Infrared small target detection in complex backgrounds is a challenging problem.Complex backgrounds often contain structured edges,unstructured clutter and noise,which have completely different properties.It is very difficult to separate small targets from these disturbances by using an attribute.To solve this problem,a LCMDM method to detect small target is proposed.Four directions of images are discussed in the process of detect the small target.A new threshold formula based on image mean and variance are used in the process of the infrared small target detection.In addition,the experimental results show that the proposed method shows better performance.