Port intelligent safety supervision and emergency response system and method

Abstract

The present invention provides a port intelligent safety supervision and emergency response system and method, the method comprising: obtaining the current port monitoring image of each port area of ​​the port, the sensor detection image of each sensor detection type, and identifying each abnormal area and abnormal information of each port area; for each port area, through the image recognition strategy corresponding to each sensor detection image, identifying each detection abnormal area and detection abnormal information corresponding to each sensor detection image, so as to analyze the target abnormal information of each target abnormal area of ​​the port area; based on each of the target abnormal areas and target abnormal information, identifying the regional abnormality detection information of the port area, and based on the regional detection type of the regional abnormality detection information, generating the emergency response strategy of the port area, sending it to the staff, and iterating the above steps. The use of this solution can improve the efficiency of safety supervision and emergency response of ports.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent safety supervision, and in particular to a port intelligent safety supervision and emergency response system and method. Background Art

[0002] Port safety supervision is achieved through the application of AI technology and intelligent map review systems, which can monitor port operations in real time and collect data, detect abnormal situations in a timely manner and issue early warnings, and effectively improve the efficiency and accuracy of port safety supervision. However, due to the large scope of ports, large port logistics volume, high logistics similarity, and many hidden areas, there will be deviations in the data monitoring and data collection of the entire port case, making the port safety supervision less accurate. In addition, due to the large area and many hidden places of the port, the timely response efficiency is also poor in the presence of emergency risks, which can easily cause property losses and safety hazards. Therefore, how to improve the efficiency of safety supervision and emergency response in port areas is the current research focus.

[0003] The traditional port safety supervision and emergency response method is to install multiple cameras in each port area of ​​the port, and add manual inspections and intelligent inspection devices for multiple inspections to ensure the port's management efficiency and be able to respond to port risk information in a timely manner. However, image detection has monitoring blind spots, and there are time intervals between human inspections and intelligent inspections. It requires a lot of resource costs and will also cause safety supervision errors and control failures caused by inspection omissions, resulting in low efficiency of port safety supervision and emergency response. Summary of the invention

[0004] The main purpose of the present invention is to provide a method for intelligent safety supervision and emergency response of ports, which aims to solve the problems in the prior art that image detection has monitoring blind spots, there are time intervals between personnel inspections and intelligent inspections, which requires a lot of resource costs, and also causes safety supervision errors and control failures caused by inspection omissions, thereby leading to low efficiency of port safety supervision and emergency response.

[0005] To achieve the above object, the present invention provides a port intelligent safety supervision and emergency response method, the method comprising:

[0006] Acquire the current port monitoring image of each port area of ​​the port and the sensor detection image of each sensor detection type of each port area, and identify each abnormal area of ​​each port area and the abnormal information of each abnormal area through an image recognition network based on the current port monitoring image of each port area;

[0007] For each port area, based on the sensor detection images of each sensor detection type of the port area, through the image recognition strategy corresponding to each sensor detection image, each detection abnormal area corresponding to each sensor detection image and the detection abnormality information of each detection abnormal area are identified; and through each abnormal area of ​​the port area, each detection abnormal area of ​​the port area, the abnormal information of each abnormal area, and the detection abnormality information of each detection abnormal area, each target abnormal area of ​​the port area and the target abnormality information of each target abnormal area are analyzed;

[0008] Based on each of the target abnormal areas and the target abnormal information of each of the target abnormal areas, identifying the regional abnormality detection information of the port area, and when the regional detection type of the regional abnormality detection information is a risk abnormality type, generating an emergency response strategy for the port area based on the regional detection type of the regional abnormality detection information;

[0009] Based on the emergency response strategy of each port area, emergency response processing is performed on the port, and when the emergency end operation of the staff is monitored, the step of returning to execute the current port monitoring image of each port area of ​​the port and the sensor detection image step of each sensor detection type of each port area is obtained.

[0010] Optionally, the identifying of each abnormal area of ​​each port area and abnormal information of each abnormal area based on the current port monitoring image of each port area through an image recognition network includes:

[0011] For each current port monitoring image, the current port monitoring image is subjected to grid splitting processing to obtain grid images of each region, and the image content type of each regional grid image is identified through an image recognition network;

[0012] Based on the image abnormality degree analysis strategy for each image content type, respectively identifying the region state corresponding to the region grid image of each image content type and the state degree of the region state of each region grid image;

[0013] The regional grid images of all regions whose states are abnormal are used as the abnormal sub-regions of the port region, and the abnormal sub-regions are clustered to obtain the abnormal regions of the port region;

[0014] The region states of each sub-abnormal region corresponding to each abnormal region and the state degrees of the region states of each sub-abnormal region corresponding to each abnormal region are used as the abnormality information of each abnormal region.

[0015] Optionally, the sensor detection images of each sensor detection type based on the port area, identifying each detection abnormality area corresponding to each sensor detection image and detection abnormality information of each detection abnormality area through an image recognition strategy corresponding to each sensor detection image, includes:

[0016] For each sensor detection type, query the database for sample abnormal images of each abnormal degree of the sensor detection type, and grid-divide the sensor detection image of the sensor detection type into multiple grid detection images;

[0017] Using an image feature recognition network, the detection image features of each grid detection image and the sample image features of each sample abnormal image are identified, and the similarity between each detection image feature and each sample image feature is calculated using a similarity algorithm;

[0018] The sample image feature corresponding to each detection image feature and having a similarity greater than a similarity threshold is used as the target sample image feature corresponding to each detection image feature, and the abnormality degree corresponding to each target sample image feature is used as the abnormality degree of the grid detection image to which the detection image feature corresponding to each target sample image feature belongs;

[0019] Clustering is performed on each of the grid detection images to obtain each detection abnormality area corresponding to the sensor detection image, and the abnormality degrees of each grid detection image contained in each detection abnormality image are distributed and arranged to obtain detection abnormality information of each detection abnormality area.

[0020] Optionally, analyzing each target abnormal area in the port area and the target abnormal information of each target abnormal area through each abnormal area in the port area, each detected abnormal area in the port area, the abnormal information of each abnormal area, and the detected abnormal information of each detected abnormal area, comprises:

[0021] Identify, among the abnormal regions, an intersecting abnormal region that has an intersecting region with the detected abnormal region, and use the detected abnormal region that has the intersecting region as the intersecting detected abnormal region;

[0022] The abnormal area without the intersection area is used as the first target abnormal area, and the abnormal information of the first target abnormal area is used as the target abnormal information of the first target abnormal area, and the detected abnormal area without the intersection area is used as the second target abnormal area, and the detected abnormal information of the second target abnormal area is used as the target abnormal information of the second target abnormal area;

[0023] Identify the abnormality type corresponding to each sensor detection type, and based on the abnormality type corresponding to each sensor detection type, add an abnormality type identifier to the detection abnormality information of the detection abnormality area corresponding to each sensor detection type to obtain target abnormality information of each detection abnormality area;

[0024] Each intersection abnormal area and all areas included in each intersection detection abnormal area are used as the third target abnormal area, and the abnormality information of the intersection abnormal area and the target abnormality information of the intersection detection abnormal area are used as the target abnormality information of the third target abnormal area.

[0025] Optionally, the identifying the regional anomaly detection information of the port area based on each of the target anomaly areas and the target anomaly information of each of the target anomaly areas includes:

[0026] Based on the target abnormality information of each of the first target abnormality areas, query the abnormality type of each of the first target abnormality areas in the database, and for each third target abnormality area, generate the sub-area abnormality distribution information of the third target abnormality area by combining the abnormality information of the intersecting abnormality area in the third target abnormality area, the abnormality type of the intersecting abnormality area, and the target abnormality information of the intersecting detection abnormality area in the third target abnormality area;

[0027] For each second target abnormal region, based on the abnormality type of the second target abnormal region and the abnormality degree of each grid detection image of the second target abnormal region, generate sub-region abnormality distribution information of the second target abnormal region, and for each first target abnormal region, based on the abnormality type of the first target abnormal region and the abnormality degree of the first target abnormal region, generate sub-region abnormality distribution information of the first target abnormal region;

[0028] The sub-region abnormal distribution information of each target abnormal region is sorted in the current port monitoring image of the port region to obtain the regional abnormality detection information of the port region.

[0029] Optionally, when the regional detection type of the regional anomaly detection information is a risk anomaly type, generating an emergency response strategy for the port area based on the regional detection type of the regional anomaly detection information includes:

[0030] Calculate the abnormal proportion of all target abnormal areas in the port area to the abnormal areas in the port area, and respectively calculate the target abnormal areas corresponding to each abnormal degree of each abnormal type to the sub-abnormal proportion of the abnormal areas in the port area;

[0031] Based on the sub-anomaly proportion values ​​of each abnormal degree of each abnormal type and the abnormal proportion value, in the database, in the abnormal database, the regional detection type of the port area is queried, and it is determined whether the regional detection type belongs to the risk abnormal type;

[0032] When the abnormality type belongs to the risk abnormality type, based on the abnormality degree of each abnormality type, the emergency treatment strategy corresponding to the abnormality degree of each abnormality type is queried in the database, and the emergency treatment strategy corresponding to the abnormality degree of each abnormality type is used as the emergency treatment strategy of the target abnormal area corresponding to the abnormality degree of each abnormality type;

[0033] The emergency handling strategies for all target abnormal areas are used as the emergency response strategies for the port area.

[0034] In addition, to achieve the above-mentioned purpose, the present invention also provides a port intelligent safety supervision and emergency response system, the port intelligent safety supervision and emergency response system comprising:

[0035] An acquisition module is used to acquire a current port monitoring image of each port area of ​​the port and a sensor detection image of each sensor detection type of each port area, and based on the current port monitoring image of each port area, identify each abnormal area of ​​each port area and abnormal information of each abnormal area through an image recognition network;

[0036] An analysis module is used for identifying, for each port area, each abnormal detection area corresponding to each sensor detection image and abnormal detection information of each abnormal detection area based on the sensor detection images of each sensor detection type of the port area through an image recognition strategy corresponding to each sensor detection image, and analyzing each target abnormal area of ​​the port area and target abnormal information of each target abnormal area through each abnormal area of ​​the port area, each abnormal detection area of ​​the port area, abnormal information of each abnormal area and detection abnormal information of each abnormal detection area;

[0037] A generating module, for identifying the regional anomaly detection information of the port area based on each of the target anomaly areas and the target anomaly information of each of the target anomaly areas, and generating an emergency response strategy for the port area based on the regional detection type of the regional anomaly detection information when the regional detection type of the regional anomaly detection information is a risk anomaly type;

[0038] The iterative module is used to perform emergency response processing on the port based on the emergency response strategy of each port area, and when the emergency end operation of the staff is monitored, return to execute the step of obtaining the current port monitoring image of each port area of ​​the port and the sensor detection image of each sensor detection type of each port area.

[0039] Optionally, the acquisition module is specifically used to:

[0040] For each current port monitoring image, the current port monitoring image is subjected to grid splitting processing to obtain grid images of each region, and the image content type of each regional grid image is identified through an image recognition network;

[0041] Based on the image abnormality degree analysis strategy for each image content type, respectively identifying the region state corresponding to the region grid image of each image content type and the state degree of the region state of each region grid image;

[0042] The regional grid images of all regions whose states are abnormal are used as the abnormal sub-regions of the port region, and the abnormal sub-regions are clustered to obtain the abnormal regions of the port region;

[0043] The region states of each sub-abnormal region corresponding to each abnormal region and the state degrees of the region states of each sub-abnormal region corresponding to each abnormal region are used as the abnormality information of each abnormal region.

[0044] Optionally, the analysis module is specifically used to:

[0045] For each sensor detection type, query the database for sample abnormal images of each abnormal degree of the sensor detection type, and grid-divide the sensor detection image of the sensor detection type into multiple grid detection images;

[0046] Using an image feature recognition network, the detection image features of each grid detection image and the sample image features of each sample abnormal image are identified, and the similarity between each detection image feature and each sample image feature is calculated using a similarity algorithm;

[0047] The sample image feature corresponding to each detection image feature and having a similarity greater than a similarity threshold is used as the target sample image feature corresponding to each detection image feature, and the abnormality degree corresponding to each target sample image feature is used as the abnormality degree of the grid detection image to which the detection image feature corresponding to each target sample image feature belongs;

[0048] Clustering is performed on each of the grid detection images to obtain each detection abnormality area corresponding to the sensor detection image, and the abnormality degrees of each grid detection image contained in each detection abnormality image are distributed and arranged to obtain detection abnormality information of each detection abnormality area.

[0049] Optionally, the analysis module is specifically used to:

[0050] Identify, among the abnormal regions, an intersecting abnormal region that has an intersecting region with the detected abnormal region, and use the detected abnormal region that has the intersecting region as the intersecting detected abnormal region;

[0051] The abnormal area without the intersection area is used as the first target abnormal area, and the abnormal information of the first target abnormal area is used as the target abnormal information of the first target abnormal area, and the detected abnormal area without the intersection area is used as the second target abnormal area, and the detected abnormal information of the second target abnormal area is used as the target abnormal information of the second target abnormal area;

[0052] Identify the abnormality type corresponding to each sensor detection type, and based on the abnormality type corresponding to each sensor detection type, add an abnormality type identifier to the detection abnormality information of the detection abnormality area corresponding to each sensor detection type to obtain target abnormality information of each detection abnormality area;

[0053] Each intersection abnormal area and all areas included in each intersection detection abnormal area are used as the third target abnormal area, and the abnormality information of the intersection abnormal area and the target abnormality information of the intersection detection abnormal area are used as the target abnormality information of the third target abnormal area.

[0054] Optionally, the generating module is specifically used for:

[0055] Based on the target abnormality information of each of the first target abnormality areas, query the abnormality type of each of the first target abnormality areas in the database, and for each third target abnormality area, generate the sub-area abnormality distribution information of the third target abnormality area by combining the abnormality information of the intersecting abnormality area in the third target abnormality area, the abnormality type of the intersecting abnormality area, and the target abnormality information of the intersecting detection abnormality area in the third target abnormality area;

[0056] For each second target abnormal region, based on the abnormality type of the second target abnormal region and the abnormality degree of each grid detection image of the second target abnormal region, generate sub-region abnormality distribution information of the second target abnormal region, and for each first target abnormal region, based on the abnormality type of the first target abnormal region and the abnormality degree of the first target abnormal region, generate sub-region abnormality distribution information of the first target abnormal region;

[0057] The sub-region abnormal distribution information of each target abnormal region is sorted in the current port monitoring image of the port region to obtain the regional abnormality detection information of the port region.

[0058] Optionally, the generating module is specifically used for:

[0059] Calculate the abnormal proportion of all target abnormal areas in the port area to the abnormal areas in the port area, and respectively calculate the target abnormal areas corresponding to each abnormal degree of each abnormal type to the sub-abnormal proportion of the abnormal areas in the port area;

[0060] Based on the sub-anomaly proportion values ​​of each abnormal degree of each abnormal type and the abnormal proportion value, in the database, in the abnormal database, the regional detection type of the port area is queried, and it is determined whether the regional detection type belongs to the risk abnormal type;

[0061] When the abnormality type belongs to the risk abnormality type, based on the abnormality degree of each abnormality type, the emergency treatment strategy corresponding to the abnormality degree of each abnormality type is queried in the database, and the emergency treatment strategy corresponding to the abnormality degree of each abnormality type is used as the emergency treatment strategy of the target abnormal area corresponding to the abnormality degree of each abnormality type;

[0062] The emergency handling strategies for all target abnormal areas are used as the emergency response strategies for the port area.

[0063] In a third aspect, the present application provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any one of the methods in the first aspect are implemented.

[0064] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0065] In a fifth aspect, the present application provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0066] The present invention provides a method and system for intelligent safety supervision and emergency response of a port, the method comprising: acquiring a current port monitoring image of each port area of ​​the port, and a sensor detection image of each sensor detection type of each port area, and based on the current port monitoring image of each port area, identifying each abnormal area of ​​each port area and abnormal information of each abnormal area through an image recognition network; for each port area, based on the sensor detection images of each sensor detection type of the port area, identifying each detection abnormal area corresponding to each sensor detection image and detection abnormal information of each detection abnormal area through an image recognition strategy corresponding to each sensor detection image, and identifying each abnormal area of ​​the port area, and each detection abnormal area of ​​the port area, and each abnormal area of ​​the abnormal area through an image recognition network. The abnormal information of the port area and the abnormal detection information of each abnormal detection area are analyzed; based on each abnormal target area and the target abnormal information of each abnormal target area, the regional abnormal detection information of the port area is identified, and when the regional detection type of the regional abnormal detection information is a risk abnormal type, the emergency response strategy of the port area is generated based on the regional detection type of the regional abnormal detection information; based on the emergency response strategy of each port area, the port is subjected to emergency response processing, and when the emergency end operation of the staff is monitored, the step of obtaining the current port monitoring image of each port area of ​​the port and the sensor detection image of each sensor detection type of each port area is returned. Based on the video surveillance image recognition, the present scheme analyzes the abnormal information of each abnormal area and the detection abnormal information of each abnormal detection area through the sensor detection images detected by multiple sensor detection types, so that not only manual inspection and intelligent inspection equipment inspection are not required, but the video surveillance blind area and the area that cannot be monitored in real time are detected in real time, and real-time image acquisition is performed, thereby ensuring the real-time and comprehensiveness of security monitoring. Then, this solution analyzes the target abnormality information of each target abnormal area in each port area by combining the abnormal information of each abnormal area and the detected abnormality information of each detected abnormal area, thereby analyzing the regional abnormality detection information of each port area, and improving the comprehensiveness and accuracy of the detection of regional abnormality detection information of the port area. After that, the terminal generates an emergency response strategy for the port area based on the regional detection type of the regional abnormality detection information, and performs emergency response processing on the port based on the emergency response strategy of each port area, thereby ensuring real-time detection and real-time response, and improving the timeliness and comprehensiveness of the emergency response. Finally, after the staff completes the emergency response processing, this solution automatically iterates the overall process to ensure the real-time and cyclic nature of the detection, thereby comprehensively improving the efficiency of safety supervision and emergency response of the port. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0068] Figure 1 is a flow chart of a method for intelligent safety supervision and emergency response of a port provided by an embodiment of the present invention;

[0069] Figure 2 It is a structural diagram of a port intelligent safety supervision and emergency response system provided by an embodiment of the present invention;

[0070] Figure 3 An internal structure diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0071] The port intelligent safety supervision and emergency response method provided by the embodiment of the present invention is applied to the port intelligent safety supervision and emergency response system. Unless otherwise defined, all technical and scientific terms used in this article have the same meaning as those generally understood by technicians in the technical field of this application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned drawings and any variations thereof are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0072] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0073] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0074] The port intelligent safety supervision and emergency response method provided by the embodiment of the present application can be applied to the application environment of port intelligent safety supervision and emergency response. The method can be applied to a terminal, a server, or a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, large computers, etc. Among them, the terminal analyzes the abnormal information of each abnormal area and the detection abnormal information of each detected abnormal area through the sensor detection images detected by multiple sensor detection types on the basis of video surveillance image recognition, so that not only manual inspection and intelligent inspection equipment inspection are not required, but the video surveillance blind area and the area that cannot be monitored in real time are detected in real time by sensor detection, and real-time image acquisition is performed, thereby ensuring the real-time nature of safety monitoring and the comprehensiveness of safety monitoring. Then this scheme analyzes the target abnormal information of each port area target abnormal area by combining the abnormal information of each abnormal area and the detection abnormal information of each detected abnormal area, thereby analyzing the regional abnormal detection information of each port area, and improving the detection comprehensiveness and detection accuracy of the regional abnormal detection information of the port area. After that, the terminal generates the emergency response strategy of the port area based on the regional detection type of the regional anomaly detection information, and performs emergency response processing on the port based on the emergency response strategy of each port area, thereby ensuring real-time detection and real-time response, and improving the timeliness and comprehensiveness of emergency response. Finally, after the staff completes the emergency response processing, this solution automatically loops and iterates the overall process to ensure the real-time and cyclic nature of the detection, thereby comprehensively improving the efficiency of port safety supervision and emergency response.

[0075] In one embodiment, Figure 1 As shown, a port intelligent safety supervision and emergency response method is provided, which is described by taking the method applied to a terminal as an example, and includes the following steps:

[0076] Step S101, obtaining the current port monitoring image of each port area of ​​the port and the sensor detection image of each sensor detection type of each port area, and based on the current port monitoring image of each port area, identifying each abnormal area of ​​each port area and the abnormal information of each abnormal area through an image recognition network.

[0077] In this embodiment, the terminal collects the monitoring images of each port area in real time through the monitoring cameras set at each position of each port area, and obtains the current port monitoring images of each port area, wherein each port area includes a cargo storage area, a cargo circulation area, a cargo port entry area, a cargo subpackaging area, a building area, and other port areas. Among them, the number of monitoring cameras set in each port area is different, and the number of the monitoring cameras is determined by the size of the port area, the number of monitoring blind areas of the port area, and the criticality of the port area. When the area of ​​the port area is large, the number of monitoring blind areas of the port area is large, and the criticality of the port area is large, the number of the monitoring cameras is large. When the area of ​​the port area is small, the number of monitoring blind areas of the port area is small, and the criticality of the port area is small, the number of the monitoring cameras is small. Then, the terminal collects the sensor detection image of each port area through the sensor data detection devices of each sensor detection type set in each port area. Among them, the sensor data detection device includes but is not limited to an infrared temperature imaging detection sensor, a radiation detection sensor, a vibration imaging detection sensor, etc. The terminal generates a sensor detection image based on the sensor values ​​detected by the detection sensors set at each position point. Among them, the infrared temperature imaging detection sensor is used to detect fire risks, the radiation detection sensor is used to detect radiation leakage risks, and the vibration imaging detection sensor is used to detect risks such as theft, floods, and typhoons. Finally, based on the current port monitoring image of each port area, the terminal identifies each abnormal area and the abnormal information of each abnormal area through an image recognition network. The image recognition network can be an image classifier based on a convolutional neural network. The specific recognition process will be described in detail later.

[0078] Step S102, for each port area, based on the sensor detection images of each sensor detection type of the port area, through the image recognition strategy corresponding to each sensor detection image, identify each detection abnormal area corresponding to each sensor detection image and the detection abnormality information of each detection abnormal area; and through each abnormal area of ​​the port area, each detection abnormal area of ​​the port area, the abnormal information of each abnormal area, and the detection abnormality information of each detection abnormal area, analyze each target abnormal area of ​​the port area and the target abnormality information of each target abnormal area.

[0079] In this embodiment, for each port area, the terminal identifies each abnormal detection area corresponding to each sensor detection image and the abnormal detection information of each abnormal detection area based on the sensor detection images of each sensor detection type in the port area through the image recognition strategy corresponding to each sensor detection image, and analyzes each target abnormal area in the port area and the target abnormal information of each target abnormal area through each abnormal area in the port area, each abnormal detection area in the port area, the abnormal information of each abnormal area, and the abnormal detection information of each abnormal detection area. Among them, the abnormal area is the area range in the port area where abnormal information exists, and the abnormal information of each abnormal area includes the degree of abnormality and the area state of the abnormal area. The area state is used to characterize whether the current state of the abnormal area is abnormal, and the area state includes the normal state of the area and the abnormal state of the area. The specific identification process will be described in detail later.

[0080] Step S103, based on each target abnormal area and the target abnormal information of each target abnormal area, the regional abnormality detection information of the port area is identified, and when the regional detection type of the regional abnormality detection information is a risk abnormality type, an emergency response strategy for the port area is generated based on the regional detection type of the regional abnormality detection information.

[0081] In this embodiment, the terminal identifies the regional anomaly detection information of the port area based on each target abnormal area and the target abnormality information of each target abnormal area, and generates an emergency response strategy for the port area based on the regional detection type of the regional anomaly detection information when the regional detection type of the regional anomaly detection information is a risk abnormality type. Among them, the regional detection type of the regional anomaly detection information includes multiple types, for example, a normal detection type, a mild abnormality type, a moderate abnormality type, and a severe abnormality type, wherein the moderate abnormality type and the severe abnormality type belong to the risk abnormality type, and the normal detection type and the mild abnormality type are non-abnormal types. The emergency response strategy of the port area includes the emergency handling strategies of all target abnormal areas. The process of identifying the emergency handling strategy of each target abnormal area will be described in detail later.

[0082] Step S104, based on the emergency response strategy of each port area, the port is subjected to emergency response processing, and when the emergency end operation of the staff is monitored, the step of returning to execute the step of obtaining the current port monitoring image of each port area of ​​the port and the sensor detection image of each sensor detection type of each port area is executed.

[0083] In this embodiment, the terminal performs emergency response processing on the port based on the emergency response strategy of each port area, and when the emergency end operation of the staff is monitored, it returns to execute the step of obtaining the current port monitoring image of each port area of ​​the port, and the sensor detection image of each sensor detection type of each port area.

[0084] Based on the above scheme, on the basis of video surveillance image recognition, the sensor detection images detected by multiple sensor detection types are used to analyze the abnormal information of each abnormal area and the detection abnormal information of each detection abnormal area, so that not only manual inspection and intelligent inspection equipment inspection are not required, but the video surveillance blind area and the area that cannot be monitored in real time are detected in real time and real-time image acquisition is performed by sensor detection, thereby ensuring the real-time and comprehensiveness of security monitoring. Then, this scheme analyzes the target abnormal information of each port area target abnormal area by combining the abnormal information of each abnormal area and the detection abnormal information of each detection abnormal area, thereby analyzing the regional abnormal detection information of each port area, and improving the comprehensiveness and accuracy of the detection of the regional abnormal detection information of the port area. After that, the terminal generates an emergency response strategy for the port area based on the regional detection type of the regional abnormal detection information, and performs emergency response processing on the port based on the emergency response strategy of each port area, thereby ensuring real-time detection and real-time response, and improving the timeliness and comprehensiveness of the emergency response. Finally, after the staff completes the emergency response, the solution automatically iterates the entire process to ensure the real-time and cyclic nature of the detection, thereby comprehensively improving the efficiency of port safety supervision and emergency response.

[0085] Optionally, based on the current port monitoring image of each port area, each abnormal area of ​​each port area and the abnormal information of each abnormal area are identified through an image recognition network, including: for each current port monitoring image, the current port monitoring image is grid-splitting processed to obtain grid images of each area, and the image content type of each regional grid image is identified through the image recognition network; based on the image abnormality degree analysis strategy of each image content type, the regional state corresponding to the regional grid image of each image content type and the state degree of the regional state of each regional grid image are respectively identified; all regional grid images whose regional states belong to abnormal states are regarded as sub-abnormal areas of the port area, and the sub-abnormal areas are clustered to obtain abnormal areas of the port area; the regional state of each sub-abnormal area corresponding to each abnormal area and the state degree of the regional state of each sub-abnormal area corresponding to each abnormal area are regarded as the abnormal information of each abnormal area.

[0086] In this embodiment, the terminal performs grid splitting processing on each current port monitoring image to obtain grid images of each region, and identifies the image content type of each regional grid image through an image recognition network. The image content type is used to characterize the image content corresponding to the regional grid image, for example, cargo image type, road image type, building image type, wood image type, vehicle image type, tree image type, sea surface image type, room image type, etc.

[0087] Then, based on the image abnormality degree analysis strategy of each image content type, the terminal identifies the regional state corresponding to the regional grid image of each image content type and the state degree of the regional state of each regional grid image. Among them, the image abnormality degree analysis strategy of each image content type is an analysis program of the image abnormality degree corresponding to each image content type constructed based on AI technology and intelligent image review system. Among them, each regional state of the same image content type corresponds to a state degree range, each regional state corresponds to a state ranking, and the state ranking of each regional state is based on the ranking position of the state degree range corresponding to each regional state in the overall state degree range. For example, the state degree is 0%-100%, wherein 0%-20% corresponds to the first regional state, 20%-40% corresponds to the second regional state, 40%-60% corresponds to the third regional state, 60%-80% corresponds to the fourth regional state, and 80%-100% corresponds to the fifth regional state. Then the state ranking of each regional state is the first regional state, the second regional state, the third regional state, the fourth regional state, and the fifth regional state. Among them, the closer the state degree range of each regional state is to the upper limit, the closer the image content is to the next regional state. The higher the regional status level is, the more normal the regional status is, and the lower the regional status level is, the more abnormal the regional status is.

[0088] Afterwards, the terminal takes the regional grid images of all regions whose status are abnormal as the abnormal sub-regions of the port area, and clusters the abnormal sub-regions to obtain the abnormal regions of the port area.

[0089] Finally, the terminal uses the region states of each sub-abnormal region corresponding to each abnormal region and the state degrees of the region states of each sub-abnormal region corresponding to each abnormal region as the abnormal information of each abnormal region.

[0090] Based on the above scheme, after the grid image areas are divided into grids, the image content type of each regional grid image is identified respectively, and then according to the different image content types, the regional status and status degree corresponding to each regional grid image are identified, thereby improving the recognition accuracy of abnormal information of different regional grid images.

[0091] Optionally, based on the sensor detection images of each sensor detection type of the port area, each detection abnormality area corresponding to each sensor detection image and the detection abnormality information of each detection abnormality area are identified through the image recognition strategy corresponding to each sensor detection image, including: for each sensor detection type, querying the sample abnormality images of each abnormality degree of the sensor detection type in the database, and gridding and splitting the sensor detection images of the sensor detection type into multiple grid detection images; identifying the detection image features of each grid detection image and the sample image features of each sample abnormality image through an image feature recognition network, and calculating the similarity between each detection image feature and each sample image feature through a similarity algorithm; using the sample image features corresponding to each detection image feature with a similarity greater than a similarity threshold as the target sample image feature corresponding to each detection image feature, and using the abnormality degree corresponding to each target sample image feature as the abnormality degree of the grid detection image to which the detection image feature corresponding to each target sample image feature belongs; clustering the grid detection images to obtain the detection abnormality areas corresponding to the sensor detection images, and distributing and arranging the abnormality degrees of each grid detection image contained in each detection abnormal image to obtain the detection abnormality information of each detection abnormality area.

[0092] In this embodiment, for each sensor detection type, the terminal queries the sample abnormal images of each abnormal degree of the sensor detection type in the database, and grids the sensor detection images of the sensor detection type into multiple grid detection images. For example, different temperature values ​​in the sensor detection image detected by the infrared temperature sensor detection device generate a sample abnormal image, and the sample abnormal image can be a color, such as dark red and dark blue. The closer the color is to dark red, the higher the temperature at the location is, and the closer the color is to dark blue, the lower the temperature at the location is.

[0093] Then, the terminal uses the image feature recognition network to identify the detection image features of each grid detection image and the sample image features of each sample abnormal image, and calculates the similarity between each detection image feature and each sample image feature through a similarity algorithm. Among them, the image feature extraction algorithm is a convolutional neural network based on the self-attention mechanism.

[0094] The terminal uses the sample image features corresponding to each detection image feature and the similarity greater than the similarity threshold as the target sample image feature corresponding to each detection image feature, and uses the abnormality degree corresponding to each target sample image feature as the abnormality degree of the grid detection image to which the detection image feature corresponding to each target sample image feature belongs. Finally, the terminal performs clustering processing on each grid detection image to obtain each detection abnormality area corresponding to the sensor detection image, and distributes and arranges the abnormality degrees of each grid detection image contained in each detection abnormality image to obtain the detection abnormality information of each detection abnormality area. That is, each

[0095] Based on the above scheme, the degree of abnormality of each detected abnormal image and the detected abnormality information of each detected abnormal area are identified by means of sample feature recognition and sample similarity determination, thereby improving the comprehensiveness and accuracy of recognition.

[0096] Optionally, each target abnormal area in the port area and the target abnormal information of each target abnormal area are analyzed through each abnormal area in the port area, each detected abnormal area in the port area, the abnormal information of each abnormal area, and the detected abnormal information of each detected abnormal area, including: identifying, among each abnormal area, an intersecting abnormal area with the detected abnormal area, and treating the detected abnormal area with the intersecting abnormal area as the intersecting detected abnormal area; treating the abnormal area without the intersecting area as the first target abnormal area, treating the abnormal information of the first target abnormal area as the target abnormal information of the first target abnormal area, and treating the detected abnormal area without the intersecting area as The second target abnormality area, taking the detection abnormality information of the second target abnormality area as the target abnormality information of the second target abnormality area; identifying the abnormality type corresponding to each sensor detection type, and based on the abnormality type corresponding to each sensor detection type, adding the abnormality type identifier to the detection abnormality information of the detection abnormality area corresponding to each sensor detection type, to obtain the target abnormality information of each detection abnormality area; taking each intersecting abnormality area and all areas contained in each intersecting detection abnormality area as the third target abnormality area, and taking the abnormality information of the intersecting abnormality area and the target abnormality information of the intersecting detection abnormality area as the target abnormality information of the third target abnormality area.

[0097] In this embodiment, the terminal identifies intersecting abnormal regions among the abnormal regions, which have intersecting regions with the detected abnormal regions, and uses the detected abnormal regions with intersecting regions as intersecting detected abnormal regions, wherein the intersecting regions include fully overlapping regions and partially overlapping regions.

[0098] The terminal will take the abnormal area where there is no intersection area as the first target abnormal area, and the abnormal information of the first target abnormal area as the target abnormal information of the first target abnormal area, and will take the detected abnormal area where there is no intersection area as the second target abnormal area, and the detected abnormal information of the second target abnormal area as the target abnormal information of the second target abnormal area.

[0099] Then, the terminal identifies the abnormal type corresponding to each sensor detection type by identifying the abnormal type identifier of the sensor detection device corresponding to each sensor detection type based on the abnormal type identifier preset in the sensor detection device corresponding to each sensor detection type.

[0100] Based on the abnormality type corresponding to each sensor detection type, the terminal adds an abnormality type identifier to the detection abnormality information of the detection abnormality area corresponding to each sensor detection type to obtain the target abnormality information of each detection abnormality area.

[0101] Finally, the terminal uses each intersection abnormal area and all areas contained in each intersection detection abnormal area as the third target abnormal area, and uses the abnormality information of the intersection abnormal area and the target abnormality information of the intersection detection abnormal area as the target abnormality information of the third target abnormal area.

[0102] Based on the above scheme, by dividing the abnormal areas identified by each method and the target abnormal information of the abnormal areas identified by each method, the recognition accuracy of each abnormal area and the comprehensiveness of the recognition of each abnormal area are improved.

[0103] Optionally, based on each target abnormal area and the target abnormality information of each target abnormal area, regional abnormality detection information of the port area is identified, including: based on the target abnormality information of each first target abnormal area, querying the abnormality type of each first target abnormal area in a database, and for each third target abnormal area, generating sub-area abnormality distribution information of the third target abnormal area based on the abnormality information of the intersecting abnormal area in the third target abnormal area, the abnormality type of the intersecting abnormal area, and the target abnormality information of the intersecting detected abnormal area in the third target abnormal area; for each second target abnormal area, based on the abnormality type of the second target abnormal area and the abnormality degree of each grid detection image of the second target abnormal area, generating sub-area abnormality distribution information of the second target abnormal area, and for each first target abnormal area, based on the abnormality type of the first target abnormal area and the abnormality degree of the first target abnormal area, generating sub-area abnormality distribution information of the first target abnormal area; and distributing and sorting the sub-area abnormality distribution information of each target abnormal area in the current port monitoring image of the port area to obtain regional abnormality detection information of the port area.

[0104] In this embodiment, the terminal queries the abnormality type of each first target abnormal area in the database based on the target abnormality information of each first target abnormal area, and for each third target abnormal area, generates the sub-area abnormal distribution information of the third target abnormal area by using the abnormality information of the intersecting abnormal area in the third target abnormal area, the abnormality type of the intersecting abnormal area, and the target abnormality information of the intersecting detection abnormal area in the third target abnormal area. Specifically, the terminal identifies the area range occupied by each abnormal degree of each abnormal type in the intersecting abnormal area, and then adds the identification information of the abnormal degree of the abnormal type corresponding to each area range in each area range. Obtain the sub-area abnormal distribution information of the intersecting abnormal area. Among them, the identification information can be a text identification, a symbol representation, a color representation, etc.

[0105] For each second target abnormal area, the terminal generates sub-area abnormal distribution information of the second target abnormal area based on the abnormal type of the second target abnormal area and the abnormal degree of each grid detection image of the second target abnormal area, and for each first target abnormal area, generates sub-area abnormal distribution information of the first target abnormal area based on the abnormal type of the first target abnormal area and the abnormal degree of the first target abnormal area. The specific generation process is the same as the generation method of the sub-area abnormal distribution information of the first target abnormal area.

[0106] Finally, the terminal sorts the sub-region abnormal distribution information of each target abnormal region in the current port monitoring image of the port area to obtain the regional abnormality detection information of the port area.

[0107] Based on the above scheme, the comprehensiveness and accuracy of the regional anomaly detection information of the port area are improved by distributing and sorting the target abnormal areas.

[0108] Optionally, when the regional detection type of the regional anomaly detection information is a risk anomaly type, an emergency response strategy for the port area is generated based on the regional detection type of the regional anomaly detection information, including: calculating all target abnormal areas in the port area, accounting for the abnormal proportion of the abnormal areas in the port area, and respectively calculating the target abnormal areas corresponding to each abnormal degree of each abnormal type, accounting for the sub-anomaly proportion of the abnormal areas in the port area; based on the sub-anomaly proportion values ​​of each abnormal degree of each abnormal type and the abnormal proportion value, querying the regional detection type of the port area in the database, and judging whether the regional detection type belongs to the risk anomaly type; when the anomaly type belongs to the risk anomaly type, based on the abnormal degree of each anomaly type, querying the emergency handling strategy corresponding to the abnormal degree of each anomaly type in the database, and using the emergency handling strategy corresponding to the abnormal degree of each anomaly type as the emergency handling strategy for the target abnormal area corresponding to the abnormal degree of each anomaly type; using the emergency handling strategies of all target abnormal areas as the emergency response strategies for the port area.

[0109] In this embodiment, the terminal calculates all target abnormal areas in the port area, the abnormal proportion of the abnormal area in the port area, and calculates the target abnormal area corresponding to each abnormal degree of each abnormal type, and the sub-abnormal proportion of the abnormal area in the port area. The proportion value is the ratio value obtained by dividing the area range of the abnormal area by the area range of the port area.

[0110] Based on the sub-anomaly ratio values ​​of each abnormal degree of each abnormal type and the abnormal ratio value, the terminal queries the regional detection type of the port area in the database and the abnormal database, and determines whether the regional detection type belongs to the risk abnormal type. Among them, the abnormal database includes the corresponding relationship between the sub-anomaly ratio value range of each abnormal degree of each abnormal type and the abnormal ratio value range and each regional detection type. The terminal identifies the regional detection type of the port area by querying the corresponding relationship.

[0111] When the abnormality type belongs to the risk abnormality type, the terminal queries the database for the emergency response strategy corresponding to the abnormality degree of each abnormality type based on the abnormality degree of each abnormality type. The corresponding relationship between the abnormality degree of each abnormality type and the emergency response strategy is stored in the database.

[0112] Finally, the terminal uses the emergency handling strategy corresponding to the abnormality degree of each abnormal type as the emergency handling strategy of the target abnormal area corresponding to the abnormality degree of each abnormal type, and uses the emergency handling strategies of all target abnormal areas as the emergency response strategy of the port area.

[0113] Based on the above scheme, the regional detection type of the port area is determined by calculating the sub-anomaly proportion value and the abnormal proportion value of each abnormality degree of each abnormality type, thereby improving the accuracy of determination, and then generating emergency handling strategies corresponding to the abnormality degree of each abnormality type respectively, ensuring that each target abnormal area can be responded to and handled in a timely manner, thereby improving the efficiency of emergency response processing.

[0114] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0115] Based on the same inventive concept, the embodiment of the present application also provides a port intelligent safety supervision and emergency response system for implementing the port intelligent safety supervision and emergency response method involved above. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the port intelligent safety supervision and emergency response system provided below can refer to the limitations of the port intelligent safety supervision and emergency response method above, and will not be repeated here.

[0116] Further references Figure 2 , as a response to the above Figure 1 The implementation of the method shown in the present application provides an embodiment of a port intelligent safety supervision and emergency response system 200, which includes an acquisition module 210, an analysis module 220, a generation module 230 and an iteration module 240, wherein:

[0117] An acquisition module 210 is used to acquire a current port monitoring image of each port area of ​​the port and a sensor detection image of each sensor detection type of each port area, and based on the current port monitoring image of each port area, identify each abnormal area of ​​each port area and abnormal information of each abnormal area through an image recognition network;

[0118] The analysis module 220 is used for identifying, for each port area, each abnormal detection area corresponding to each sensor detection image and abnormal detection information of each abnormal detection area based on the sensor detection images of each sensor detection type of the port area through an image recognition strategy corresponding to each sensor detection image, and analyzing each target abnormal area of ​​the port area and target abnormal information of each target abnormal area through each abnormal area of ​​the port area, each abnormal detection area of ​​the port area, abnormal information of each abnormal area and detection abnormal information of each abnormal detection area;

[0119] A generating module 230, configured to identify the regional anomaly detection information of the port area based on each of the target anomaly areas and the target anomaly information of each of the target anomaly areas, and generate an emergency response strategy for the port area based on the regional detection type of the regional anomaly detection information when the regional detection type of the regional anomaly detection information is a risk anomaly type;

[0120] The iteration module 240 is used to perform emergency response processing on the port based on the emergency response strategy of each port area, and when the emergency end operation of the staff is monitored, return to execute the step of obtaining the current port monitoring image of each port area of ​​the port and the sensor detection image of each sensor detection type of each port area.

[0121] Optionally, the acquisition module 210 is specifically configured to:

[0122] For each current port monitoring image, the current port monitoring image is subjected to grid splitting processing to obtain grid images of each region, and the image content type of each regional grid image is identified through an image recognition network;

[0123] Based on the image abnormality degree analysis strategy for each image content type, respectively identifying the region state corresponding to the region grid image of each image content type and the state degree of the region state of each region grid image;

[0124] The regional grid images of all regions whose states are abnormal are used as the abnormal sub-regions of the port region, and the abnormal sub-regions are clustered to obtain the abnormal regions of the port region;

[0125] The region states of each sub-abnormal region corresponding to each abnormal region and the state degrees of the region states of each sub-abnormal region corresponding to each abnormal region are used as the abnormality information of each abnormal region.

[0126] Optionally, the analysis module 220 is specifically used for:

[0127] For each sensor detection type, query the database for sample abnormal images of each abnormal degree of the sensor detection type, and grid-divide the sensor detection image of the sensor detection type into multiple grid detection images;

[0128] Using an image feature recognition network, the detection image features of each grid detection image and the sample image features of each sample abnormal image are identified, and the similarity between each detection image feature and each sample image feature is calculated using a similarity algorithm;

[0129] The sample image feature corresponding to each detection image feature and having a similarity greater than a similarity threshold is used as the target sample image feature corresponding to each detection image feature, and the abnormality degree corresponding to each target sample image feature is used as the abnormality degree of the grid detection image to which the detection image feature corresponding to each target sample image feature belongs;

[0130] Clustering is performed on each of the grid detection images to obtain each detection abnormality area corresponding to the sensor detection image, and the abnormality degrees of each grid detection image contained in each detection abnormality image are distributed and arranged to obtain detection abnormality information of each detection abnormality area.

[0131] Optionally, the analysis module 220 is specifically used for:

[0132] Identify, among the abnormal regions, an intersecting abnormal region that has an intersecting region with the detected abnormal region, and use the detected abnormal region that has the intersecting region as the intersecting detected abnormal region;

[0133] The abnormal area without the intersection area is used as the first target abnormal area, and the abnormal information of the first target abnormal area is used as the target abnormal information of the first target abnormal area, and the detected abnormal area without the intersection area is used as the second target abnormal area, and the detected abnormal information of the second target abnormal area is used as the target abnormal information of the second target abnormal area;

[0134] Identify the abnormality type corresponding to each sensor detection type, and based on the abnormality type corresponding to each sensor detection type, add an abnormality type identifier to the detection abnormality information of the detection abnormality area corresponding to each sensor detection type to obtain target abnormality information of each detection abnormality area;

[0135] Each intersection abnormal area and all areas included in each intersection detection abnormal area are used as the third target abnormal area, and the abnormality information of the intersection abnormal area and the target abnormality information of the intersection detection abnormal area are used as the target abnormality information of the third target abnormal area.

[0136] Optionally, the generating module 230 is specifically configured to:

[0137] Based on the target abnormality information of each of the first target abnormality areas, query the abnormality type of each of the first target abnormality areas in the database, and for each third target abnormality area, generate the sub-area abnormality distribution information of the third target abnormality area by combining the abnormality information of the intersecting abnormality area in the third target abnormality area, the abnormality type of the intersecting abnormality area, and the target abnormality information of the intersecting detection abnormality area in the third target abnormality area;

[0138] For each second target abnormal region, based on the abnormality type of the second target abnormal region and the abnormality degree of each grid detection image of the second target abnormal region, generate sub-region abnormality distribution information of the second target abnormal region, and for each first target abnormal region, based on the abnormality type of the first target abnormal region and the abnormality degree of the first target abnormal region, generate sub-region abnormality distribution information of the first target abnormal region;

[0139] The sub-region abnormal distribution information of each target abnormal region is sorted in the current port monitoring image of the port region to obtain the regional abnormality detection information of the port region.

[0140] Optionally, the generating module 230 is specifically configured to:

[0141] Calculate the abnormal proportion of all target abnormal areas in the port area to the abnormal areas in the port area, and respectively calculate the target abnormal areas corresponding to each abnormal degree of each abnormal type to the sub-abnormal proportion of the abnormal areas in the port area;

[0142] Based on the sub-anomaly proportion values ​​of each abnormal degree of each abnormal type and the abnormal proportion value, in the database, in the abnormal database, the regional detection type of the port area is queried, and it is determined whether the regional detection type belongs to the risk abnormal type;

[0143] When the abnormality type belongs to the risk abnormality type, based on the abnormality degree of each abnormality type, the emergency treatment strategy corresponding to the abnormality degree of each abnormality type is queried in the database, and the emergency treatment strategy corresponding to the abnormality degree of each abnormality type is used as the emergency treatment strategy of the target abnormal area corresponding to the abnormality degree of each abnormality type;

[0144] The emergency handling strategies for all target abnormal areas are used as the emergency response strategies for the port area.

[0145] Each module in the above-mentioned port intelligent safety supervision and emergency response system can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0146] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input system connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for intelligent safety supervision and emergency response of a port is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input system of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the shell of the computer device, or an external keyboard, touchpad or mouse.

[0147] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0148] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of any one of the methods in the first aspect are implemented.

[0149] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0150] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the steps of any one of the methods in the first aspect.

[0151] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0152] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0153] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0154] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A port intelligent safety supervision and emergency response method, characterized in that: The method comprises: Acquire a current port monitoring image of each port area of ​​the port, and a sensor detection image of each sensor detection type of each port area; For each current port monitoring image, the current port monitoring image is subjected to grid splitting processing to obtain grid images of each region, and the image content type of each regional grid image is identified through an image recognition network; Based on the image abnormality degree analysis strategy for each image content type, respectively identifying the region state corresponding to the region grid image of each image content type and the state degree of the region state of each region grid image; The regional grid images of all regions whose states are abnormal are used as the abnormal sub-regions of the port region, and the abnormal sub-regions are clustered to obtain the abnormal regions of the port region; Using the regional states of each sub-abnormal region corresponding to each abnormal region and the state degrees of the regional states of each sub-abnormal region corresponding to each abnormal region as the abnormal information of each abnormal region; For each sensor detection type in each port area, query the database for sample abnormal images of each abnormal degree of the sensor detection type, and grid-divide the sensor detection image of the sensor detection type into multiple grid detection images; Using an image feature recognition network, the detection image features of each grid detection image and the sample image features of each sample abnormal image are identified, and the similarity between each detection image feature and each sample image feature is calculated using a similarity algorithm; The sample image feature corresponding to each detection image feature and having a similarity greater than a similarity threshold is used as the target sample image feature corresponding to each detection image feature, and the abnormality degree corresponding to each target sample image feature is used as the abnormality degree of the grid detection image to which the detection image feature corresponding to each target sample image feature belongs; Performing clustering processing on each of the grid detection images to obtain each detection abnormality area corresponding to the sensor detection image, and performing distribution and arrangement processing on the abnormality degree of each grid detection image contained in each detection abnormality image to obtain detection abnormality information of each detection abnormality area; Identify, among the abnormal regions, an intersecting abnormal region that has an intersecting region with the detected abnormal region, and use the detected abnormal region that has the intersecting region as the intersecting detected abnormal region; The abnormal area without the intersection area is used as the first target abnormal area, and the abnormal information of the first target abnormal area is used as the target abnormal information of the first target abnormal area, and the detected abnormal area without the intersection area is used as the second target abnormal area, and the detected abnormal information of the second target abnormal area is used as the target abnormal information of the second target abnormal area; Identify the abnormality type corresponding to each sensor detection type, and based on the abnormality type corresponding to each sensor detection type, add an abnormality type identifier to the detection abnormality information of the detection abnormality area corresponding to each sensor detection type to obtain target abnormality information of each detection abnormality area; Taking each intersection abnormal area and all areas included in each intersection detection abnormal area as a third target abnormal area, and taking the abnormality information of the intersection abnormal area and the target abnormality information of the intersection detection abnormal area as the target abnormality information of the third target abnormal area; Based on each of the target abnormal areas and the target abnormal information of each of the target abnormal areas, identifying the regional abnormality detection information of the port area; Calculate the abnormal proportion of all target abnormal areas in the port area to the abnormal areas in the port area, and respectively calculate the target abnormal areas corresponding to each abnormal degree of each abnormal type to the sub-abnormal proportion of the abnormal areas in the port area; Based on the sub-anomaly proportion values ​​of each abnormal degree of each abnormal type and the abnormal proportion value, in the database, in the abnormal database, the regional detection type of the port area is queried, and it is determined whether the regional detection type belongs to the risk abnormal type; When the abnormality type belongs to the risk abnormality type, based on the abnormality degree of each abnormality type, the emergency treatment strategy corresponding to the abnormality degree of each abnormality type is queried in the database, and the emergency treatment strategy corresponding to the abnormality degree of each abnormality type is used as the emergency treatment strategy of the target abnormal area corresponding to the abnormality degree of each abnormality type; Using the emergency handling strategies of all target abnormal areas as the emergency response strategy of the port area; When the regional detection type of the regional anomaly detection information is a risk anomaly type, generating an emergency response strategy for the port area based on the regional detection type of the regional anomaly detection information; Based on the emergency response strategy of each port area, emergency response processing is performed on the port, and when the emergency end operation of the staff is monitored, the step of returning to execute the current port monitoring image of each port area of ​​the port and the sensor detection image step of each sensor detection type of each port area is obtained.

2. The method according to claim 1, characterized in that The identifying the regional anomaly detection information of the port area based on each of the target anomaly areas and the target anomaly information of each of the target anomaly areas includes: Based on the target abnormality information of each of the first target abnormality areas, query the abnormality type of each of the first target abnormality areas in the database, and for each third target abnormality area, generate the sub-area abnormality distribution information of the third target abnormality area by combining the abnormality information of the intersecting abnormality area in the third target abnormality area, the abnormality type of the intersecting abnormality area, and the target abnormality information of the intersecting detection abnormality area in the third target abnormality area; For each second target abnormal region, based on the abnormality type of the second target abnormal region and the abnormality degree of each grid detection image of the second target abnormal region, generate sub-region abnormality distribution information of the second target abnormal region, and for each first target abnormal region, based on the abnormality type of the first target abnormal region and the abnormality degree of the first target abnormal region, generate sub-region abnormality distribution information of the first target abnormal region; The sub-region abnormal distribution information of each target abnormal region is sorted in the current port monitoring image of the port region to obtain the regional abnormality detection information of the port region.

3. A port intelligent safety supervision and emergency response system, characterized in that: The system comprises: An acquisition module is used to acquire the current port monitoring image of each port area of ​​the port and the sensor detection images of each sensor detection type of each port area; for each current port monitoring image, the current port monitoring image is subjected to grid splitting processing to obtain grid images of each area, and the image content type of each regional grid image is identified through an image recognition network; based on the image abnormality degree analysis strategy of each image content type, the regional state corresponding to the regional grid image of each image content type and the state degree of the regional state of each regional grid image are respectively identified; all regional grid images whose regional states belong to abnormal states are taken as sub-abnormal areas of the port area, and the sub-abnormal areas are clustered to obtain abnormal areas of the port area; the regional state of each sub-abnormal area corresponding to each abnormal area and the state degree of the regional state of each sub-abnormal area corresponding to each abnormal area are taken as abnormal information of each abnormal area; The analysis module is used to query the sample abnormal images of each abnormal degree of each sensor detection type in each port area in the database, and grid the sensor detection images of the sensor detection type into multiple grid detection images; identify the detection image features of each grid detection image and the sample image features of each sample abnormal image through the image feature recognition network, and calculate the similarity between each detection image feature and each sample image feature through the similarity algorithm; use the sample image features corresponding to each detection image feature with a similarity greater than the similarity threshold as the target sample image feature corresponding to each detection image feature, and use the abnormal degree corresponding to each target sample image feature as the abnormal degree of the grid detection image to which the detection image feature corresponding to each target sample image feature belongs; perform clustering processing on each of the grid detection images to obtain each detection abnormal area corresponding to the sensor detection image, and perform distribution and arrangement processing on the abnormal degrees of each grid detection image contained in each detection abnormal image to obtain the detection abnormality of each detection abnormal area. Information; identifying, among the abnormal areas, the intersecting abnormal areas that have an intersection with the detected abnormal area, and taking the detected abnormal area with the intersection as the intersecting detected abnormal area; taking the abnormal area without an intersection as the first target abnormal area, taking the abnormal information of the first target abnormal area as the target abnormal information of the first target abnormal area, and taking the detected abnormal area without an intersection as the second target abnormal area, taking the detected abnormal area as the target abnormal information of the second target abnormal area; identifying the abnormal type corresponding to each sensing detection type, and based on the abnormal type corresponding to each sensing detection type, adding an abnormal type identifier to the detected abnormal information of the detected abnormal area corresponding to each sensing detection type to obtain the target abnormal information of each detected abnormal area; taking each intersecting abnormal area and all areas contained in each intersecting detected abnormal area as the third target abnormal area, and taking the abnormal information of the intersecting abnormal area and the target abnormal information of the intersecting detected abnormal area as the target abnormal information of the third target abnormal area; A generation module is used to identify the regional anomaly detection information of the port area based on each of the target anomaly areas and the target anomaly information of each of the target anomaly areas; calculate the anomaly proportion values ​​of all the target anomaly areas of the port area to the anomaly areas of the port area, and respectively calculate the sub-anomaly proportion values ​​of the target anomaly areas corresponding to each anomaly degree of each anomaly type to the anomaly areas of the port area; based on the sub-anomaly proportion values ​​of each anomaly degree of each anomaly type and the anomaly proportion values, query the regional detection type of the port area in the database and the anomaly database, and determine whether the regional detection type belongs to the risk anomaly type; when the anomaly type belongs to the risk anomaly type, based on the anomaly degree of each anomaly type, query the emergency treatment strategy corresponding to the anomaly degree of each anomaly type in the database, and use the emergency treatment strategy corresponding to the anomaly degree of each anomaly type as the emergency treatment strategy of the target anomaly area corresponding to the anomaly degree of each anomaly type; use the emergency treatment strategy of all target anomaly areas as the emergency response strategy of the port area; The iterative module is used to perform emergency response processing on the port based on the emergency response strategy of each port area, and when the emergency end operation of the staff is monitored, return to execute the step of obtaining the current port monitoring image of each port area of ​​the port and the sensor detection image of each sensor detection type of each port area.

4. The system according to claim 3, characterized in that The generating module is specifically used for: Based on the target abnormality information of each of the first target abnormality areas, query the abnormality type of each of the first target abnormality areas in the database, and for each third target abnormality area, generate the sub-area abnormality distribution information of the third target abnormality area by combining the abnormality information of the intersecting abnormality area in the third target abnormality area, the abnormality type of the intersecting abnormality area, and the target abnormality information of the intersecting detection abnormality area in the third target abnormality area; For each second target abnormal region, based on the abnormality type of the second target abnormal region and the abnormality degree of each grid detection image of the second target abnormal region, generate sub-region abnormality distribution information of the second target abnormal region, and for each first target abnormal region, based on the abnormality type of the first target abnormal region and the abnormality degree of the first target abnormal region, generate sub-region abnormality distribution information of the first target abnormal region; The sub-region abnormal distribution information of each target abnormal region is sorted in the current port monitoring image of the port region to obtain the regional abnormality detection information of the port region.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 2 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 2 are implemented.

7. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 2 are implemented.

Images