CN-122001999-A - Oilfield video monitoring and early warning method, device and storage medium

CN122001999ACN 122001999 ACN122001999 ACN 122001999ACN-122001999-A

Abstract

The invention discloses an oilfield video monitoring and early warning method, which comprises the following steps of acquiring a monitoring video shot by monitoring equipment connected with an analysis server in real time, identifying abnormal behaviors in the monitoring video, determining the type of an abnormal event, generating alarm data according to the type of the abnormal event, the occurrence time and the place of the abnormal behaviors, and sending the alarm data to a service server, wherein the alarm data are used for indicating the service server to determine an early warning level and sending early warning information to a user level corresponding to the early warning level. The invention also discloses an oilfield video monitoring and early warning device and a storage medium. The invention solves the problems that the number of video monitoring devices in the oilfield video monitoring system is large and the service servers cannot be adapted in the prior art.

Inventors

MA SHIQING
HUANG TIANHU
CHEN PENG
LI ERYANG
WANG ZHONGHAI
REN YANBING
ZHANG RUI
HE ZHANYOU
XU BIN
GAO LEI
HUANG BIN
ZHANG LIANG
REN XIAORONG

Assignees

中国石油天然气股份有限公司

Dates

Publication Date: 20260508
Application Date: 20241101

Claims (10)

1. The oilfield video monitoring and early warning method is characterized by comprising the following specific steps of: Acquiring a monitoring video shot by monitoring equipment connected with an analysis server in real time; Identifying abnormal behaviors in the monitoring video and determining the type of an abnormal event; generating alarm data according to the type of the abnormal event, the occurrence time and the place of the abnormal behavior; and sending alarm data to the service server, wherein the alarm data is used for indicating the service server to determine the early warning level and sending early warning information to a user level corresponding to the early warning level.
2. The oilfield video monitoring and early warning method of claim 1, wherein identifying abnormal behavior in the monitoring video and determining an abnormal event type comprises: selecting a key frame in the monitoring video; judging whether each behavior in the key frame belongs to abnormal behavior; matching the abnormal behavior with each event type in the abnormal type database, and determining the abnormal event type; the anomaly type database comprises image data of unsafe behaviors of field personnel and unsafe states of equipment of an oilfield enterprise.
3. The oilfield video monitoring and early warning method according to claim 2, wherein the selecting the key frame in the monitoring video comprises: Performing optical flow analysis on each frame of the monitoring video to obtain a motion trail of each moving target; Constructing a motion vector based on the motion trail of each moving object; mapping the motion vector as a function of time; and taking a frame at the moment corresponding to the minimum value of the function as a key frame.
4. The oilfield video monitoring and early warning method according to claim 3, wherein selecting key frames in the monitoring video before performing optical flow analysis on each frame of the monitoring video further comprises: Identifying the outline of the object and the figure in each frame of the monitoring video; The image of the object and the person detected in the fixed time interval T is cut out, and the image is converted into fixed size and recorded in the original image Where { x i ,y i } represents the point in the upper left corner of the identification box and { w i ,h i } represents the width and height of the identification box.
5. The oilfield video monitoring and pre-warning method of claim 4, further comprising: Intercepting video clips with abnormal behaviors; transmitting and intercepting video clips with abnormal behaviors to a training server for identification; and acquiring a training result of the training server, and updating and identifying.
6. The oilfield video monitoring and early warning method according to claim 5, wherein the intercepting the video segment in which the abnormal behavior occurs comprises: starting to intercept from the key frame with abnormal behavior; and when the key frame for identifying the end of the abnormal behavior is reached or the preset time is reached, cutting off the video to form a video fragment.
7. The oilfield video monitoring and pre-warning method of claim 6, further comprising: binding information is acquired, wherein the binding information comprises identity information and calculation force request of video monitoring equipment; Establishing a data communication link with the video monitoring equipment according to the identity information; And distributing computing power to the video monitoring equipment according to the computing power request.
8. The oilfield video monitoring pre-warning method of claim 7, further comprising: acquiring a monitoring requirement; The method comprises the steps of sending a monitoring demand to a storage server, wherein the monitoring demand is used for indicating the storage server to send an identification algorithm corresponding to the monitoring demand to an analysis server; and acquiring the identification algorithm and configuring.
9. Oilfield video monitoring and early warning device, its characterized in that includes: The acquisition module is used for acquiring the monitoring video shot by the monitoring equipment connected with the analysis server in real time; The identification module is used for identifying abnormal behaviors in the monitoring video by adopting an identification algorithm and determining the type of the abnormal event; the generation module is used for generating corresponding alarm data according to the type of the abnormal event, the occurrence time and the place of the abnormal behavior; the sending module is used for sending the alarm data to the service server, wherein the alarm data is used for indicating the service server to determine the early warning level and sending early warning information to a user level corresponding to the early warning level.
10. A computer readable storage medium, wherein the computer readable storage medium stores computer readable instructions that, when executed by a processor, implement the oilfield video surveillance method of any one of claims 1-8.

Description

Oilfield video monitoring and early warning method, device and storage medium Technical Field The invention belongs to the technical field of video monitoring, in particular to an oilfield video monitoring and early warning method, and further relates to an oilfield video monitoring and early warning device and a storage medium. Background The oilfield video monitoring system is one of key technologies in the safety production management of the oil and gas industry, and utilizes modern information technology to realize real-time monitoring and management of oil production stations, metering stations, combined stations and the like. And by integrating the video monitoring system and the informatization platform in the oil field, a unified monitoring and management platform is constructed, and the efficiency and effect of safety management of the oil field are improved. However, the video processing processes of the current oilfield video monitoring system such as video analysis, abnormal behavior identification and early warning information generation are all carried out on a service server, a large number of video monitoring devices are arranged in the oilfield video monitoring system, a large amount of video data is generated at each moment, the service server is required to have a relatively strong computing capacity, and after the number of the video monitoring devices is continuously increased to exceed the processing capacity of the service server, the service server is required to be expanded and upgraded. Disclosure of Invention The invention aims to provide an oilfield video monitoring and early warning method, which solves the problems that in the prior art, the number of video monitoring devices in an oilfield video monitoring system is large and a service server cannot be adapted. The invention further aims to provide an oilfield video monitoring and early warning device and a storage medium. The first technical scheme adopted by the invention is that the oilfield video monitoring and early warning method comprises the following steps: Acquiring a monitoring video shot by monitoring equipment connected with an analysis server in real time; adopting a pre-configured recognition algorithm to recognize abnormal behaviors in the monitoring video and determining the type of the abnormal event; generating alarm data according to the type of the abnormal event, the occurrence time and the place of the abnormal behavior; and sending alarm data to the service server, wherein the alarm data is used for indicating the service server to determine the early warning level and sending early warning information to a user level corresponding to the early warning level. The first aspect of the present invention is also characterized in that, Identifying abnormal behavior in the monitoring video and determining the type of the abnormal event comprises the following steps: selecting a key frame in the monitoring video; judging whether each behavior in the key frame belongs to abnormal behavior; matching the abnormal behavior with each event type in the abnormal type database, and determining the abnormal event type; The anomaly type database contains image data of unsafe behaviors of field personnel and unsafe states of equipment of the oilfield enterprise. The selecting of the key frames in the monitoring video comprises the following steps: Performing optical flow analysis on each frame of the monitoring video to obtain a motion trail of each moving target; Constructing a motion vector based on the motion trail of each moving object; mapping the motion vector as a function of time; and taking a frame at the moment corresponding to the minimum value of the function as a key frame. Before optical flow analysis is performed on each frame of the monitoring video, selecting a key frame in the monitoring video further includes: Identifying the outline of the object and the figure in each frame of the monitoring video; The image of the object and the person detected in the fixed time interval T is cut out, and the image is converted into fixed size and recorded in the original image Where { x i,yi } represents the point in the upper left corner of the identification box and { w i,hi } represents the width and height of the identification box. The method further comprises the steps of: Intercepting video clips with abnormal behaviors; transmitting and intercepting video clips with abnormal behaviors to a training server for identification; and acquiring a training result of the training server, and updating and identifying. Intercepting the video clips in which the abnormal behavior occurs comprises: starting to intercept from the key frame with abnormal behavior; and when the key frame for identifying the end of the abnormal behavior is reached or the preset time is reached, cutting off the video to form a video fragment. The method further comprises the steps of: binding information is acquired, wherein the binding info