A review of multi-class change detection for satellite remote sensing imagery

Zhu, Qiqi; Guo, Xi; Li, Ziqi; Li, Deren

doi:10.1080/10095020.2022.2128902

Cited by 56 publications

(15 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the fiveway one-shot scenario, the top three performers achieved accuracy rates of 86.03 ± 0.13% for CS 2 TFSL, 85.41 ± 0.35% for Ji et al [26], and 81.06 ± 0.60% for SPNet [25]. More specifically, CS 2 TFSL not only outperformed the second-place method by a margin of 0.62% in terms of accuracy but also significantly surpassed the other competitors. The outstanding performance of the CS 2 TFSL is also evident in the five-way five-shot scenario.…”

Section: Results On the Whu-rs19 Datasetmentioning

confidence: 91%

“…Remote sensing scene classification refers to the task of identifying and categorizing different objects or scenes in remote sensing imagery [1]. It plays a crucial role in geological exploration [2], environmental monitoring [3], urban planning [4], and disaster monitoring [5]. By performing scene classification, we can obtain vital geographic information to support decision making and resource management, thereby promoting sustainable development and the construction of smart cities [6].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Collaborative Self-Supervised Transductive Few-Shot Learning for Remote Sensing Scene Classification

Han,

Huang,

Wang

2023

Electronics

View full text Add to dashboard Cite

With the advent of deep learning and the accessibility of massive data, scene classification algorithms based on deep learning have been extensively researched and have achieved exciting developments. However, the success of deep models often relies on a large amount of annotated remote sensing data. Additionally, deep models are typically trained and tested on the same set of classes, leading to compromised generalization performance when encountering new classes. This is where few-shot learning aims to enable models to quickly generalize to new classes with only a few reference samples. In this paper, we propose a novel collaborative self-supervised transductive few-shot learning (CS2TFSL) algorithm for remote sensing scene classification. In our approach, we construct two distinct self-supervised auxiliary tasks to jointly train the feature extractor, aiming to obtain a powerful representation. Subsequently, the feature extractor’s parameters are frozen, requiring no further training, and transferred to the inference stage. During testing, we employ transductive inference to enhance the associative information between the support and query sets by leveraging additional sample information in the data. Extensive comparisons with state-of-the-art few-shot scene classification algorithms on the WHU-RS19 and NWPU-RESISC45 datasets demonstrate the effectiveness of the proposed CS2TFSL. More specifically, CS2TFSL ranks first in the settings of five-way one-shot and five-way five-shot. Additionally, detailed ablation experiments are conducted to analyze the CS2TFSL. The experimental results reveal significant and promising performance improvements in few-shot scene classification through the combination of self-supervised learning and direct transductive inference.

show abstract

Section: Results On the Whu-rs19 Datasetmentioning

confidence: 91%

Section: Introductionmentioning

confidence: 99%

Collaborative Self-Supervised Transductive Few-Shot Learning for Remote Sensing Scene Classification

Han,

Huang,

Wang

2023

Electronics

View full text Add to dashboard Cite

show abstract

“…Specifically, deep learning-based multi-class change detection methods can be categorized into two primary types: direct classification and post-classification comparison. Unlike binary change detection, which solely determines the presence or absence of changes, multi-class change detection offers granular insights into specific "from-to" change types [23].…”

Section: A Deep Learning-based Change Detectionmentioning

confidence: 99%

Land Cover Change Detection Based on Vector Polygons and Deep Learning With High-Resolution Remote Sensing Images

Zhang,

Liu,

Niu

et al. 2024

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

Vector Polygons are valuable survey data, serving as crucial outputs of national geographical censuses and a fundamental data source for detecting changes in geographical conditions. Current remote-sensing image change detection methods rely on comparing images but overlook abundant historical vector results, struggle with model generalization, and lack adequate samples. Consequently, change detection remains a manual process primarily, unable to meet the requirements for automated and efficient monitoring of standardized geographical conditions. Hence, this paper proposes a change detection method for land cover vector polygons based on high-resolution remote sensing images and deep learning. Initially, the enhanced simple linear iterative clustering (SLIC) algorithm is applied to segment dual-temporal images from identical regions. Subsequently, an annotated dataset is generated using a multi-scale extraction, cropping-with-inpainting approach. Next, datasets derived from pre-and post-temporal images are used for training and testing, respectively, and the training set is purified by using twoclassifier cross-validation. Finally, an improved object-oriented convolutional neural network (CNN) model performs finegrained scene classification. The change rules and postprocessing method are then integrated to identify changed vector polygons. To validate the effectiveness and superiority of the proposed method, we conducted experiments on land cover change detection using datasets from two study areas. The results indicate that the proposed method achieves precision and recall rates of 91.89% and 94.44% on dataset-1, respectively. Similarly, in dataset-2, the precision and recall rates reach 87.59% and 91.41%, respectively. These findings demonstrate the method's efficacy in detecting changed vector polygons, reducing manual intervention, and enhancing detection efficiency.

show abstract

“…In contrast to BCD, multi-class change detection (MCD) involves identifying two or more categories of changes and requires detecting both the extent and type of change. MCD provides richer change information than BCD; however, the complexity of the MCD task also makes it more challenging than BCD [30]. The schematic diagrams of BCD and MCD are shown in Figure 1, where BCD represents areas of change and no change with two colors, and MCD uses different colors to represent changes in multiple categories, such as from vegetation to bare land or from water to buildings, etc.…”

Section: Introductionmentioning

confidence: 99%

“…Although research on MCD has appeared earlier, the datasets and algorithms for MCD have not been well developed compared to BCD [30]. Some scholars have attempted to extract multi-class change information using convolutional neural networks (CNN), but their application still has many limitations due to the lack of semantic information [59].…”

Section: Introductionmentioning

confidence: 99%

SIGNet: A Siamese Graph Convolutional Network for Multi-Class Urban Change Detection

et al. 2023

View full text Add to dashboard Cite

Detecting changes in urban areas presents many challenges, including complex features, fast-changing rates, and human-induced interference. At present, most of the research on change detection has focused on traditional binary change detection (BCD), which becomes increasingly unsuitable for the diverse urban change detection tasks as cities grow. Previous change detection networks often rely on convolutional operations, which struggle to capture global contextual information and underutilize category semantic information. In this paper, we propose SIGNet, a Siamese graph convolutional network, to solve the above problems and improve the accuracy of urban multi-class change detection (MCD) tasks. After maximizing the fusion of change differences at different scales using joint pyramidal upsampling (JPU), SIGNet uses a graph convolution-based graph reasoning (GR) method to construct static connections of urban features in space and a graph cross-attention method to couple the dynamic connections of different types of features during the change process. Experimental results show that SIGNet achieves state-of-the-art accuracy on different MCD datasets when capturing contextual relationships between different regions and semantic correlations between different categories. There are currently few pixel-level datasets in the MCD domain. We introduce a new well-labeled dataset, CNAM-CD, which is a large MCD dataset containing 2508 pairs of high-resolution images.

show abstract

A review of multi-class change detection for satellite remote sensing imagery

Cited by 56 publications

References 69 publications

Collaborative Self-Supervised Transductive Few-Shot Learning for Remote Sensing Scene Classification

Collaborative Self-Supervised Transductive Few-Shot Learning for Remote Sensing Scene Classification

Land Cover Change Detection Based on Vector Polygons and Deep Learning With High-Resolution Remote Sensing Images

SIGNet: A Siamese Graph Convolutional Network for Multi-Class Urban Change Detection

Contact Info

Product

Resources

About