An effective Hough Transform based track initiation

作者:Jin Shu Ling*; Liang Yan; He Peng; Pan Guang Lin; Pan Quan; Cheng Yong Mei
来源:5th International Conference on Machine Learning and Cybernetics, 2006-08-13 to 2006-08-16.

摘要

In this paper, a Hough Transform (HT) based track initiation method is proposed to detect targets in 3L (low signal-to-clutter-ratio, low signal-to-noise-ratio, and low detection probability) environment using a new and effective accumulation method. In the new accumulation method, the contributions of each cell's votes are determined by the neighbors of the resolving cell, instead of just the cell itself so that the disturbance of sampled probability density, due to finite samples, can be smoothed. Simulation results show that our method not only makes significant improvement in reducing false tracks but also has stronger robustness to measurement noise and clutters, compared with standard HT based track initiation with binary accumulation.