CN115603466B - Ship shore power system based on artificial intelligence visual identification - Google Patents
Ship shore power system based on artificial intelligence visual identification Download PDFInfo
- Publication number
- CN115603466B CN115603466B CN202211598566.XA CN202211598566A CN115603466B CN 115603466 B CN115603466 B CN 115603466B CN 202211598566 A CN202211598566 A CN 202211598566A CN 115603466 B CN115603466 B CN 115603466B
- Authority
- CN
- China
- Prior art keywords
- ship
- shore power
- intelligent
- port
- algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000000007 visual effect Effects 0.000 title claims abstract description 25
- 238000013473 artificial intelligence Methods 0.000 title claims description 14
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000004891 communication Methods 0.000 claims abstract description 8
- 238000001514 detection method Methods 0.000 claims description 70
- 239000011159 matrix material Substances 0.000 claims description 30
- 238000004364 calculation method Methods 0.000 claims description 23
- 230000033001 locomotion Effects 0.000 claims description 22
- 238000012360 testing method Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000003064 k means clustering Methods 0.000 claims description 5
- 239000013307 optical fiber Substances 0.000 claims description 4
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000012544 monitoring process Methods 0.000 description 14
- 238000000034 method Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 101150012579 ADSL gene Proteins 0.000 description 1
- 102100020775 Adenylosuccinate lyase Human genes 0.000 description 1
- 108700040193 Adenylosuccinate lyases Proteins 0.000 description 1
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00001—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00006—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00006—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment
- H02J13/00022—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment using wireless data transmission
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00006—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment
- H02J13/00022—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment using wireless data transmission
- H02J13/00026—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment using wireless data transmission involving a local wireless network, e.g. Wi-Fi, ZigBee or Bluetooth
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/7072—Electromobility specific charging systems or methods for batteries, ultracapacitors, supercapacitors or double-layer capacitors
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02T90/10—Technologies relating to charging of electric vehicles
- Y02T90/12—Electric charging stations
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Human Computer Interaction (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a ship shore power system based on artificial intelligent visual identification, which comprises a camera, an intelligent charging pile, an intelligent gateway, a data switch, an intelligent algorithm server and an intelligent shore power platform, wherein the camera is connected with the intelligent charging pile; the intelligent charging pile and the camera are connected with the data switch through the intelligent gateway; the data exchanger is connected with the intelligent algorithm server, and the data exchanger is connected with the intelligent shore power platform; the intelligent shore power platform is in communication connection with a traffic management shore power management terminal, a port operation management terminal and a ship mobile terminal, the invention recognizes and analyzes the ship state of berthing at a wharf through an artificial intelligent visual recognition analysis technology, monitors the berthing state of the ship in real time, detects the access state of shore power equipment and the ship in real time, warns and reminds the ship which is not accessed to the shore power according to the specified time, can prompt and reminds the berthing of the ship to access to the shore power, improves the low utilization rate of a ship shore power system, and further improves the utilization rate of the shore power equipment.
Description
Technical Field
The invention belongs to the technical field of ship shore power systems, and particularly relates to a ship shore power system based on artificial intelligent visual identification.
Background
The port shore power facilities are widely used in ports as important components for port new energy transformation and construction, and when ships are berthed, the ships are required to be connected into a shore power system according to regulations of traffic departments at all levels, and various operations are performed by using the shore power system to supply power so as to reduce carbon emission when the ships lean against ports.
Patent number CN102333203a discloses a shore power wireless video monitoring system, which is mainly aimed at performing remote video monitoring on the state of a shore power box during movement through wireless video monitoring, so as to realize remote operation without monitoring the ship berthing state; the patent number CN106026046B discloses a marine shore power safety system based on visual Internet of things and a marine shore power compensation method, which utilize various visual sensors of ultraviolet, visible light and infrared lamps to monitor shore power equipment, and the main purpose of the monitoring is that the health and safety state (based on temperature) of the shore power equipment are monitored, so that the power compensation of the shore power is realized, and whether the shore power is effectively accessed and used according to the stipulation is not monitored; patent number CN106773979B discloses a control system and method suitable for power supply and information of a ship shore power system, which mainly collects voltage and frequency information of shore power in the shore power information collection, is used for performing shore power closing control, and does not collect use information of shore power equipment.
The existing ship shore power system can remotely monitor shore power, acquire information, but does not have the function of automatically detecting whether a port-leaning ship is connected to the shore power system or not, when a plurality of ships are in port-leaning operation at present, the shore power system is not connected to the shore power system according to regulations, but the existing ship shore power system can only provide power supply for the ships, and can not judge whether the ship is connected to the shore power after berthing, so that the utilization rate of the ship shore power system is low, and shore power equipment is idle.
Disclosure of Invention
The invention aims to provide a ship shore power system based on artificial intelligence visual identification, which aims to solve the problems in the background technology.
In order to achieve the above purpose, the present invention provides the following technical solutions: a ship shore power system based on artificial intelligent visual identification comprises a camera, an intelligent charging pile, an intelligent gateway, a data switch, an intelligent algorithm server and an intelligent shore power platform;
the intelligent charging pile and the camera are connected with the data switch through the intelligent gateway;
the intelligent algorithm server is connected with the data switch and automatically recognizes and records ship identity information, a port entering state, a berthing time and a ship shore power access state through the camera;
The data exchanger is connected with the intelligent shore power platform;
the intelligent shore power platform is in communication connection with a traffic management shore power management terminal, a port operation management terminal and a ship mobile terminal and is used for transmitting shore power use data to the traffic management shore power management terminal, the port operation management terminal and the ship mobile terminal.
Preferably, the intelligent charging pile is arranged at a port and a dock where the ship is berthed, the intelligent charging pile is used for providing a shore power function for the ship berthed at the port and the dock, and transmitting the shore power use data of the ship berthed at the port and the dock to the data exchange through the intelligent gateway.
Preferably, the cameras are arranged at ports and docks corresponding to the intelligent charging piles, and the cameras are used for collecting images or video information of ships entering the ports and docks in real time and transmitting the images or video information to the intelligent algorithm server through the intelligent gateway and the data switch.
Preferably, the intelligent algorithm server is internally provided with a ship target recognition algorithm and a ship target tracking algorithm, the ship target recognition algorithm is used for recognizing and analyzing ships in the image or video information transmitted by the camera, the ship target tracking algorithm is used for carrying out target tracking on the ships in the image or video information transmitted by the camera, and the intelligent algorithm server transmits the ship identity information, the harbor entering state, the berthing time and the ship shore power access state recognized and recorded by the ship target recognition algorithm and the ship target tracking algorithm to the data switch.
Preferably, the data exchange is used for transmitting the received ship identity information, the port entering state, the berthing time, the ship shore power access state and the ship shore power use data to the intelligent shore power platform.
Preferably, the intelligent shore power platform is used for deploying a cloud end, storing the received ship identity information, the port entering state, the berthing time, the ship shore power access state and the ship shore power use data, pushing the ship shore power use data to the traffic management shore power management terminal, the port operation management terminal and the corresponding ship mobile terminal, and sending warning information to the ship which is not accessed to the shore power system beyond the specified time.
Preferably, the intelligent gateway is one or more of a 3G network, a 4G network, a 5G network, WIFI or network cable optical fibers.
A control method of a ship shore power system based on artificial intelligence visual identification comprises the following steps:
identifying and tracking the ship by port: shooting image or video information of a ship target entering a port through a camera arranged at a port and a dock, transmitting the image or video information to an intelligent algorithm server through an intelligent gateway and a data switch, carrying out identity recognition analysis on the ship target in the image or video information transmitted by the camera by a ship target recognition algorithm built in the intelligent algorithm server so as to obtain identity information of the ship target, and tracking the ship target in the image or video information transmitted by the camera by a ship target tracking algorithm built in the intelligent algorithm server so as to obtain a port entering state, a port stopping time and a ship shore power access state of the ship target, and transmitting the recognized and recorded ship identity information, port entering state, stopping time and ship shore power access state to the data switch;
And (3) identifying a shore power access state: recording the time of a ship target entering a port berth through a camera and an intelligent algorithm server, timing the ship target berthing time, judging whether the ship target is accessed to shore power according to the use state of an intelligent charging pile corresponding to the port berth of the ship target after the ship target berthing time reaches the specified access shore power time, and transmitting the ship shore power use data berthed at the port berthing to the data switch through the intelligent gateway;
reporting and warning ship target information which is not accessed into a shore power system according to the specified time: when the ship target berthing time exceeds the specified time and is not connected to the shore power system, the data exchange machine reports the ship target information to the intelligent shore power platform, the intelligent shore power platform pushes the ship target information to the traffic management shore power management terminal and the port operation management terminal, and the ship information is broadcast through the port and the corresponding ship mobile terminal to send out warning and reminding;
shore power utilization rate: counting each berth of a port, the number of berthed ships at the berth and the number of times of using shore power according to a specified rule through a camera, an intelligent algorithm server and an intelligent charging pile, and analyzing the utilization rate of the shore power of the port;
Counting illegal ships and setting a blacklist of illegal ships: and counting the ship information which is illegal to access the shore power, wherein when the number of times that the ship is illegal to access the shore power exceeds the set number of times, the intelligent shore power platform can set a blacklist for the ship, and when the ship enters a port to stop, warning prompt is immediately sent out through port broadcasting.
Preferably, the ship target recognition algorithm in the port-by-port recognition tracking of the ship improves the ship target recognition by a K-mean clustering algorithm on the basis of a YOLO v4 algorithm, and the specific calculation steps of the ship target recognition algorithm are as follows:
establishing ship and berth data sets through ship and port berth images acquired in real time by a camera, marking the data sets, randomly extracting the data sets, dividing the data sets into training sets and test sets, and carrying out ship target recognition training and testing through a YOLO v4 algorithm;
the K-means clustering algorithm clusters the data set, and the YOLO v4 algorithm is evaluated through two evaluation indexes of mAP and F1-score;
the calculation formula of mAP is shown in formula (1):
wherein ,X1 Represents the integral lower limit, X 2 Representing an upper integral limit;
p represents Precision, rec represents Recall rate Recall, d represents differential sign, and no variable is made;
mAP represents the average value of average precision AP of the ship and berth detection targets, and when the precision P is evaluated, the integral is carried out from 0 to 1, and mAP=AP;
the calculation formula of the average precision AP of the ship and berth detection targets is shown as formula (2):
the calculation formula of F1-score is shown in formula (3):
f1_score is an index for measuring the quality of a detection model, and gives consideration to Precision and Recall, and is a harmonic mean of the Precision and Recall;
precision is the detection result index of each category, namely, the percentage of the ship and berth is correctly detected in all detection areas, and the calculation formula is shown as formula (4):
the Recall ratio Recall is the percentage of the ship and berth correctly detected in all detection results, and the calculation formula is shown in formula (5):
TP is the number of accurately detected ship and berth detection targets;
FP is the number of detected ship and berth targets detected in the background;
FN is the number of detected ship as background and berth detection targets.
Preferably, the ship target tracking algorithm in the port-by-port identification tracking of the ship improves the ship target tracking by a deep algorithm on the basis of the SORT algorithm, and the specific steps of the SORT algorithm are as follows:
Prediction model: the SORT algorithm distributes an identity ID to the tracked ship target, associates the identity ID with the next video frame, performs motion modeling, and approximates the frame-to-frame displacement of each object by using a linear constant speed model which is completely independent of the motion of other objects and cameras;
and (3) data association: when the detection results are distributed to the existing ship targets, the boundary frame geometric shape of each ship target is estimated by predicting the new position of each ship target in the current frame, the distribution cost matrix is calculated as the intersection ratio IOU distance between each detection result and all the prediction frames of the existing ship targets, the distribution cost matrix is optimally solved by adopting the Hungary algorithm, and the distribution that the overlapping of the detection result and the ship target is smaller than the minimum value of the intersection ratio is refused;
creation and deletion of track identifications: when a ship target in an image enters or leaves the image in a certain frame, an identity ID (identity) of the tracked ship target needs to be established or removed, any detection result with the overlap smaller than the minimum value of the cross ratio indicates that an untracked ship target exists, and a new tracked ship target identity ID needs to be established for the untracked ship target;
the Deepsort algorithm comprises the following specific steps:
Deepsort: the Deepsort algorithm assumes the environment in which the tracked vessel target is located on a state space (u v q h x y r h) containing the center position (u, v) of the bounding box, the aspect ratio q, the height h, and the respective velocity information of these parameters in the image coordinate system, and uses a standard kalman filter with constant velocity motion and a linear observation model, wherein the boundary coordinates (u, v, q, h) are used as direct observations of the object state;
matching problem: the Deepsort algorithm comprehensively considers the motion information and the appearance information of the ship target, solves the association problem, and uses the Mars distance between the Kalman prediction result and the detection result of the existing motion ship target motion state to associate the motion information, wherein the Mars distance calculation formula is as formula (6):
wherein ,
representing the Margaret distance of the ship target obtained by jointly calculating the predicted position of the ith track on the ship target and the jth ship target detection frame;
wherein r represents a feature vector,
represents a set of eigenvectors r, T represents a matrix transpose symbol, k represents one of the numbers 0-i, λ represents +.>
L represents the ship position;
cascade matching: when a certain ship target is shielded by an object for a long time, matching the same track in the time from small to large, ensuring that the most recently appeared target is given the greatest priority, and matching the unmatched track of unconfirmed and the unmatched track of the age=1 and the detected ship target in the final matching stage.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, a camera is deployed on a port and a dock through an artificial intelligent visual recognition analysis technology, monitoring image information of the port and the dock is acquired in real time, the state of a ship entering the port and berthing at the dock is recognized and analyzed, when the ship enters the port, the ship runs to a region close to the dock, the camera recognizes the identity information of the ship, the ship starts to be tracked, after the ship berthes, the berthing state of the ship is judged through an intelligent algorithm in an intelligent algorithm server, the ship number is judged, and the identity of the ship is recognized; in the process of berthing the ship at the wharf, monitoring the berthing state of the ship in real time, detecting the access state of shore power equipment and the ship in real time, and judging whether the ship is not accessed with shore power for more than a specified time through real-time monitoring of the access state of the ship and calculation of the port-leaning duration of the ship; if the ship is not connected with the shore power according to the requirements, information such as the ship number, the dock berth, the shore power equipment number, the ship berthing time and the like is reported to a shore power system management platform, warning reminding is carried out on the ship which is not connected with the shore power according to the specified time, supervising reminding can be carried out on the ship which is berthed with the shore power, the utilization rate of a ship shore power system is improved, and the utilization rate of the shore power equipment is further improved.
Drawings
FIG. 1 is a schematic diagram of the system principle of the present invention;
FIG. 2 is a schematic structural diagram of a detection and alarm flow of a berthing access shore power state of a ship according to the invention;
FIG. 3 is a schematic diagram of a first layout of cameras according to the present invention;
FIG. 4 is a schematic diagram of a second layout of cameras according to the present invention;
fig. 5 is a schematic diagram of a third layout manner of the camera according to the present invention.
In the figure: 1. a camera; 2. an intelligent charging pile; 3. an intelligent gateway; 4. a data exchange; 5. an intelligent algorithm server; 6. an intelligent shore power platform; 7. a traffic management shore power management terminal; 8. a port operation management terminal; 9. and the ship mobile terminal.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1-5, the ship shore power system based on artificial intelligent visual identification provided by the invention comprises a camera 1, an intelligent charging pile 2, an intelligent gateway 3, a data switch 4, an intelligent algorithm server 5 and an intelligent shore power platform 6;
The intelligent charging pile 2 and the camera 1 are connected with the data switch 4 through the intelligent gateway 3, the intelligent charging pile 2 is arranged at a port and a dock where a ship is parked, the intelligent charging pile 2 is used for providing a shore power function for the ship parked at the port and the dock, and transmitting the shore power use data of the ship parked at the port and the dock to the data switch 4 through the intelligent gateway 3, and the intelligent gateway 3 is one or more of a 3G network, a 4G network, a 5G network, WIFI or network cable optical fibers;
the intelligent algorithm server 5 is connected with the data switch 4, the intelligent algorithm server 5 automatically recognizes and records the ship identity information, the port entering state, the berthing time and the ship shore power access state through the camera 1, the camera 1 is arranged at the port and dock corresponding to the intelligent charging pile 2, the camera 1 is used for acquiring the image or video information of the ship entering the port and dock in real time, and the image or video information is transmitted to the intelligent algorithm server 5 through the intelligent gateway 3 and the data switch 4;
the arrangement of the camera 1 includes, but is not limited to, the following three ways:
in the first mode, as shown in fig. 3, a high pole is erected at a dock, a camera 1 is arranged above the high pole, a camera is a dome camera, the maximum view field is 100 degrees, the camera 1 is vertically and downwardly overlooked, the installation height of the camera 1 is determined according to the identification range of the camera 1, the video images in the coverage range of the camera 1 are ensured to be clearly visible, ship characteristics can be identified through machine vision analysis, the arrangement quantity of the cameras 1 is determined according to the dock length, the coverage range between adjacent cameras 1 is 10% overlapped, for example, the coverage range of the camera 1 is 25 meters, 2.5 meters of monitoring areas between two adjacent cameras 1 are overlapped and covered, for example, the dock with the dock length of 200 meters, the camera erection height is 15 meters, and when the view field coverage range is 25 meters, 10-12 cameras 1 need to be arranged;
In the second mode, as shown in fig. 4, a high pole is erected at a dock, the cameras 1 are installed above the high pole, the cameras 1 are vertically and downwardly overlooked, the cameras are spliced by the wide-angle double cameras 1, the maximum field angle is 180 degrees, the installation height of each camera 1 is determined according to the identification range of each camera 1, the video image in the coverage range of each camera 1 is ensured to be clearly visible, ship characteristics can be identified through machine vision analysis, the arrangement quantity of each camera 1 is determined according to the dock length, the coverage range between every two adjacent cameras 1 is 10% overlapped, for example, the coverage range of each camera 1 is 30 meters, 3 meters of monitoring areas between every two adjacent cameras 1 are overlapped and covered, for example, the dock with the dock length of 200 meters, the erection height of each camera is 15 meters, and 8-10 cameras need to be arranged when the field coverage range is 30 meters;
in the third mode, as shown in fig. 5, cameras 1 are arranged on the quay, the cameras 1 are ball cameras, the heights of the cameras 1 and the quay are consistent, the video direction of the cameras 1 is parallel to the ground and the water surface, the view field angle of the cameras 1 is 120 degrees, the coverage distance of the view field of the cameras 1 is about 20-40 meters, the video images in the coverage area of the cameras 1 are ensured to be clearly visible, ship characteristics can be identified through machine vision analysis, the arrangement quantity of the cameras 1 is determined according to the quay length, the coverage areas between the adjacent cameras 1 are 50% overlapped, for example, the view field distance of the cameras 1 is 40 meters, 20 m monitoring areas overlap and cover the quay with the berth length of 200 m, for example, the view field distance of the cameras is 40 m, and 6 cameras 1 are required to be arranged;
The intelligent algorithm server 5 is internally provided with a ship target recognition algorithm and a ship target tracking algorithm, the ship target recognition algorithm is used for recognizing and analyzing ships in the image or video information transmitted by the camera 1, the ship target tracking algorithm is used for tracking targets of the ships in the image or video information transmitted by the camera 1, and the intelligent algorithm server 5 transmits ship identity information, harbor entering state, berthing time and ship shore power access state recognized and recorded by the ship target recognition algorithm and the ship target tracking algorithm to the data switch 4;
the data exchanger 4 is connected with the intelligent shore power platform 6, and the data exchanger 4 is used for transmitting the received ship identity information, the port entering state, the berthing time, the ship shore power access state and the ship shore power use data to the intelligent shore power platform 6;
the intelligent shore power platform 6 is in communication connection with a traffic management shore power management terminal 7, a port operation management terminal 8 and a ship mobile terminal 9, and is used for transmitting shore power usage data to the traffic management shore power management terminal 7, the port operation management terminal 8 and the ship mobile terminal 9, the intelligent shore power platform 6 is used for deploying a cloud end, storing received ship identity information, a port entering state, a berthing time, a ship shore power access state and ship shore power usage data, pushing the ship shore power usage data to the traffic management shore power management terminal 7, the port operation management terminal 8 and the corresponding ship mobile terminal 9, and sending warning information to a ship which is not accessed to a shore power system beyond a specified time;
According to the invention, through an artificial intelligent visual recognition analysis technology, a camera 1 is deployed on a port and a dock, monitoring image information of the port and the dock is acquired in real time, the state of a ship entering the port and berthing at the dock is recognized and analyzed, when the ship enters the port, the ship runs to a region close to the dock, the camera 1 recognizes the identity information of the ship, the ship starts to be tracked, after the ship berthes, the berthing state of the ship is judged through an intelligent algorithm in an intelligent algorithm server 5, the ship number is judged, and the identity of the ship is recognized; in the process of berthing the ship at the wharf, monitoring the berthing state of the ship in real time, detecting the access state of shore power equipment and the ship in real time, and judging whether the ship is not accessed with shore power for more than a specified time through real-time monitoring of the access state of the ship and calculation of the port-leaning duration of the ship; if the ship is not connected with the shore power according to the requirements, information such as the ship number, the dock berth, the shore power equipment number, the ship berthing time and the like is reported to a shore power system management platform, warning reminding is carried out on the ship which is not connected with the shore power according to the specified time, supervising reminding can be carried out on the ship which is berthed with the shore power, the utilization rate of a ship shore power system is improved, and the utilization rate of the shore power equipment is further improved.
The control method of the ship shore power system based on artificial intelligence visual identification provided by the embodiment comprises the following steps:
identifying and tracking the ship by port: shooting image or video information of a ship target entering a port through a camera arranged at a port and a dock, transmitting the image or video information to an intelligent algorithm server through an intelligent gateway and a data switch, carrying out identity recognition analysis on the ship target in the image or video information transmitted by the camera by a ship target recognition algorithm built in the intelligent algorithm server so as to obtain identity information of the ship target, and tracking the ship target in the image or video information transmitted by the camera by a ship target tracking algorithm built in the intelligent algorithm server so as to obtain a port entering state, a port entering time and a ship shore power access state of the ship target, and transmitting the recognized and recorded ship identity information, the port entering state, the port entering time and the ship shore power access state to the data switch;
the ship target recognition algorithm in ship port-by-port recognition tracking is based on a YOLO v4 algorithm, the YOLO algorithm is a single-stage visual recognition algorithm based on a convolutional neural network algorithm, classification and regression are realized in one process, the algorithm directly predicts the class probability and the boundary frames of objects from an input image by using a simple CNN, the model divides the input image into a fixed number of grids, each grid predicts the fixed number of boundary frames and carries a confidence index, the confidence index is to multiply the probability of detecting the objects by the intersection ratio between the predicted frames and the real frames, the boundary frames exceeding the class probability of a threshold value are selected to position the objects in the image, the ship target recognition is improved by a K-mean clustering algorithm, and the specific calculation steps of the ship target recognition algorithm are as follows:
Establishing ship and berth data sets through ship and port berth images acquired in real time by a camera, marking the data sets, randomly extracting the data sets, dividing the data sets into training sets and test sets, and carrying out ship target recognition training and testing through a YOLO v4 algorithm;
the K-means clustering algorithm clusters the data set, and the YOLO v4 algorithm is evaluated through two evaluation indexes of mAP and F1-score;
the calculation formula of mAP is shown in formula (1):
wherein ,X1 Represents the integral lower limit, X 2 Representing an upper integral limit;
p represents Precision, rec represents Recall rate Recall, d represents differential sign, and no variable is made;
mAP represents the average value of average precision AP of the ship and berth detection targets, and when the precision P is evaluated, the integral is carried out from 0 to 1, and mAP=AP;
the calculation formula of the average precision AP of the ship and berth detection targets is shown as formula (2):
the calculation formula of F1-score is shown in formula (3):
f1_score is an index for measuring the quality of a detection model, and gives consideration to Precision and Recall, and is a harmonic mean of the Precision and Recall;
precision is the detection result index of each category, namely, the percentage of the ship and berth is correctly detected in all detection areas, and the calculation formula is shown as formula (4):
The Recall ratio Recall is the percentage of the ship and berth correctly detected in all detection results, and the calculation formula is shown in formula (5):
TP is the number of accurately detected ship and berth detection targets;
FP is the number of detected ship and berth targets detected in the background;
FN is the detected ship as background and berth detection target number;
according to the invention, the YOLO v4 algorithm is improved through the K-means clustering algorithm, the detection model obtains better precision and recall rate, the problem of target detection of the berth of the ship is solved, the target recognition rate is greatly improved, and the practicability in the port environment is greatly improved;
the ship target tracking algorithm in the ship port-leaning identification tracking is a multi-target tracking algorithm taking TLD as a frame and is more commonly used in the tracking field, the current position of a target is predicted by utilizing Kalman filtering of a traditional state prediction algorithm on the basis of a detection result, the target algorithm of a target detection frame is associated by a Hungary algorithm, the intersection ratio between two frames is used as a target relation measurement index between the front frame and the rear frame of a video, the ship target tracking is improved by a deep algorithm, and the specific steps of the SORT algorithm are as follows:
Prediction model: the SORT algorithm distributes an identity ID to the tracked ship target, associates the identity ID with the next video frame, performs motion modeling, and approximates the frame-to-frame displacement of each object by using a linear constant speed model which is completely independent of the motion of other objects and cameras;
and (3) data association: when the detection results are distributed to the existing ship targets, the boundary frame geometric shape of each ship target is estimated by predicting the new position of each ship target in the current frame, the distribution cost matrix is calculated as the intersection ratio IOU distance between each detection result and all the prediction frames of the existing ship targets, the distribution cost matrix is optimally solved by adopting the Hungary algorithm, and the distribution that the overlapping of the detection result and the ship target is smaller than the minimum value of the intersection ratio is refused;
creation and deletion of track identifications: when a ship target in an image enters or leaves the image in a certain frame, an identity ID (identity) of the tracked ship target needs to be established or removed, any detection result with the overlap smaller than the minimum value of the cross ratio indicates that an untracked ship target exists, and a new tracked ship target identity ID needs to be established for the untracked ship target;
the Deepsort algorithm comprises the following specific steps:
Deepsort: the Deepsort algorithm assumes the environment in which the tracked vessel target is located on a state space (u v q h x y r h) containing the center position (u, v) of the bounding box, the aspect ratio q, the height h, and the respective velocity information of these parameters in the image coordinate system, and uses a standard kalman filter with constant velocity motion and a linear observation model, wherein the boundary coordinates (u, v, q, h) are used as direct observations of the object state;
matching problem: the Deepsort algorithm comprehensively considers the motion information and the appearance information of the ship target, solves the association problem, and uses the Mars distance between the Kalman prediction result and the detection result of the existing motion ship target motion state to associate the motion information, wherein the Mars distance calculation formula is as formula (6):
wherein ,
representing the Margaret distance of the ship target obtained by jointly calculating the predicted position of the ith track on the ship target and the jth ship target detection frame;
wherein r represents a feature vector,
represents a set of eigenvectors r, T represents a matrix transpose symbol, k represents one of the numbers 0-i, λ represents +.>
L represents the ship position;
cascade matching: when a certain ship target is shielded by an object for a long time, matching the same track in the time from small to large, ensuring that the most recently appeared target is given the greatest priority, and matching the unmatched track of unconfirmed and the unmatched track of the age=1 and the detected ship target in the final matching stage.
And (3) identifying a shore power access state: recording the time of a ship target entering a port berth through a camera and an intelligent algorithm server, timing the berthing time of the ship target, judging whether the ship target is accessed to the shore power according to the use state of an intelligent charging pile corresponding to the port berth of the ship target after the specified access shore power time is reached, and transmitting the use data of the ship shore power berthed at the port berth to a data switch through an intelligent gateway;
Reporting and warning ship target information which is not accessed into a shore power system according to the specified time: when the ship target berthing time exceeds the specified time and is not connected to the shore power system, the data exchange machine reports the ship target information to the intelligent shore power platform, the intelligent shore power platform pushes the ship target information to the traffic management shore power management terminal and the port operation management terminal, and the ship information is broadcast through the port and the corresponding ship mobile terminal to send out warning and reminding;
shore power utilization rate: counting each berth of a port, the number of berthed ships at the berth and the number of times of using shore power according to a specified rule through a camera, an intelligent algorithm server and an intelligent charging pile, and analyzing the utilization rate of the shore power of the port;
counting illegal ships and setting a blacklist of illegal ships: counting the information of the ships which are illegal to access the shore power, setting a blacklist for the ships by the intelligent shore power platform after the number of times that the ships are illegal to access the shore power exceeds the set number of times, and immediately sending out warning and reminding through port broadcasting when the ships enter a port to stop;
the multi-target tracking algorithm adopting the Deepsort algorithm increases cascade matching and target feature matching on the basis of the SORT algorithm, reduces target track jump, improves tracking effect, and solves the influence caused by overlapping of tracking targets to a certain extent.
The method for systematically controlling the shore power of the ship based on the visual identification of the artificial intelligence, provided by the embodiment, comprises the following specific implementation steps:
firstly, deploying cameras in a port, wherein the deployment modes of the cameras are shown in fig. 3, 4 and 5, and the main principle is that the cameras can completely cover a wharf area;
deploying an intelligent gateway in a port, and selecting to deploy a wireless gateway or a wired gateway according to the actual situation of the port, wherein cameras connected through a wireless network are required to be provided with wireless communication modules including but not limited to wifi, 5G, 4G or 3G, and cameras connected through a wired network are provided with ADSL network interfaces or optical fiber interfaces;
thirdly, deploying a data exchanger in a port data machine room (data center), wherein the data exchanger ensures that a wharf deploys data communication between a camera and shore power equipment, and a remote control instruction is issued;
step four, an intelligent algorithm server is deployed in a port data machine room (data machine room), intelligent algorithm software is installed in the intelligent algorithm server, and the intelligent algorithm server realizes data communication with a camera through a data switch; the method comprises the steps that image or video data collected by a camera in real time are extracted from 24 frames of images per second in a frame extraction mode, n pictures are transmitted to an intelligent algorithm server, n is determined by network communication bandwidth, and under the condition that the network is smooth and the data delay is less than 100 milliseconds, the larger the n value is, the better the n value is, and the value range of n is 1-24;
Fifthly, the intelligent algorithm server identifies the ship berthing state through an independently developed YOLO v4 algorithm improved by K-means clustering, when the ship does not leave after entering a port for more than 5 minutes, the ship berthing state is identified as the ship berthing operation, the intelligent algorithm identifies the ship berthing state every 1 minute, and after the ship berthes for 2 hours, the system detects the shore power use state of the ship berthing position;
fifth, the ship approaches the port for 2 hours, shore power is connected into the ship, and if the ship is normally used, the system continuously detects the berthing state of the ship until the ship leaves the port;
step six, when the ship approaches the port for 2 hours and the shore power is not connected into the ship, judging that the ship breaks rules and works, and transmitting information such as the ship number, the dock berth number, the ship berthing time, the ship port entering time and the like to an intelligent shore power platform without using the shore power;
and seventhly, the intelligent shore power platform can forward the information of the illegal non-accessed shore power ships to a traffic management terminal, a port operation management terminal and a ship mobile terminal, and the information can be forwarded to a computer, a mobile phone and the like through a wired network and a wireless network.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. The ship shore power system based on the artificial intelligent visual identification is characterized by comprising a camera, an intelligent charging pile, an intelligent gateway, a data switch, an intelligent algorithm server and an intelligent shore power platform;
the intelligent charging pile and the camera are connected with the data switch through the intelligent gateway;
the intelligent algorithm server is connected with the data switch and automatically recognizes and records ship identity information, a port entering state, a berthing time and a ship shore power access state through the camera;
the data exchanger is connected with the intelligent shore power platform;
the intelligent shore power platform is in communication connection with a traffic management shore power management terminal, a port operation management terminal and a ship mobile terminal and is used for transmitting shore power use data to the traffic management shore power management terminal, the port operation management terminal and the ship mobile terminal;
identifying and tracking the ship by port: shooting image or video information of a ship target entering a port through a camera arranged at a port and a dock, transmitting the image or video information to an intelligent algorithm server through an intelligent gateway and a data switch, carrying out identity recognition analysis on the ship target in the image or video information transmitted by the camera by a ship target recognition algorithm built in the intelligent algorithm server so as to obtain identity information of the ship target, and tracking the ship target in the image or video information transmitted by the camera by a ship target tracking algorithm built in the intelligent algorithm server so as to obtain a port entering state, a port stopping time and a ship shore power access state of the ship target, and transmitting the recognized and recorded ship identity information, port entering state, stopping time and ship shore power access state to the data switch;
And (3) identifying a shore power access state: recording the time of a ship target entering a port berth through a camera and an intelligent algorithm server, timing the ship target berthing time, judging whether the ship target is accessed to shore power according to the use state of an intelligent charging pile corresponding to the port berth of the ship target after the ship target berthing time reaches the specified access shore power time, and transmitting the ship shore power use data berthed at the port berthing to the data switch through the intelligent gateway;
reporting and warning ship target information which is not accessed into a shore power system according to the specified time: when the ship target berthing time exceeds the specified time and is not connected to the shore power system, the data exchange machine reports the ship target information to the intelligent shore power platform, the intelligent shore power platform pushes the ship target information to the traffic management shore power management terminal and the port operation management terminal, and the ship information is broadcast through the port and the corresponding ship mobile terminal to send out warning and reminding;
shore power utilization rate: counting each berth of a port, the number of berthed ships at the berth and the number of times of using shore power according to a specified rule through a camera, an intelligent algorithm server and an intelligent charging pile, and analyzing the utilization rate of the shore power of the port;
Counting illegal ships and setting a blacklist of illegal ships: and counting the ship information which is illegal to access the shore power, and setting a blacklist for the ship by the intelligent shore level platform after the number of times that the ship is illegal to access the shore power exceeds the set number of times, and immediately sending out warning and reminding through port broadcasting when the ship enters a port to stop.
2. The ship shore power system based on artificial intelligence visual identification according to claim 1, wherein: the intelligent charging pile is arranged at a port and a dock where the ship is stopped, the intelligent charging pile is used for providing a shore power function for the ship stopped at the port and the dock, and transmitting the shore power use data of the ship stopped at the port and the dock to the data exchange through the intelligent gateway.
3. The ship shore power system based on artificial intelligence visual identification according to claim 2, wherein: the camera is arranged at the port and dock corresponding to the intelligent charging pile, and is used for collecting the image or video information of the ship entering the port and dock in real time and transmitting the image or video information to the intelligent algorithm server through the intelligent gateway and the data switch.
4. A marine shore power system based on artificial intelligence visual identification according to claim 3, wherein: the intelligent algorithm server is internally provided with a ship target recognition algorithm and a ship target tracking algorithm, the ship target recognition algorithm is used for recognizing and analyzing ships in the image or video information transmitted by the camera, the ship target tracking algorithm is used for tracking targets of the ships in the image or video information transmitted by the camera, and the intelligent algorithm server transmits the ship identity information, the port entering state, the berthing time and the ship shore power access state which are recognized and recorded by the ship target recognition algorithm and the ship target tracking algorithm to the data switch.
5. The ship shore power system based on artificial intelligence visual identification according to claim 4, wherein: the data exchanger is used for transmitting the received ship identity information, the port entering state, the berthing time, the ship shore power access state and the ship shore power use data to the intelligent shore power platform.
6. The ship shore power system based on artificial intelligence visual identification according to claim 5, wherein: the intelligent shore power platform is used for deploying a cloud end, storing the received ship identity information, the port entering state, the berthing time, the ship shore power access state and the ship shore power use data, pushing the ship shore power use data to the traffic management shore power management terminal, the port operation management terminal and the corresponding ship mobile terminal, and sending warning information to the ship which is not accessed to the shore power system beyond the specified time.
7. The ship shore power system based on artificial intelligence visual identification according to claim 1, wherein: the intelligent gateway is one or more of a 3G network, a 4G network, a 5G network, WIFI or network cable optical fibers.
8. The ship shore power system based on artificial intelligence visual identification according to claim 1, wherein: the ship target recognition algorithm in the port approaching recognition tracking of the ship is improved through a K-mean clustering algorithm on the basis of a YOLO v4 algorithm, and the specific calculation steps of the ship target recognition algorithm are as follows:
establishing ship and berth data sets through ship and port berth images acquired in real time by a camera, marking the data sets, randomly extracting the data sets, dividing the data sets into training sets and test sets, and carrying out ship target recognition training and testing through a YOLO v4 algorithm;
the K-means clustering algorithm clusters the data set, and the YOLO v4 algorithm is evaluated through two evaluation indexes of mAP and F1-score;
the calculation formula of mAP is shown in formula (1):
wherein ,X1 Represents the integral lower limit, X 2 Representing an upper integral limit;
p represents Precision, rec represents Recall rate Recall, d represents differential sign, and no variable is made;
mAP represents the average value of average precision AP of the ship and berth detection targets, and when the precision P is evaluated, the integral is carried out from 0 to 1, and mAP=AP;
the calculation formula of the average precision AP of the ship and berth detection targets is shown as formula (2):
the calculation formula of F1-score is shown in formula (3):
f1_score is an index for measuring the quality of a detection model, and gives consideration to Precision and Recall, and is a harmonic mean of the Precision and Recall;
precision is the detection result index of each category, namely, the percentage of the ship and berth is correctly detected in all detection areas, and the calculation formula is shown as formula (4):
the Recall ratio Recall is the percentage of the ship and berth correctly detected in all detection results, and the calculation formula is shown in formula (5):
TP is the number of accurately detected ship and berth detection targets;
FP is the number of detected ship and berth targets detected in the background;
FN is the number of detected ship as background and berth detection targets.
9. The ship shore power system based on artificial intelligence visual identification according to claim 1, wherein: the ship target tracking algorithm in the port-by-port identification tracking of the ship is based on a SORT algorithm, the ship target tracking is improved through the Deepsort algorithm, and the SORT algorithm comprises the following specific steps:
Prediction model: the SORT algorithm distributes an identity ID to the tracked ship target, associates the identity ID with the next video frame, performs motion modeling, and approximates the frame-to-frame displacement of each object by using a linear constant speed model which is completely independent of the motion of other objects and cameras;
and (3) data association: when the detection results are distributed to the existing ship targets, the boundary frame geometric shape of each ship target is estimated by predicting the new position of each ship target in the current frame, the distribution cost matrix is calculated as the intersection ratio IOU distance between each detection result and all the prediction frames of the existing ship targets, the distribution cost matrix is optimally solved by adopting the Hungary algorithm, and the distribution that the overlapping of the detection result and the ship target is smaller than the minimum value of the intersection ratio is refused;
creation and deletion of track identifications: when a ship target in an image enters or leaves the image in a certain frame, an identity ID (identity) of the tracked ship target needs to be established or removed, any detection result with the overlap smaller than the minimum value of the cross ratio indicates that an untracked ship target exists, and a new tracked ship target identity ID needs to be established for the untracked ship target;
the Deepsort algorithm comprises the following specific steps:
Deepsort: the Deepsort algorithm assumes the environment in which the tracked vessel target is located on a state space (u v q h x y r h) containing the center position (u, v) of the bounding box, the aspect ratio q, the height h, and the respective velocity information of these parameters in the image coordinate system, and uses a standard kalman filter with constant velocity motion and a linear observation model, wherein the boundary coordinates (u, v, q, h) are used as direct observations of the object state;
matching problem: the Deepsort algorithm comprehensively considers the motion information and the appearance information of the ship target, solves the association problem, and uses the Mars distance between the Kalman prediction result and the detection result of the existing motion ship target motion state to associate the motion information, wherein the Mars distance calculation formula is as formula (6):
wherein ,
representing the Margaret distance of the ship target obtained by jointly calculating the predicted position of the ith track on the ship target and the jth ship target detection frame;
wherein r represents a feature vector,
represents a set of eigenvectors r, T represents a matrix transpose symbol, k represents one of the numbers 0-i, λ represents +.>
L represents the ship position;
cascade matching: when a certain ship target is shielded by an object for a long time, matching the same track in the time from small to large, ensuring that the most recently appeared target is given the greatest priority, and matching the unmatched track of unconfirmed and the unmatched track of the age=1 and the detected ship target in the final matching stage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211598566.XA CN115603466B (en) | 2022-12-14 | 2022-12-14 | Ship shore power system based on artificial intelligence visual identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211598566.XA CN115603466B (en) | 2022-12-14 | 2022-12-14 | Ship shore power system based on artificial intelligence visual identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115603466A CN115603466A (en) | 2023-01-13 |
| CN115603466B true CN115603466B (en) | 2023-05-09 |
Family
ID=84854201
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211598566.XA Active CN115603466B (en) | 2022-12-14 | 2022-12-14 | Ship shore power system based on artificial intelligence visual identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115603466B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116247824B (en) * | 2023-03-30 | 2023-11-17 | 国网河南省电力公司安阳供电公司 | Control method and system for power equipment |
| CN117040110B (en) * | 2023-07-07 | 2024-03-26 | 广州航海学院 | Grid-connected system, method, computer equipment and medium of portable ship electric controller |
| CN116683451B (en) * | 2023-08-03 | 2023-10-13 | 合肥三物信息技术有限公司 | Wisdom shore power management system |
| CN117096704A (en) * | 2023-08-21 | 2023-11-21 | 连云港远洋流体装卸设备有限公司 | Medium voltage cable shore power intelligent docking control system |
| CN116977435A (en) * | 2023-09-15 | 2023-10-31 | 广州志正电气有限公司 | Shore power system for automatic identification of ship on shore |
| CN116994458B (en) * | 2023-09-28 | 2023-12-01 | 亿海蓝（北京）数据技术股份公司 | Ship arrival/departure monitoring method, device and readable storage medium |
| CN117191246B (en) * | 2023-11-07 | 2024-02-06 | 江苏航运职业技术学院 | Ship berthing impact force testing system and method based on artificial intelligence |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107769381B (en) * | 2017-10-27 | 2020-11-03 | 中远海运科技股份有限公司 | Ship shore-based power supply information monitoring system and method |
| CN113991842A (en) * | 2021-10-22 | 2022-01-28 | 江苏科技大学 | Ship shore power integrated monitoring system |
| CN114530938A (en) * | 2022-02-10 | 2022-05-24 | 南京港龙潭集装箱有限公司 | Port intelligent ship shore-based power supply system |
-
2022
- 2022-12-14 CN CN202211598566.XA patent/CN115603466B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115603466A (en) | 2023-01-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115603466B (en) | Ship shore power system based on artificial intelligence visual identification | |
| KR102129893B1 (en) | Ship tracking method and system based on deep learning network and average movement | |
| CN108053427B (en) | Improved multi-target tracking method, system and device based on KCF and Kalman | |
| CN108062349B (en) | Video monitoring method and system based on video structured data and deep learning | |
| CN103258427B (en) | Urban expressway traffic real-time monitoring system and method based on information physical network | |
| CN108052859B (en) | Abnormal behavior detection method, system and device based on clustering optical flow characteristics | |
| CN105744232B (en) | A kind of method of the transmission line of electricity video external force damage prevention of Behavior-based control analytical technology | |
| KR102122859B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| KR102095555B1 (en) | Analysis of illegal activities and monitoring based on recognition using unmanned aerial vehicle and artificial intelligence deep running that can monitor illegal activities in the field farm | |
| CN201766663U (en) | Residential property monitoring system | |
| CN111225189B (en) | Middle and small-sized channel bridge monitoring device | |
| CN201278180Y (en) | Automatic tracking recognition system | |
| CN108802758B (en) | Intelligent security monitoring device, method and system based on laser radar | |
| CN107818326A (en) | A kind of ship detection method and system based on scene multidimensional characteristic | |
| CN106816038A (en) | A kind of inland waters abnormal behaviour ship automatic identification system and method | |
| CN103456024B (en) | A kind of moving target gets over line determination methods | |
| KR102122850B1 (en) | Solution for analysis road and recognition vehicle license plate employing deep-learning | |
| CN108500992A (en) | A kind of multi-functional mobile security robot | |
| CN107590834A (en) | A kind of road traffic accident video detecting method and system | |
| CN116208745A (en) | Dock ship berthing monitoring and recording system and method | |
| CN102254394A (en) | Antitheft monitoring method for poles and towers in power transmission line based on video difference analysis | |
| CN111783579B (en) | Unmanned aerial vehicle visual analysis-based detection system for crossing fence by constructors | |
| CN110867046A (en) | Intelligent car washer video monitoring and early warning system based on cloud computing | |
| CN105574468A (en) | Video flame detection method, device and system | |
| CN111163290A (en) | Device and method for detecting and tracking night navigation ship |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230419 Address after: 210000 No. 223, Shui Simen street, Nanjing, Jiangsu. Applicant after: JIANGSU ENERGY TECH DEVELOPMENT Co.,Ltd. Applicant after: Wang Qin Address before: 210000 No. 223, Shui Simen street, Nanjing, Jiangsu. Applicant before: JIANGSU ENERGY TECH DEVELOPMENT Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |