Land-cover classification with high-resolution remote sensing images using transferable deep models

By Xin-Yi Tong, Guisong Xia, Qikai Lu, Huanfeng Shen, Shengyang Li, Shucheng You. Remote Sensing of Environment, 2018

979 citations recorded by the source corpus.

Abstract In recent years, large amount of high spatial-resolution remote sensing (HRRS) images are available for land-cover mapping. However, due to the complex information brought by the increased spatial resolution and the data disturbances caused by different conditions of image acquisition, it is often difficult to find an efficient method for achieving accurate land-cover classification with high-resolution and heterogeneous remote sensing images. In this paper, we propose a scheme to apply deep model obtained from labeled land-cover dataset to classify unlabeled HRRS images. The main idea is to rely on deep neural networks for presenting the contextual information contained in different types of land-covers and propose a pseudo-labeling and sample selection scheme for improving the transferability of deep models. More precisely, a deep Convolutional Neural Networks (CNNs) is first pre-trained with a well-annotated land-cover dataset, referred to as the source data. Then, given a target image with no labels, the pre-trained CNN model is utilized to classify the image in a patch-wise manner. The patches with high confidence are assigned with pseudo-labels and employed as the queries to retrieve related samples from the source data. The pseudo-labels confirmed with the retrieved results are regarded as supervised information for fine-tuning the pre-trained deep model. To obtain a pixel-wise land-cover classification with the target image, we rely on the fine-tuned CNN and develop a hybrid classification by combining patch-wise classification and hierarchical segmentation. In addition, we create a large-scale land-cover dataset containing 150 Gaofen-2 satellite images for CNN pre-training. Experiments on multi-source HRRS images, including Gaofen-2, Gaofen-1, Jilin-1, Ziyuan-3, Sentinel-2A, and Google Earth platform data, show encouraging results and demonstrate the applicability of the proposed scheme to land-cover classification with multi-source HRRS images.

Subfields: Atmospheric Science, Ecology, Media Technology

View Paper: https://www.semanticscholar.org/paper/a954015bd31711737b3dda7db34ea8ccf5a16e03

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atmospheric Science

1. Qian Du
2. Wei Li
3. Danfeng Hong
4. Jing Yao
5. Yanfei Zhong
6. Bing Zhang
7. Xiaoxiang Zhu
8. Yushi Chen

In Ecology

1. Jing Yao
2. Bing Zhang
3. Danfeng Hong
4. Qian Du
5. Trevor Darrell
6. Yanfei Zhong
7. Xiaoxiang Zhu
8. Wei Li

In Media Technology

1. Danfeng Hong
2. Qian Du
3. Jing Yao
4. Bing Zhang
5. Wei Li
6. Xiaoxiang Zhu
7. Yanfei Zhong
8. Chenyu Li

Citations in the corpus

• Cross-City Matters: A Multimodal Remote Sensing Benchmark Dataset for Cross-City Semantic Segmentation using High-Resolution Domain Adaptation Networks (2023, methodology): “Tong et al. [30] first pre-trained a deep CNN with a well-annotated Gaofen-2 land cover dataset, and transferred the trained deep model for the unlabeled RS image classification in the target domain.”
• Scale-MAE: A Scale-Aware Masked Autoencoder for Multiscale Geospatial Representation Learning (2022, methodology): “In Table 5, we further evaluate Scale-MAE , SatMAE, and ConvMAE across SpaceNet v1, SpaceNet v2 [53], INRIA Aerial Image [44], and GID-15 [59] remote sensing datasets at native resolution.”
• UNetFormer: A UNet-like transformer for efficient semantic segmentation of remote sensing urban scene imagery (2021, neutral): “Since then, CNN-based methods have dominated the semantic segmentation task in the remote sensing field (Kemker et al., 2018; Kotaridis and Lazaridou, 2021; Ma et al., 2019; Tong et al., 2020; Zhao and Du, 2016; Zhu et al., 2017).”
• LoveDA: A Remote Sensing Land-Cover Dataset for Domain Adaptive Semantic Segmentation (2021, methodology): “Recently, adversarial training [13, 35] and self-training [38] have been studied for UDA land-cover semantic segmentation .”
• The Segment Anything Model (SAM) for Remote Sensing Applications: From Zero to One Shot (2023, methodology): “With transfer learning, a pre-trained model on a large dataset is adapted for a different but related task [49].”
• Single-Source Domain Expansion Network for Cross-Scene Hyperspectral Image Classification (2022, background): “University [46], which contains multispectral images (MSI) taken at different times in many regions of China.”