Effective Pulsar Dispersion Measure Estimation Based on Progressive Probabilistic Hough Transform

In pulsar search pipelines, a great deal of observational data needs to be processed and produce prospective candidates. Typically, because of unknown dispersion measure(DM), a brute-force attempt with different values in a range will be performed. Aiming at providing reference values of DM to mitigate a severe waste of time and space spent on DM trials, a novel and straightforward method was introduced in this paper. Based on the prior physical knowledge of the dispersion effect, the estimation of the pulsar DM value is converted into the problem of line detection in two-dimensional images. First, the original two-dimensional image is processed by coordinate transformation and in which the dispersion curve will be transformed into a linear trajectory. Then the image is preprocessed to enhance the sharpness of line contour and reduce the interference of background noise. Finally, the Sobel operator is used for edge detection, and Progressive Probabilistic Hough Transform(PPHT) is employed to realize the estimation algorithm of pulsar DM. The experimental results show that the divergence between the estimated DM value and the actual catalog value is small, which can provide an effective way to speed up the pulsar search process and has high application value.