Aneurysm growth evaluation and detection: a computer-assisted follow-up MRA analysis.

Growing intracranial aneurysms pose a high risk of rupture, making the detection and quantification of the growth crucial for timely treatment strategy adoption. In this paper we propose a computer-assisted approach based on the extraction of IA shapes from associated baseline and follow-up angiographic scans and non-rigid morphing of the two shapes. From the obtained shape deformations we computed four novel features, including differential volume (dV), surface area (dSA), aneurysm-size normalized median deformation path length (dMPL), and integral of cumulative deformation distances (dICDD). An experienced neuroradiologist manually extracted the IA shape models from the baseline and follow-up MRAs and, by utilizing size change and visual assessments, classified each aneurysm into stable with morphology changes, stable or growing. We investigated the classification performance and found that three of the novel and one cross-sectional feature exhibited significantly different mean values (p-value < 0.05 ; Tukey's HSD test) between the stable and growing IA groups, while the mean dICDD was significantly different between all the three groups. The cross-sectional features has sensitivity to growing IAs in range 0.05-0.86, while novel features had generally higher sensitivity in range 0.81-0.90, making them promising candidates as surrogate follow-up imaging-based biomarkers for IA growth detection. These findings may offer valuable information for clinical management of patients with IAs based on follow-up imaging.

Evolution of waterborne diseases: A case study of Khyber Pakhtunkhwa, Pakistan.

Objectives: In Pakistan, the degradation of drinking water quality is exacerbated by the increasing population size and rapid industrialization. Contaminated water serves as the predominant source of numerous diseases, including diarrhea, gastroenteritis, and typhoid. This article explores the evolution of waterborne diseases across 21 districts of the Khyber Pakhtunkhwa province in Pakistan by monitoring changes in the clustering solutions. Methods: The data employed in this study were sourced from 21 districts of KP by the Director-General Health Services. Cluster analysis was utilized to uncover patterns in waterborne disease incidence, while principal component analysis was employed to reveal underlying patterns and reduce dimensionality. Additionally, the MONItoring Clusters (MONIC) framework was applied for change detection, facilitating the identification of significant shifts in disease patterns over time and aiding in the understanding of temporal dynamics. Results: Our analysis indicates that two clusters survived consistently over time, while other clusters exhibited inconsistency. Profiling of the surviving clusters (C12 → C24 → C32 → C43) suggests a gradual increase in cases of bloody diarrhea in the Swat Valley, Hangu, Karak, and Lakki Marwat regions. Similarly, profiling of the surviving clusters (⊙→ C22 → C34 → C44) suggests an increase in the acute watery diarrhea (non-cholera) and typhoid fever in the regions of Peshawar, Nowshera, and Swabi. Conclusion: The findings of this study hold significant importance as they pinpoint the most vulnerable regions for various waterborne diseases. These insights offer valuable guidance to policymakers and health officials, empowering them to implement effective measures for controlling waterborne diseases in the respective regions of Khyber Pakhtunkhwa, Pakistan.

Mixing novel and familiar cues modifies representations of familiar visual images and affects behavior.

While visual responses to familiar and novel stimuli have been extensively studied, it is unknown how neuronal representations of familiar stimuli are affected when they are interleaved with novel images. We examined a large-scale dataset from mice performing a visual go/no-go change detection task. After training with eight images, six novel images were interleaved with two familiar ones. Unexpectedly, we found that the behavioral performance in response to familiar images was impaired when they were mixed with novel images. When familiar images were interleaved with novel ones, the dimensionality of their representation increased, indicating a perturbation of their neuronal responses. Furthermore, responses to familiar images in the primary visual cortex were less predictive of responses in higher-order areas, indicating less efficient communication. Spontaneous correlations between neurons were predictive of responses to novel images, but less so to familiar ones. Our study demonstrates the modification of representations of familiar images by novelty.

AC-ASPECTS, ACh-ASPECTS, and H-ASPECTS: new imaging scales to assess territorial and total cerebral hemispheric ischemic injury.

Background: The extent of ischemic injury in acute stroke is assessed in clinical practice using the Acute Stroke Prognosis Early CT Score (ASPECTS) rating system. However, current ASPECTS semi-quantitative topographic scales assess only the middle cerebral artery (MCA) (original ASPECTS) and posterior cerebral (PC-ASPECTS) territories. For treatment decision-making in patients with anterior cerebral artery (ACA) occlusions and internal carotid artery (ICA) occlusions with large ischemic cores, measures of all hemispheric regions are desirable. Methods: In this cohort study, anatomic rating systems were developed for the anterior cerebral (AC-ASPECTS, 3 points) and anterior choroidal artery (ACh-ASPECTS, 1 point) territories. In addition, a total supratentorial hemisphere (H-ASPECTS, 16 points) score was calculated as the sum of the MCA ASPECTS (10 regions), supratentorial PC-ASPECTS (2 regions), AC-ASPECTS (3 regions), and ACh-ASPECTS (1 region). Three raters applied these scales to initial and 24 h CT and MR images in consecutive patients with ischemic stroke (IS) due to ICA, M1-MCA, and ACA occlusions. Results: Imaging ratings were obtained for 96 scans in 50 consecutive patients with age 74.8 (±14.0), 60% female, NIHSS 15.5 (9.25-20), and occlusion locations ICA 34%; M1-MCA 58%; and ACA 8%. Treatments included endovascular thrombectomy +/- thrombolysis in 72%, thrombolysis alone in 8%, and hemicraniectomy in 4%. Among experienced clinicians, inter-rater reliability for AC-, ACh-, and H-ASPECTS scores was substantial (kappa values 0.61-0.80). AC-ASPECTS abnormality was present in 14% of patients, and ACh-ASPECTS abnormality in 2%. Among patients with ACA and ICA occlusions, H-ASPECTS scores compared with original ASPECTS scores were more strongly associated with disability level at discharge, ambulatory status at discharge, discharge destination, and combined inpatient mortality and hospice discharge. Conclusion: AC-ASPECTS, ACh-ASPECTS, and H-ASPECTS expand the scope of acute IS imaging scores and increase correlation with functional outcomes. This additional information may enhance prognostication and decision-making, including endovascular thrombectomy and hemicraniectomy.

Combining Google Earth historical imagery and UAV photogrammetry for urban development analysis.

Rural-urban migration often triggers additional demand for housing and infrastructural development to cater for the growing population in urban areas. Consequently, town planners and urban development authorities need to understand the urban development trend to make sustainable urban planning decisions. Yet, methods to analyse changes and trends in urban spatial development are often complex and require costly data collection. This article thus presents a simplified method to analyse the urban development trend in an area. The method integrates Google Earth (GE) historical imagery (baseline data) and unmanned aerial vehicle (UAV) photogrammetry (recent data) to quantify the changes over time. This approach can be applied to study the urban development trends in low-income countries with budget constraints. The method is discussed under four main headings: (1) background, (2) method details, (3) limitations, and (4) conclusion.•Google Earth historical image can be extracted with its associated world file.•The population of an area can be estimated by using average household size data and the number of residential buildings in the area.•The building height ratio can be used to ascertain if the land is being used parsimoniously.

Vegetation change detection and recovery assessment based on post-fire satellite imagery using deep learning.

Wildfires are uncontrolled fires fuelled by dry conditions, high winds, and flammable materials that profoundly impact vegetation, leading to significant consequences including noteworthy changes to ecosystems. In this study, we provide a novel methodology to understand and evaluate post-fire effects on vegetation. In regions affected by wildfires, earth-observation data from various satellite sources can be vital in monitoring vegetation and assessing its impact. These effects can be understood by detecting vegetation change over the years using a novel unsupervised method termed Deep Embedded Clustering (DEC), which enables us to classify regions based on whether there has been a change in vegetation after the fire. Our model achieves an impressive accuracy of 96.17%. Appropriate vegetation indices can be used to evaluate the evolution of vegetation patterns over the years; for this study, we utilized Enhanced Vegetation Index (EVI) based trend analysis showing the greening fraction, which ranges from 0.1 to 22.4 km2 while the browning fraction ranges from 0.1 to 18.1 km2 over the years. Vegetation recovery maps can be created to assess re-vegetation in regions affected by the fire, which is performed via a deep learning-based unsupervised method, Adaptive Generative Adversarial Neural Network Model (AdaptiGAN) on post-fire data collected from various regions affected by wildfire with a training error of 0.075 proving its capability. Based on the results obtained from the study, our approach tends to have notable merits when compared to pre-existing works.

Transformers for Remote Sensing: A Systematic Review and Analysis.

Research on transformers in remote sensing (RS), which started to increase after 2021, is facing the problem of a relative lack of review. To understand the trends of transformers in RS, we undertook a quantitative analysis of the major research on transformers over the past two years by dividing the application of transformers into eight domains: land use/land cover (LULC) classification, segmentation, fusion, change detection, object detection, object recognition, registration, and others. Quantitative results show that transformers achieve a higher accuracy in LULC classification and fusion, with more stable performance in segmentation and object detection. Combining the analysis results on LULC classification and segmentation, we have found that transformers need more parameters than convolutional neural networks (CNNs). Additionally, further research is also needed regarding inference speed to improve transformers' performance. It was determined that the most common application scenes for transformers in our database are urban, farmland, and water bodies. We also found that transformers are employed in the natural sciences such as agriculture and environmental protection rather than the humanities or economics. Finally, this work summarizes the analysis results of transformers in remote sensing obtained during the research process and provides a perspective on future directions of development.

Precancerous Change Detection Technique on Mammography Breast Cancer Images based on Mean Ratio and Log Ratio using Fuzzy c Mean Classification with Gabor Filter

BACKGROUND: The growing rate of breast cancer necessitates immediate global attention. Mammography images are used to determine the stage of malignancy. Breast cancer stages must be identified in order to save a person's life. OBJECTIVE: This article's main goal is to identify different techniques to obtain the difference between two breast cancer mammography images taken of the same individual at different times. This is the first effort to identify breast cancer in mammography images using change detection techniques. The Mammogram Image Change Detection (ICD) technique is also a recent advancement to prevent breast cancer in the early stage and precancerous level in medical images. METHODS: The main purpose of this work is to observe the changes between breast cancer images in different screening periods using different techniques. Mammogram Breast Cancer Image Change Detection (MBCICD) methods usually start with a Difference Image (DI) and classify the pixels in the DI into changed and unaffected classes using unsupervised fuzzy c means (FCM) clustering methods based on texture features taken from the log and mean ratio difference pictures. Two operators, mean ratio and log ratio, were used to check the changes in the images. The Gabor wavelet is utilized as a feature extraction technique among several standards. Using the Gabor wavelet ratio operators is a useful method for altering the detection of breast cancer in mammography images. Currently, it is challenging to obtain real malignant images of the same person for testing or training. In this study, two images are utilized. To clearly see the changes, one is an image from the MIAS breast cancer mammography images dataset, and the other is a self-generated change image. RESULTS: The research aims to examine the image results and other quantitative analysis results of proposed change detection methods on cancer images. The Mean Ratio Accuracy result is 0.9738, and the Log ratio PCC is 0.9737. The classification results are the Log Ratio + Gabor Filter + FCM is 0.9737, and Mean Ratio +Gabor Filter + FCM is 0.9719. The mean Ratio Accuracy result is 0.9738, Log ratio is 0.9737. Log Ratio + Gabor Filter + FCM is 0.9737, Mean Ratio +Gabor Filter + FCM is 0.9719. Comparing the PCC of proposed change detection methods with the FDA-RMG method on the same dataset, the accuracy is 0.9481 only. CONCLUSION: The study concludes that variations in mammography breast cancer images could be successfully identified using the ratio operators with Gabor wavelet features.

Auditory change detection and visual selective attention: association between MMN and N2pc.

While the auditory and visual systems each provide distinct information to our brain, they also work together to process and prioritize input to address ever-changing conditions. Previous studies highlighted the trade-off between auditory change detection and visual selective attention; however, the relationship between them is still unclear. Here, we recorded electroencephalography signals from 106 healthy adults in three experiments. Our findings revealed a positive correlation at the population level between the amplitudes of event-related potential indices associated with auditory change detection (mismatch negativity) and visual selective attention (posterior contralateral N2) when elicited in separate tasks. This correlation persisted even when participants performed a visual task while disregarding simultaneous auditory stimuli. Interestingly, as visual attention demand increased, participants whose posterior contralateral N2 amplitude increased the most exhibited the largest reduction in mismatch negativity, suggesting a within-subject trade-off between the two processes. Taken together, our results suggest an intimate relationship and potential shared mechanism between auditory change detection and visual selective attention. We liken this to a total capacity limit that varies between individuals, which could drive correlated individual differences in auditory change detection and visual selective attention, and also within-subject competition between the two, with task-based modulation of visual attention causing within-participant decrease in auditory change detection sensitivity.

DASUNet: a deeply supervised change detection network integrating full-scale features.

The change detection (CD) technology has greatly improved the ability to interpret land surface changes. Deep learning (DL) methods have been widely used in the field of CD due to its high detection accuracy and application range. DL-based CD methods usually cannot fuse the extracted feature information at full scale, leaving out effective information, and commonly use transfer learning methods, which rely on the original dataset and training weights. To address the above issues, we propose a deeply supervised (DS) change detection network (DASUNet) that fuses full-scale features, which adopts a Siamese architecture, fuses full-scale feature information, and realizes end-to-end training. In order to obtain higher feature information, the network uses atrous spatial pyramid pooling (ASPP) module in the coding stage. In addition, the DS module is used in the decoding stage to exploit feature information at each scale in the final prediction. The experimental comparison shows that the proposed network has the current state-of-the-art performance on the CDD and the WHU-CD, reaching 94.32% and 90.37% on F1, respectively.

Employing post classification comparison to detect land use cover change patterns and quantify conversions in Abakaliki LGA Nigeria from 2000 to 2022.

Rapid urbanization is restructuring landscapes across sub-Saharan Africa. This study employed post-classification comparison of multi-temporal Landsat imagery to characterize land cover changes in Abakaliki Local Government Area, Ebonyi State, Nigeria between 2000 and 2022, addressing the need for empirical baselines to guide sustainable planning. Four classes were considered and images classified with overall accuracy of 95% for the year 2000 and 97% for the year 2022. Notably, 21,000 hectares of vegetation were lost, while built-up and bare land increased by 7500 and 13,700 hectares respectively. Spatial patterns revealed built-up encroachment from vegetation and bare land; this establishes the first standardized quantification of Abakaliki LGA's shifting landscape, with results supporting compact development models while conserving ecological services under ongoing transformations. The study makes a significant contribution by establishing an empirical baseline characterizing Nigeria's urbanization trajectory essential for evidence-based stewardship of regional resources and livelihoods in a period of accelerating change.

The Motion-Silencing Illusion Depends on Object-Centered Representation.

Motion silencing is a striking and unexplained visual illusion wherein changes that are otherwise salient become difficult to perceive when the changing elements also move. We develop a new method for quantifying illusion strength (Experiments 1a and 1b), and we demonstrate a privileged role for rotational motion on illusion strength compared with highly controlled stimuli that lack rotation (Experiments 2a to 3b). These contrasts make it difficult to explain the illusion in terms of lower-level detection limits. Instead, we explain the illusion as a failure to attribute changes to locations. Rotation exacerbates the illusion because its perception relies upon structured object representations. This aggravates the difficulty of attributing changes by demanding that locations are referenced relative to both an object-internal frame and an external frame. Two final experiments (4a and 4b) add support to this account by employing a synchronously rotating external frame of reference that diminishes otherwise strong motion silencing. All participants were Johns Hopkins University undergraduates.

LIRRN: Location-Independent Relative Radiometric Normalization of Bitemporal Remote-Sensing Images.

Relative radiometric normalization (RRN) is a critical pre-processing step that enables accurate comparisons of multitemporal remote-sensing (RS) images through unsupervised change detection. Although existing RRN methods generally have promising results in most cases, their effectiveness depends on specific conditions, especially in scenarios with land cover/land use (LULC) in image pairs in different locations. These methods often overlook these complexities, potentially introducing biases to RRN results, mainly because of the use of spatially aligned pseudo-invariant features (PIFs) for modeling. To address this, we introduce a location-independent RRN (LIRRN) method in this study that can automatically identify non-spatially matched PIFs based on brightness characteristics. Additionally, as a fast and coregistration-free model, LIRRN complements keypoint-based RRN for more accurate results in applications where coregistration is crucial. The LIRRN process starts with segmenting reference and subject images into dark, gray, and bright zones using the multi-Otsu threshold technique. PIFs are then efficiently extracted from each zone using nearest-distance-based image content matching without any spatial constraints. These PIFs construct a linear model during subject-image calibration on a band-by-band basis. The performance evaluation involved tests on five registered/unregistered bitemporal satellite images, comparing results from three conventional methods: histogram matching (HM), blockwise KAZE, and keypoint-based RRN algorithms. Experimental results consistently demonstrated LIRRN's superior performance, particularly in handling unregistered datasets. LIRRN also exhibited faster execution times than blockwise KAZE and keypoint-based approaches while yielding results comparable to those of HM in estimating normalization coefficients. Combining LIRRN and keypoint-based RRN models resulted in even more accurate and reliable results, albeit with a slight lengthening of the computational time. To investigate and further develop LIRRN, its code, and some sample datasets are available at link in Data Availability Statement.

Actively Driven Light-Addressable Sensor/Actuator System for Automated pH Control for the Integration in Lab-On-A-Chip (LoC) Platforms.

The miniaturization of microfluidic systems usually comes at the cost of more difficult integration of sensors and actuators inside the channel. As an alternative, this work demonstrates the embedding of semiconductor-based sensor and actuator technologies that can be spatially and temporally controlled from outside the channel using light. The first element is a light-addressable potentiometric sensor, consisting of an Al/Si/SiO2/Si3N4 structure, that can measure pH changes at the Si3N4/electrolyte interface. The pH value is a crucial factor in biological and chemical systems, and besides measuring, it is often important to bring the system out of equilibrium or to adjust and control precisely the surrounding medium. This can be done photoelectrocatalytically by utilizing light-addressable electrodes. These consist of a glass/SnO2:F/TiO2 structure, whereby direct charge transfer between the TiO2 and the electrolyte leads to a pH change upon irradiation. To complement the advantages of both, we integrated a light-addressable sensor with a pH sensitivity of 41.5 mV·pH-1 and a light-addressable electrode into a microfluidic setup. Here, we demonstrated a simultaneous operation with the ability to generate and record pH gradients inside a channel under static and dynamic flow conditions. The results show that dependent on the light-addressable electrode (LAE)-illumination conditions, pH changes up to ΔpH of 2.75 and of 3.52 under static and dynamic conditions, respectively, were spatially monitored by the light-addressable potentiometric sensor. After flushing with fresh buffer solution, the pH returned to its initial value. Depending on the LAE illumination, pH gradients with a maximum pH change of ΔpH of 1.42 were tailored perpendicular to the flow direction. In a final experiment, synchronous LAE illumination led to a stepwise increase in the pH inside the channel.

A Deep Learning Based Platform for Remote Sensing Images Change Detection Integrating Crowdsourcing and Active Learning.

Remote sensing images change detection technology has become a popular tool for monitoring the change type, area, and distribution of land cover, including cultivated land, forest land, photovoltaic, roads, and buildings. However, traditional methods which rely on pre-annotation and on-site verification are time-consuming and challenging to meet timeliness requirements. With the emergence of artificial intelligence, this paper proposes an automatic change detection model and a crowdsourcing collaborative framework. The framework uses human-in-the-loop technology and an active learning approach to transform the manual interpretation method into a human-machine collaborative intelligent interpretation method. This low-cost and high-efficiency framework aims to solve the problem of weak model generalization caused by the lack of annotated data in change detection. The proposed framework can effectively incorporate expert domain knowledge and reduce the cost of data annotation while improving model performance. To ensure data quality, a crowdsourcing quality control model is constructed to evaluate the annotation qualification of the annotators and check their annotation results. Furthermore, a prototype of automatic detection and crowdsourcing collaborative annotation management platform is developed, which integrates annotation, crowdsourcing quality control, and change detection applications. The proposed framework and platform can help natural resource departments monitor land cover changes efficiently and effectively.

Assessing the impact of anthropogenic activities on land use and land cover changes in the semi-arid and arid regions of Algeria.

Land use and land cover are critical factors that influence the environment and human societies. The dynamics of LULC have been constantly changing over the years, and these changes can be analyzed at different spatial and temporal scales to evaluate their impact on the natural environment. This study employs multitemporal satellite data to investigate the spatial and temporal transformations that occurred in Sidi Bel Abbes province, situated in the northwestern region of Algeria, spanning from the early 1990s to 2020. Notably, this province is marked by semi-arid and arid climates and hosts a wide range of areas susceptible to gravitational hazards, especially concerning alterations in land use and forest fires. The interactive supervised classification tool utilized multiple machine learning algorithms including Random Forest, Support Vector Machine, Classification and Regression Tree, and Naïve Bayes to produce land cover maps with six main classes: forest, shrub, agricultural, pasture, water, and built-up. The findings showed that the LULC in the research area is undergoing continuous change, particularly in the forest and agricultural lands. The forest area has decreased significantly from 10.80% in 1990 to 5.25% in 2020, mainly due to repeated fires. Agricultural land has also undergone fluctuations, with a decrease between 1990 and 2000, followed by a fast increase and near stabilization in 2020. At the same time, pasture lands and built-up areas grew steadily, increasing by 11% and 13% respectively. This research highlights the significant impact of anthropogenic activities on LULC changes in the study area and can provide valuable insights for promoting sustainable land use policies.

Shortening fire return interval predisposes west-central Canadian boreal peatlands to more rapid vegetation growth and transition to forest cover.

Climate change in northern latitudes is increasing the vulnerability of peatlands and the riparian transition zones between peatlands and upland forests (referred to as ecotones) to greater frequency of wildland fires. We examined early post-fire vegetation regeneration following the 2011 Utikuma complex fire (central Alberta, Canada). This study examined 779 peatlands and adjacent ecotones, covering an area of ~182 km2 . Based on the known regional fire history, peatlands that burned in 2011 were stratified into either long return interval (LRI) fire regimes of >80 years (i.e., no recorded prior fire history) or short fire return interval (SRI) of 55 years (i.e., within the boundary of a documented severe fire in 1956). Data from six multitemporal airborne lidar surveys were used to quantify trajectories of vegetation change for 8 years prior to and 8 years following the 2011 fire. To date, no studies have quantified the impacts of post-fire regeneration following short versus long return interval fires across this broad range of peatlands with variable environmental and post-fire successional trajectories. We found that SRI peatlands demonstrated more rapid vascular and shrub growth rates, especially in peatland centers, than LRI peatlands. Bogs and fens burned in 1956, and with little vascular vegetation (classified as "open peatlands") prior to the 2011 fire, experienced the greatest changes. These peatlands tended to transition to vascular/shrub forms following the SRI fire, while open LRI peatlands were not significantly different from pre-fire conditions. The results of this study suggest the emergence of a positive feedback, where areas experiencing SRI fires in southern boreal peatlands are expected to transition to forested vegetation forms. Along fen edges and within bog centers, SRI fires are expected to reduce local peatland groundwater moisture-holding capacity and promote favorable conditions for increased fire frequency and severity in the future.

A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis.

Introduction: Multiple sclerosis (MS) is a chronic neurological disorder characterized by the progressive loss of myelin and axonal structures in the central nervous system. Accurate detection and monitoring of MS-related changes in brain structures are crucial for disease management and treatment evaluation. We propose a deep learning algorithm for creating Voxel-Guided Morphometry (VGM) maps from longitudinal MRI brain volumes for analyzing MS disease activity. Our approach focuses on developing a generalizable model that can effectively be applied to unseen datasets. Methods: Longitudinal MS patient high-resolution 3D T1-weighted follow-up imaging from three different MRI systems were analyzed. We employed a 3D residual U-Net architecture with attention mechanisms. The U-Net serves as the backbone, enabling spatial feature extraction from MRI volumes. Attention mechanisms are integrated to enhance the model's ability to capture relevant information and highlight salient regions. Furthermore, we incorporate image normalization by histogram matching and resampling techniques to improve the networks' ability to generalize to unseen datasets from different MRI systems across imaging centers. This ensures robust performance across diverse data sources. Results: Numerous experiments were conducted using a dataset of 71 longitudinal MRI brain volumes of MS patients. Our approach demonstrated a significant improvement of 4.3% in mean absolute error (MAE) against the state-of-the-art (SOTA) method. Furthermore, the algorithm's generalizability was evaluated on two unseen datasets (n = 116) with an average improvement of 4.2% in MAE over the SOTA approach. Discussion: Results confirm that the proposed approach is fast and robust and has the potential for broader clinical applicability.

Siamese Transformer-Based Building Change Detection in Remote Sensing Images.

To address the challenges of handling imprecise building boundary information and reducing false-positive outcomes during the process of detecting building changes in remote sensing images, this paper proposes a Siamese transformer architecture based on a difference module. This method introduces a layered transformer to provide global context modeling capability and multiscale features to better process building boundary information, and a difference module is used to better obtain the difference features of a building before and after a change. The difference features before and after the change are then fused, and the fused difference features are used to generate a change map, which reduces the false-positive problem to a certain extent. Experiments were conducted on two publicly available building change detection datasets, LEVIR-CD and WHU-CD. The F1 scores for LEVIR-CD and WHU-CD reached 89.58% and 84.51%, respectively. The experimental results demonstrate that when utilized for building change detection in remote sensing images, the proposed method exhibits improved robustness and detection performance. Additionally, this method serves as a valuable technical reference for the identification of building damage in remote sensing images.

A Siamese Swin-Unet for image change detection.

The problem of change detection in remote sensing image processing is both difficult and important. It is extensively used in a variety of sectors, including land resource planning, monitoring and forecasting of agricultural plant health, and monitoring and assessment of natural disasters. Remote sensing images provide a large amount of long-term and fully covered data for earth environmental monitoring. A lot of progress has been made thanks to deep learning's quick development. But the majority of deep learning-based change detection techniques currently in use rely on the well-known Convolutional neural network (CNN). However, considering the locality of convolutional operation, CNN unable to master the interplay between global and distant semantic information. Some researches has employ Vision Transformer as a backbone in remote sensing field. Inspired by these researches, in this paper, we propose a network named Siam-Swin-Unet, which is a Siamesed pure Transformer with U-shape construction for remote sensing image change detection. Swin Transformer is a hierarchical vision transformer with shifted windows that can extract global feature. To learn local and global semantic feature information, the dual-time image are fed into Siam-Swin-Unet which is composed of Swin Transformer, Unet Siamesenet and two feature fusion module. Considered the Unet and Siamesenet are effective for change detection, We applied it to the model. The feature fusion module is designed for fusion of dual-time image features, and is efficient and low-compute confirmed by our experiments. Our network achieved 94.67 F1 on the CDD dataset (season varying).