Built-Up Area Extraction Combing Densely Connected Dual-Attention Network and Multiscale Context

Abstract

Built-up areas are the main areas for human production and life. The wide availability of high-resolution satellite images provides new opportunities for fine-scale detection and mapping of built-up areas. However, great challenges remain because built-up areas are large-scale composite geographical objects, which contain diverse object classes and complex scenes, and have extremely large feature heterogeneity across space. For the automatic recognition of built-up areas in high spatial resolution satellite images, a block-level built-up area extraction framework combing densely connected dual-attention network and multiscale context is proposed. The proposed method first divides an image into multiscale blocks with a certain proportion of overlap through multisize grids and their multistep offsets. It then uses the constructed lightweight network integrating dense connection and dual attention to realize the feature representation and discrimination of image blocks. Finally, it achieves the refined detection of built-up areas by integrating the prediction results under different divisions through pixel-level multilabel voting. GaoFen-2 satellite images covering Shenzhen city, China, are used to verify the effectiveness of the proposed method. In the five selected test areas, the F1 score of the proposed method ranges from 0.8720 to 0.8983. Results visually preserve the integrity of internal morphology and have a well-defined boundary. The proposed method shows better performance than the state-of-the-art built-up area Manuscript

Lightweight Multilevel Feature-Fusion Network for Built-Up Area Mapping from Gaofen-2 Satellite Images

“…According to the quantitative evaluation results of the five test areas described in Section 3, their average F1 score and IoU value are 0.9121 and 0.8386, respectively. Compared with previous similar studies [12], this result also demonstrates superior performance in the accuracy and shape integrity of urban built-up area detection.…”

“…Traditional built-up-area extraction methods mainly use artificially designed algorithms to extract the spectrum, texture, and local features of an image and then identify the built-up area by thresholding the built-up index/saliency map [5][6][7][8][9][10] or by using a supervised classifier [11]. Built-up areas are a composite target, covering a large geographical range, and their image features have an extremely large spatial heterogeneity that makes the design of a feature extraction (FE) algorithm with strong adaptability and robustness very difficult [4,12]. Thus, these methods can achieve good detection results for low-and medium-resolution satellite images or HR images of simple scenes, but their detection performance is often poor when they are applied to the HR images of large-scale complex scenes [13].…”

“…However, due to the effect of image segmentation quality, expressing HR complex scenes is difficult for image objects. Recently, block-based deep learning methods have been applied to extract built-up areas from HR images [12,[35][36][37]. An image block typically contains multiple objects and their spatial distribution patterns.…”

Deep Learning for Built-up Fractional Mapping using Sentinel-2 images: A Case Study in Delhi, India

“…Wu et al [26] use attention-based CNN for delineating built-up areas using TerraSAR-X SAR imagery where attention block (i.e., a combination of convolution and activation layers intended to highlight only the relevant activations during training by giving higher or large weights to relevant parts of the image and less relevant parts get small weights) helped minimize the higher and lower false alarm rate caused by weighted cross-entropy loss. A densely linked dual-attention network with a multiscale context was suggested by Chen et al [27] to extract buildings at the block level. The recommended method divides SAR into multi-scale blocks that partially overlap, utilizing several size grids and their multiple-step offsets.…”

Untitled

“…The principle of DenseNet [36][37][38][39] is to use the outputs of all previous layers as inputs for the next layer, enhancing the data flow by obtaining features from multiple levels. Compared to ResNet, DenseNet can acquire more feature maps with fewer filters, which can reduce the number of parameters to some extent.…”