CN-116312055-B - Ship movement tag tracking method based on video identification and AIS data fusion

CN116312055BCN 116312055 BCN116312055 BCN 116312055BCN-116312055-B

Abstract

The invention discloses a ship moving label tracking method based on video identification and AIS data fusion, which comprises the steps of 1 converting AIS longitude and latitude uploaded by a ship into Cartesian coordinate data, 2 transmitting the Cartesian coordinate data to a video identification device, 3 predicting the position of the ship, 4 converting the predicted coordinate position into image coordinates, 5 converting the image coordinates into screen coordinates, 6 obtaining the position of the ship in the screen, and 7 binding the moving label with the ship screen coordinates to realize tracking and listing of the moving label. The invention provides the ship moving label tracking method based on the video identification and AIS data fusion, which can monitor the near-distance marine ships, can also integrate the video identification and AIS data, simplifies the monitoring tracking and implementation difficulty, reduces the cost and realizes the tracking and monitoring of the past ships in an automatic moving label tracking mode.

Inventors

LIU YUESHENG
ZHONG ZIYANG
TIAN CHI
SUN XUETAO
Yi Zhongli

Assignees

中华人民共和国深圳海事局

Dates

Publication Date: 20260505
Application Date: 20230221

Claims (2)

1. A ship movement tag tracking method based on video identification and AIS data fusion is characterized by comprising the following steps: Step 1) converting AIS longitude and latitude uploaded by a ship into Cartesian coordinate data under a coordinate system of a holder installation position; Step 1) uploading AIS data through a ship, wherein the AR eagle eye identification equipment identifies the ship entering a monitoring area; step 2) transmitting the Cartesian coordinate data to a video recognition device coordinate; Step 3) predicting the position of the ship according to the coordinate data of the video recognition device; the step 3) comprises the steps of predicting the coordinate position of the ship at the next moment under the coordinate of the AR hawk eye camera, and predicting the camera coordinate position of the ship at the next moment; Step 4) converting the predicted coordinate position into image coordinates; step 5) converting the image coordinates into screen coordinates, and calculating the position of the ship in the screen according to the AIS longitude and latitude data of the ship; step 6) carrying out linear fusion calculation on the position of the ship in the screen calculated by the AIS data and the position of the ship in the screen obtained by the video recognition device to obtain the position of the ship in the screen; The step 6) comprises the steps of carrying out fusion calculation on the coordinates of the ship in the screen, which are detected by the AR eagle eye identification equipment, and the coordinates of the ship in the screen, which are calculated by the longitude and latitude coordinates of the AIS, so as to obtain the real-time position of the ship in the screen; the fusion calculation includes step A, B, C, D, E: And A, calculating the position of the ship in a coordinate system with an AR eagle eye as a coordinate origin by using AIS coordinates: latitude eccentricity coefficient: ; Distance in latitudinal direction: ; Distance in longitudinal direction: ; Wherein: D is a latitude eccentricity coefficient; ER is the equatorial radius of the earth; PR is the earth polar radius; Lat is latitude; Lon is longitude; AIS is the ship position; IL is the mounting position of the cradle head; step B, coordinate transformation relation exists between the AR eagle eye and the mounting position of the cradle head, the transformation relation is assumed to be R, the translation relation is assumed to be T, The coordinate position of the ship with respect to the AR hawk eye is: ; and C, predicting the next position of the ship by using the Hall index: ; Wherein, the T' is the uploading of the ship to the AR hawk eye The frequency of the data; For time series t-th period Smoothing the value; actual value of time series t period ); Is time series t phase Trend values of (2); For predicting the predicted ship position in the m-th period on the basis of the t-th period of the current time sequence ; Step D, through the above process, the real-time predicted position of the ship in the AR hawk-eye coordinate system can be obtained through AIS information of the ship, and the predicted coordinate is further converted into an image coordinate: ; and E, converting the image coordinates into pixel coordinates: ; the linear fusion calculation comprises the steps of obtaining the position of the ship in the screen calculated by ship AIS data, setting the position of the ship in the screen calculated by the brains AI as # ) The position of the ship predicted by AIS in the screen is set as% ) Linear fusion is carried out, and the final result is : ; The step 6) comprises the steps of carrying out fusion calculation on the longitude and latitude coordinates of the ship and the longitude and latitude coordinates of the AIS, which are detected by the AR eagle eye identification equipment, so as to avoid the condition that the label is abnormal or drifts; and 7) binding the mobile tag with the ship screen coordinates after acquiring the real-time ship screen coordinates, and realizing tracking and listing of the mobile tag.
2. The method for tracking the ship moving label based on the fusion of video identification and AIS data according to claim 1, wherein the step 7) is characterized in that after the screen coordinates of the ship are calculated, the moving label is bound with the ship by limiting the Euclidean distance between the screen coordinates of the moving label and the screen coordinates of the ship, and the ship attribute is displayed in the moving label by acquiring the ship attribute data in the AIS data.

Description

Ship movement tag tracking method based on video identification and AIS data fusion Technical Field The invention relates to the technical field of ship monitoring, in particular to a ship movement tag tracking method based on video identification and AIS data fusion. Background At present, two methods for monitoring ships at sea generally comprise a marine ship detection method based on fusion of machine vision and multi-source data. The method combines Yolov target detection algorithm, radar, common monitoring equipment and AIS system to realize offshore monitoring of ships; another marine ship monitoring method is an adaptive weighted data fusion method based on vision and multi-source radar. The scheme is similar to the scheme I, and the position in the image information of the target ship and the position in the data information of the target ship are obtained by utilizing the vision sensor and the radar sensor and are fused, so that the accuracy of fusion of the vision information with the radar and AIS position information is improved. The two marine vessel monitoring methods have the following defects: 1. the scheme needs to fuse the data of a plurality of devices, the system deployment is complex, and the cost is high; 2. The video monitoring equipment of the scheme can only monitor ships at a close range, has poor monitoring capability and cannot acquire remote target information; 3. the radar cannot judge the type of the obstacle, so that the tracking and positioning of the remote ship cannot be realized; 4. The accuracy of the method is difficult to ensure, and when the distance between ships is short, the ship target with short distance is easy to be identified as one target. 5. The AIS data updating frequency is slower, and the ship tracking is easy to drift. Disclosure of Invention The invention aims to overcome the technical defects, and provides the ship moving tag tracking method based on the fusion of the video identification and the AIS data, which can monitor the near-distance marine ship, can also be fused with the AIS data through the video identification, simplifies the monitoring tracking and implementation difficulty, reduces the cost and realizes the tracking monitoring of the past ship in an automatic tracking mode of the moving tag. In order to achieve the above purpose, the invention provides a ship movement tag tracking method based on video identification and AIS data fusion, which comprises the following steps: Step 1), converting AIS longitude and latitude uploaded by a ship into Cartesian coordinate data under a coordinate system of a holder installation position; step 2) transmitting the Cartesian coordinate data to a video recognition device coordinate; Step 3) predicting the position of the ship according to the coordinate data of the video recognition device; Step 4) converting the predicted coordinate position into image coordinates; step 5) converting the image coordinates into screen coordinates, and calculating the position of the ship in the screen according to the AIS longitude and latitude data of the ship; step 6) carrying out linear fusion calculation on the position of the ship in the screen calculated by the AIS data and the position of the ship in the screen obtained by the video recognition device to obtain the position of the ship in the screen; and 7) binding the mobile tag with the ship screen coordinates after acquiring the real-time ship screen coordinates, and realizing tracking and listing of the mobile tag. The video recognition device comprises an AR eagle eye recognition device. And 1) uploading AIS data through the ship, and identifying the ship entering the monitoring area by the AR eagle eye identification equipment. And 3) predicting the coordinate position of the ship at the next moment under the coordinate of the AR hawk eye camera, and predicting the camera coordinate position of the ship at the next moment. And 6) performing fusion calculation on the coordinates of the ship in the screen, which are detected by the AR eagle eye identification equipment, and the coordinates of the ship in the screen, which are calculated by the longitude and latitude coordinates of the AIS, so as to obtain the real-time position of the ship in the screen. The step 6) comprises the steps of carrying out fusion calculation on the longitude and latitude coordinates of the ship and the longitude and latitude coordinates of the AIS, which are detected by the AR eagle eye identification equipment, so that the condition that the label is abnormal or drifts is avoided. And 7) after the screen coordinates of the ship are obtained through calculation, binding the mobile tag with the ship by limiting the Euclidean distance between the screen coordinates of the mobile tag and the screen coordinates of the ship, and displaying the ship attribute in the mobile tag by acquiring the ship attribute data in the AIS data. The technical scheme has the advantages that the marin