In this paper, multi-value hit-or-miss transformation (MVHMT) is introduced into a joint transform correlator (JTC) for binary image recognition. Considering fuzzy concepts are lost MVHMT improves the ability of the ordinary morphological hit-or-miss transformation (HMT) because it reduces the decision effect of the pixels that is easily disturbed by rotation and scalling distortion. Not only MVHMT can realize the precision recognition between two images because no information of the input images is lost, but also it has a better distortion-invariant ability for scaling and rotation than the ordinary HMT. Using a novel multi-value complementary encoding method, MVHMT is performed on a JTC in one step. With a high-level thresholding value, the correct justification can be achieved from the correlated resultant image. The simulation results are given.