Integrated bridge site flood real-time automatic identification and alarm system and method

Abstract

The invention discloses an integrated bridge site flood real-time automatic identification and alarm system and method. The integrated bridge site flood real-time automatic identification and alarm system comprises a flood data acquisition and analysis layer, a flood data analysis layer and an alarm information distribution layer, the flood data acquisition and analysis layer is used for acquiring a river channel video stream, and decoding and restoring the river channel video stream to obtain an uncompressed standardized picture stream; the standardized picture stream comprises a plurality of standardized pictures which are sequentially numbered according to time; the flood data analysis layer is used for performing average hue calculation on a plurality of obtained standardized pictures, real-time online flood automatic identification and alarm are realized through the flood data acquisition and analysis layer, the flood data analysis layer and the alarm information distribution layer, the defect that the existing monitoring equipment cannot realize flood monitoring is avoided, and the monitoring efficiency is improved. The bridge site flood on-line monitoring, identification and alarm are really realized, and the technical problem that the bridge site flood monitoring method in the existing bridge safety monitoring technology cannot realize real-time automatic monitoring is solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of bridge safety monitoring, and relates to an automatic identification and alarm system, in particular to a system and method for integrated real-time automatic identification and alarm of bridge site flood. BACKGROUND

[0002] River-crossing bridges are the key nodes of road traffic, and account for a large proportion of mountainous highway bridges and are an important part of mountainous highway bridges. Flood disasters often pose a serious threat to the safety of bridges. Traditional flood monitoring methods, such as manual inspection and fixed-point water level monitoring, have problems such as low monitoring efficiency, poor real-time performance, and incomplete data acquisition, and are difficult to accurately determine the flood conditions at the bridge site and issue early warnings at the first time of flood occurrence. River-crossing bridge disasters caused by floods have sounded the alarm for the safe operation and maintenance of river-crossing bridges. Monitoring the flood conditions of river-crossing bridges and timely alarming bridge management personnel when large floods occur on river-crossing bridges are important needs to ensure the safe operation and maintenance of river-crossing bridges.

[0003] At present, bridge monitoring mainly relies on bridge health monitoring systems. The sensors of the health monitoring system are divided into two categories: the first category is structural response sensors such as acceleration sensors and displacement sensors, which are used to collect the responses of the structure under different environments, extract structural state evolution indicators from the structural responses through data processing methods, and then realize structural state evaluation. The second category is environmental sensors such as wind speed and direction sensors and temperature sensors, which are used to collect bridge site environmental data and provide information for traffic control and manual work through real-time monitoring of environmental changes. These two types of monitoring hardware are large in number and have two requirements in actual application: (1) the bridge site must be powered to provide power for sensors and data acquisition devices; (2) the bridge site must have a network to provide a transmission channel for the large amount of data collected by the sensors and transmit the data to the monitoring center. The above two types of sensors cannot realize bridge site flood monitoring, and for many river-crossing bridges, the two requirements of power supply and network supply are basically difficult to meet. Therefore, the existing monitoring system cannot realize the monitoring, identification and alarm of bridge site floods.

[0004] The basic feature of bridge monitoring is real-time online, and flood identification also needs to be realized in real time, that is, flood data collection, data analysis and alarm are realized at the same time, which requires the flood monitoring system to realize the full automation of data analysis, data analysis and data distribution, and the monitoring system must be independent of city power supply and network, realizing self-power supply, data self-processing and information self-sending. SUMMARY

[0005] In view of the deficiencies of the prior art, the purpose of the present application is to provide a system and method for integrated real-time automatic identification and alarm of bridge site flood, so as to solve the technical problem that the bridge site flood monitoring method in the existing bridge safety monitoring technology cannot realize real-time automatic monitoring.

[0006] In order to solve the above technical problems, the present application adopts the following technical solutions:

[0007] The system for integrated real-time automatic identification and alarm of bridge site flood comprises a flood data acquisition and analysis layer, a flood data analysis layer and an alarm information distribution layer.

[0008] The flood data acquisition and analysis layer is used to acquire a river video stream, decode and restore the river video stream, and obtain an uncompressed standardized picture stream; the standardized picture stream comprises a plurality of standardized pictures numbered in time sequence.

[0009] The flood data analysis layer is used to respectively calculate the average hue of the obtained plurality of standardized pictures, and sequentially judge whether flood occurs according to the calculation results; if flood occurs, generate an alarm information, and transmit the alarm information to the alarm information distribution layer; the alarm information comprises flood prompt information and flood pictures.

[0010] The alarm information distribution layer is used to transmit the alarm information to an external cloud platform.

[0011] Further, the method for integrated real-time automatic identification and alarm of bridge site flood is based on the system for integrated real-time automatic identification and alarm of bridge site flood, and comprises the following steps:

[0012] Step one, a camera is used to acquire a river video stream, decode and restore the river video stream, and obtain an uncompressed standardized picture stream; the standardized picture stream comprises a plurality of standardized pictures numbered in time sequence.

[0013] Step two, respectively calculate the average hue of the plurality of standardized pictures obtained in step one, and sequentially judge whether flood occurs according to the calculation results; if flood occurs, generate an alarm information, and transmit the alarm information to the alarm information distribution layer; the alarm information comprises flood prompt information and flood pictures.

[0014] Step three, transmit the alarm information obtained in step two to an external cloud platform.

[0015] The present application further comprises the following technical features:

[0016] Step two specifically comprises the following steps:

[0017] Q1, sequentially read the standardized pictures obtained in step one according to the numbering, determine the starting point coordinates (x, y) of the cropping area according to the position of the river area on the plurality of standardized pictures, and crop the river area from the standardized pictures to obtain a plurality of river pictures; Q2, calculate the average hue of the plurality of river pictures, and sequentially judge whether flood occurs according to the calculation results; if flood occurs, generate an alarm information, and transmit the alarm information to the alarm information distribution layer; the alarm information comprises flood prompt information and flood pictures.s ,y s ) and the width d and the height h of the clipping region;

[0018] Q2, clipping the uncompressed standard image according to the starting point coordinates (x s ,y s ), the width d and the height h of the clipping region to obtain a river region picture; the river region picture is a rectangle formed by the starting point coordinates (x s ,y s ) and the end point coordinates (x e ,y e );

[0019] x e =x s +d

[0020] y e =y s +h

[0021] Q3, converting the river region picture obtained by Q2 to HSV space, extracting the hue H value of the HSV space, calculating the average hue value, judging whether the average hue value is less than the alarm threshold, if yes, triggering the flood alarm, generating an alarm information, transmitting the alarm information to the alarm information distribution layer, if not, deleting the uncompressed standard image.

[0022] In Q3, the alarm threshold is 0.1.

[0023] The communication device is used for transmitting the alarm information to an external cloud platform.

[0024] Compared with the prior art, the present application has the beneficial technical effects that:

[0025] (I) In the present application, through the flood data collection and analysis layer, the flood data analysis layer and the alarm information distribution layer, real-time online flood automatic identification and alarm are realized, the shortcomings that the existing monitoring equipment cannot realize flood monitoring are avoided, and real-time online bridge site flood monitoring, identification and alarm are truly realized, and the technical problem that the bridge site flood monitoring method in the existing bridge safety monitoring technology cannot realize real-time automatic monitoring is solved.

[0026] (II) In the present application, the real-time online integrated bridge site flood automatic identification and alarm edge computing method is deployed on the real-time online integrated bridge site flood automatic identification and alarm edge computing device, hardware and software integration is realized, data collection, analysis and analysis integration are realized, the dependence on power supply and network of the traditional monitoring system is eliminated, and the flood monitoring system is suitable for any bridge site environment. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 is a system overall architecture diagram of the present application;

[0028] Figure 2 The data flow conversion architecture diagram of the application.

[0029] The specific content of the application is further explained in detail in combination with the embodiments. DETAILED DESCRIPTION

[0030] It should be noted that all parts in the application, unless otherwise specified, use parts known in the art.

[0031] The specific embodiments of the application are given below, it should be noted that the application is not limited to the following specific embodiments, any equivalent transformation made on the basis of the technical scheme of the application falls within the protection scope of the application.

[0032] The application provides an integrated bridge site flood real-time automatic identification and alarm system, including a flood data acquisition and analysis layer, a flood data analysis layer, and an alarm information distribution layer.

[0033] The flood data acquisition and analysis layer is used to collect river video streams, decode and restore the river video streams to obtain uncompressed standardized picture streams; the standardized picture streams include a plurality of standardized pictures numbered in time sequence.

[0034] The flood data analysis layer is used to calculate the average hue of the obtained plurality of standardized pictures respectively, and sequentially judge whether a flood occurs according to the calculation results; if a flood occurs, generate an alarm information, and transmit the alarm information to the alarm information distribution layer; the alarm information includes flood prompt information and flood pictures.

[0035] The alarm information distribution layer is used to transmit the alarm information to an external cloud platform.

[0036] In the above technical scheme, through the flood data acquisition and analysis layer, the flood data analysis layer, and the alarm information distribution layer, real-time online flood automatic identification and alarm are realized, the shortcomings that the existing monitoring equipment cannot realize flood monitoring are avoided, and online monitoring, identification and alarm of bridge site flood are truly realized, solving the technical problem that the existing bridge safety monitoring technology cannot realize real-time automatic monitoring.

[0037] The flood data acquisition and analysis layer uses the decoding function in the FFmpeg open source library to decode according to the encoding format of the video, parses the video stream into one frame of image, and stores it in RGB format. Using a bilinear interpolation image scaling algorithm, the new position and value of each pixel point are calculated according to the target size to make the image adapt to the size requirements of standardization; the parsed standard picture is sent to an edge computing device and stored locally according to time.

[0038] The method for real-time automatic identification and alarm of integrated bridge site flood comprises the following steps:

[0039] Step one, a camera is used to collect a river channel video stream, the river channel video stream is decoded and restored to obtain an uncompressed standardized picture stream; the standardized picture stream comprises a plurality of standardized pictures numbered in sequence according to time;

[0040] Step two, average hue calculation is respectively performed on the plurality of standardized pictures obtained in step one, and whether flood is generated is sequentially judged according to the calculation results; if flood is generated, alarm information is generated, and the alarm information is transmitted to an alarm information distribution layer; the alarm information comprises flood prompt information and a flood picture;

[0041] Step three, the alarm information obtained in step two is transmitted to an external cloud platform.

[0042] In the above technical solution, the real-time online integrated bridge site flood automatic identification and alarm edge computing method is deployed on the system composed of a camera device, an edge computing device and a communication device, hardware and software integration is realized, data collection, analysis and integration are realized, the dependence of a traditional monitoring system on power supply and network is eliminated, and the flood monitoring system is applicable to any bridge site environment.

[0043] The present application also comprises the following technical features:

[0044] Step two specifically comprises the following steps:

[0045] Q1, the standardized pictures obtained in step one are sequentially read according to the numbers, and the starting point coordinates (x s ,y s ) of a cutting region, the width d and the height h of the cutting region are determined according to the positions of the river channel regions on the plurality of standardized pictures;

[0046] Q2, the starting point coordinates (x s ,y s ), the width d and the height h of the cutting region obtained in Q1 are used to automatically cut the uncompressed standard image to obtain a river channel region picture; the river channel region picture is a rectangle formed by the starting point coordinates (x s ,y s ) and the end point coordinates (x e ,y e );

[0047] x e =x s +d

[0048] y e =y s +h

[0049] Q3, converting the river region picture obtained in Q2 to HSV space, extracting hue H value in HSV space, calculating average hue value, judging whether the average hue value is less than an alarm threshold, if yes, triggering a flood alarm, generating an alarm information, transmitting the alarm information to an alarm information distribution layer, and if not, deleting the uncompressed standard image.

[0050] In Q3, the alarm threshold is 0.1.

[0051] In the above technical solution, the average value of hue is generally used to determine the flood alarm level. The color of normal water flow is light green, and the H value distribution range of the corresponding picture is greater than 0.2. The color of flood is yellow, and the H value range of the corresponding picture is less than 0.1.

Claims

1. An integrated bridge site flood real-time automatic identification and alarm system, characterized by: It includes flood data collection and parsing layer, flood data analysis layer, and alarm information distribution layer; The flood data collection and analysis layer is used to collect river video streams, decode and restore the river video streams, and obtain uncompressed standardized image streams; the standardized image streams include multiple standardized images numbered in chronological order; The flood data analysis layer is used to calculate the average hue of the obtained multiple standardized images, and determine whether a flood has occurred based on the calculation results; if a flood has occurred, an alarm message is generated and transmitted to the alarm information distribution layer; the alarm message includes flood prompt information and flood images; The alarm information distribution layer is used to transmit the alarm information to the external cloud platform.

2. An integrated bridge site flood real-time automatic identification and alarm method, characterized in that: The integrated bridge site flood real-time automatic identification and alarm system according to claim 1 comprises the following steps: Step 1: using a camera to capture a river video stream, decoding and restoring the river video stream to obtain an uncompressed standardized image stream; the standardized image stream includes a plurality of standardized images numbered in chronological order; Step 2: Calculate the average hue of the multiple standardized images obtained in step 1, and determine whether a flood has occurred based on the calculation results; if a flood has occurred, generate an alarm message, and transmit the alarm message to the alarm information distribution layer; the alarm message includes flood prompt information and flood images; Step 3: Transmit the alarm information obtained in step 2 to the external cloud platform.

3. The integrated bridge site flood real-time automatic identification and alarm method according to claim 2 is characterized in that: Step 2 specifically includes the following steps: Q1, read the standardized images obtained in step 1 in sequence according to the number, and determine the starting coordinates (x s ,y s ) and the width d and height h of the cropping area; Q2, the starting coordinates of the clipping area obtained according to Q1 (x s ,y s ), width d and height h are used to automatically crop the uncompressed standard image to obtain a river area picture; the river area picture is the starting point coordinate (x s ,y s ) and the end point coordinates (x e ,y e ) formed by the rectangle; x e =x s +d y e =y s +h Q3 converts the river area image obtained in Q2 into the HSV space, extracts the hue H value of the HSV space, and calculates the average hue value to determine whether the average hue value is less than the alarm threshold. If so, a flood alarm is triggered, an alarm message is generated, and the alarm message is transmitted to the alarm information distribution layer. If not, the uncompressed standard image is deleted.

4. The integrated bridge site flood real-time automatic identification and alarm method according to claim 3 is characterized in that: In Q3, the alarm threshold is selected as 0.1.

Images