基于大数据挖掘模型的船舶破碎尾迹智能跟踪方法

Intelligent tracking method of ship’s broken wake based on big data mining model

为了解决传统船舶破碎尾迹跟踪方法跟踪轨迹点离散性过大的问题,设计基于大数据挖掘模型,提出一种新的船舶破碎尾迹智能跟踪方法。采用GIBS抽样算法作为数据挖掘模型的内核,将传统单一的马尔科夫数据链转化为多链,建立一种多链结构的数据挖掘挖掘模型,进行船舶当前尾迹数据挖掘,利用尾迹状态方程和尾迹测量方程,将尾迹数据图像化,确定尾迹实际像素并进行补偿处理,对处理后的图像进行回波特征转化,根据图像像素的回波特征值,评判像素点是否属于船舶尾迹,从而实现当前船舶尾迹跟踪。实验数据表明,该方法输出的船舶尾迹跟踪结果,其轨迹回波有效间距提高了32%,补偿间距提高了29%,可以有效降低轨迹离散。

in order to solve the problem of excessively large discretization of tracking track points in the traditional tracking method of ship’s broken wake,a new intelligent tracking method of ship’s broken wake was proposed based on the big data mining model.Using GIBS sampling algorithm,as the kernel of data mining model,the traditional single markov chain for the chain data,set up more than a chain structure model of data mining,mining ship wake current data mining,using wake wake measuring equation and state equation,the trail data visualization,and compensate for trail actual pixel processing,echo characteristics of the processed image transformation,according to the echo characteristics of image pixel value,judging whether the pixel belongs to ship wake,so as to realize the current trail tracking of the ship.Experimental data show that the tracking results of ship wake output by this method can reduce the echo interval by 32%and the compensating interval by 29%,which can effectively reduce the track dispersion.