Synthetic aperture radar images are characterized by salt & pepper noise that affects change detection accuracy. To solve this problem, an improved method based on intuitionistic fuzzy C-core mean clustering is applied to synthetic aperture radar image change detection. First, we use algebraic operation methods, such as the difference, ratio, and image regression methods, to construct three types of spectral variation difference images, and three different images are grouped to column vectors. Second, the principal component analysis algorithm is used to extract features from column vectors of the difference images. Finally, the change map is obtained using intuitionistic fuzzy C-core mean clustering. Experimental results show that the proposed method can reduce the influence of salt & pepper noise in synthetic aperture radar images, retain feature information, and improve the accuracy of change detection in the image.