The invention discloses a multi-temporal remote sensing image change detection method based on FCM and the evidence theory, and the method is characterized in that the method comprises the steps: firstly solving the corresponding band difference of remote sensing images at two time phases, the amplitude values of variable vectors of the two time phases, and the cosine values of spectrum inclined angles of the two time phases; secondly taking the above values as the input of the FCM, and respectively obtaining a corresponding fuzzy division matrix, and enabling the fuzzy degree of each type of fuzzy division matrixes to serve as a quality function of the evidence theory; finally carrying out the fusion of the above three division matrixes through employing the evidence theory, obtaining a new fuzzy division matrix, and obtaining the final change detection result according to the above. The beneficial effects of the invention are that the method is based on the FCM and the D-S evidence theory, achieves the fusion of the band difference, the amplitude values of variable vectors and the cosine values of spectrum inclined angles through employing the evidence theory, inputs the above values into the FCM model and obtains a detection result, eliminates the uncertainty in change detection through the detection result, and enables the result of change detection to be more reliable and robust.