Port space safety monitoring and early warning method and device

Abstract

The present invention discloses a port space safety monitoring and early warning method and device, the method includes substituting the first element image and the second element image of the target area into the safety detection model, and then judging whether there is a safety hazard. The present application connects various areas through the Internet of Things, installs color cameras and infrared thermal cameras for each area that needs to be monitored, and realizes safety monitoring of the fire field by changing the temperature and image of the target area. When the image and / or temperature of the target area is abnormal, an alarm is promptly issued to the user to remind the user to take targeted measures, thereby improving the automation of the Internet of Things technology in the field of port security and improving port safety.

Description

Technical Field

[0001] The present invention relates to the technical field of port space operation safety monitoring and surveillance, and in particular to a port space safety monitoring and early warning method and device. Background Art

[0002] With the continuous acceleration of urbanization and the reconstruction of old cities, the application of integrated pipeline corridors in modern urban construction projects is becoming more and more extensive, including the laying and operation of many 220 kV and 110 kV power cables in integrated pipeline corridors and tunnels. As a limited space, the integrated pipeline corridor has narrow entrances and exits and is relatively isolated from the outside environment. It is easy to cause toxic and harmful substances due to poor natural ventilation. When carrying out construction work in the confined space of the port, it is necessary to ensure that the internal environmental quality and gas are in normal state. Although general monitoring equipment can monitor the internal area in many aspects, it is impossible to notify multiple departments to deal with it in time when a fire accident occurs, and it is impossible to guarantee the safety of personnel to the greatest extent. Safety hazards are prone to occur during construction work.

[0003] With the development of technology, most port facilities have image monitoring conditions. Modern security monitoring and data acquisition and transmission technologies based on the Internet of Things have greatly facilitated the security monitoring of port facilities. However, monitoring technology can only obtain images, but cannot effectively process images and cannot effectively determine whether there are security risks at each node. Summary of the invention

[0004] In order to make up for the above-mentioned deficiencies in port confined space security monitoring and improve the automated application of monitoring images in the field of port security monitoring, the present application proposes a port space security monitoring and early warning method and device.

[0005] The technical solution includes: a port space security monitoring and early warning method, which substitutes the first element image and the second element image of the target area into the security detection model to determine whether there are safety hazards, including: a security detection model construction step and an image substitution judgment step,

[0006] The safety detection model building step includes:

[0007] The collection frequency is set, and the collection range is 0-24 hours of a natural day. The historical first element image set and the historical second element image set are respectively obtained from the historical monitoring images collected by the first element collector and the second element collector preset in the target area, and the collection angle and collection range of the first element collector and the second element collector are consistent;

[0008] Extracting first characteristic factors of each image in the historical first element image set in chronological order, and constructing a first element model according to the first characteristic factors, wherein the first element model is used to express the relationship between the first characteristic factors in the first element image set and changes over time;

[0009] Extracting the second characteristic factors of each image in the historical second element image set in chronological order, and constructing a second element model according to the second characteristic factors, wherein the second element model is used to express the relationship between the second characteristic factors in the second element image set and changes over time;

[0010] The first element model is coupled with the second element model to form a safety detection model of the target area;

[0011] The image substitution judgment step comprises:

[0012] The first characteristic factor and the second characteristic factor of the newly collected first element image and the second element image are extracted respectively, and the first characteristic factor, the second characteristic factor and the corresponding time period are substituted into the safety detection model to determine whether the substitution result is within the normal threshold. If so, it is judged to be normal, otherwise an alarm is issued.

[0013] Furthermore, the first element model is coupled with the second element model to form a safety detection model for the target area, and the process includes: taking the time point as X, the first characteristic factor of the corresponding time as the Y value, and the second characteristic factor of the corresponding time as the Z value, constructing a three-dimensional sweeping path, sweeping with a preset value of the corresponding time as the sweeping radius to form a three-dimensional solid model, taking the surface and inner side of the three-dimensional solid model as the normal threshold, and taking the outer side of the three-dimensional solid model as the abnormal threshold.

[0014] Furthermore, the first element image is a color image, and the second element image is an infrared thermal image.

[0015] Furthermore, the first characteristic factor is a positive square root of the grayscale values ​​of the upwind point, the downwind point, the center point and the device point within the acquisition range of the first element image.

[0016] Furthermore, the second characteristic factor is the temperature value of the upwind point, the downwind point, the center point and the equipment point within the acquisition range of the second element image.

[0017] Furthermore, the preset value corresponding to the time is a value f(t) that changes with time, and the change rule of f(t) is:

[0018] When the acquisition time t is greater than 14 o'clock, f(t)=24-|1-t|;

[0019] When the acquisition time t is less than or equal to 14 o'clock, f(t)=t;

[0020] Wherein, t is the collection time under the 24-hour counting method of the natural day.

[0021] Furthermore, the safety monitoring model also includes a self-iteration process, which includes: counting the first characteristic factor and the second characteristic factor of each collection time node of the natural day without an alarm and accumulating the statistics to the last day of the ten-day period, and calculating the average values ​​of the first characteristic factor and the second characteristic factor of all the natural days without an alarm in the ten-day period on the last day of the ten-day period, using the average value of the first characteristic factor as the first iteration value, and using the average value of the second characteristic factor as the second iteration value, and replacing the Y value and Z value of the corresponding time point with the first special iteration factor and the second iteration factor respectively to realize iteration.

[0022] A safety monitoring and early warning device, which adopts any one of the above methods to perform safety monitoring and early warning.

[0023] The present application establishes a safety detection model by monitoring images and makes corresponding judgments on the temperature changes and image changes in the monitored area. Generally speaking, as a closed or generally uninhabited area, the target area has corresponding ranges and regularities for image and temperature changes. Once the temperature and / or image changes exceed the range, there may be safety hazards, such as fire, line disconnection, dust mixing, etc. Therefore, the present application monitors the temperature and image grayscale to determine whether there are changes exceeding the threshold in the target area, thereby improving the safety of the monitored area.

[0024] This application connects the monitoring images of various areas, installs color cameras and infrared thermal cameras for each area that needs to be monitored, and realizes safety monitoring of the fire field by monitoring the temperature changes and image changes in the target area. When the image and / or temperature of the target area is abnormal, an alarm is promptly issued to the user to remind the user to take targeted measures, thereby improving the automation of Internet of Things technology in the field of port security and improving port safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of a construction process of a safety detection model in one embodiment of the method of the present invention;

[0026] Figure 2 The corresponding relationship between the first characteristic factor and time in an embodiment of the method of the present invention;

[0027] Figure 3 The corresponding relationship between the second characteristic factor and time in one embodiment of the method of the present invention;

[0028] Figure 4 A schematic diagram of a safety detection model in an embodiment of the method of the present invention;

[0029] Figure 5Schematic diagram of an abnormal situation in one embodiment of the method of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0032] Reference Figure 1, an implementation method of the present application. A port space security monitoring and early warning method of the present application pre-connects the first element collector and the second element collector of the area to be monitored with the user background through the Internet of Things, and the first element collector and the second element collector obtain the first element image and the second element image respectively. When the method of the present application is implemented, the first element image and the second element image of the target area are substituted into the safety detection model to determine whether there are safety hazards. This embodiment uses an image comparison method to perform safety monitoring, which is mainly aimed at corresponding areas such as electric wires and pipelines that are closed or have less external interference. The image elements in the above-mentioned areas are relatively simple, with less external interference. In most cases, the changes in image elements are relatively regular or change within a specific threshold range. In this embodiment, the image comparison relationship of the same time period is obtained by establishing a dual-image feature factor of the corresponding area, and different types of images are compared to determine whether there are safety hazards in the view. It includes: a safety detection model construction step and an image substitution judgment step. The safety detection model construction step includes: setting the acquisition frequency, taking 0-24 hours of a natural day as the acquisition range, and obtaining the historical first element image set and the historical second element image set from the historical monitoring images collected by the preset first element collector and the second element collector of the target area, respectively. The acquisition angle and acquisition range of the first element collector and the second element collector are consistent. In this implementation step, generally, 0-24 hours of a natural day is taken as a collection cycle. Within a collection cycle, the collection frequency is set according to monitoring needs. Generally, it is 5-15 times per minute according to the fire risk or the burning speed of the corresponding material, that is, the interval interval is set to 4-12 seconds for collection to avoid missing the fire source or drastic temperature changes. At the same time, it is not necessary to collect every second to reduce the information transmission bandwidth. The areas that need to be monitored in this application are generally divided into dozens or even dozens. If the acquisition frequency is too high, it will have higher requirements on the transmission channel. At this stage, the Internet of Things transmission of ports has corresponding tasks, and it is impossible to match too many loans. In addition, too many collected images put a lot of pressure on background processing. Reference Figure 2 and Figure 3 , extract the first characteristic factor of each image in the historical first element image set in chronological order, and construct a first element model based on the first characteristic factor, wherein the first element model is used to express the relationship between the first characteristic factor in the first element image set and the time change; extract the second characteristic factor of each image in the historical second element image set in chronological order, and construct a second element model based on the second characteristic factor, wherein the second element model is used to express the relationship between the second characteristic factor in the second element image set and the time change; couple the first element model with the second element model to form a safety detection model for the target area (such as Figure 4); In this embodiment, the safety detection model generally includes a three-dimensional function consisting of the first characteristic factor and the second characteristic factor of the corresponding time node and the corresponding threshold value. In the later stage, the first characteristic factor and the second characteristic factor of the new image are compared to determine whether it is normal. The image substitution judgment step includes: extracting the first characteristic factor and the second characteristic factor of the newly collected first element image and the second element image respectively, substituting the first characteristic factor, the second characteristic factor and the corresponding time period into the safety detection model, and judging whether the substitution result is within the normal threshold value. If so, it is judged to be normal, otherwise, an alarm is issued. Reference Figure 5 , one of the abnormal situations judged by the security detection model. Figure 5 In the example, the normal threshold is inside the security detection model, and there are no isolated points outside. Figure 5 The presence of isolated points in the data indicates that the point value exceeds the normal threshold and an alarm is required.

[0033] On the basis of the above embodiment, further, the first element model is coupled with the second element model to form a safety detection model of the target area, including: taking the time point as X, the first characteristic factor of the corresponding time as the Y value, and the second characteristic factor of the corresponding time as the Z value, constructing a three-dimensional sweeping path, sweeping with the preset value of the corresponding time as the sweeping radius to form a three-dimensional solid model, taking the surface and inner side of the three-dimensional solid model as the normal threshold, and taking the outer side of the three-dimensional solid model as the abnormal threshold. In this embodiment, constructing a three-dimensional swept solid model visualizes the data value, and the normal range value on the surface and inner side of the three-dimensional solid model more quickly and intuitively shows the suspicious direction of the abnormal situation, which is convenient for users to judge. At the same time, after having three coordinates and time nodes as reference coordinates, the three-dimensional model is convenient and fast to build, saving development time.

[0034] On the basis of one or more of the above embodiments, further, the first element image is a color image, and the second element image is an infrared thermal image. In this embodiment, the color image and the infrared thermal grayscale image are used as references, the color image is used to determine whether there is external interference, and the infrared thermal image is used to determine the temperature change trend, which is convenient for users to eliminate false alarms after abnormal alarms in the later stage.

[0035] On the basis of one or more of the above embodiments, further, the first characteristic factor is the positive square root of the grayscale value of the upwind point, downwind point, center point and equipment point within the acquisition range of the first element image. In this embodiment, the image is collected based on the upwind point and downwind point that are most likely to change, the center point that is most easily affected, and the equipment point of the high-risk object to improve the monitoring efficiency. When the device is a cable, the general equipment point is the cable trisection point of the acquisition range. Since the grayscale value range is 0-255, if the direct grayscale value is directly used as the Y-axis, the difference between the temperature value (generally -10-70) under normal circumstances during the modeling process is too large, resulting in the slope of the path curve of the three-dimensional model being too large, which is prone to false alarms in the later stage. Therefore, this embodiment adopts the grayscale value square root method to try to control the grayscale value within a range that is not much different from the temperature value range.

[0036] On the basis of one or more of the above embodiments, further, the second characteristic factor is the temperature value of the upwind point, downwind point, center point and equipment point within the acquisition range of the second element image. In this embodiment, the image is acquired at the upwind point and downwind point where temperature changes are most likely to occur, the center point that is most easily affected by temperature, and the equipment point of the high-risk object to improve the monitoring efficiency. When the device is a cable, the general equipment point is the point where the cable is divided into three equal parts within the acquisition range.

[0037] In the above two acquisition points, due to the limitation of computing power, it is impossible to extract and calculate the entire image. The above two embodiments both adopt point extraction to save computing power.

[0038] On the basis of one or more of the above embodiments, further, the preset value corresponding to the time is a value f(t) that changes with time, and the change rule of f(t) is:

[0039] When the acquisition time t is greater than 14 o'clock, f(t)=24-|1-t|;

[0040] When the acquisition time t is less than or equal to 14 o'clock, f(t)=t;

[0041] Among them, t is the collection time under the 24-hour counting method of the natural day. In this application, most of my country's ports are in the eastern region, and Beijing time is used for timing. Therefore, the daily temperature change pattern follows the highest between 14:00 and 15:00, and the temperature changes relatively quickly before 14:00 and after 15:00. Therefore, the application is verified and judged through experiments and in the model, with 14:00 as the node and as the maximum threshold range, and 24:00-2:00 in the evening as the threshold relatively small range, to maximize the high temperature change range during the day and the detection accuracy of the low temperature area at night.

[0042] Based on the above embodiments, Figure 2 and Figure 3 In one implementation flow of the present application, the first element image processing process includes obtaining a grayscale image of the first element image, collecting grayscale blocks of upwind points, downwind points, center points and equipment points within the range of the first element image according to the first characteristic factor focus point, and the grayscale block of each collection point is generally n*n pixels (3<n≤10). After obtaining the grayscale block, the average grayscale value of the pixels in the grayscale block is used as the value of the first characteristic factor of the focus point, and finally the grayscale value of each focus point is obtained, and the grayscale value of each point is used as the first characteristic factor. The second characteristic factor is the temperature average value of the temperature block of the upwind point, downwind point, center point and equipment point (the temperature block is generally m*m pixels, 2<m≤5)). The first element image and the second element image are collected simultaneously at the same time, thereby obtaining each first characteristic factor and the second characteristic factor corresponding to the same time point and the same focus point. On the basis of having the same corresponding X-coordinate axis (time node), a path curve in three-dimensional space is constructed with the first characteristic factor as the Y value and the second characteristic factor as the Z value, referring to Figure 4 , and then use the preset value as the sweep radius to complete the modeling of the safety detection model.

[0043] On the basis of one or more of the above embodiments, further, the safety monitoring model also includes a self-iteration process, which includes: counting the first characteristic factor and the second characteristic factor of each collection time node of the natural day without alarm and accumulating the statistics to the last day of the ten-day period, and calculating the average value of the first characteristic factor and the second characteristic factor of all the days without alarm in the ten-day period on the last day of the ten-day period, taking the average value of the first characteristic factor as the first iteration value, and taking the average value of the second characteristic factor as the second iteration value, and replacing the Y value and Z value of the corresponding time point with the first special iteration factor and the second iteration factor to achieve iteration. In this embodiment, although the temperature or climate of the port does not change much every day under normal weather, especially the confined space is not sensitive to external weather and temperature changes, its internal temperature will still change with the change of seasons. In order to avoid the interference of seasonal temperature and humidity changes on the safety detection model of this community, this application uses the corresponding characteristic factors collected in each natural ten-day period to iterate in a ten-day iteration manner to ensure that the model of this application can be detected normally regardless of spring, summer, autumn and winter.

[0044] The present application also proposes a safety monitoring and early warning device, which uses the method described in any one of the above embodiments to perform safety monitoring and early warning.

[0045] The construction and arrangement of the present application shown in a plurality of different exemplary embodiments are only exemplary. Although only a few embodiments are described in detail in this disclosure, it should be readily understood by those who refer to this disclosure that many modifications are possible (e.g., the size, scale, structure, shape and proportion of various elements, and parameter values ​​(e.g., temperature, pressure, etc.), mounting arrangements, use of materials, color, directional changes, etc.) without substantially departing from the novel teachings and advantages of the subject matter described in the application. For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of the element may be inverted or otherwise changed, and the nature or number or position of the discrete element may be changed or changed. Therefore, all such modifications are intended to be included within the scope of the present invention. The order or sequence of any process or method steps may be changed or reordered according to an alternative embodiment. In the claims, any "device plus function" clause is intended to cover the structure described herein that performs the function, and is not only structurally equivalent but also equivalent structure. Other substitutions, modifications, changes and omissions may be made in the design, operating conditions and arrangement of the exemplary embodiments without departing from the scope of the present invention. Therefore, the invention is not limited to a specific embodiment, but extends to numerous modifications still falling within the scope of the appended claims.

[0046] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment may not be described (ie, those features that are not relevant to the best mode presently contemplated for carrying out the invention or those features that are not relevant to implementing the invention).

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the essence and scope of the technical solution of the present invention.

Claims

1. A port space security monitoring and early warning method, characterized in that: Substituting the first element image and the second element image of the target area into the safety monitoring model, and then judging whether there is a safety hazard, which includes: a safety monitoring model construction step and an image substitution judgment step, The safety monitoring model building step includes: The collection frequency is set, and the collection range is 0-24 hours of a natural day. The historical first element image set and the historical second element image set are respectively obtained from the historical monitoring images collected by the first element collector and the second element collector preset in the target area, and the collection angle and collection range of the first element collector and the second element collector are consistent; Extracting first characteristic factors of each image in the historical first element image set in chronological order, and constructing a first element model according to the first characteristic factors, wherein the first element model is used to express the relationship between the first characteristic factors in the first element image set and changes over time; Extracting the second characteristic factors of each image in the historical second element image set in chronological order, and constructing a second element model according to the second characteristic factors, wherein the second element model is used to express the relationship between the second characteristic factors in the second element image set and changes over time; The first element model is coupled with the second element model to form a safety monitoring model of the target area; The process of coupling the first element model with the second element model to form a safety monitoring model of the target area includes: taking the time point as X, the first characteristic factor corresponding to the time as the Y value, and the second characteristic factor corresponding to the time as the Z value, constructing a three-dimensional sweeping path, sweeping with a preset value corresponding to the time as the sweeping radius to form a three-dimensional solid model, taking the surface and the inner side of the three-dimensional solid model as the normal threshold, and taking the outer side of the three-dimensional solid model as the abnormal threshold; The image substitution judgment step comprises: The first characteristic factor and the second characteristic factor of the newly collected first element image and the second element image are extracted respectively, and the first characteristic factor, the second characteristic factor and the corresponding time period are substituted into the safety monitoring model to determine whether the substitution result is within the normal threshold. If so, it is judged to be normal, otherwise an alarm is issued.

2. The method according to claim 1, characterized in that The first element image is a color image, and the second element image is an infrared thermal image.

3. The method according to claim 2, characterized in that The first characteristic factor is the positive square root of the grayscale values ​​of the upwind point, the downwind point, the center point and the device point within the acquisition range of the first element image.

4. The method according to claim 2, characterized in that: The second characteristic factor is the temperature value of the upwind point, the downwind point, the center point and the equipment point within the acquisition range of the second element image.

5. The method according to claim 1, characterized in that The preset value of the corresponding time is a value f(t) that changes with time, and the change rule of f(t) is: When the acquisition time t is greater than 14 o'clock, f(t)=24-|1-t|; When the acquisition time t is less than or equal to 14 o'clock, f(t)=t; Wherein, t is the collection time under the 24-hour counting method of the natural day.

6. The method according to claim 1, characterized in that The safety monitoring model also includes a self-iteration process, which includes: counting the first characteristic factor and the second characteristic factor of each collection time node of the natural day without an alarm and accumulating the statistics to the last day of the ten-day period, and calculating the average values ​​of the first characteristic factor and the second characteristic factor of all the natural days without an alarm in the ten-day period on the last day of the ten-day period, using the average value of the first characteristic factor as the first iteration value, and using the average value of the second characteristic factor as the second iteration value, and replacing the Y value and Z value of the corresponding time point with the first special iteration factor and the second iteration factor respectively to realize iteration.

7. A safety monitoring and early warning device, characterized in that: The device adopts the method described in any one of claims 1 to 6 to perform safety monitoring and early warning.

Images