Abstract: The discontinuity information obtained by a combination of conventional, coherent attributes, and ant tracking technology is not sufficient for the fine analysis of medium-thin sandstone reservoirs in river and delta facies.To better detect the internal discontinuity of thin sandstone reservoirs, we present a unique method for attaining these discontinuity features.First, the gray-level co-occurrence matrix (GLCM) homogeneity is calculated using the root mean square (RMS) amplitude attributes extracted from the thin layer.Then, according to the distribution characteristics of the thin sandstone discontinuity structure, the path curvature is used to restrict the movement of artificial ants and optimize the plane enhancement effect of the ant colony algorithm.The interference is then suppressed, and discontinuity characteristics are highlighted.Thus, the combined technology can enable thin sandstone reservoir discontinuity detection.Through a model experiment and field data test, the edges of a thin sandstone body, and its internal small-scale discontinuity structure, could be identified, verifying the effectiveness of the technology for detecting the internal discontinuity of thin sandstone reservoirs, and allowing the use of the detection results for improving the accuracy of sandstone thickness prediction.This can provide technical support for the subsequent fine characterization of the internal structure of the sand body.