Gravity correlation imaging with a moving data window

•The resolution of correlation imaging is improved.•The result is stable when gravity data is corrupted with noise.•Correlation imaging is easy to calculate.

As a kind of imaging technique, correlation imaging of potential data can be used to determine the shape and position of sources approximately. In order to improve the resolution of correlation imaging results, I adopted a strategy similar to Euler deconvolution method. This method includes three steps. First, calculate the vertical gradient of gravity anomaly. Then add a moving data window to the correlation imaging procedure. At last, search the local maximum or minimum of the results and output the extreme points that meet the acceptance criterion. In the synthetic model tests, the resolution of correlation imaging is improved. Compared with the Euler deconvolution, the results are closer to the centers of the models. In a case study of a sylvite mine gravity survey in Yunnan Province, China, the positions of extreme points of correlation imaging are in the range of the ore zone defined by drills.