Object detection is the core part of optical remote sensing image interpretation,which has wide applications in intelligence reconnaissance,land resource utilization,urban planning and other fields.In recent years,the maturation of deep learning has led to significant improvements in the accuracy and efficiency of optical remote sensing object detection.In this paper,the general target detection algorithms based on deep learning is reviewed first.Secondly,the current commonly used optical remote sensing image object detection data set is introduced and the development trend of the data set based on the data characteristics is analyzed.Then,according to the difficulties of remote sensing image detection,the optimization scheme of the algorithm is sorted out in detail from five aspects:arbitrary direction,multi scale,small target,dense distribution,and complex background.Finally,the development direction of optical remote sensing image object detection research is prospected.