基于膨胀运算的移动对象兴趣点检测方法

针对传统兴趣点检测算法在准确性和效率方面的不足，提出基于膨胀运算的移动对象兴趣点检测方法（DMDO）。通过矩阵二值化操作滤除停留点噪声，提高预测准确率，并用膨胀运算替代传统方法中的聚类算法提高算法效率。将DMDO在开放空间数据集AMSA和IMIS3Days上进行仿真实验，结果表明：DMDO相比基于密度的空间聚类算法，在数据集AMSA上准确率平均提高17.94%，算法效率提高6.63倍；在数据集IMIS3Days上准确率平均提高19.98%，算法效率提高9.13倍；相比以聚类点排序结果确定聚类结构算法，DMDO在数据集AMSA上准确率平均提高20.04%，算法效率提高14.61倍；在数据集IMIS3Days上准确率平均提高16.60%，算法效率提高42.19倍；DMDO相比传统方法均表现出较高的预测准确性、较低的时间开销，适用于解决大数据背景下的移动对象兴趣点检测问题。

Interest Point Detection Method Based on Dilation Operation

An interest point detection method based on dilation operation (DMDO) is proposed to improve the efficiency and accuracy of interest point detection,in which binarization is used to filter the noise,and the dilation operation is used to replace the clustering approach to enhance the efficiency of algorithm.DMDO is applied to two datasets of open space-AMSA and IMIS3Days.Compared to DensityBased Spatial Clustering of Applications with Noise (DBSCAN),the accuracy of DMDO is increased by 17.94％ on dataset AMSA,and by 19.98％ on dataset IMIS3Days,while the efficiency is improved by 6.63 times on dataset AMSA,and by 9.13 times on dataset IMIS3Days.Compared to Ordering Point To Identify the Cluster Structure (OPTICS),the accuracy of DMDO is increased by 20.04％ on dataset AMSA,and by 16.60％ on dataset IMIS3Days,while the efficiency is improved by 14.61 times on dataset AMSA,and by 42.19 times on dataset IMIS3Days.Experimental results demonstrate that,compared with traditional methods,DMDO has higher accuracy with less time overhead.DMDO is applicable to detect the interest points in the era of big data.