Object contour plays an important role in fields such as semantic segmentation and image classification. However, the extraction of the contour is a difficult task, especially when the contour is incomplete or unclosed. In this paper, existing contour detection approaches are  reviewed and roughly divided into three categories: pixel-based, edge-based, and region-based. In addition, while the traditional contour detection approaches have achieved a high degree of sophistication, deep convolutional neural networks (DCNNs) have demonstrated a good performance in image recognition, and therefore, the DCNNs based contour detection approaches are also covered in this paper. Furthermore, the future development of contour detection is analyzed and predicted.