Aiming at the problem of low detection rate of small infrared target under complex background and heavy clutter interference, an infrared small target detection method based on local multidirectional gradient is proposed. Firstly, Laplace operator is used to sharpen the image and enhance the target. Then, the local multidirectional gradient function is redefined by the gradient direction feature analysis of the image. And it improves the anti-jamming ability for strong clutter. Finally, the real target is extracted by adaptive threshold segmentation. The experimental results show that the method can effectively improve the target detection ability under complex background and strong clutter interference.