Given a secret image I, a threshold r, and a set of n (≥ r) participants P = {1, 2, … , n} with a set of weights W = {w 1 , w 2 , … , w n } where w i is the weight (which indicates the degree/rank of importance) of participant i and we assume that w 1 ≤ w 2 ≤ … ≤ w n. The idea of weighted threshold secret image sharing encodes I into n shadows S 1 , S 2 , … , S n with sizes |S 1 | ≤ |S 2 | ≤ … ≤ |S n | in which S i is distributed to participant i such that only when a group of r participants can reconstruct I by using their shadows, while any group of less than r participants cannot. We propose a novel weighted threshold secret image sharing scheme based upon Chinese remainder theorem in this paper. As compared to the conventional Shamir's and recent Thien-Lin's schemes, which produce shadows with the same size, our scheme is more flexible due to the reason that the dealer is able to distribute various-sized shadows to participants with different degrees/ranks of importance in terms of practical concerns.