A system for detecting drowning incidents based on edge computing

Abstract

The present invention provides a system for detecting drowning incidents based on edge computing, comprising a camera device, an edge computing system and a management terminal, wherein: the camera device and the edge computing system are arranged at preset positions of a target water area; the camera device is used to shoot a video of the target water area to obtain video data of the target water area; the edge computing system is used to determine whether a drowning incident may have occurred in the target water area based on the video data of the target water area; when it is determined that a drowning incident may have occurred in the target water area, a reminder is sent to the management terminal; and the management terminal is used to execute a reminder operation when receiving the reminder.

Description

Technical Field

[0001] The present invention relates to the field of edge computing technology, and in particular to a system for detecting drowning incidents based on edge computing. Background Art

[0002] Generally, roadsides are equipped with roadside cameras to monitor the surrounding conditions of the river, but they are only effective for monitoring. The cameras are not fully functional and cannot identify whether people have fallen into the river and drowned, thus failing to provide a complete safety function. Summary of the Invention

[0003] An embodiment of the present invention provides a system for detecting drowning events based on edge computing.

[0004] An embodiment of the present invention provides a system for detecting drowning events based on edge computing, including a camera device, an edge computing system, and a management terminal, wherein:

[0005] The camera device and the edge computing system are arranged at a preset position in the target water area;

[0006] The camera device is used to capture a video of the target water area to obtain video data of the target water area;

[0007] The edge computing system is configured to determine, based on the video data of the target water area, whether a drowning incident may have occurred in the target water area; and when it is determined that a drowning incident may have occurred in the target water area, send a reminder to the management terminal;

[0008] The management terminal is configured to execute a reminder operation upon receiving the reminder.

[0009] Preferably, the preset position of the target water area includes above the center of the water system of the target water area or above the water bank of the target water area;

[0010] The camera device shoots a video of the target water area from above the center of the water system of the target water area or above the bank of the target water area.

[0011] Preferably, judging whether a drowning incident may have occurred in the target water area based on the video data of the target water area includes:

[0012] Perform image analysis on each video frame in the video data of the target water area in the order of shooting time, from the earliest to the last, to obtain the image content included in each video frame, wherein the following steps B1-B3 are performed on each video frame:

[0013] Step B1: performing a first recognition of the image content of the current video frame, wherein the first recognition is used to determine whether the current video frame includes a water area and a non-water area, wherein the non-water area includes an image area whose pixel color is not the color of water; when the current video frame includes both the water area and the non-water area, proceeding to step B2; when the current video frame includes only the water area but not the non-water area, proceeding to step B3;

[0014] Step B2: When the current video frame includes a water area and a non-water area, identifying the content included in the non-water area in the current video frame, and extracting content feature information in the non-water area in the current video frame; determining whether the content feature information in the non-water area in the current video frame includes human organ features; if so, determining whether a drowning incident may have occurred in the target water area based on the human organ features in the content feature information in the non-water area in the current video frame and the contents of N subsequent video frames after the current video frame; N is a positive integer equal to or greater than 2; if not, returning to step B1 with the next video frame of the current video frame as the new current video frame;

[0015] Step B3: When the current video frame includes only the water area but not the non-water area, the next video frame of the current video frame is used as a new current video frame and the process returns to step B1.

[0016] Preferably, judging whether a drowning incident may have occurred in the target water area based on the human organ features in the content feature information in the non-water area in the current video frame and the contents in N consecutive subsequent video frames after the current video frame includes steps C1-C3:

[0017] Step C1: performing the following steps A1-A3 on each of the N subsequent video frames:

[0018] Step A1: performing a first recognition of the image content of the current subsequent video frame; if it is determined based on the recognition result of the first recognition that the current subsequent video frame includes only the water area and no non-water area, proceeding to step A2; if it is determined based on the recognition result of the first recognition that the current subsequent video frame includes both the water area and the non-water area, proceeding to step A3;

[0019] Step A2: when it is determined according to the recognition result of the first recognition that the current subsequent video frame includes only the water area but not the non-water area, recording the current subsequent video frame as the first valid reference video frame of the current video frame;

[0020] Step A3: when it is determined that the current subsequent video frame includes a water area and a non-water area according to the recognition result of the first recognition, the content included in the non-water area in the current subsequent video frame is identified, and content feature information in the non-water area in the current subsequent video frame is extracted; it is determined whether the content feature information in the non-water area in the current subsequent video frame includes a human organ feature; if so, it is determined whether the distance between the position of the human organ feature in the current subsequent video frame and the position of the human organ feature in the current video frame is equal to or less than a preset distance; if so, the current subsequent video frame is recorded as a second valid reference video frame of the current video frame; if not, the current subsequent video frame is recorded as an invalid reference video frame of the current video frame;

[0021] Step C2: determining whether the number of the first valid reference video frames is equal to or greater than 1, and determining whether the number of the second valid reference video frames is equal to or greater than 1;

[0022] If the number of the first valid reference video frames is equal to or greater than 1, and the number of the second valid reference video frames is equal to or greater than 1, determining that the possibility of a drowning incident occurring in the target water area is a first possibility;

[0023] If the number of the first valid reference video frames is equal to 0 and the number of the second valid reference video frames is equal to or greater than 1, then determining that the possibility of a drowning incident occurring in the target water area is a second possibility;

[0024] If the number of the first valid reference video frames is equal to or greater than 1, and the number of the second valid reference video frames is equal to 0, then determining that the probability of a drowning incident occurring in the target water area is a third probability;

[0025] Among them, the probability of the first possibility is less than the probability of the second possibility; the probability of the second possibility is less than or equal to the probability of the third possibility.

[0026] Preferably, when the edge computing system determines that a drowning incident may have occurred in the target water area, sending a reminder to the management terminal includes: sending the possibility of a drowning incident in the target water area to the management terminal;

[0027] When the management terminal receives the reminder, it performs a reminder operation, including: performing a corresponding reminder operation according to the possibility of a drowning incident occurring in the target water area, and the greater the possibility, the greater the reminder level.

[0028] Preferably, N is a positive integer less than or equal to 720.

[0029] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purposes and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0030] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0032] Figure 1 This is a structural diagram of a system for detecting drowning incidents based on edge computing in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0034] The embodiment of the present invention provides a system for detecting drowning events based on edge computing, such as Figure 1 As shown, it includes camera equipment, edge computing system and management terminal, among which:

[0035] The camera device and the edge computing system are arranged at a preset position of a target water area; wherein the target water area may be, for example, a river, a lake, a reservoir, or other water area;

[0036] The camera device is used to capture a video of the target water area to obtain video data of the target water area;

[0037] The edge computing system is configured to determine, based on the video data of the target water area, whether a drowning incident may have occurred in the target water area; and when it is determined that a drowning incident may have occurred in the target water area, send a reminder to the management terminal;

[0038] The management terminal is configured to execute a reminder operation upon receiving the reminder.

[0039] Among them, the above-mentioned camera equipment and edge computing system can be independently set up in each target water area, thereby realizing one-to-one targeted processing, improving the timeliness of drowning incident monitoring in the water area, and improving monitoring efficiency.

[0040] The beneficial effects of the above technical solution are: the above-mentioned camera equipment and edge computing system can be independently set up in the target water area, and by performing intelligent analysis on the video data of the target water area captured by the camera equipment, drowning incidents that may occur in the target water area can be judged in a timely manner, thereby improving the timeliness of monitoring drowning incidents in the water area and improving monitoring efficiency.

[0041] In one embodiment, the preset position of the target water area includes above the center of the water system of the target water area or above the water bank of the target water area;

[0042] The camera device shoots a video of the target water area from above the center of the water system of the target water area or above the bank of the target water area.

[0043] When implementing this technical solution, the specific location of the camera equipment can be set according to needs and the actual conditions of the water area.

[0044] In one embodiment, determining whether a drowning incident may have occurred in the target water area based on the video data of the target water area includes:

[0045] Perform image analysis on each video frame in the video data of the target water area in the order of shooting time, from the earliest to the last, to obtain the image content included in each video frame, wherein the following steps B1-B3 are performed on each video frame:

[0046] Step B1: performing a first recognition of the image content of the current video frame, wherein the first recognition is used to determine whether the current video frame includes a water area and a non-water area, wherein the non-water area includes an image area whose pixel color is not the color of water; when the current video frame includes both the water area and the non-water area, proceeding to step B2; when the current video frame includes only the water area but not the non-water area, proceeding to step B3;

[0047] Step B2: When the current video frame includes a water area and a non-water area, the content included in the non-water area in the current video frame is identified, and content feature information in the non-water area in the current video frame is extracted; it is determined whether the content feature information in the non-water area in the current video frame includes human organ features, and if so, whether a drowning incident may have occurred in the target water area based on the human organ features in the content feature information in the non-water area in the current video frame and the contents in N subsequent video frames after the current video frame; N is a positive integer equal to or greater than 2; if not, the next video frame of the current video frame is used as the new current video frame and returned to step B1; preferably, in one embodiment, N is a positive integer less than or equal to 720, and of course it can also be other values ​​that the implementer considers reasonable, and this application does not make specific limitations;

[0048] Step B3: When the current video frame includes only the water area but not the non-water area, the next video frame of the current video frame is used as a new current video frame and the process returns to step B1.

[0049] In one embodiment, judging whether a drowning incident may have occurred in the target water area based on the human organ features in the content feature information in the non-water area in the current video frame and the content in N consecutive subsequent video frames after the current video frame includes steps C1-C3:

[0050] Step C1: performing the following steps A1-A3 on each of the N subsequent video frames:

[0051] Step A1: performing a first recognition of the image content of the current subsequent video frame; if it is determined based on the recognition result of the first recognition that the current subsequent video frame includes only the water area and no non-water area, proceeding to step A2; if it is determined based on the recognition result of the first recognition that the current subsequent video frame includes both the water area and the non-water area, proceeding to step A3;

[0052] Step A2: when it is determined according to the recognition result of the first recognition that the current subsequent video frame includes only the water area but not the non-water area, recording the current subsequent video frame as the first valid reference video frame of the current video frame;

[0053] Step A3: when it is determined that the current subsequent video frame includes a water area and a non-water area according to the recognition result of the first recognition, the content included in the non-water area in the current subsequent video frame is identified, and content feature information in the non-water area in the current subsequent video frame is extracted; it is determined whether the content feature information in the non-water area in the current subsequent video frame includes a human organ feature; if so, it is determined whether the distance between the position of the human organ feature in the current subsequent video frame and the position of the human organ feature in the current video frame is equal to or less than a preset distance; if so, the current subsequent video frame is recorded as a second valid reference video frame of the current video frame; if not, the current subsequent video frame is recorded as an invalid reference video frame of the current video frame;

[0054] Step C2: determining whether the number of the first valid reference video frames is equal to or greater than 1, and determining whether the number of the second valid reference video frames is equal to or greater than 1;

[0055] If the number of the first valid reference video frames is equal to or greater than 1, and the number of the second valid reference video frames is equal to or greater than 1, determining that the possibility of a drowning incident occurring in the target water area is a first possibility;

[0056] If the number of the first valid reference video frames is equal to 0 and the number of the second valid reference video frames is equal to or greater than 1, then determining that the possibility of a drowning incident occurring in the target water area is a second possibility;

[0057] If the number of the first valid reference video frames is equal to or greater than 1, and the number of the second valid reference video frames is equal to 0, then determining that the probability of a drowning incident occurring in the target water area is a third probability;

[0058] Among them, the probability of the first possibility is less than the probability of the second possibility; the probability of the second possibility is less than or equal to the probability of the third possibility.

[0059] The beneficial effect of the above technical solution is that through the analysis of the above steps, it is possible to intelligently determine whether a drowning incident has occurred in the target water area and the degree of possibility of the drowning incident.

[0060] In one embodiment, when the edge computing system determines that a drowning incident may have occurred in the target water area, sending a reminder to the management terminal includes: sending a probability of a drowning incident occurring in the target water area to the management terminal;

[0061] When the management terminal receives the reminder, it performs a reminder operation, including: performing a corresponding reminder operation according to the possibility of a drowning incident occurring in the target water area, and the greater the possibility, the greater the reminder level.

[0062] The beneficial effects of the above technical solution are: the management terminal can be used by managers of the target water area, for example, it can be a portable terminal device or a computer, tablet computer, etc.; the corresponding reminder operation is performed according to the possibility of a drowning incident. The greater the possibility, the greater the degree of reminder, thereby realizing differentiated reminders for managers and improving the reminder effect.

[0063] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A system for detecting drowning incidents based on edge computing, characterized in that: It includes camera equipment, edge computing system and management terminal, including: The camera device and the edge computing system are arranged at a preset position in the target water area; The camera device is used to capture a video of the target water area to obtain video data of the target water area; The edge computing system is configured to determine, based on the video data of the target water area, whether a drowning incident may have occurred in the target water area; and when it is determined that a drowning incident may have occurred in the target water area, send a reminder to the management terminal; The management terminal is configured to execute a reminder operation upon receiving the reminder; The determining, based on the video data of the target water area, whether a drowning incident may have occurred in the target water area includes: Perform image analysis on each video frame in the video data of the target water area in the order of shooting time, from the earliest to the last, to obtain the image content included in each video frame, wherein the following steps B1-B3 are performed on each video frame: Step B1: performing a first recognition of the image content of the current video frame, wherein the first recognition is used to determine whether the current video frame includes a water area and a non-water area, wherein the non-water area includes an image area whose pixel color is not the color of water; when the current video frame includes both the water area and the non-water area, proceeding to step B2; when the current video frame includes only the water area but not the non-water area, proceeding to step B3; Step B2: When the current video frame includes a water area and a non-water area, identifying the content included in the non-water area in the current video frame, and extracting content feature information in the non-water area in the current video frame; determining whether the content feature information in the non-water area in the current video frame includes human organ features; if so, determining whether a drowning incident may have occurred in the target water area based on the human organ features in the content feature information in the non-water area in the current video frame and the contents of N subsequent video frames after the current video frame; N is a positive integer equal to or greater than 2; if not, returning to step B1 with the next video frame of the current video frame as the new current video frame; Step B3: When the current video frame includes only the water area but not the non-water area, the next video frame of the current video frame is used as a new current video frame and the process returns to step B1; The method of determining whether a drowning incident may have occurred in the target water area based on the human organ features in the content feature information in the non-water area in the current video frame and the content in N consecutive subsequent video frames after the current video frame includes steps C1-C3: Step C1: performing the following steps A1-A3 on each of the N subsequent video frames: Step A1: performing a first recognition of the image content of the current subsequent video frame; if it is determined based on the recognition result of the first recognition that the current subsequent video frame includes only the water area and no non-water area, proceeding to step A2; if it is determined based on the recognition result of the first recognition that the current subsequent video frame includes both the water area and the non-water area, proceeding to step A3; Step A2: when it is determined according to the recognition result of the first recognition that the current subsequent video frame includes only the water area but not the non-water area, recording the current subsequent video frame as the first valid reference video frame of the current video frame; Step A3: when it is determined that the current subsequent video frame includes a water area and a non-water area according to the recognition result of the first recognition, the content included in the non-water area in the current subsequent video frame is identified, and content feature information in the non-water area in the current subsequent video frame is extracted; it is determined whether the content feature information in the non-water area in the current subsequent video frame includes a human organ feature; if so, it is determined whether the distance between the position of the human organ feature in the current subsequent video frame and the position of the human organ feature in the current video frame is equal to or less than a preset distance; if so, the current subsequent video frame is recorded as a second valid reference video frame of the current video frame; if not, the current subsequent video frame is recorded as an invalid reference video frame of the current video frame; Step C2: determining whether the number of the first valid reference video frames is equal to or greater than 1, and determining whether the number of the second valid reference video frames is equal to or greater than 1; If the number of the first valid reference video frames is equal to or greater than 1, and the number of the second valid reference video frames is equal to or greater than 1, determining that the possibility of a drowning incident occurring in the target water area is a first possibility; If the number of the first valid reference video frames is equal to 0 and the number of the second valid reference video frames is equal to or greater than 1, then determining that the possibility of a drowning incident occurring in the target water area is a second possibility; If the number of the first valid reference video frames is equal to or greater than 1, and the number of the second valid reference video frames is equal to 0, then determining that the probability of a drowning incident occurring in the target water area is a third probability; Among them, the probability of the first possibility is less than the probability of the second possibility; the probability of the second possibility is less than or equal to the probability of the third possibility.

2. The system according to claim 1, wherein The preset position of the target water area includes above the center of the water system of the target water area or above the water bank of the target water area; The camera device shoots a video of the target water area from above the center of the water system of the target water area or above the bank of the target water area.

3. The system according to claim 1, wherein: When the edge computing system determines that a drowning incident may have occurred in the target water area, sending a reminder to the management terminal includes: sending a probability that a drowning incident may have occurred in the target water area to the management terminal; When the management terminal receives the reminder, it performs a reminder operation, including: performing a corresponding reminder operation according to the possibility of a drowning incident occurring in the target water area, and the greater the possibility, the greater the reminder level.

4. The system according to claim 1, wherein: The N is a positive integer less than or equal to 720.

Images