DenseNet-Based Land Cover Classification Network With Deep Fusion

Recently, fully convolutional network (FCN)-based (Long et al., 2015) networks have made impressive success in semantic segmentation, and these approaches achieve satisfactory results in natural images. However, in the field of high-resolution remote sensing image segmentation, the accuracy has a considerable huge gap compared with that of natural images. Through the development process of semantic segmentation, we found that the key to accurate segmentation is the context. Effective networks can always obtain large contexts, which means that context is the key to one successful segmentation network. For high-resolution remote sensing images, their elements always extend to large scope and they have no clear or regular boundaries. As a result, it needs more context to correctly classify each pixel. However, the networks designed for natural images obviously do not meet this requirement, and thus, they achieve poor segmentation results for high-resolution images. Therefore, we do some targeted improvements. Based on one powerful backbone, we add two new fusions called unit fusion and cross-level fusion, respectively. Unit fusion makes the connection from the encoder part to the decoder part not only occur in the final output of each dense block but also in the middle feature layers inside one dense block. These added fusions make feature fusion in the same level more complete, which is of great significance for complex and zigzag boundary areas. As a complement for unit fusion, cross-level fusion aims to enhance the fusion of different dense blocks. Specifically, cross-level fusion learns from the internal structure of the dense block and applies the design to the whole network level. It can incorporate nonadjacent features and rapidly increase the receptive field and context, which is very effective for the segmentation of targets with very large sizes. Experiments on Deepglobe (Demir et al., 2018) show significant improvements in our work.

Automated summary

Recently, FCN-based networks have achieved impressive success in semantic segmentation of natural images. However, in high-resolution remote sensing image segmentation, there is a considerable gap in accuracy compared to natural images. The key to accurate segmentation is context, and effective networks can obtain large contexts. To address the limitations of networks designed for natural images, targeted improvements including unit fusion and cross-level fusion were proposed. Experimental results on Deepglobe dataset showed significant improvements in segmentation performance.