This paper presents a novel change detection approach for synthetic aperture radar images based on deep learning. The approach accomplishes the detection of the changed and unchanged areas by designing a deep neural network. The main guideline is to produce a change detection map directly from two images with the trained deep neural network. The method can omit the process of generating a difference image (DI) that shows difference degrees between multitemporal synthetic aperture radar images. Thus, it can avoid the effect of the DI on the change detection results. The learning algorithm for deep architectures includes unsupervised feature learning and supervised fine-tuning to complete classification. The unsupervised feature learning aims at learning the representation of the relationships between the two images. In addition, the supervised fine-tuning aims at learning the concepts of the changed and unchanged pixels. Experiments on real data sets and theoretical analysis indicate the advantages, feasibility, and potential of the proposed method. Moreover, based on the results achieved by various traditional algorithms, respectively, deep learning can further improve the detection performance.

Change Detection in Synthetic Aperture Radar Images Based on Deep Neural Networks

Multi-spatial-resolution change detection is a newly proposed issue and it is of great significance in remote sensing, environmental and land use monitoring, etc. Though multi-spatial-resolution image-pair are two kinds of representations of the same reality, they are often incommensurable superficially due to their different modalities and properties. In this paper, we present a novel multi-spatial-resolution change detection framework, which incorporates deep-architecture-based unsupervised feature learning and mapping-based feature change analysis. Firstly, we transform multi-resolution image-pair into the same pixel-resolution through co-registration, followed by details recovery, which is designed to remedy the spatial details lost in the registration. Secondly, the denoising autoencoder is stacked to learn local and high-level representation/feature from the local neighborhood of the given pixel, in an unsupervised fashion. Thirdly, motivated by the fact that multi-resolution image-pair share the same reality in the unchanged regions, we try to explore the inner relationships between them by building a mapping neural network. And it can be used to learn a mapping function based on the most-unlikely-changed feature-pairs, which are selected from all the feature-pairs via a coarse initial change map generated in advance. The learned mapping function can bridge the different representations and highlight changes. Finally, we can build a robust and contractive change map through feature similarity analysis, and the change detection result is obtained through the segmentation of the final change map. Experiments are carried out on four real datasets, and the results confirmed the effectiveness and superiority of the proposed method.

Change detection based on deep feature representation and mapping transformation for multi-spatial-resolution remote sensing images

The expected increasing availability of remote sensing satellite hyperspectral (HS) images provides an important and unique data source for Earth observation (EO). HS images are characterized by a detailed spectral sampling (i.e., very high spectral resolution) over a wide spectral wavelength range, which makes it possible to monitor land-cover dynamics at a fine spectral scale. This is due to its capability of detecting subtle spectral variations in multitemporal images associated with land-cover changes that are not detectable in traditional multispectral (MS) images because of their limited spectral resolution (i.e., sufficient for representing only abrupt, strong changes in the spectral signature, as a rule). To fully exploit the available multitemporal HS images and their rich information content in change detection (CD), it is necessary to develop advanced automatic techniques that can address the complexity of the extraction of change information in an HS space. This article provides a comprehensive overview of the CD problem in HS images, as well as a survey on the main CD techniques available for multitemporal HS images. We review both widely used methods and new techniques proposed in the recent literature. The basic concepts, categories, open issues, and challenges related to CD in HS images are discussed and analyzed in detail. Experimental results obtained using state-of-the-art approaches are shown, to illustrate relevant concepts and problems.

A Review of Change Detection in Multitemporal Hyperspectral Images: Current Techniques, Applications, and Challenges

Urban growth and environmental impacts in Jing-Jin-Ji, the Yangtze, River Delta and the Pearl River Delta

As a significant role for traffic management, city planning, road monitoring, GPS navigation and map updating, the technology of road extraction from a remote sensing (RS) image has been a hot research topic in recent years. In this paper, after analyzing different road features and road models, the road extraction methods were classified into the classification-based methods, knowledge-based methods, mathematical morphology, active contour model, and dynamic programming. Firstly, the road features, road model, existing difficulties and interference factors for road extraction were analyzed. Secondly, the principle of road extraction, the advantages and disadvantages of various methods and research achievements were briefly highlighted. Then, the comparisons of the different road extraction algorithms were performed, including road features, test samples and shortcomings. Finally, the research results in recent years were summarized emphatically. It is obvious that only using one kind of road features is hard to get an excellent extraction effect. Hence, in order to get good results, the road extraction should combine multiple methods according to the real applications. In the future, how to realize the complete road extraction from a RS image is still an essential but challenging and important research topic.

A review of road extraction from remote sensing images

The objective of this research is to examine effective methods for urban change detection using multitemporal spaceborne SAR data in two rapid expanding cities in China. One scene of ERS-2 SAR C-VV image was acquired in Beijing in 1998 and in shanghai in 1999 respectively and one scene of ENVISAT ASAR C-VV image was acquired in near-anniversary dates in 2008 in Beijing and Shanghai. To compare the SAR images from different dates, a modified ratio operator that takes into account both positive and negative changes was developed to derive a change image. A generalized version of Kittler-Illingworth minimum-error thresholding algorithm was then tested to automatically classify the change image into two classes, change and no change. Various probability density functions such as Log normal, Generalized Gaussian, Nakagami ratio, and Weibull ratio were investigated to model the distribution of the change and no change classes. The results showed that Kittler-Illingworth algorithm applied to the modified ratio image is very effective in detecting temporal changes in urban areas using SAR images. Log normal and Nakagami density models achieved the best results. The Kappa coefficients of these methods were of 0.82 and 0.71 for Beijing and Shanghai respectively while the false alarm rates were 2.7% and 4.75%. The findings indicated that the change accuracies obtained using Kittler-Illingworth algorithm vary depending on how the assumed conditional class density function fits the histograms of change and no change classes.

Multitemporal Spaceborne SAR Data for Urban Change Detection in China

Multitemporal synthetic aperture radar (SAR) images have been increasingly used in change detection studies. However, the presence of speckle is the main disadvantage of this type of data. To reduce speckle, many local adaptive filters have been developed. Although these filters are effective in reducing speckle in homogeneous areas, their use is often accompanied with the degradation of spatial details and fine structures. In this paper, we investigate a nonlocal means (NLM) denoising algorithm that combines local structures with a global averaging scheme in the context of change detection using multitemporal SAR images. First, the ratio image is logarithmically scaled to convert the multiplicative noise model to an additive model. A multidimensional change image is then constructed using image neighborhood feature vectors. Principle component analysis is then used to reduce the dimensionality of the neighborhood feature vectors. Recursive linear regression combined with fitting-accuracy assessment strategy is developed to determine the number of significant PC components to be retained for similarity weight computation. An intuitive method to estimate the unknown noise variance (necessary to run the NLM algorithm) based on the discarded PC components is also proposed. The efficiency of the method has been assessed using two different bitemporal SAR datasets acquired in Beijing and Shanghai, respectively. For comparison purposes, the algorithm is also tested against some of the most commonly used local adaptive filters. Qualitative and quantitative analyses of the algorithm have demonstrated the efficiency of the algorithm in recovering the noise-free change image while preserving the complex structures in urban areas.

Improving Urban Change Detection From Multitemporal SAR Images Using PCA-NLM

In remote sensing change detection, Markov random field (MRF) has been used successfully to model the prior probability using class-labels dependencies MRF has played an important role in the detection of complex urban changes using optical images However, the preservation of details in urban change analysis turns out to be a highly complex task if multitemporal SAR images with their speckle are to be used Here, the ability of MRF to preserve geometric details and to combat speckle effect at the same time becomes questionable Blob-region phenomenon and fine structures removal are common consequences of the application of traditional MRF-based change detection algorithm To overcome these limitations, the iterated conditional modes (ICM) framework for the optimization of the maximum a posteriori (MAP-MRF) criterion function is extended to include a nonlocal probability maximization step This probability model, which characterizes the relationship between pixels’ class-labels in a nonlocal scale, has the potential to preserve spatial details and to reduce speckle effects Two multitemporal SAR datasets were used to assess the proposed algorithm Experimental results using three density functions [ie, the log normal (LN), generalized Gaussian (GG), and normal distributions (ND)] have demonstrated the efficiency of the proposed approach in terms of detail preservation and noise suppression Compared with the traditional MRF algorithm, the proposed approach proved to be less-sensitive to the value of the contextual parameter and the chosen density function The proposed approach has also shown less sensitivity to the quality of the initial change map when compared with the ICM algorithm

Improving SAR-Based Urban Change Detection by Combining MAP-MRF Classifier and Nonlocal Means Similarity Weights

Mapping Earth’s surface and its rapid changes with remotely sensed data is a crucial task to understand the impact of an increasingly urban world population on the environment. However, the impressive amount of available Earth observation data is only marginally exploited in common classifications. In this study, we use the computational power of Google Earth Engine and Google Cloud Platform to generate an oversized feature set in which we explore feature importance and analyze the influence of dimensionality reduction methods to object-based land cover classification with Support Vector Machines. We propose a methodology to extract the most relevant features and optimize an SVM classifier hyperparameters to achieve higher classification accuracy. The proposed approach is evaluated in two different urban study areas of Stockholm and Beijing. Despite different training set sizes in the two study sites, the averaged feature importance ranking showed similar results for the top-ranking features. In particular, Sentinel-2 NDVI, NDWI, and Sentinel-1 VV temporal means are the highest ranked features and the experiment results strongly indicated that the fusion of these features improved the separability between urban land cover and land use classes. Overall classification accuracies of 94% and 93% were achieved in Stockholm and Beijing study sites, respectively. The test demonstrated the viability of the methodology in a cloud-computing environment to incorporate dimensionality reduction as a key step in the land cover classification process, which we consider essential for the exploitation of the growing Earth observation big data. To encourage further research and development of reliable workflows, we share our datasets and publish the developed Google Earth Engine and Python scripts as free and open-source software.

Dimensionality Reduction and Feature Selection for Object-Based Land Cover Classification based on Sentinel-1 and Sentinel-2 Time Series Using Google Earth Engine

With its synoptic view and the repeatability, satellite remote sensing can provide timely, accurate and consistent information about earth’s surface for cost-effective monitoring of environmental changes. In this chapter, recent development in change detection techniques using multitemporal remotely sensed images were reviewed. The chapter covers change detection methods for both optical and SAR images. Various aspects of change detection processes were presented including data preprocessing, change image generation and change detection algorithms such as unsupervised and supervised change detection as well as pixel-based and object-based change detection. The review shows that significant progress has been made in the field of change detection and innovative methods have been developed for change detection using both multitemporal SAR and optical data. Attempts have been made for change detection using multitemporal multisensor/cross-sensor images. The review also identified a number of challenges and opportunities in change detection.

Osama A. Yousif

Papers

Multitemporal Spaceborne SAR Data for Urban Change Detection in China

Improving Urban Change Detection From Multitemporal SAR Images Using PCA-NLM

Improving SAR-Based Urban Change Detection by Combining MAP-MRF Classifier and Nonlocal Means Similarity Weights

Dimensionality Reduction and Feature Selection for Object-Based Land Cover Classification based on Sentinel-1 and Sentinel-2 Time Series Using Google Earth Engine

Change Detection Techniques: A Review