism, respectively, including: the panoramic imaging mechanism has a built-in panoramic lens, a filter and an image sensor, and the image sensor is a CCD sensor or a CMOS sensor.

[0072] Third embodiment

[0073] Figure 4 1 is an internal structure diagram of a video surveillance and fire alarm linkage system according to a third embodiment of the present invention.

[0074] like Figure 4 As shown, compared with Figure 2 , the video surveillance and fire alarm linkage system also includes:

[0075] A content storage mechanism, connected to the step-by-step training mechanism, for completing the model storage of the intelligent fire analysis model by storing various model parameters of the intelligent fire analysis model;

[0076] For example, a content storage mechanism is connected to the step-by-step training mechanism, and is used to complete the model storage of the intelligent fire analysis model by storing various model parameters of the intelligent fire analysis model, including: selecting a CF storage chip or a static storage chip to implement the content storage mechanism, and is used to complete the model storage of the intelligent fire analysis model by storing various model parameters of the intelligent fire analysis model.

[0077] Fourth embodiment

[0078] Figure 5 1 is an internal structure diagram of a video surveillance and fire alarm linkage system according to a fourth embodiment of the present invention.

[0079] like Figure 5 As shown, compared with Figure 2 , the video surveillance and fire alarm linkage system also includes:

[0080] A real-time display mechanism, connected to the fire analysis mechanism, for displaying the fire warning duration in real time as the fire alarm linkage information of the fire monitoring area when the fire warning duration is received;

[0081] Specifically, the real-time display mechanism is connected to the fire analysis mechanism and is used to display the fire warning duration as the fire alarm linkage information of the fire monitoring area in real time upon receiving the fire warning duration, including: the real-time display mechanism is a giant screen display mechanism.

[0082] Fifth embodiment

[0083] Figure 6 1 is an internal structure diagram of a video surveillance and fire alarm linkage system according to a fifth embodiment of the present invention.

[0084] like Figure 6 As shown, compared with Figure 2 , the video surveillance and fire alarm linkage system also includes:

[0085] An on-site warning mechanism, connected to the fire analysis mechanism, for executing an on-site fire warning operation related to the fire monitoring area when the received fire warning duration is less than or equal to a set duration threshold;

[0086] Wherein, the on-site warning mechanism is connected to the fire analysis mechanism and is used to perform on-site fire warning operations related to the fire monitoring area when the received fire warning duration is less than or equal to the set duration threshold, including: the on-site warning mechanism is an acoustic warning mechanism or an optical warning mechanism;

[0087] Specifically, the on-site warning mechanism is connected to the fire analysis mechanism and is used to perform on-site fire warning operations related to the fire monitoring area when the received fire warning duration is less than or equal to the set duration threshold, and also includes: the acoustic warning mechanism or the optical warning mechanism adopts an acoustic broadcast mode or an optical display mode to complete the on-site fire warning operation related to the fire monitoring area.

[0088] Next, various embodiments of the present invention will be further described.

[0089] In each of the above embodiments, optionally, in the video surveillance and fire alarm linkage system:

[0090] Acquiring a panoramic image of a fire monitoring area, and outputting the Y component value, U component value and V component value of each foreground pixel point of the panoramic image, as well as the depth of field value and coordinate value of each foreground pixel point of the panoramic image as multiple foreground information of the panoramic image, including: identifying the background imaging area of ​​the fire monitoring area based on the imaging features of the background wall of the fire monitoring area, and using the panoramic image after removing the background imaging area as the foreground imaging area of ​​the panoramic image, wherein the foreground imaging area of ​​the panoramic image is composed of each foreground pixel point of the panoramic image;

[0091] Likewise, in the panoramic image, the background imaging area of ​​the panoramic image is composed of various background pixel points of the panoramic image;

[0092] Wherein, obtaining a panoramic image of the fire monitoring area, and outputting the Y component value, U component value and V component value of each foreground pixel point of the panoramic image and the depth of field value and coordinate value of each foreground pixel point of the panoramic image as multiple foreground information of the panoramic image further comprises: the value range of any one of the Y component value, U component value and V component value of each foreground pixel point of the panoramic image is between 0 and 255;

[0093] And in each of the above embodiments, optionally, in the video surveillance and fire alarm linkage system:

[0094] Inputting the area of ​​the fire monitoring area, multiple foreground information of the panoramic image, and various content data of the panoramic image into the intelligent fire analysis model in parallel, running the intelligent fire analysis model to obtain the interval time from the shooting time corresponding to the panoramic image to the occurrence time of the fire in the fire monitoring area output by the intelligent fire analysis model, and outputting it as the fire warning time, including: performing numerical normalization processing on the area of ​​the fire monitoring area, multiple foreground information of the panoramic image, and various content data of the panoramic image, and then inputting them into the intelligent fire analysis model in parallel;

[0095] And wherein, the area of ​​the fire monitoring area, multiple foreground information of the panoramic image and various content data of the panoramic image are input into the intelligent fire analysis model in parallel, and the intelligent fire analysis model is run to obtain the interval duration from the shooting moment corresponding to the panoramic image to the occurrence moment of the fire in the fire monitoring area output by the intelligent fire analysis model, and the output as the fire warning duration also includes: the interval duration from the shooting moment corresponding to the panoramic image to the occurrence moment of the fire in the fire monitoring area output by the intelligent fire analysis model is a numerically normalized representation.

[0096] Sixth embodiment

[0097] Figure 7 The present invention is a flowchart showing the steps of a method for linking video surveillance and fire alarm according to a sixth embodiment of the present invention.

[0098] like Figure 7 As shown, the video surveillance and fire alarm linkage method includes the following steps:

[0099] Step STEP 1: Obtain a panoramic image of the fire monitoring area, and output the Y component value, U component value and V component value of each foreground pixel point of the panoramic image and the depth of field value and coordinate value of each foreground pixel point of the panoramic image as multiple foreground information of the panoramic image;

[0100] For example, the resolution of the acquired panoramic image of the fire monitoring area is fixed, that is, the number of pixel rows and the number of pixel columns of the acquired panoramic image of the fire monitoring area are both fixed values;

[0101] Step STEP 2: Obtain a panoramic image of the fire monitoring area, and output the overall grayscale value, overall depth of field value, and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image;

[0102] Specifically, each target in the panoramic image refers to each object corresponding to each block image area distribution except the background area in the picture, such as an electrical meter box, an instrument or a pipeline;

[0103] Step STEP 3: Performing a set number of trainings on the BP neural network to obtain the BP neural network after each training, and outputting the BP neural network after each training as an intelligent fire analysis model, wherein the value of the set number is positively correlated with the area occupied by the fire monitoring area and the number of pixels of the panoramic image;

[0104] For example, a set number of trainings are performed on the BP neural network to obtain the BP neural network after each training, and the BP neural network after each training is output as an intelligent fire analysis model, and the value of the set number is positively correlated with the floor area of ​​the fire monitoring area and positively correlated with the number of pixels of the panoramic image, including: when the floor area of ​​the fire monitoring area remains unchanged, the more the number of pixels of the panoramic image is, the larger the value of the set number is; and when the number of pixels of the panoramic image remains unchanged, the larger the floor area of ​​the fire monitoring area is, the larger the value of the set number is;

[0105] In this way, corresponding intelligent fire analysis models can be customized for different fire monitoring areas and panoramic images with different resolutions;

[0106] Step STEP 4: Inputting the area of ​​the fire monitoring area, multiple foreground information of the panoramic image and various content data of the panoramic image into the intelligent fire analysis model in parallel, running the intelligent fire analysis model to obtain the interval time from the shooting time corresponding to the panoramic image to the occurrence time of the fire in the fire monitoring area output by the intelligent fire analysis model, and outputting it as the fire warning time;

[0107] Step STEP 5: wirelessly transmitting the fire warning duration as the fire alarm linkage information of the fire monitoring area to a remote fire management service network element;

[0108] Specifically, wirelessly transmitting the fire warning duration as the fire alarm linkage information of the fire monitoring area to the remote fire management service network element includes: wirelessly transmitting the fire warning duration as the fire alarm linkage information of the fire monitoring area to the remote fire management service network element through a time division duplex communication link or a frequency division duplex communication link;

[0109] Wherein, outputting the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image comprises: determining whether the pixel point is an edge pixel point based on the pixel value gradient of each pixel point of the panoramic image, performing fitting processing on each edge pixel point in the panoramic image to obtain each target imaging area respectively occupied by each target in the panoramic image;

[0110] The BP neural network is trained for a set number of times to obtain the BP neural network after each training, and the BP neural network after each training is output as an intelligent fire analysis model, and the value of the set number is positively correlated with the area of ​​the fire monitoring area and the number of pixels of the panoramic image, including: in each training performed on the BP neural network, the area of ​​a certain fire monitoring area, multiple foreground information of the panoramic image corresponding to the certain fire monitoring area, and various content data of the panoramic image corresponding to the certain fire monitoring area are used as various input contents of the BP neural network, and the interval time from the shooting time corresponding to the panoramic image corresponding to the certain fire monitoring area to the occurrence time of the fire in the certain fire monitoring area is used as the single output content of the BP neural network to complete this training action;

[0111] The BP neural network is trained for a set number of times to obtain the BP neural network after each training, and the BP neural network after each training is output as an intelligent fire analysis model, wherein the value of the set number is positively correlated with the area of ​​the fire monitoring area and the number of pixels of the panoramic image, and further comprises: using a double-input single-output function to represent the numerical mapping relationship that the value of the set number is positively correlated with the area of ​​the fire monitoring area and the number of pixels of the panoramic image;

[0112] The use of a double-input single-output function to represent a numerical mapping relationship in which the value of the set number is positively correlated with the area occupied by the fire monitoring area and is positively correlated with the number of pixels in the panoramic image includes: in the double-input single-output function, the value of the set number is a single output information, and the area occupied by the fire monitoring area and the number of pixels in the panoramic image are two input information;

[0113] Wherein, outputting the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image further comprises: for each target in the panoramic image, taking the arithmetic mean of each grayscale value corresponding to each pixel point of the target imaging area occupied by the target as its overall grayscale value;

[0114] Wherein, outputting the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image further comprises: for each target in the panoramic image, taking the arithmetic mean of each depth of field value corresponding to each pixel point of the target imaging area occupied by the target as its overall depth of field value;

[0115] And wherein, the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio in the panoramic image exceeds the limit are output as various content data of the panoramic image, which also includes: for each target in the panoramic image, the total number of each pixel point of the target imaging area occupied by it is taken as its number of occupied pixels.

[0116] In addition, in the video surveillance and fire alarm linkage system and method according to the present invention:

[0117] For each target in the panoramic image, taking the arithmetic mean of the grayscale values ​​corresponding to the pixels in the target imaging area occupied by the target as its overall grayscale value includes: for each target in the panoramic image, taking the arithmetic mean of the remaining multiple grayscale values ​​obtained after removing the maximum value and the minimum value of the grayscale values ​​corresponding to the pixels in the target imaging area occupied by the target as its overall grayscale value;

[0118] Specifically, for each target in the panoramic image, after removing the maximum and minimum values ​​of the grayscale values ​​corresponding to the pixels of the target imaging area occupied by the target, the arithmetic mean of the remaining multiple grayscale values ​​is obtained as the overall grayscale value thereof, including: the grayscale value corresponding to each pixel is between 0 and 255;

[0119] And wherein, for each target in the panoramic image, taking the arithmetic mean of each depth of field value corresponding to each pixel point of the target imaging area occupied by it as its overall depth of field value includes: for each target in the panoramic image, removing the maximum value and the minimum value of each depth of field value corresponding to each pixel point of the target imaging area occupied by it, and obtaining the arithmetic mean of the remaining multiple depth of field values ​​as its overall depth of field value.

[0120] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0121] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the relevant technology, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A video surveillance and fire alarm linkage system, characterized in that: The system comprises: A first capturing mechanism is used to obtain a panoramic image of a fire monitoring area, and output a Y component value, a U component value, and a V component value of each foreground pixel point of the panoramic image, as well as a depth value and a coordinate value of each foreground pixel point of the panoramic image as multiple foreground information of the panoramic image; The second capturing mechanism is used to obtain a panoramic image of the fire monitoring area, and output the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image; A step-by-step training mechanism, used for performing a set number of trainings on the BP neural network to obtain the BP neural network after each training, and outputting the BP neural network after each training as an intelligent fire analysis model, wherein the value of the set number is positively correlated with the area occupied by the fire monitoring area and is positively correlated with the number of pixels of the panoramic image; a fire analysis mechanism, connected to the first capture mechanism, the second capture mechanism and the step-by-step training mechanism respectively, for inputting the area of ​​the fire monitoring area, the multiple foreground information of the panoramic image and the various content data of the panoramic image in parallel into the intelligent fire analysis model, running the intelligent fire analysis model to obtain the interval time from the shooting time corresponding to the panoramic image to the occurrence time of the fire in the fire monitoring area output by the intelligent fire analysis model, and outputting it as the fire warning time; A wireless transmission mechanism, connected to the fire analysis mechanism, for wirelessly transmitting the fire warning duration as the fire alarm linkage information of the fire monitoring area to a remote fire management service network element; Wherein, outputting the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image comprises: determining whether the pixel point is an edge pixel point based on the pixel value gradient of each pixel point of the panoramic image, performing fitting processing on each edge pixel point in the panoramic image to obtain each target imaging area respectively occupied by each target in the panoramic image; Wherein, for each target in the panoramic image, the total number of pixels in the target imaging area occupied by the target is taken as the number of occupied pixels.

2. The video surveillance and fire alarm linkage system according to claim 1, characterized in that: Performing a set number of trainings on the BP neural network to obtain the BP neural network after each training, and outputting the BP neural network after each training as an intelligent fire analysis model, wherein the value of the set number is positively correlated with the area occupied by the fire monitoring area and the number of pixels of the panoramic image, including: in each training performed on the BP neural network, taking the area occupied by a certain fire monitoring area, multiple foreground information of the panoramic image corresponding to the certain fire monitoring area, and various content data of the panoramic image corresponding to the certain fire monitoring area as various input contents of the BP neural network, and taking the interval time from the shooting time corresponding to the panoramic image corresponding to the certain fire monitoring area to the occurrence time of the fire in the certain fire monitoring area as the single output content of the BP neural network to complete this training action; The BP neural network is trained for a set number of times to obtain the BP neural network after each training, and the BP neural network after each training is output as an intelligent fire analysis model, wherein the value of the set number is positively correlated with the area of ​​the fire monitoring area and the number of pixels of the panoramic image, and further comprises: using a double-input single-output function to represent the numerical mapping relationship that the value of the set number is positively correlated with the area of ​​the fire monitoring area and the number of pixels of the panoramic image; Among them, the use of a double-input single-output function to represent the numerical mapping relationship in which the value of the set number is positively correlated with the area occupied by the fire monitoring area and is positively correlated with the number of pixels of the panoramic image includes: in the double-input single-output function, the value of the set number is a single output information, and the area occupied by the fire monitoring area and the number of pixels of the panoramic image are two input information.

3. The video surveillance and fire alarm linkage system according to claim 2, characterized in that: Outputting the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image further includes: for each target in the panoramic image, taking the arithmetic mean of each grayscale value corresponding to each pixel point of the target imaging area occupied by the target as its overall grayscale value; Among them, outputting the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image also includes: for each target in the panoramic image, taking the arithmetic mean of each depth of field value corresponding to each pixel point of the target imaging area occupied by it as its overall depth of field value.

4. The video surveillance and fire alarm linkage system according to claim 3, characterized in that: The system further comprises: The panoramic imaging mechanism is connected to the first capturing mechanism and the second capturing mechanism respectively, and is used to send the captured panoramic images of the fire monitoring area to the first capturing mechanism and the second capturing mechanism respectively.

5. The video surveillance and fire alarm linkage system according to claim 3, characterized in that: The system further comprises: The content storage mechanism is connected to the step-by-step training mechanism and is used to complete the model storage of the intelligent fire analysis model by storing various model parameters of the intelligent fire analysis model.

6. The video surveillance and fire alarm linkage system according to claim 3, characterized in that: The system further comprises: The real-time display mechanism is connected to the fire analysis mechanism and is used to display the fire warning duration as the fire alarm linkage information of the fire monitoring area in real time when the fire warning duration is received.

7. The video surveillance and fire alarm linkage system according to claim 3, characterized in that: The system further comprises: An on-site warning mechanism, connected to the fire analysis mechanism, for executing an on-site fire warning operation related to the fire monitoring area when the received fire warning duration is less than or equal to a set duration threshold; Among them, the on-site warning mechanism is connected to the fire analysis mechanism, and is used to perform on-site fire warning operations related to the fire monitoring area when the received fire warning duration is less than or equal to the set duration threshold, including: the on-site warning mechanism is an acoustic warning mechanism or an optical warning mechanism.

8. The video surveillance and fire alarm linkage system according to claim 7, characterized in that: Acquiring a panoramic image of a fire monitoring area, and outputting the Y component value, U component value and V component value of each foreground pixel point of the panoramic image, as well as the depth of field value and coordinate value of each foreground pixel point of the panoramic image as multiple foreground information of the panoramic image, including: identifying the background imaging area of ​​the fire monitoring area based on the imaging features of the background wall of the fire monitoring area, and using the panoramic image after removing the background imaging area as the foreground imaging area of ​​the panoramic image, wherein the foreground imaging area of ​​the panoramic image is composed of each foreground pixel point of the panoramic image; Among them, obtaining a panoramic image of the fire monitoring area, and outputting the Y component value, U component value and V component value of each foreground pixel point of the panoramic image and the depth of field value and coordinate value of each foreground pixel point of the panoramic image as multiple foreground information of the panoramic image also includes: the value range of any component value among the Y component value, U component value and V component value of each foreground pixel point of the panoramic image is between 0-255.

9. The video surveillance and fire alarm linkage system according to claim 7, characterized in that: Inputting the area of ​​the fire monitoring area, multiple foreground information of the panoramic image, and various content data of the panoramic image into the intelligent fire analysis model in parallel, running the intelligent fire analysis model to obtain the interval time from the shooting time corresponding to the panoramic image to the occurrence time of the fire in the fire monitoring area output by the intelligent fire analysis model, and outputting it as the fire warning time, including: performing numerical normalization processing on the area of ​​the fire monitoring area, multiple foreground information of the panoramic image, and various content data of the panoramic image, and then inputting them into the intelligent fire analysis model in parallel; Among them, the area occupied by the fire monitoring area, multiple foreground information of the panoramic image and various content data of the panoramic image are input into the intelligent fire analysis model in parallel, and the intelligent fire analysis model is run to obtain the interval duration from the shooting time corresponding to the panoramic image to the occurrence time of the fire in the fire monitoring area output by the intelligent fire analysis model, and the output as the fire warning duration also includes: the interval duration from the shooting time corresponding to the panoramic image to the occurrence time of the fire in the fire monitoring area output by the intelligent fire analysis model is a numerically normalized representation.

10. A method for linking video surveillance and fire alarm, characterized in that: The method comprises: Acquire a panoramic image of the fire monitoring area, and output the Y component value, U component value and V component value of each foreground pixel point of the panoramic image and the depth of field value and coordinate value of each foreground pixel point of the panoramic image as multiple foreground information of the panoramic image; Obtain a panoramic image of the fire monitoring area, and output the overall grayscale value, overall depth of field value, and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image; Performing a set number of trainings on the BP neural network to obtain the BP neural network after each training, and outputting the BP neural network after each training as an intelligent fire analysis model, wherein the value of the set number is positively correlated with the area of ​​the fire monitoring area and the number of pixels of the panoramic image; Inputting the area of ​​the fire monitoring area, multiple foreground information of the panoramic image and various content data of the panoramic image into the intelligent fire analysis model in parallel, running the intelligent fire analysis model to obtain the interval time from the shooting time corresponding to the panoramic image to the occurrence time of the fire in the fire monitoring area output by the intelligent fire analysis model, and outputting it as the fire warning time; Wirelessly transmit the fire warning duration as the fire alarm linkage information of the fire monitoring area to a remote fire management service network element; Wherein, outputting the overall grayscale value, overall depth of field value and number of occupied pixels of each target whose occupied area ratio exceeds the limit in the panoramic image as various content data of the panoramic image comprises: determining whether the pixel point is an edge pixel point based on the pixel value gradient of each pixel point of the panoramic image, performing fitting processing on each edge pixel point in the panoramic image to obtain each target imaging area respectively occupied by each target in the panoramic image; Wherein, for each target in the panoramic image, the total number of pixels in the target imaging area occupied by the target is taken as the number of occupied pixels.

Images