Mineralized rock alteration anomaly is an important indication of ore forming. It plays an important role in the orefinding using remote sensing as a main parameter of fastevaluating and predicting the mineral resources. On the basis of analyzing spectral features used to recognizing the target objects of ETM image data, this paper introduces a method of extracting the alterative anomalies information, which combined wavelet adaptive threshold enhancing with the principal component analysis integrating MATLAB developing flat and remote sensing image processing software as ERDAS IMAGINE 8.6. Its effect of alterative abnormal information extracting is compared to the traditional methods that are based on the selective band principal component extraction and the combination of principal component and ratio band respectively. The results show that the coincidence rate between the known deposits and the ore (mineralized) spots and remote sensing alteration anomalies extracted using the method in this paper is 80 percent. At the same time, a new mineralized spot and many periphery clues of orefinding. This result is quite satisfactory. Moreover, it is of the features such as rapid, effective and economic etc., and the much more superiority than the traditional methods.