Automatic picking of dispersion curves based on region splitting and merging method

Surface wave dispersion curve picking is one of the important steps in surface wave data processing. At present, the dispersion curve picking mainly relies on a large number of manual processing and is time consuming when a large number of dispersion curves need to be picked up, and it is difficult to meet the demand of efficient processing of modern surface wave exploration. How to efficiently and accurately realize picking of surface wave dispersion curves automatically has become an urgent problem. In this paper, the picking of dispersion curves is regarded as an image segmentation problem, and a region splitting and merging algorithm is proposed to realize picking of dispersion curves from the dispersion energy map quickly and accurately. The algorithm uses the standard deviation and average value of the grayscale value of the pixels in the dispersion energy map region to establish the consistency criterion of region splitting and merging, uses the consistency criterion to split and merge them continuously to split the disperse energy region, and finally realizes the automatic picking of the dispersion curve by tracking the peaks of different frequencies in the disperse energy region. To verify the effectiveness of the algorithm, a soft sandwich-containing model and a velocity-increasing model are used for comprehensive testing. The results show that the automatically picking dispersion curves are highly consistent with the theoretical dispersion curves. Meanwhile, the actual surface wave data test further confirms the accuracy of the region splitting and merging algorithm.