This article discusses the use of different algorithms for analyzing data obtained by magnetic measurements in archaeological excavations. The aim of this work is the automatic detection of possible anomalies in data that may indicate the presence of archaeological sites. The article discusses two kinds of anomalies - useful (caused by the presence of archaeological sites) and spot anomalies (caused by the presence of various metal debris). Spot anomalies create local noise that interfere with the determination of the spatial anomalies caused by the remains of ancient buildings, such as walls of houses, wells, dugouts, etc. The paper proposes algorithms that allow to exclude from the data spot anomalies and to concentrate on identifying spatial anomalies. Also discusses innovative methods that allow to determine the contours of the spatial anomalies (archaeological sites). The algorithms were tested on many real data sets, obtained during excavations in the Crimea, and showed good results.