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.

Although widely exploited in recent decades, road extraction is still a very significant and challenging research in the field of remote sensing image processing due to the complex background and road distribution. Among the existing CNN-based methods, U-shape architectures composed of encoders and decoders have shown their effectiveness. In this letter, we propose an improved encoder–decoder method, named DBRANet, for extracting roads from remote sensing images. In the encoding phase, we present a dual-branch network module (DBNM) to construct more effective features, thus improving the fusion feature maps of different scales. One branch utilizes the residual block, and the other branch utilizes the refined asymmetric block, which effectively increases the feature extraction capability of the backbone. In the decoding phase, considering the sinuous shape and the unbalanced distribution of roads in remote sensing images, we design a novel attention module, named the regional attention network module (RANM), to automatically learn the importance of each channel according to the regional information. Extensive experiments on several public remote sensing road data sets show that our DBRANet achieves higher segmentation [ $F1$ score and Intersection over Union (IoU)] and connectivity [average path length similarity (APLS)] accuracy, which verifies the effectiveness of our approach.

Incremental learning using neural networks achieves great success in semantic segmentation but still suffers from catastrophic forgetting. In this article, we propose an effective class-incremental segmentation method without storing old data. To alleviate the issue of forgetting, we present two important modules, i.e., the deep feature distillation (DFD) module and the label mixed (LM) module. The DFD module is established to learn a good feature representation of old classes by distilling a new compact feature representation from different layers of networks. The proposed LM module first identifies the examples (pixels) of old classes with high confidences utilizing the output of old models, and then, they are combined with examples of new classes to supervise the training of new models, which can achieve a good balance between learning new classing and avoiding forgetting old ones. Our ablation studies show that the DFD module and the LM module can make the learning network obtain 6.2% and 15% performance gains [mean Intersection over Union (mIOU)], respectively. Furthermore, by introducing the supervision of output distillation loss, we compare our method with several state-of-the-art methods in the extensive experiments, and the experimental results all show that our method is significantly superior to them on the dataset of aerial images.

Deep neural networks achieve significant progress in semantic segmentation but still suffer from the catastrophic forgetting problem, i.e., networks will forget old classes as they learn new ones. In this article, we propose an effective class-incremental segmentation framework without storing old data. Specifically, to alleviate the issue of catastrophic forgetting, we present two important modules, i.e., the Pixel-level Feature Generation (PFG) module, and the Task-wise Knowledge distillation (TKD) module. The PFG module is designed to constantly generate any number of features of the old classes to keep the old memory. The PFG module is the first attempt to use the generative method in class-incremental segmentation of aerial images, and it abandons the previous image generation approach but to generate pixel-level features, which is more suitable for the segmentation task. Meanwhile, the proposed TKD module is specially designed for class incremental tasks, and it only compares classes in the same learning step (task), thus avoiding the squeezing of new classes to old classes when the output is normalized (softmax), making distillation more effective. Sufficient experiments show that our method is remarkably effective and achieves more than 4.5% gains compared with state-of-the-art methods, and more than 13% compared with baselines, on all learning conditions. The ablation studies show that the PFG module and the TKD module are both indispensables. Besides, the proposed framework can be well combined with any existing class incremental learning method to achieve better performance.