With the popularization of biometric recognition technology, palmprint recognition has attracted widespread attention because of its high uniqueness and non-replicability. However, recent studies have shown that various image-level attacks on palmprint recognition systems can interfere with the decision-making process of the model, leading to a decline in both the accuracy and security of the system. In this paper, we first summarize the applications of palmprint recognition technology and then describe two main types of image-level attacks: adversarial attacks and reconstruction attacks. Finally, we summarize existing works on image-level attacks targeting palmprint recognition, highlighting key methods and strategies.