Building Extraction from Very-High-Resolution Remote Sensing Images Using Semi-Supervised Semantic Edge Detection

Abstract

The automated detection of buildings in remote sensing images enables understanding the distribution information of buildings, which is indispensable for many geographic and social applications, such as urban planning, change monitoring and population estimation. The performance of deep learning in images often depends on a large number of manually labeled samples, the production of which is time-consuming and expensive. Thus, this study focuses on reducing the number of labeled samples used and proposing a semi-supervised deep learning approach based on an edge detection network (SDLED), which is the first to introduce semi-supervised learning to the edge detection neural network for extracting building roof boundaries from high-resolution remote sensing images. This approach uses a small number of labeled samples and abundant unlabeled images for joint training. An expert-level semantic edge segmentation model is trained based on labeled samples, which guides unlabeled images to generate pseudo-labels automatically. The inaccurate label sets and manually labeled samples are used to update the semantic edge model together. Particularly, we modified the semantic segmentation network D-LinkNet to obtain high-quality pseudo-labels. Specifically, the main network architecture of D-LinkNet is retained while the multi-scale fusion is added in its second half to improve its performance on edge detection. The SDLED was tested on high-spatial-resolution remote sensing images taken from Google Earth. Results show that the SDLED performs better than the fully supervised method. Moreover, when the trained models were used to predict buildings in the neighboring counties, our approach was superior to the supervised way, with line IoU improvement of at least 6.47% and F1 score improvement of at least 7.49%.

Self-Filtered Learning for Semantic Segmentation of Buildings in Remote Sensing Imagery With Noisy Labels

“…[11] first pre-trained a Fully Convolutional Network [47] (FCN) with raw OSM and then fine-tuned the model with a relatively small amount of clean labels. Although [11] significantly reduced labeling cost by applying a common practice of transfer learning, it requires some portion of clean verified labels, similar to the assumption of the semi-supervised learning set-up [12][13][14]. [23] also confirmed that pre-training with noisy OSM labels benefits deep learning performance and further found that a large amount of OSM labels can be beneficial compared to a small number of clean labels.…”

“…However, we still witnessed a performance gap of SFL compared to the strong supervision case. Although not investigated in this study, integrating SFL with labelrefinement [41][42][43], pre-trained models [11,23], or semisupervised methods [12][13][14] may further close the gap to the strong supervision in practical use.…”

“…The map by experts is highly accurate but has less diverse building features; while the crowd-sourced map has good diversity but is less accurate. Hence, challenges in utilizing publicly available building maps as a training source can be summarized into two problems: 1) the problem of limited To address the problem of limited representation, many studies have addressed this issue through domain adaptation [10,27], transfer learning [11], and semi-supervised learning [12][13][14]. Those approaches generally assume two different domains, one of which is the region where the training samples are collected, the source domain, and the other one is a new region to map, the target domain.…”

Extraction of typical urban buildings in China based on K-Net

“…It is very important to update the information of features in a timely manner and grasp the development and changes of urban space to guide the planning of cities. Among them, the traditional method mainly relies on the relevant staff's industry experience and professional knowledge to collect the feature information, which is time-consuming and labor-intensive 1 . With the rapid development of remote sensing technology, it has achieved wide application in urban planning 2 , crop extraction 3 , and resource monitoring 4 .…”

DEANet: Dual Encoder with Attention Network for Semantic Segmentation of Remote Sensing Imagery

“…In terms of environmental indicator monitoring, machine learning algorithms such as Support Vector Machine (SVM) have been used to assess the percentage of agricultural land contaminated by plastic [7]. For detection of buildings, a semi-supervised deep learning method based on edge detection network D-LinkNet has been designed to understand the distribution of the buildings [8], which is useful for urban planning, change monitoring and population estimation.…”