Unsupervised Change Detection for Remote Sensing Images Based on Principal Component Analysis and Differential Evolution

Numerous land-cover change detection techniques have been developed with varying opinions about their appropriateness and success. Decisions on the selection of the most suitable change detection method is often influenced by the study region landscape complexity and the type of data used for analysis. For different climatic areas, the method that suits best the seasonal land-cover change identification remains uncertain. In this study, 11 different binary change detection methods were tested and compared with respect to their capability in detecting land-cover change/no-change information in different seasons. The methods include image differencing (I_Diff), Improved image differencing (Imp_Diff), principal component image differencing (PC_Diff), vegetation index differencing (VI_Diff), change vector analysis (CVA), image ratioing (IR), improved image ratioing (Imp_IR), vegetation index image ratioing (VI_R), multi-date principal component analysis (PCA) using all bands (M_PCA), two-date bands PCA (B_PCA), and two-date vegetation index images PCA (VI_PCA). Multi-Date Thematic Mapper (TM) data were used for a wide set of change image generation. A relatively new approach was applied for optimal threshold value determination for the separation of change/no-change areas. Research results indicated that any methods involving TM Band 4 performed better than those using TM Band 3 or 5 on each of the change periods. However, irrespective of the method used, the accuracy assessment and change/no-change validation results from normalized difference vegetation index (NDVI)-based techniques outperformed all other tested techniques in the change detection process (overall accuracy >90% and kappa value >0.85 for all six change periods). The image differencing technique was found to be marginally better than PCA and IR in most cases and any of these techniques can be used for change detection. However, because of the simplistic nature and relative ease in identifying both negative and positive changes from difference images, the NDVI differencing technique is recommended for seasonal land-cover change identification in the study region.

ABSTRACTIn this article, we propose a novel difference image (DI) creation method for unsupervised change detection in multi-temporal multi-spectral remote-sensing images based on deep learning theory. First, we apply deep belief network to learn local and high-level features from the local neighbour of a given pixel in an unsupervised manner. Second, a back propagation algorithm is improved to build a DI based on selected training samples, which can highlight the difference on changed regions and suppress the false changes on unchanged regions. Finally, we get the change trajectory map using simple clustering analysis. The proposed scheme is tested on three remote-sensing data sets. Qualitative and quantitative evaluations show its superior performance compared to the traditional pixel-level and texture-level-based approaches.

AN ASSESSMENT OF SPATIAL VARIATION OF LAND SURFACE CHARACTERISTICS OF MINNA, NIGER STATE NIGERIA FOR SUSTAINABLE URBANIZATION USING GEOSPATIAL TECHNIQUES

Change detection is usually treated as a problem of explicitly detecting land cover transitions in satellite images obtained at different times, and helps with emergency response and government management. This study presents an unsupervised change detection method based on the image fusion of multi-temporal images. The main objective of this study is to improve the accuracy of unsupervised change detection from high-resolution multi-temporal images. Our method effectively reduces change detection errors, since spatial displacement and spectral differences between multi-temporal images are evaluated. To this end, a total of four cross-fused images are generated with multi-temporal images, and the iteratively reweighted multivariate alteration detection (IR-MAD) method—a measure for the spectral distortion of change information—is applied to the fused images. In this experiment, the land cover change maps were extracted using multi-temporal IKONOS-2, WorldView-3, and GF-1 satellite images. The effectiveness of the proposed method compared with other unsupervised change detection methods is demonstrated through experimentation. The proposed method achieved an overall accuracy of 80.51% and 97.87% for cases 1 and 2, respectively. Moreover, the proposed method performed better when differentiating the water area from the vegetation area compared to the existing change detection methods. Although the water area beneath moderate and sparse vegetation canopy was captured, vegetation cover and paved regions of the water body were the main sources of omission error, and commission errors occurred primarily in pixels of mixed land use and along the water body edge. Nevertheless, the proposed method, in conjunction with high-resolution satellite imagery, offers a robust and flexible approach to land cover change mapping that requires no ancillary data for rapid implementation.

The purpose of this study was to examine the influence of image properties on the accuracy of remote sensing change detection methods. Spectral class separability, radiometric normalization and image band correlation were evaluated through experiments with simulated data. The experimental results were then evaluated as a tool for predicting the relative accuracy of change detection results obtained from Landsat TM satellite image pairs of three U.S. cities: Las Vegas, Nevada; Phoenix, Arizona; and Atlanta, Georgia. The change detection methods used were post-classification comparison, direct classification, image differencing, principal component analysis, and change vector analysis. Results of the simulated experiments confirmed that the relative accuracy of the change detection methods varied with changes in image properties. For the class separability experiments, post-classification comparison, direct classification, image differencing, and PCA with a large number of the principal components, were found generally to have higher accuracies than CVA and PCA with a small number of the principal components. For classes with very good separability, image differencing is an excellent method; for classes with poor spectral separability, image differencing was found to have the lowest accuracy. The influence of the error in radiometric normalization on the accuracy of change detection techniques varied greatly with different degrees of class separability. This can be seen particularly well with image differencing, which showed the highest sensitivity to large changes in radiometric error and very poor class separability. Image differencing and PCA were also found to be more sensitive to band correlation. The classification of the real change detection data from the Landsat pairs showed complex and varying patterns, depending on whether complete (mapping all unchanged and changed transitions) or partial (grouping all unchanged transitions into a single class) change analysis was conducted. However, image differencing was relatively consistent in producing good results for the real data. Also, PCA produced satisfactory results for all three cities. On the other hand, the CVA was found to have amongst the lowest accuracies of the partial change detection methods, though this was higher than most of the complete change accuracy. The variation of accuracy results obtained from the change detection methods used in this study suggests that the contradictory results found in previous change detection studies is likely at least partially a result of varying image properties.

In recent decades, high-resolution (HR) remote sensing images have shown considerable potential for providing detailed information for change detection. The traditional change detection methods based on HR remote sensing images mostly only detect a single land type or only the change range, and cannot simultaneously detect the change of all object types and pixel-level range changes in the area. To overcome this difficulty, we propose a new coarse-to-fine deep learning-based land-use change detection method. We independently created a new scene classification dataset called NS-55, and innovatively considered the adaptation relationship between the convolutional neural network (CNN) and the scene complexity by selecting the CNN that best fit the scene complexity. The CNN trained by NS-55 was used to detect the category of the scene, define the final category of the scene according to the majority voting method, and obtain the changed scene by comparison to obtain the so-called coarse change result. Then, we created a multi-scale threshold (MST) method, which is a new method for obtaining high-quality training samples. We used the high-quality samples selected by MST to train the deep belief network to obtain the pixel-level range change detection results. By mapping coarse scene changes to range changes, we could obtain fine multi-type land-use change detection results. Experiments were conducted on the Multi-temporal Scene Wuhan dataset and aerial images of a particular area of Dapeng New District, Shenzhen, where promising results were achieved by the proposed method. This demonstrates that the proposed method is practical, easy-to-implement, and the NS-55 dataset is physically justified. The proposed method has the potential to be applied in the large scale land use fine change detection problem and qualitative and quantitative research on land use/cover change based on HR remote sensing data.

ABSTRACTThis letter presents a novel method for unsupervised change detection (CD) from remote sensing images using level set evolution with local uncertainty constraints (LSELUC). Uncertainty analysis of pixel labels was implemented as prior information to guide the evolution of level curves. Then, local uncertainty and gradient information of level curves were incorporated into the level set energy function to construct local energy constraints. The proposed method can reduce noise, while preserving details in change regions. Furthermore, an advanced regularization strategy of the level set function was adopted to improve the computational efficiency. The performance of the proposed method was validated on two remote sensing data sets. Experimental results show that the proposed method can produce satisfactory CD results.

Updating geographical database is an important task which can be realized by change detection. Objects of change detection usually include residential areas, roads, rivers, and so on. Because of different geometrical shapes and physical properties, different objects have different change detection methods. This paper aims at river change detection and introduces a compositive method composed of wall is filter, pattern classifying, region growing, morphology, etc

Land cover change detection (LCCD) based on bi-temporal remote sensing images plays an important role in the inventory of land cover change. Due to the benefit of having spatial dependency properties within the image space while using remote sensing images for detecting land cover change, many contextual information-based change detection methods have been proposed in past decades. However, there is still a space for improvement in accuracies and usability of LCCD. In this paper, a LCCD method based on adaptive contextual information is proposed. First, an adaptive region is constructed by gradually detecting the spectral similarity surrounding a central pixel. Second, the Euclidean distance between pairwise extended regions is calculated to measure the change magnitude between the pairwise central pixels of bi-temporal images. All the bi-temporal images are scanned pixel by pixel so the change magnitude image (CMI) can be generated. Then, the Otsu or a manual threshold is employed to acquire the binary change detection map (BCDM). The detection accuracies of the proposed approach are investigated by three land cover change cases with Landsat bi-temporal remote sensing images and aerial images with very high spatial resolution (0.5 m/pixel). In comparison to several widely used change detection methods, the proposed approach can produce a land cover change inventory map with a competitive accuracy.

In the context of multi-temporal SAR change detection for earth monitoring applications, one critical issue is to generate accurate change map. A common method to generate change map is to apply logarithm to the ratio image. However, due to the speckle effect and without consideration of contextual information, it is usually not efficient for accurate change detection. In this paper, an unsupervised change detection method in wavelet domain based on statistical wavelet subband modeling is proposed. The motivation is to capture textures efficiently in wavelet domain. Wavelet transform is applied to decompose the image into multiple scales and probability density function of the coefficient magnitudes of each subband assumed to be Generalized Gaussian Distribution (GGD) and Generalized Gamma Distribution $({rm G}Gamma{rm D})$ are obtained by fast parameter estimation. Closed-form expression of Kullback-Leibler divergence between two corresponding subbands of the same scale is computed and used to generate the change map. This approach is comprehensively evaluated and compared using different parameter setting, different scales, window sizes and estimators. The proposed SAR change detection in wavelet domain shows promising results as texture can be better characterized in wavelet domain than in spatial domain. Through this study, we conclude that the accuracy depends heavily on the estimation methods although the model is important. Both parameter estimation for GGD based on shape equation and parameter estimation for ${rm G}Gamma{rm D}$ using method of log-cumulants (MoLC) in wavelet domain performs quite well.

Preface The objective of this report is to review the requirements for disturbance information, the manifestation of damage patterns that may be encountered, and to provide an overview of remote sensing sensors and change detection methods that have been, or could be applied to mapping of insect defoliation and aspen dieback. It was developed with financial support from the Canadian Space Agency (CSA) by the Canadian Forest Service (CFS) and the Canada Centre for Remote Sensing (CCRS) as part of a Government-Related Initiatives Program (GRIP) project entitled "Gauging the Health of Canada's Forests: Accounting for Insect Defoliation and Dieback in the Indicators of Sustainability for Canadians". The report was first submitted as a deliverable to the CSA in 2007 and reviews the utility of both optical and radar remote sensing sensors and change detection approaches. For this release of the report as a Geomatics Canada Open File, the text was revised in part to relay major developments regarding the availability of satellite sensors and change detection methods in particular. A comprehensive review of literature published after 2007 was beyond the scope of the revision. A recent review paper by Hall et al. (2016) draws from this report but is limited to a discussion of the utility of optical sensor systems and change detection methods.

Remote Sensing (RS) images capture spatial-temporal data on the Earth's surface that is valuable for understanding geographical changes over time. Change detection (CD) is applied in monitoring land use patterns, urban development, evaluating disaster impacts among other applications. Traditional CD methods often face challenges in distinguishing between changes and irrelevant variations in data, arising from comparison of pixel values, without considering their context. Deep feature based methods have shown promise due to their content extraction capabilities. However, feature extraction alone might not be enough for accurate CD. This study proposes incorporating spatial-temporal dependencies to create contextual understanding by modelling relationships between images in space and time dimensions. The proposed model processes dual time points using parallel encoders, extracting highly representative deep features independently. The encodings from the dual time points are then concatenated and passed through Long Short Term Memory (LTSM) layers and a decoder. The output from the LSTM is then concatenated with that from the decoder in a space-time feature fusion. This optimizes information representation of spectral, spatial and temporal details, in RS images before further analysis of change. This approach aims to maximize authentic information while reducing noise interference by introducing the context of change. Compared to conventional CD methods, the proposed technique achieves higher overall accuracy of 97.4%, an F1 Score of 89% and an intersection over union (IoU) of 86.7 when evaluated on the EGY-BCD dataset. The results demonstrate the potential of incorporating spatial-temporal dependencies in CD tasks for RS images.

Characterized by tremendous spectral information, hyperspectral image is able to detect subtle changes and discriminate various change classes for change detection. The recent research works dominated by hyperspectral binary change detection, however, cannot provide fine change classes information. And most methods incorporating spectral unmixing for hyperspectral multiclass change detection (HMCD), yet suffer from the neglection of temporal correlation and error accumulation. In this study, we proposed an unsupervised Binary Change Guided hyperspectral multiclass change detection Network (BCG-Net) for HMCD, which aims at boosting the multiclass change detection result and unmixing result with the mature binary change detection approaches. In BCG-Net, a novel partial-siamese united-unmixing module is designed for multi-temporal spectral unmixing, and a groundbreaking temporal correlation constraint directed by the pseudo-labels of binary change detection result is developed to guide the unmixing process from the perspective of change detection, encouraging the abundance of the unchanged pixels more coherent and that of the changed pixels more accurate. Moreover, an innovative binary change detection rule is put forward to deal with the problem that traditional rule is susceptible to numerical values. The iterative optimization of the spectral unmixing process and the change detection process is proposed to eliminate the accumulated errors and bias from unmixing result to change detection result. The experimental results demonstrate that our proposed BCG-Net could achieve comparative or even outstanding performance of multiclass change detection among the state-of-the-art approaches and gain better spectral unmixing results at the same time.

It is quite an important and challenging problem for change detection (CD) from heterogeneous remote sensing images. The images obtained from different sensors (i.e., synthetic aperture radar (SAR) & optical camera) characterize the distinct properties of objects. Thus, it is impossible to detect changes by direct comparison of heterogeneous images. In this article, a new unsupervised change detection (USCD) method is proposed based on image translation. The cycle-consistent adversarial networks (CycleGANs) are employed to learn the subimage to subimage mapping relation using the given pair (i.e., before and after the event) of heterogeneous images from which the changes will be detected. Then, we can translate one image (e.g., SAR) from its original feature space (e.g., SAR) to another space (e.g., optical). By doing this, the pair of images can be represented in a common feature space (e.g., optical). The pixels with close pattern values in the before-event image may have quite different values in the after-event image if the change happens on some ones. Thus, we can generate the difference map between the translated before-event image and the original after-event image. Then, the difference map is divided into changed and unchanged parts. However, these detection results are not very reliable. We will select some significantly changed and unchanged pixel pairs from the two parts with the clustering technique (i.e., <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> -means). These selected pixel pairs are used to learn a binary classifier, and the other pixel pairs will be classified by this classifier to obtain the final CD results. Experimental results on different real datasets demonstrate the effectiveness of the proposed USCD method compared with several other related methods.

The paper describes the use of Principal Component Analysis (PCA) of remote sensing images as a method of change detection for the Kafue Flats, an inland wetland system in southern Zambia. The wetland is under human and natural pressures but is also an important wildlife habitat. A combination of Landsat MSS and TM images were used. The images used were from 24 September 1984 (MSS), 3 September 1988 (MSS), 12 September 1991 (TM) and 20 September 1994 (TM). They were geometrically co‐registered and, in the process, the 80m resolution MSS images were resampled to 30m using nearest neighbour resampling. Preliminary PCA revealed that for the MSS images most of the data variance was in near infrared reflectance while for the TM images it was in mid and thermal infrared bands. Holding sensor type constant, separate inter‐band correlation analysis for each image could indicate whether the wetland was drier or wetter on one date versus another. The 1994 image was made the reference image and equivalent green, red and near infrared bands from the other images were radiometrically normalised with those on the reference image. All the bands, three from each date, were then merged into a twelve‐band image on which PCA for change detection was undertaken. A colour composite of eigen images from the resulting principal components was used in change detection. Hydrological data, indicating long‐term reduced inflow of water into the wetland due to human regulation, help explain some of the wetland change detected. Compared to a classification comparison approach to change detection for this area, PCA was found to be very useful in indicating where change had occurred, though interpretation of the changes was difficult without reference to the input images. The methodology appears to have potential use in habitat monitoring for this wetland area.