Green land monitoring and alarming system based on edge computing

Abstract

The application provides a green land monitoring alarm system based on edge computing. The system comprises: a camera configured to monitor and photograph a green land to obtain a monitoring video; an edge computing system connected with the camera through a short-distance network, configured to analyze whether a preset event occurs on the green land according to the monitoring video; when the preset event occurs on the green land, sending a reminder to a reminding device; wherein the preset event comprises a trampling event and / or an article leaving event; and the reminding device is configured to perform a reminding operation when receiving the reminder from the edge computing system.

Description

Technical Field

[0001] This invention relates to the field of monitoring and alarm technology, and in particular to a green space monitoring and alarm system based on edge computing. Background Technology

[0002] Currently, the only way to control visitors from entering green areas within parks is by setting up directional signs. When there are large numbers of visitors, park staff monitor them and constantly remind them, but this method is very resource-intensive and has limited effectiveness.

[0003] Therefore, there is an urgent need to design a technology that can intelligently monitor and alert visitors' behavior on green spaces. Summary of the Invention

[0004] This invention provides a green space monitoring and alarm system based on edge computing.

[0005] This invention provides a green space monitoring and alarm system based on edge computing, comprising:

[0006] Cameras are used to monitor and capture images of green spaces to obtain surveillance video.

[0007] An edge computing system, having a short-range network connection with the camera, is used to analyze the surveillance video to determine if a preset event has occurred on the green area; when the preset event occurs on the green area, an alert is sent to an alert device; wherein the preset event includes trampling events and / or object leaving events.

[0008] The reminder device is used to perform a reminder operation when it receives a reminder from the edge computing system.

[0009] In one embodiment, analyzing whether a preset event has occurred on the green space based on the surveillance video includes:

[0010] Extract video frames from the surveillance video when no living organisms are present in the green area;

[0011] The video frames are analyzed to determine the composition of the green space, which includes green areas where green plants are present and non-green areas where no green plants are present.

[0012] The surveillance video is monitored in real time to determine whether any living organisms appear in the surveillance video.

[0013] When a living organism is present, it is determined whether a trampling event has occurred on the green area based on the size and / or movement trajectory of the organism.

[0014] In one embodiment, determining whether a trampling event has occurred on the green area based on the size and / or movement trajectory of the organism when it is present includes:

[0015] The size of the organism is determined based on the image of the organism in the surveillance video, and the size of the organism includes the height of the organism;

[0016] When the size of the organism is smaller than a preset size, it is determined that no trampling event has occurred on the green area.

[0017] When the size of the organism is equal to or greater than a preset size, the movement trajectory of the organism on the green area is determined. When the movement trajectory passes through the green area, it is determined that a trampling event has occurred on the green area; when the movement trajectory does not pass through the green area, it is determined that no trampling event has occurred on the green area.

[0018] In one embodiment, analyzing whether a preset event has occurred on the green space based on the surveillance video includes:

[0019] The video frames of the monitoring video are analyzed in real time to determine the first video frame in which a living organism appears.

[0020] Based on the image of the organism in the first video frame, the species of the organism is determined, and the species of the organism includes humans and animals;

[0021] When the species of the organism is human, each subsequent video frame after the first video frame in which the organism appears is analyzed frame by frame. For each subsequent video frame, the analysis method includes:

[0022] Compare the current frame with the next frame;

[0023] Determine whether an object that did not appear in the current frame or subsequent video frame appears on the ground within a preset distance range of the creature in the next video frame;

[0024] If so, then continue to determine whether the object still exists on the ground in the next N subsequent video frames; if so, determine that an object leaving event has occurred on the green ground, and determine whether the object has moved by observing the existence status of the object on the ground in the next N subsequent video frames. The reason for the object moving is that the object is light and the wind blows it. If movement occurs, add an object movement marker to the object leaving event log of the organism. The specific steps include:

[0025] Step A1: Using formula (1), compare the current video frame with the next video frame to determine whether an object that did not appear in the current video frame appears on the ground within a preset distance range of the creature in the next video frame:

[0026]

[0027] Where E represents the determination value of whether an object not present in the current frame or subsequent video frame appears on the ground within a preset distance range of the creature in the next video frame; (i,j) represents the coordinates of the i-th row and j-th column position in the video frame; (I,J) represents the position of the creature in the video frame, i.e., the coordinates of the I-th row and J-th column position; L[(i,j),(I,J)] represents the actual distance between the coordinates of the i-th row and j-th column position and the coordinates of the I-th row and J-th column position in the video frame. L0 represents the preset distance; f{} represents the judgment function, which evaluates to 1 if the expression within the parentheses is true and 0 if the expression within the parentheses is false; H0(i,j) represents the pixel value of the pixel at the i-th row and j-th column in the previous frame and subsequent video frames; H1(i,j) represents the pixel value of the pixel at the i-th row and j-th column in the next frame and subsequent video frames; || represents the absolute value; m represents the total number of pixels in any row of the video frame; n represents the total number of pixels in any column of the video frame.

[0028] If E = 0, it means that no object that did not appear in the current frame or subsequent video frame has appeared on the ground within the preset distance range of the creature in the next video frame.

[0029] If E=1, it means that an object that did not appear in the current frame appears on the ground within a preset distance range of the creature in the next video frame.

[0030] Step A2: If an object that did not appear in the current video frame appears on the ground within a preset distance range of the creature in the next video frame, then use formula (2) to determine whether the object has moved by checking the existence status of the object on the ground in the subsequent N video frames:

[0031]

[0032] Where Y represents the determination value of whether the object has moved; N represents N frames; S(a) represents the state value of the object on the ground in the subsequent video frames after frame a.

[0033] If Y = 1, it indicates that the object is moving;

[0034] If Y = 0, it means that the object has not moved.

[0035] Step A3: Using formula (3), add an item movement marker to the item legacy event log of the creature based on the judgment that the object has moved:

[0036] R′ 16 =R 16 >>len(W 16 )<<1+(Y) 16 < <len(W 16 )+W 16 (3)

[0037] Where R′ 16 This represents the hexadecimal form of the log data after adding an item movement identifier to the item legacy event log of the creature; R 16 This represents the hexadecimal form of the item legacy event log for the creature without an item movement identifier; W 16 This represents the hexadecimal format of the end of the log data frame; len() indicates the number of bits in the hexadecimal data within the parentheses; >> indicates right shift; << indicates left shift; (Y) 16 This indicates that the numerical value Y is converted to hexadecimal form.

[0038] The beneficial effects of the above technical solution are as follows: The formula (1) in step A1 is used to compare the current frame with the next frame to determine whether an object that did not appear in the current frame appears on the ground within the preset distance range of the creature in the next frame, thereby making an automatic, intelligent, and reliable judgment to ensure the reliability and accuracy of the judgment; Then, the formula (2) in step A2 is used to determine whether the object has moved by observing the existence status of the object on the ground in the subsequent N frames of video, thereby knowing the status of the object and facilitating the identification of the subsequent log; Finally, the formula (3) in step A3 is used to add an object movement marker to the creature's object leaving event log according to the judgment of the object movement phenomenon, thereby recording the object leaving movement marker situation, which facilitates the purposeful search for the object in the future.

[0039] In one embodiment, the alerting device includes:

[0040] Terminal devices that have a short-range network connection with the camera;

[0041] and / or

[0042] Sound and light reminder devices are installed around the green space.

[0043] In one embodiment, after determining that an object has been left on the green space, the method further includes:

[0044] The system controls all cameras used to monitor the green space to take pictures of the organisms, obtain their biometric information, and send the biometric information, the time of the event where an item was left on the green space, and the geographical location of the green space to the background control center for storage as an item leaving event log.

[0045] In one embodiment, the short-range network connection includes any one or more of Wi-Fi connection, RFID connection, and ZigBee connection.

[0046] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0047] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0048] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0049] Figure 1 This is a schematic diagram of a green space monitoring and alarm system based on edge computing in an embodiment of the present invention. Detailed Implementation

[0050] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0051] This invention provides a green space monitoring and alarm system based on edge computing, such as... Figure 1 As shown, the system includes:

[0052] Cameras are used to monitor and capture images of green spaces to obtain surveillance video.

[0053] An edge computing system, having a short-range network connection with the camera, is used to analyze whether a preset event has occurred on the green area based on the surveillance video; when the preset event occurs on the green area, an alert is sent to an alert device; wherein, the preset event includes trampling events and / or object leaving events; the objects here can be garbage and any other items;

[0054] The reminder device is used to perform a reminder operation when it receives a reminder from the edge computing system.

[0055] The short-range network connection includes one or more of Wi-Fi connection, RFID connection, and ZigBee connection; the edge computing system acquires surveillance video captured by the camera in real time through the short-range network connection.

[0056] The beneficial effects of the above technical solution are as follows: In the above technical solution, setting up an independent edge computing system for each green space can improve the monitoring and alarm efficiency of each green space; in addition, it can intelligently analyze whether a stampede event (referring to an event of trampling on the green space) or an event of leaving objects on the green space has occurred based on the monitoring video of the green space, thereby realizing intelligent monitoring and alarm of the green space, reducing the investment of human monitoring, and improving the monitoring and alarm efficiency.

[0057] In one embodiment, analyzing whether a preset event has occurred on the green space based on the surveillance video includes:

[0058] Extract video frames from the surveillance video when no living organisms are present in the green area;

[0059] The video frames are analyzed to determine the composition of the green space, which includes green areas where green plants are present and non-green areas where no green plants are present.

[0060] The surveillance video is monitored in real time to determine whether any living organisms appear in the surveillance video.

[0061] When a living organism is present, it is determined whether a trampling event has occurred on the green area based on the size and / or movement trajectory of the organism.

[0062] In one embodiment, determining whether a trampling event has occurred on the green area based on the size and / or movement trajectory of the organism when it is present includes:

[0063] The size of the organism is determined based on the image of the organism in the surveillance video, and the size of the organism includes the height of the organism;

[0064] When the size of the organism is smaller than a preset size, it is determined that no trampling event has occurred on the green area; if the size of the organism is small, it means that the organism will not cause too much damage to the green area, and at this time, it can be determined that no trampling event has occurred on the green area.

[0065] When the size of the organism is equal to or greater than a preset size, the movement trajectory of the organism on the green area is determined. When the movement trajectory passes through the green area, it is determined that a trampling event has occurred on the green area; when the movement trajectory does not pass through the green area, it is determined that no trampling event has occurred on the green area.

[0066] The beneficial effects of the above technical solution are: it can intelligently and accurately determine whether a trampling event has occurred on the green area based on the size of the organism and whether the organism's movement trajectory has passed through the green area, thus ensuring the accuracy of the judgment result.

[0067] In one embodiment, analyzing whether a preset event has occurred on the green space based on the surveillance video includes:

[0068] The video frames of the monitoring video are analyzed in real time to determine the first video frame in which a living organism appears.

[0069] Based on the image of the organism in the first video frame, the species of the organism is determined, and the species of the organism includes humans and animals;

[0070] When the species of the organism is human, each subsequent video frame after the first video frame in which the organism appears is analyzed frame by frame. For each subsequent video frame, the analysis method includes:

[0071] Compare the current frame with the next frame;

[0072] Determine whether an object that did not appear in the current frame or subsequent video frame appears on the ground within a preset distance range of the creature in the next video frame;

[0073] If so, then continue to determine whether the object still exists on the ground in the next N subsequent video frames; if so, determine that an object leaving event has occurred on the green ground, and determine whether the object has moved by observing the existence status of the object on the ground in the next N subsequent video frames. The reason for the object moving is that the object is light and the wind blows it. If movement occurs, add an object movement marker to the object leaving event log of the organism. The specific steps include:

[0074] Step A1: Using formula (1), compare the current video frame with the next video frame to determine whether an object that did not appear in the current video frame appears on the ground within a preset distance range of the creature in the next video frame:

[0075]

[0076] Where E represents the determination value of whether an object not present in the current frame or subsequent video frame appears on the ground within a preset distance range of the creature in the next video frame; (i,j) represents the coordinates of the i-th row and j-th column position in the video frame; (I,J) represents the position of the creature in the video frame, i.e., the coordinates of the I-th row and J-th column position; L[(i,j),(I,J)] represents the actual distance between the coordinates of the i-th row and j-th column position and the coordinates of the I-th row and J-th column position in the video frame. L0 represents the preset distance; f{} represents the judgment function, which evaluates to 1 if the expression within the parentheses is true and 0 if the expression within the parentheses is false; H0(i,j) represents the pixel value of the pixel at the i-th row and j-th column in the previous frame and subsequent video frames; H1(i,j) represents the pixel value of the pixel at the i-th row and j-th column in the next frame and subsequent video frames; || represents the absolute value; m represents the total number of pixels in any row of the video frame; n represents the total number of pixels in any column of the video frame.

[0077] If E = 0, it means that no object that did not appear in the current frame or subsequent video frame has appeared on the ground within the preset distance range of the creature in the next video frame.

[0078] If E=1, it means that an object that did not appear in the current frame appears on the ground within a preset distance range of the creature in the next video frame.

[0079] Step A2: If an object that did not appear in the current video frame appears on the ground within a preset distance range of the creature in the next video frame, then use formula (2) to determine whether the object has moved by checking the existence status of the object on the ground in the subsequent N video frames:

[0080]

[0081] Where Y represents the determination value of whether the object has moved; N represents N frames; S(a) represents the state value of the object on the ground in the subsequent video frames after frame a.

[0082] If Y = 1, it indicates that the object is moving;

[0083] If Y = 0, it means that the object has not moved.

[0084] Step A3: Using formula (3), add an item movement marker to the item legacy event log of the creature based on the judgment that the object has moved:

[0085] R′ 16 =R 16 >>len(W 16)<<1+(Y) 16 < <len(W 16 )+W 16 (3)

[0086] Where R′ 16 This represents the hexadecimal form of the log data after adding an item movement identifier to the item legacy event log of the creature; R 16 This represents the hexadecimal form of the item legacy event log for the creature without an item movement identifier; W 16 This represents the hexadecimal format of the end of the log data frame; len() indicates the number of bits in the hexadecimal data within the parentheses; >> indicates right shift; << indicates left shift; (Y) 16 This indicates that the numerical value Y is converted to hexadecimal form.

[0087] The beneficial effects of the above technical solution are as follows: The formula (1) in step A1 is used to compare the current frame with the next frame to determine whether an object that did not appear in the current frame appears on the ground within the preset distance range of the creature in the next frame, thereby making an automatic, intelligent, and reliable judgment to ensure the reliability and accuracy of the judgment; Then, the formula (2) in step A2 is used to determine whether the object has moved by observing the existence status of the object on the ground in the subsequent N frames of video, thereby knowing the status of the object and facilitating the identification of the subsequent log; Finally, the formula (3) in step A3 is used to add an object movement marker to the creature's object leaving event log according to the judgment of the object movement phenomenon, thereby recording the object leaving movement marker situation, which facilitates the purposeful search for the object in the future.

[0088] In one embodiment, the alerting device includes:

[0089] A terminal device with a short-range network connection to the camera; the terminal device can be carried by park management personnel in real time. Because of the short-range network connection with the camera, management personnel who are close to the green space can take timely measures to deal with trampling or leaving objects on the green space, such as reminding tourists, thus improving the efficiency of handling.

[0090] and / or

[0091] The sound and light reminder devices are installed around the green space. These devices can be located at a distance equal to or less than a preset distance from the camera, or they can be specifically designed for green spaces. The sound and light reminder devices can use both sound alerts and flashing lights to enhance the reminder effect. Sound alerts can play reminder messages such as "Please do not step on the green space" or "Please take your belongings with you." This type of sound and light reminder device can more directly remind visitors, improving reminder efficiency.

[0092] In one embodiment, after determining that an object has been left on the green space, the method further includes:

[0093] The system controls all cameras used to monitor the green space to take pictures of the organisms, obtain their biometric information, and send the biometric information, the time of the event where an item was left on the green space, and the geographical location of the green space to the background control center for storage as an item leaving event log.

[0094] The beneficial effects of the above technical solution are: when people leave items on green spaces, the relevant information can be recorded in a timely and intelligent manner, making it easier to find the owner of the left-behind items and improving the efficiency of park management.

[0095] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A green space monitoring and alarm system based on edge computing, characterized in that, include: Cameras are used to monitor and capture images of green spaces to obtain surveillance video. An edge computing system, which has a short-range network connection with the camera, is used to analyze whether a preset event has occurred on the green area based on the surveillance video. When the preset event occurs on the green space, an alert is sent to the alerting device; wherein, the preset event includes a trampling event and / or an object leaving event. The alerting device is used to perform an alert operation when it receives an alert from the edge computing system. The step of analyzing whether a preset event has occurred on the green space based on the surveillance video includes: The video frames of the monitoring video are analyzed in real time to determine the first video frame in which a living organism appears. Based on the image of the organism in the first video frame, the species of the organism is determined, and the species of the organism includes humans and animals; When the species of the organism is human, each subsequent video frame after the first video frame in which the organism appears is analyzed frame by frame. For each subsequent video frame, the analysis method includes: Compare the current frame with the next frame; Determine whether an item that did not appear in the current frame or subsequent video frame has appeared on the ground within a preset distance range of the creature in the next video frame; If so, then continue to determine whether the item still exists on the ground in the next N subsequent video frames; if so, determine that an item was left on the green ground, and determine whether the item has moved by checking the status of the item on the ground in the next N subsequent video frames. The reason for the item moving is that the item is lightweight and the wind blows it. If movement occurs, add an item movement flag to the item's item leaving event log. The specific steps include: Step A1: Using formula (1), compare the current video frame with the next video frame to determine whether an item that did not appear in the current video frame appears on the ground within a preset distance range of the creature in the next video frame: Where E represents the determination value of whether an item not present in the current frame or subsequent video frame appears on the ground within a preset distance range of the creature in the next video frame; (i,j) represents the coordinates of the i-th row and j-th column position in the video frame; (I,J) represents the position of the creature in the video frame, i.e., the coordinates of the I-th row and J-th column position; L[(i,j),(I,J)] represents the actual distance between the coordinates of the i-th row and j-th column position and the coordinates of the I-th row and J-th column position in the video frame. L0 represents the preset distance; f{} represents the judgment function, which evaluates to 1 if the expression within the parentheses is true and 0 if the expression within the parentheses is false; H0(i,j) represents the pixel value of the pixel at the i-th row and j-th column in the previous frame and subsequent video frames; H1(i,j) represents the pixel value of the pixel at the i-th row and j-th column in the next frame and subsequent video frames; || represents the absolute value; m represents the total number of pixels in any row of the video frame; n represents the total number of pixels in any column of the video frame. If E = 0, it means that no item that did not appear in the current frame or subsequent video frame has appeared on the ground within the preset distance range of the creature in the next video frame. If E=1, it means that an item that did not appear in the current frame appears on the ground within a preset distance range of the creature in the next video frame. Step A2: If an object that did not appear in the current video frame appears on the ground within a preset distance range of the creature in the next video frame, then use formula (2) to determine whether the object has moved by checking the presence status of the object on the ground in the following N video frames: Where Y represents the determination value of whether the item has moved; N represents N frames; S(a) represents the state value of the item on the ground in the subsequent video frames after frame a; If Y = 1, it indicates that the item has moved; If Y = 0, it means that the item has not moved. Step A3: Using formula (3), add an item movement marker to the item's item legacy event log of the creature based on the judgment of the item's movement phenomenon: R′ 16 =R 16 >>len(W 16 )<<1+(Y) 16 <<len(W 16 )+W 16 (3) Where R′ 16 This represents the hexadecimal form of the log data after adding an item movement identifier to the item legacy event log of the creature; R 16 This represents the hexadecimal form of the item legacy event log for the creature without an item movement identifier; W 16 This represents the hexadecimal format of the end of the log data frame; len() indicates the number of bits in the hexadecimal data within the parentheses; >> indicates right shift; << indicates left shift; (Y) 16 This indicates that the numerical value Y is converted to hexadecimal form.

2. The system as described in claim 1, characterized in that, The step of analyzing whether a preset event has occurred on the green space based on the surveillance video includes: Extract video frames from the surveillance video when no living organisms are present in the green area; The video frames are analyzed to determine the composition of the green space, which includes green areas where green plants are present and non-green areas where no green plants are present. The surveillance video is monitored in real time to determine whether any living organisms appear in the surveillance video. When a living organism is present, it is determined whether a trampling event has occurred on the green area based on the size and / or movement trajectory of the organism.

3. The system as described in claim 2, characterized in that, When a living organism is present, determining whether a trampling event has occurred on the green area based on the size and / or movement trajectory of the organism includes: The size of the organism is determined based on the image of the organism in the surveillance video, and the size of the organism includes the height of the organism; When the size of the organism is smaller than a preset size, it is determined that no trampling event has occurred on the green area. When the size of the organism is equal to or greater than a preset size, the movement trajectory of the organism on the green area is determined. When the movement trajectory passes through the green area, it is determined that a trampling event has occurred on the green area; when the movement trajectory does not pass through the green area, it is determined that no trampling event has occurred on the green area.

4. The system as described in claim 1, characterized in that... The reminder device includes: Terminal devices that have a short-range network connection with the camera; and / or Sound and light reminder devices are installed around the green space.

5. The system as described in claim 1, characterized in that, After determining that an event of an object being left on the green space has occurred, the method further includes: The system controls all cameras used to monitor the green space to take pictures of the organisms, obtain their biometric information, and send the biometric information, the time of the event where an item was left on the green space, and the geographical location of the green space to the background control center for storage as an item leaving event log.

6. The system as described in claim 1, characterized in that, The short-range network connection includes any one or more of Wi-Fi connection, RFID connection, and ZigBee connection.

Images