Abstract: Oil and gas leakages in the Jimusar area are spatially associated with salinization. In this study, the airborne hyperspectral image of the Jimusar region with a total area of 2500 km² in Xinjiang was obtained by using CASI/SASI imagery of the National Key Laboratory of Science and Technology on Remote Sensing Information and Image analysis. The dataset was used to extract oil leakage information by using newly developed classification method. Seven oil leakage areas were detected. In addition, the extraction model for soil salt was also developed and applied to study the distribution of soil salt in the Jimasar region. Based on the superposition analysis of oil and gas anomaly area and salinization anomaly area, the authors detected a high degree of coincidence in the spatial location of the two ares. The authors believe that this hyperspectral remote sensing detection not only has theoretical significance but also has practical value. Its practical value is mainly reflected in the hyperspectral remote sensing oil and gas exploration. Through the mutual confirmation of hydrocarbon information, one can reduce redundant information of oil and gas leakage and improve the accuracy of oil and gas leakage identification. Additionally, in the absence of aerospace hyperspectral remote sensing data sources, satellite multispectral remote sensing data can also be used to extract the salinization anomalies and delineate the anomalous areas of salinization and alkalinization, thereby helping to narrow the target areas for oil and gas exploration. The results obtained by the authors indicate that the salinization of nature may not be wholly attributed to "dry evaporation", and there may be oil and gas related salinization. The distinguishing between these two kinds of salinization is a scientific problem worthy of in-depth discussion, especially in the field of hyperspectral remote sensing.