CN-122002058-A - Ship lock monitoring multimedia linkage method and electronic equipment

CN122002058ACN 122002058 ACN122002058 ACN 122002058ACN-122002058-A

Abstract

The invention relates to the technical field of navigation operation safety guarantee, in particular to a ship lock monitoring multimedia linkage method and electronic equipment. The invention relates to the technical field of computer vision and target detection, in particular to a method and a terminal for automatically generating training data of a target detection model. The method comprises the steps of receiving a measuring point signal, determining a target scene related to the action of the measuring point signal based on a preset mapping relation, acquiring a target video data stream for presenting the target scene, pushing the target video data stream to a display terminal for display, realizing automatic and accurate linkage of a monitoring video, reducing manual operation links and response time, and improving automation level and emergency disposal efficiency of ship lock monitoring.

Inventors

HUANG KAI
Lin Shanyin
CHEN CHENG
SHEN GUOJIAN
ZHANG SHUYUE
He Congyi
Gao Yidian
LIU QINGYONG
ZHU YANHUA
LIAO WANYING
WANG KAICHENG
Qing Bite

Assignees

国网福建省电力有限公司
福建水口发电集团有限公司

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (10)

1.A method for monitoring multimedia linkage of a ship lock, comprising: receiving a measuring point signal; determining a target scene associated with the action of the measuring point signal based on a preset mapping relation; And acquiring a target video data stream for presenting the target scene, and pushing the target video data stream to a display terminal for display.
2. The ship lock monitoring multimedia linkage method according to claim 1, further comprising: Determining broadcast audio associated with the target scene; pushing the broadcast audio to a broadcasting device for playing.
3. The method according to claim 1, wherein the determining a target scene associated with the action of the station signal based on a preset mapping relation comprises: inquiring the mapping relation according to the measuring point signals to obtain candidate scenes; Acquiring user authority information; And screening the candidate scenes according to the user permission information, and determining the scenes obtained after screening as the target scenes.
4. The ship lock monitoring multimedia linkage method according to claim 1, further comprising, after pushing the target video data stream to a display terminal for display: image analysis is carried out on the target video data stream so as to identify whether a target risk object exists in a preset risk area; If the target risk object is identified, triggering an alarm operation, wherein the alarm operation comprises at least one of the following operations: The display terminal is touched with an audible and visual alarm; Controlling broadcasting equipment to play warning voice corresponding to the target risk object; And performing screenshot archiving on the image containing the target risk object.
5. The ship lock monitoring multimedia linkage method according to claim 4, wherein pushing the target video data stream to a display terminal for display comprises: dividing the display terminal into a plurality of display areas; pushing the target video data stream to a display area corresponding to the target scene for display.
6. A ship lock monitoring multimedia linkage method according to claim 4, characterized by further comprising: Acquiring ship passing images under different target scenes, different working states and different time periods, and marking target types in the ship passing images, wherein the target types comprise personnel, ships, gate equipment, foreign matters and risk areas so as to construct a training data set; Acquiring a public data set and a ship lock general data set, and pre-training an initial target detection model to obtain a pending target detection model; training the target detection model by adopting the training data set, and obtaining an optimized target detection model by taking the minimum error between the target type identified by the target detection model and the target type marked by the target detection model as a target; The image analysis of the target video data stream comprises: And carrying out image analysis on the target video data stream through the optimized target detection model.
7. The method for monitoring a ship lock according to claim 6, wherein, the training the pending target detection model using the training dataset includes: Dividing the whole target area into a plurality of subareas, and acquiring a training data subset corresponding to each subarea; Fixing basic parameters of a to-be-determined target detection model, and sequentially carrying out parameter adjustment on the to-be-determined target detection model based on each training data subset; Combining the training data of the sub-areas to obtain continuous multi-frame images serving as the intensified training data of the to-be-determined target monitoring model; And merging the training data of all the subareas, and carrying out overall training on the undetermined target detection model according to all the training data.
8. The method for monitoring a ship lock according to claim 6, wherein, the training the pending target detection model using the training dataset includes: According to a preset difficult case judgment standard, a difficult case sample is screened out from a training set based on the reasoning result of the target detection model to be determined; checking the screened difficult sample to obtain a checked effective difficult sample; Mixing the effective difficult sample with a conventional sample according to a preset proportion, constructing a training subset, preferentially loading the effective difficult sample for training, and distributing a loss function weight higher than that of the conventional sample for the effective difficult sample; After the training of the target detection model reaches the preset iteration times, returning to execute the step of screening the difficult sample from the training set based on the reasoning result of the target detection model by using the updated target detection model according to the preset difficult judgment standard, removing the old difficult sample from the training subset, updating the training subset, and carrying out the next round of model training.
9. The method for monitoring a ship lock according to claim 6, further comprising, after obtaining the optimized target detection model: Loading the optimized target detection model, and freezing a network layer close to an output end in a backbone network of the optimized target detection model; Carrying out channel-based structured pruning on a non-backbone network part of the optimization target detection model; And carrying out parameter adjustment on the optimized target detection model after pruning by using the training data set.
10. An electronic device comprising a memory and a processor, the memory storing a computer program, wherein the processor, when executing the computer program, implements a lock monitoring multimedia linkage method as claimed in any one of claims 1 to 9.

Description

Ship lock monitoring multimedia linkage method and electronic equipment Technical Field The invention relates to the technical field of navigation operation safety guarantee, in particular to a ship lock monitoring multimedia linkage method and electronic equipment. Background With the rapid development of inland and offshore shipping industry, the throughput of water transportation is greatly improved, various ship locks bear huge pressure of channel management, safety and quick passing, so that a more perfect monitoring means is needed to ensure operation safety, the prior art scheme mainly adopts a mode of paired display of a camera and a monitor to synchronously display a lock chamber and nearby conditions to a monitoring hall, but the scheme is only suitable for single-stage single-line ship locks, video linkage pictures are insufficient when multi-stage multi-line ship locks are combined to operate, dead angles exist in monitoring, thereby threatening navigation safety, and meanwhile, the prior art scheme does not have a sound prompt function, can not provide enough warning for operators and ships, and further increases operation risks. Disclosure of Invention The invention aims to solve the technical problem of providing a ship lock monitoring multimedia linkage method, a device and electronic equipment, and realizing multistage multi-line ship lock video linkage. In order to solve the technical problems, the invention adopts a technical scheme that: A ship lock monitoring multimedia linkage method, comprising: receiving a measuring point signal; determining a target scene associated with the action of the measuring point signal based on a preset mapping relation; And acquiring a target video data stream for presenting the target scene according to the target scene, and pushing the target video data stream to a display terminal for display. In order to solve the technical problems, the invention adopts another technical scheme that: an electronic device comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of a lock monitoring multimedia linkage method as claimed in the preceding claims when the computer program is executed. The invention has the beneficial effects that: The application realizes the acquisition of the real-time data of the ship lock running state by receiving the measuring point signals. And determining a target scene associated with the measuring point signal action based on a preset mapping relation, and mapping the discrete measuring point event to a specific monitoring scene through a predefined logic association. And acquiring a corresponding target video data stream according to the target scene and pushing the corresponding target video data stream to the display terminal, so that automatic scheduling and presentation of video resources are realized. By establishing the mapping relation between the action of the measuring point signal and the monitoring scene, when a specific measuring point signal is received, the target scene video stream associated with the specific measuring point signal can be automatically and accurately positioned and called for display. Compared with the mode of manually searching and switching the monitoring picture in the related art, the method realizes automatic and accurate linkage of the monitoring video, reduces manual operation links and response time, and improves the automation level and emergency disposal efficiency of ship lock monitoring. Drawings FIG. 1 is a flow chart of steps of a method for monitoring multimedia linkage of a ship lock according to an embodiment of the present invention; fig. 2 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. Detailed Description In order to make the technical problems, technical schemes and beneficial effects to be solved more clear, the application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application. The relevant terms in the present invention are presented below: UDP-User Datagram Protocol user datagram protocol. IMC INTELLIGENT MONITORING AND CONTROL, intelligent monitoring and control. MySQL is an open source relational database management system. RTSP, REAL TIME STREAMING Protocol, real-time streaming Protocol. H.264/h.265 video coding standard (h.264 is advanced video coding and h.265 is efficient video coding). WebSocket, a network communication protocol, supports full duplex communication. HTTP Hypertext Transfer Protocol, hypertext transfer protocol. RTMP Real-TIME MESSAGING Protocol, real-time message Protocol. AI ARTIFICIAL INTELLIGENCE, artificial intelligence. IoU: intersection over Union, cross ratio, used for measuring the overlapping degree of the prediction