 Image Augmentation is a powerful tool for increasing the size of datasets and machine learning applications. This paper reviews the most common image augmentation techniques and classifies them according to their implementation strategies. Experiments were conducted to evaluate the effectiveness of these techniques on the segmentation task of two classical material microscopic images. The unit model was selected as a representative benchmark model for image segmentation tasks as it is the classic and most widely used model in this field. The results show that the use of image augmentation techniques can significantly improve the segmentation performance of the unit model. Furthermore, the advantages and applicability of various image augmentation techniques in the material microscopic image segmentation task are discussed. This article was authored by Jin Chow Ma, Chen Fei Hu, Ping Zhou, and others.