Covariance intersection algorithm is a conservative distributed fusion estimation method obtained by optimizing certain objective functionwithout needing to compute the correlation among local estimation errorswhich provids a new idea for image fusion enhancement.Firstlythe 1D covariance intersection algorithm was introduced;then this method was extended to 2D signal and image fusion for the first timeand an image fusion method based on covariance intersection was put forward.Finallythe fused image was compared with the existing fusion methodsand the results showed that fusion effect of it was superior to that of wavelet methodempirical mode decomposition methodand non-negative matrix factorization method.