[ACM Press the 2nd International Conference - Chengdu, China (2018.06.16-2018.06.18)] Proceedings of the 2nd International Conference on Advances in Image Processing - ICAIP '18 - MIFNet

摘要： Sea-land segmentation is of great significance to coastline extraction and ship detection. Due to the complicated texture and intensity distribution of high resolution remote sensing images, traditional methods based on threshold and artificial features are difficult to perform well. This paper presents a new multi-information fusion network (MIFNet) based on convolutional neural network. MIFNet not only considers multi-scale edges and multi-scale segmentation information, but also introduces global context information, and fuses different scales and types of information through network learning. Experiments on a set of natural-colored images from Google Earth show that our model achieves better performance than the state-of-the-art methods.