Change Detection Algorithm Research Articles

Continuous monitoring and sub-annual change detection in high-latitude forests using Harmonized Landsat Sentinel-2 data

Spatially and temporally precise monitoring of forest resources is becoming increasingly vital in rapidly changing environmental and economic conditions. However, recent developments in moderate resolution satellite imagery have led to spectrally consistent and temporally dense datasets capable of capturing both seasonal variation and long-term trajectories. Previous studies have demonstrated the ability of continuous monitoring algorithms to detect transitional changes such as timber harvesting at sub-annual scales. However, less is known about the ability to use these algorithms in areas such as high-latitude forests which have large data gaps over winter due to prolonged snow cover. This study investigated the use of a time series of 7-day Harmonized Landsat Sentinel-2 (HLS) composites along with the Bayesian Estimator of Abrupt Change, Seasonality, and Trend (BEAST) algorithm for change detection and monitoring in two managed forest areas in Canada. On average, both sites had no data periods (over winter) lasting ∼200 days. Despite this, BEAST successfully detected change, with overall accuracies of 88.7% and 97.3% for both sites when compared to an independent validation dataset derived from visual interpretation of PlanetScope images. A simulation of change detection in near-real time revealed that approximately 75% of changes were detected within 7 and 11 valid images for each site. Furthermore, monitoring the trend slope of the spectral index showed that areas with a new declining slope were typically about twice as likely to occur in photointerpreted polygons of non-stand replacing disturbance, indicating that such a technique could be used as an early warning signal of fine-scale changes such as those due to insect defoliation or drought. Overall, this study demonstrated the efficacy of a continuous change detection and monitoring framework, which could be used to provide insights at previously unavailable temporal and spatial scales, thereby allowing forest managers the ability to quickly react to changes in uncertain future conditions.

Mapping Cropland Abandonment in Mountainous Areas in China Using the Google Earth Engine Platform

Knowledge about the spatial-temporal pattern of cropland abandonment is the premise for the management of abandoned croplands. Traditional mapping approaches of abandoned croplands usually utilize a multi-date classification-based land cover change trajectory. It requires quality training samples for land cover classification at each epoch, which is challenging in regions of smallholder agriculture in the absence of high-resolution imagery. Facing these challenges, a theoretical model is proposed to recognize abandoned croplands based on post-abandonment secondary succession. It applies the continuous change detection and classification (CCDC) temporal segmentation algorithm to Landsat time series (1986~2021) to obtain disjoint segments, representing croplands’ status. The post-abandonment secondary succession showing a greening trend is recognized using NDVI-based harmonic analysis, so as to capture its preceding abandonment. This algorithm is applied to a mountainous area in southwest China, where cropland abandonments are widespread. Validation based on stratified random samples referenced by a vegetation index time series and satellite images shows that the detected abandoned croplands have user accuracy, producer accuracy and an F1 score ranging from 43% to 71%, with variation among abandonment year. The study area has a potential cropland extent of 22,294 km2, within which 9252 km2 of the cropland was abandoned. The three peak years of abandonment were 1994, 2000, and 2011. The algorithm is suitable to be applied to large-scale mapping due to its automatic manner.

A novel approach towards continuous monitoring of forest change dynamics in fragmented landscapes using time series Landsat imagery

Carbon emissions from forest ecosystems are greatly impacted by the acceleration of fragmentation and edge effects. Understanding these effects requires accurate monitoring of changes in fragmented forest landscapes. However, these changes are often low-intensity and small-scale, making it difficult to detect them using medium spatial resolution satellite images (e.g., Landsat). To address this challenge, this study developed the Pure Forest Index (PFI), which uses a combination of the existing vegetation index (VI) and spectral mixture analysis (SMA) to more effectively detect and characterize the contribution of forests to the observed spectral response of a pixel. The PFI was applied to detect forest changes in the Amazon rainforest from 1986 to 2020 using the Continuous Change Detection and Classification (CCDC) algorithm (hereafter referred to as the CCDC-PFI algorithm). The results showed reliable performance in mapping forest changes, with an overall accuracy of 0.94 (±0.03) at the spatial scale and a temporal accuracy of 91.1 % (within a two-year window). Comparison with other indices revealed that the PFI improves the ability to monitor forest dynamics with an increased overall accuracy of 0.02–0.35. The PFI also demonstrated advantages in enhancing sub-pixel forest information and suppressing non-forest backgrounds in various scenes compared to conventional VIs. The proposed approach is expected to benefit further research on forests and ecosystems.

Application of Satellite Remote Sensing, UAV-Geological Mapping, and Machine Learning Methods in the Exploration of Podiform Chromite Deposits

The irregular and sporadic occurrence of chromite pods in podiform chromite deposits (PCD), especially in mountainous terranes with rough topography, necessitates finding innovative methods for reconnaissance and prospecting. This research combines several remote sensing methods to discriminate the highly serpentinized peridotites hosting chromite pods from the other barren ultramafic and mafic cumulates. The case study is the area of the Sabzevar Ophiolite (NE Iran), which hosts several known chromite and other mineral deposits. The integration of satellite images [e.g., Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) satellite sensor, Landsat series, and Sentinel-2] coupled with change detection, band rationing, and target detection algorithms [including the Spectral Angle Mapper (SAM)] were used to distinguish potential lithological units hosting chromites. Results have been verified by an initial on-field checking and compared with the high-resolution (GSD ~6 cm) orthomosaic images obtained by the processing of photographs taken from an Unmanned Aerial Vehicle (UAV) at a promising area of 35 km2. The combination of visual interpretation and supervised classification by machine learning methods [Support Vector Machine (SVM)] yielded the production of a geological map, in which the lithological units and structures are outlined, including the crust-mantle transition zone units, mafic cumulates, crosscutting dykes, and mantle sequences. The validation of the results was performed through a second phase, made up of field mapping, sampling, chemical analysis, and microscopic studies, leading to the discovery of new chromite occurrences and mineralized zones. All ultramafic units were classified into four groups based on the degree of serpentinization, represented by the intensity of their average spectral reflectance. Based on their presumed protolith, the highly serpentinized ultramafics and serpentinites were classified into two main categories (dunite or harzburgite). The serpentinite with probable dunitic protolith, discriminated for a peculiar Fe-rich Ni-bearing lateritic crust, is more productive for chromite prospecting. This is particularly true at the contact with mafic dykes, akin to some worldwide chromite deposits. The results of our work highlight the potential of multi-scale satellite and UAV-based remote sensing to find footprints of some chromite mineral deposits.

Globally vs. Locally Trained Machine Learning Models for Landslide Detection: A Case Study of a Glacial Landscape

Landslide risk mitigation is limited by data scarcity; however, this could be improved using continuous landslide detection systems. To investigate which image types and machine learning models are most useful for landslide detection in a Norwegian setting, we compared the performance of five different machine learning models, for the Jølster case study (30 July 2019), in Western Norway. These included three globally pre-trained models; (i) the continuous change detection and classification (CCDC) algorithm, (ii) a combined k-means clustering and random forest classification model, and (iii) a convolutional neural network (CNN), and two locally trained models, including; (iv) classification and regression Trees and (v) a U-net CNN model. Images used included Sentinel-1, Sentinel-2, as well as digital elevation model (DEM) and slope. The globally trained models performed poorly in shadowed areas and were all outperformed by the locally trained models. A maximum Matthew’s correlation coefficient (MCC) score of 89% was achieved with a CNN U-net deep learning model, using combined Sentinel-1 and -2 images as input. This is one of the first attempts to apply deep learning to detect landslides with both Sentinel-1 and -2 images. Using Sentinel-1 images only, the locally-trained deep-learning model significantly outperformed the conventional machine learning model. These findings contribute to developing a national continuous monitoring system for landslides.

Temperature is positively associated with tree mortality in California subalpine forests containing whitebark pine

AbstractWhitebark pine (Pinus albicaulis Engelm.) is a keystone high‐elevation tree species occurring across much of western North America, yet it has been listed as threatened under the US Endangered Species Act due to rapid population declines and extensive ongoing pressures from white pine blister rust (Cronartium ribicola), mountain pine beetle (Dendroctonus ponderosae), and increasing temperature and aridity associated with climate change. Past research has shown that whitebark pine mortality is more likely in hotter and drier sites, but no broad‐extent analyses spanning multiple ecoregions have directly quantified the relationship between mortality probability and its environmental drivers at a fine spatial scale. To address this gap, we used spatially continuous high‐resolution (30 m) maps of mortality in California produced by a remote sensing‐based change detection algorithm, and we developed generalized additive models to explain this mortality variability using relevant biophysical variables. We found that maximum daily temperature was strongly positively associated with mortality intensity and that mortality was also generally more intense in sites with greater solar exposure (e.g., south‐facing slopes). Our results support the interpretation that as climate warms and aridity increases, populations in “trailing edge” (i.e., hot, dry) sites are most vulnerable. The environmental relationships we quantify will aid in identifying the locations and environments with the greatest risk of future mortality as well as those likely to serve as refugia.

Flexible strain sensor with a hat-shaped structure for in situ measurement of 3D deformation

Flexible strain sensors that are currently available are mainly used in human motion recognition and medical health detection applications, and there is still an urgent need for sensors to realize real-time monitoring of the 3D deformation of industrial and agricultural products. In this work, a flexible strain sensor with a hat-shaped structure was fabricated using a molding technique to perform in situ measurement of 3D deformation. An algorithm for resistance change detection and linear calibration equations were proposed to enable analysis of the deformation data and calculation of local shape changes. The sensor was applied to monitor the growth deformation of a kumquat fruit, and the results were highly consistent with the algorithm. The proposed technique has great potential for application to 3D deformation detection of flexible objects.

Slope deformation detection using subpixel offset tracking and an unsupervised learning technique based on unmanned aerial vehicle photogrammetry data

Detecting slope deformation is an important issue in engineering. Timely deformation detection can effectively avoid catastrophic slope failure and ensure the safety of a project and engineering personnel. In this study, deformation detection for a quarry slope is implemented using the integration of subpixel offset tracking (sPOT) and unsupervised change detection algorithms based on unmanned aerial vehicle (UAV) image datasets. The sPOT algorithm is used to give the surface displacement field of the slope with subpixel accuracy, and the unsupervised change detection algorithm yields the ground object reconstruction area of the slope to verify and explain the sPOT result. The integrated analysis method in this paper is highly applicable and only requires a minimum of two UAV datasets as raw data. Combining the advantages of the sPOT and unsupervised change detection algorithms, the proposed method has the ability to detect and analyse slow and rapid slope deformation with good accuracy.

A novel framework for vegetation change characterization from time series landsat images

Understanding terrestrial ecosystem dynamics requires a comprehensive examination of vegetation changes. Remote sensing technology has been established as an effective approach to reconstructing vegetation change history, investigating change properties, and evaluating the ecological effects. However, current remote sensing techniques are primarily focused on break detection but ignore long-term trend analysis. In this study, we proposed a novel framework based on a change detection algorithm and a trend analysis method that could integrate both short-term disturbance detection and long-term trends to comprehensively assess vegetation change. With this framework, we characterized the vegetation changes in Zhejiang Province from 1990 to 2020 using Landsat and landcover data. Benefiting from combining break detection and long-term trend analysis, the framework showcased its capability of capturing a variety of dynamics and trends of vegetation. The results show that the vegetation was browning in the plains while greening in the mountains, and the overall vegetation was gradually greening during the study period. By comparison, detected vegetation disturbances covered 57.71% of the province's land areas (accounting for 66.92% of the vegetated region) which were mainly distributed around the built-up areas, and most disturbances (94%) occurred in forest and cropland. There were two peak timings in the frequency of vegetation disturbances: around 2003 and around 2014, and the proportions of more than twice disturbances in a single location were low. The results illustrate that this framework is promising for the characterization of regional vegetation growth, including long-term trends and short-term features. The proposed framework enlightens a new direction for the continuous monitoring of vegetation dynamics.

Attention-guided siamese networks for change detection in high resolution remote sensing images

Understanding surface changes requires the ability to identify changes in high resolution remote sensing images. Because current deep learning-based change detection algorithms are not able to accurately discriminate between altered and unmodified areas, which leads to the problem of edge uncertainty and small target missing in the detection process. To identify changes in high resolution remote sensing images, this research proposes an unique Attention-Guided Siamese Network (SAGNet). In this network, bitemporal images’ highly representative deep semantic features are retrieved using a fully convolutional dual-stream architecture, and the extracted deep semantic features are then used to extract semantic variation data from the Global Semantic Aggregation Module (GSAM). In the feature decoding stage, the extracted features are refined layer by layer through the Attention Fusion Module (AFM) for change map reconstruction. In addition, we propose two other auxiliary modules: Cross-scale Fusion Module (CFM) and Bilateral Feature Fusion Module (BFFM), which enable the network to remove background noise while improving the recognition accuracy of changing object boundaries and small-changing targets in the output change map. A public dataset called LEVIR-CD and a challenging dataset called BICD made up of bitemporal images from Google Earth covering various parts of China are used to experimentally test SAGNet. Finally, experimental evidence shows that our approach outperforms current cutting-edge change detection techniques.

Monitoring of 35-Year Mangrove Wetland Change Dynamics and Agents in the Sundarbans Using Temporal Consistency Checking

Mangrove wetlands are rapidly being lost due to anthropogenic disturbances and natural processes, such as sea-level rise (SLR), but are also recovering as a result of conservation efforts. Accurate and contemporary mangrove maps to detect their distribution and changes are urgently needed to understand how mangroves respond to global change and develop effective conservation projects. Here, we developed a new change detection algorithm called temporal consistency checking combining annual classification and spectral time series (TCC-CS) for tracking mangrove losses and gains. Specifically, mangrove change events were determined by measuring the deviation of greenness and wetness of candidate change segments from automatically collected mangrove reference samples. By applying to the world’s largest mangrove patches, we monitored the 35-year mangrove trajectory in the Sundarbans from 1988 to 2022 using all available Landsat images on the Google Earth Engine platform. In the Sundarbans, 18,501.89 ha of mangroves have been gained, but these have been offset by losses of 27,009.79 ha, leading to a net mangrove loss of 1.42% (8507.9 ha) in the past 35 years. We further mapped the pixel-level change agents and found that SLR-induced erosion and degradation, instead of human activities, were the major drivers of losses in the Sundarbans. Trend analysis on loss agents indicates that mangrove losses caused by human activities, such as the expansion of croplands and aquaculture ponds, have declined, but SLR is still a persistent threat to mangrove wetlands in this iconic mangrove area. Our study provides a computationally efficient methodology for examining large-scale mangrove changes, and the resultant annual mangrove maps provide strong support for mangrove conservation in the Sundarbans.

Performance Monitoring of Wind Turbines Gearbox Utilising Artificial Neural Networks — Steps toward Successful Implementation of Predictive Maintenance Strategy

Manufacturing and energy sectors provide vast amounts of maintenance data and information which can be used proactively for performance monitoring and prognostic analysis which lead to improve maintenance planning and scheduling activities. This leads to reduced unplanned shutdowns, maintenance costs and any fatal events that could affect the operations of the overall system. Performance and condition monitoring are among the most used strategies for prognostic and health management (PHM), in which different methods and techniques can be implemented to analyse maintenance and online data. Offshore wind turbines (WTs) are complex systems increasingly needing maintenance. This study proposes a performance monitoring system to monitor the performance of the WT power generation process by exploiting artificial neural networks (ANN) composed of different network designs and training algorithms, using simulated supervisory control and data acquisition (SCADA) data. The performance monitoring is based on different operating modes of the same type of wind turbine. The degradation models were developed based on the generated active power resulting from different degradation levels of the gearbox, which is a critical component of the WTs. The deviations of the wind power curves for all operating modes over time are monitored in terms of the resulting power residuals and are modelled using ANN with a unique network architecture. The monitoring process uses the recursive form of the cumulative summation (CUSUM) change detection algorithm to detect the state change point in which the gearbox efficiency is degraded by evaluating the power residuals predicted by the ANN model. To increase the monitoring effectiveness, a second ANN model was developed to predict the gearbox efficiency to monitor any failure that could happen once the efficiency degrades below a threshold. The results show a high degree of accuracy in power and efficiency prediction in addition to monitoring the abnormal state or deviations of the power generation process resulting from the degraded gearbox efficiency and their corresponding time slots. The developed monitoring method can be a valuable tool to provide maintenance experts with alarms and insights into the general state of the power generation process, which can be used for further maintenance decision-making.

Satellite Image Compression Guided by Regions of Interest

Small satellites empower different applications for an affordable price. By dealing with a limited capacity for using instruments with high power consumption or high data-rate requirements, small satellite missions usually focus on specific monitoring and observation tasks. Considering that multispectral and hyperspectral sensors generate a significant amount of data subjected to communication channel impairments, bandwidth constraint is an important challenge in data transmission. That issue is addressed mainly by source and channel coding techniques aiming at an effective transmission. This paper targets a significant further bandwidth reduction by proposing an on-the-fly analysis on the satellite to decide which information is effectively useful before coding and transmitting. The images are tiled and classified using a set of detection algorithms after defining the least relevant content for general remote sensing applications. The methodology makes use of the red-band, green-band, blue-band, and near-infrared-band measurements to perform the classification of the content by managing a cloud detection algorithm, a change detection algorithm, and a vessel detection algorithm. Experiments for a set of typical scenarios of summer and winter days in Stockholm, Sweden, were conducted, and the results show that non-important content can be identified and discarded without compromising the predefined useful information for water and dry-land regions. For the evaluated images, only 22.3% of the information would need to be transmitted to the ground station to ensure the acquisition of all the important content, which illustrates the merits of the proposed method. Furthermore, the embedded platform's constraints regarding processing time were analyzed by running the detection algorithms on Unibap's iX10-100 space cloud platform.

A Hybrid Algorithm for Urban LULC Change Detection for Building Smart-city by Using WorldView Images

Technological advancement in smart cities can have adverse effects on the environment. Timely monitoring of smart cities to preserve environmental sustainability is a thrust area of research. It can be done by using change detection with multi-temporal satellite data. The success of these methods solely depends on the calibre of the backend image segmentation and Land-use Land-cover classification technique. The limitation of using cutting-edge classification algorithms is the availability of a proper dataset and identification of the edge structure of different LULC classes. In contrast, a segmentation algorithm cannot detect LULC classes automatically. In this research, we eliminated these shortcomings by considering a hybrid approach. We proposed a multi-class Support Vector Machine (SVM) and ISODATA-embedded large-scale change detection method. This method can automatically segment, detect, and perform LULC change analysis. We have considered the multi-sensor dataset of Barasat, West Bengal, India, obtained from the WorldView satellite sensor for the experimental study. The proposed method is validated concerning three different cutting-edge methods.

SAR Image Change Detection in Spatial-Frequency Domain Based on Attention Mechanism and Gated Linear Unit

Change detection based on synthetic aperture radar (SAR) images is an important application in the remote sensing technology field. However, the lack of labeled data has been a difficult problem in SAR image detection, especially for pixel-level change detection. In this letter, we propose a novel unsupervised change detection algorithm, which improves the detection accuracy by exploring features from both spatial and frequency domains of SAR images. In particular, firstly clustering is used as pre-classification to obtain pseudo-labels, and then by incorporating classifiers and pseudo-labels in terms of feature learning, a novel unsupervised detection algorithm is proposed. In order to improve the sensitivity of the algorithm to changed details and enhance the anti-noise ability of the change detection network, the attention mechanism (AM) is integrated into the network to fully extract important spatial structure information. Moreover, a multi-domain fusion module is proposed to integrate spatial and frequency domain features into complementary feature representations. This module contains multi-region features weighted by the channel-spatial AM and deep features filtered out by the gated linear units (GLU) in the frequency domain. To verify the effectiveness of the proposed algorithm, it is compared against the other four SAR image change detection algorithms using three real datasets. The experimental results show that the proposed method outperforms the other four algorithms in terms of percent correct classification (PCC) and Kappa coefficient (KC).

Обнаружение дрейфа концепта при классификации мобильных приложений с использованием автокодировщиков

The study observes the task of concept drift detection in multiclass applications classification tasks on the example of collected data set of network traffic in the form of IP packets from Purpose of the study: development and software implementation of an algorithm for a concept change detection in tasks of multiclass mobile application traffic classification using ANNs of the autoencoder type (AC). Novelty of the study consists in drift detection of one or several mobile applications based on changes in the statistical characteristics of one or several attributes without usage of true class labels implying ANNs of the autoencoder type. Results: The study developed an algorithm for concepts drift of application detection based on the analysis of changes in the statistical characteristics of attributes or a noticeable decrease in the quality of the analyzed applications classification. As for fundamental model of concept drift detector of analyzed applications, the study used autoencoders. The research contains basic theoretical positions of the algorithm creation. The study shows that in case of trained AC only on high-quality prototypes, it will be able to reconstruct normal observations but not abnormal observations (unknown concepts). As a result, when the autoencoder detects a significant reconstruction error, it classifies the observation data as abnormal. Estimation of reconstruction errors of the analyzed applications and excess of threshold value assess the presence of drift. The Python software environment provides the implementation of the presented solution.

Axial Cross Attention Meets CNN: Bibranch Fusion Network for Change Detection

In the previous years, vision transformer has demonstrated a global information extraction capability in the field of computer vision that convolutional neural network (CNN) lacks. Due to the lack of inductive bias in vision transformer, it requires a large amount of data to support its training. In the field of remote sensing, it costs a lot to obtain a significant number of high-resolution remote sensing images. Most existing change detection networks based on deep learning rely heavily on the CNN, which cannot effectively utilize the long-distance dependence between pixels for difference discrimination. Therefore, this work aims to use a high-performance vision transformer to conduct change detection research with limited data. A bibranch fusion network based on axial cross attention (ACABFNet) is proposed. The network extracts local and global information of images through the CNN branch and transformer branch, respectively, and then, fuses local and global features by the bidirectional fusion approach. In the upsampling stage, similar feature information and difference feature information of the two branches are explicitly generated by feature addition and feature subtraction. Considering that the self-attention mechanism is not efficient enough for global attention over small datasets, we propose the axial cross attention. First, global attention along the height and width dimensions of images is performed respectively, and then cross attention is used to fuse the global feature information along two dimensions. Compared with the original self-attention, the structure is more graphics processing unit friendly and efficient. Experimental results on three datasets reveal that the ACABFNet outperforms existing change detection algorithms.

An Extremely Lightweight Change Detection Algorithm Based on Light Global-Local Feature Enhancement Module

An Extremely Lightweight Change Detection Algorithm Based on Light Global-Local Feature Enhancement Module

Advanced Evaluation Method for Video Retrieval System

n the past few years, the multimedia storage has increased and costs of storing multi-media data has become cheaper. Which is why, there are large numbersof videos that are available in the video repositories. With development of the multi-media data types and available bandwidths.The proposed method transferseach video of database into scenes using color histogrambased scene change detection algorithm and key framesare extracted. For key frames multiple features areobtained using straight forward rules. A new framework depending on CNNs (convolutional neural networks) is recommended to perform the content based video retrieval with the less storage cost also with higher search capability. The recommended framework subsists of extraction algorithm with respect to key-frame and the feature collection strategies. Particularly, the extraction algorithm of key-frame takes benefit of clustering idea; so in that case excessive information is taken out from video data and also the storage cost is highly shortened. This work present a method uses the extracted features with convolutional neural network (CNN) for classification tasks. In this research paper, different types of videos will be used and the important features will be extracted using SIFT, after that we use the deep learning(CNN) process will be performed. The experimental results showed the efficiency and effectiveness of suggested approach

Maximum Power Extraction for PMVG-Based WECS Using Q-Learning MPPT Algorithm With Finite-Time Control Scheme

This study is concerned with the problem of maximum power extraction for a permanent magnet vernier generator (PMVG)-based wind energy conversion system (WECS) using the Q-learning maximum power point tracking (MPPT) algorithm with the finite-time control (FTC) scheme. To do this, the model-free sensorless reinforcement Q-learning algorithm-based MPPT method is firstly proposed. At this time, the reward and learning rate are updated by the Q-values for each state-action pair. Next, the proposed learning algorithm is used to construct an optimal speed–power curve for achieving the fast and steady MPPT operation for the WECS using the learned action values. Besides, the wind-speed change detection algorithm is added to the proposed method so that the PMVG-based WECS can work in various wind speed conditions. And then, the FTC method is proposed to track the reference speed of PMVG which support to achieving the maximum power extraction, and Lyapunov stability theory is derived to ensure the overall system's stability. Finally, the simulation and experimental results from the 5 kW PMVG-based WECS are presented to demonstrate the applicability and superiority of the proposed control method.