Multi-Source Fusion Enhanced Feature Segmentation in Remote Sensing Imagery

Convolutional Neural Network (CNN) is widely used for semantic segmentation and land-use and land-cover (LULC) mapping of very high-resolution (VHR) remote sensing images. The convolution operation is a powerful method for VHR classification, but the loss of high-frequency detail information caused during its operation decreases the classification accuracy, particularly in the boundary. Thus, it is necessary to supply additional boundary information to the CNN for alleviating this situation. In the classification task (and in LULC mapping), providing more effective information generates a better classification result. Current methods regard the boundary of images as the same category object and process it uniformly, which loses a notable amount of useful information because of the different properties, such as ambiguity and transition, between remote sensing images and their boundaries. Thus, a semantic segmentation method with category boundary for LULC mapping is proposed in this paper. First, a multi-task CNN called the category boundary detection network (CBDN) is designed to extract the boundary information of different category objects. Second, this category boundary and VHR images are used for initial semantic segmentation. Finally, the category boundary and the initial semantic segmentation result (ISSR) are fused to obtain the final LULC map by a two-step strategy, including the explicit fusion and the boundary attention loss function. To verify whether category boundary improved the classification accuracy, a set of comparative experiments were conducted on the International Society for Photogrammetry and Remote Sensing (ISPRS) Vaihingen and Potsdam datasets. The method in this paper was compared with a semantic segmentation method with no boundary information and a semantic segmentation method with global boundary. The results showed that the proposed method in this paper achieved good performances in the Vaihingen (overall accuracy (OA) = 0.924, Kappa coefficient (K) = 0.898, mean F1 score (mF1) = 0.896 and mean Intersection over Union (mIoU) = 0.817) and Potsdam datasets (OA = 0.890, K = 0.857, mF1 = 0.923, and mIoU = 0.860) based on the eroded labels.

Land use refers to the development of an area due to human activities, such as agriculture, industry, and residential. Land use and land cover (LULC) dynamics significantly impact agricultural productivity and food security, necessitating a comprehensive understanding and prediction of these changes. This study addresses the importance of accurate land use predictions in the context of rapid urbanisation, population growth, and climate change. This research aims to bridge the gap between land management and food security by utilising remote sensing, machine learning, and GIS technologies to predict dynamic LULC patterns. The transition potential modelling module (MOLUSCE) was used as a plugin in QGIS software to create land use and land cover (LULC) in Shah Alam City using Landsat 8 satellite images and dual sensors from 2014-2023. The ANN model was chosen to predict LULC in 2032. LULC are classified as water, developed, bare land, forests and vegetation. As a result, the developed area was the largest in 2014 and 2023, occupying 51.50% and 63.54% of the total area, respectively. Water was the smallest land use in both years. The study integrated spatial variables such as Digital Elevation Model (DEM) data and road network maps to enhance land use predictions. The findings show a fair agreement (kappa value of 0.619) between predicted and observed land-use changes, highlighting the potential for evidence-based decision-making in sustainable urban development and food security. This research contributes to the field by providing insights into future land use patterns, supporting informed policy decisions, and promoting sustainable agricultural practices for global food security efforts.

In Earth observation, cloud severely affects the interpretation of optical satellites generated high-resolution images. Cloud-free optical images are vital for downstream tasks such as semantic segmentation and object detection. Thus, the elimination of clouds from optical imagery has emerged as a significant topic in remote sensing. Currently, most existing methods are proposed to leverage the texture information from auxiliary synthetic aperture radar (SAR) images to restore cloud-free images via direct channel merging. However, such a unified feature extraction approach often neglects the inherent distribution disparity between SAR and optical images—the result of differing imaging principles-potentially leading to significant feature loss. To this end, we introduce a network by jointing SAR and optical images multimodal and cross-attention network (MMCANet) to effectively extract multiscale contextual features from SAR imagery and integrate them with optical features. Specifically, instead of simple concatenation of the channels of SAR and optical images, we obtain high-dimensional features from them through independent feature extractors. The integration of these features is facilitated by a cross-attention mechanism that provides a more fine-grained amalgamation of information. Meanwhile, an atrous spatial pyramid pooling (ASPP) module is introduced into the integration of high-level features, which captures multiscale contextual information around clouded areas. In addition, we propose four advanced remote sensing image restoration algorithms that approach image restoration as a series of subtasks, gradually eliminating clouds to enhance performance. Comprehensive assessments show that MMCANet performs well on the SEN 12 MS-CR dataset with peak signal-to-noise ratio (PSNR) of 39.8871, structural similarity index (SSIM) of 0.9672, mean absolute error (MAE) of 0.0081, and spectral angle mapper (SAM) of 2.9884.

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

The use of deep neural networks for semantic segmentation of RS (remote sensing) images is an important research topic in the field of remote sensing intelligent interpretation, which has significant application value in urban planning, disaster assessment, carbon estimation, and other fields. However, high-resolution RS images bring problems of large data scale, high computational complexity, diverse scales, and irregular shapes. Moreover, there may be distant spatial relationships between target objects, requiring the utilization of long-range contextual information for accurate semantic segmentation. Inspired by the powerful global modelling capability of the Swin Transformer, we propose a new RS image semantic segmentation network. Swin Transformer is used as an encoder to extract global features. To prevent the loss of local details in the encoding stage, one LKA-PM (LKA-Patch Merging) module is designed to extract local detail information by decomposing large kernel convolutions, thereby enhancing the feature representation ability of small objects. Furthermore, to decode global and local features and obtain multi-scale information, one WORC (Weighted Operation Re-parameterization Convolution) module is designed to improve the adaptability and accuracy of the model to different RS images. Finally, to preserve spatial details during decoding and reduce the problem of boundary blurring caused by occlusion in RS imagery, the Coordswin module is designed to provide coordinate information and improve the discrimination ability of similar objects. Experimental results demonstrate that our CW-SwinUNet achieves excellent segmentation performance (91.8 % F1 and 85.2 % MIoU) on the Vaihingen dataset and (93.8 % F1 and 87.9 % MIoU) on Potsdam datasets, surpassing other state-of-the-art methods. The source code of the proposed CW-SwinUNet is available at https://github.com/xmy1135/cwswin/tree/main.

Semantic segmentation is a crucial approach for remote sensing interpretation. High-precision semantic segmentation results are obtained at the cost of manually collecting massive pixelwise annotations. Remote sensing imagery contains complex and variable ground objects and obtaining abundant manual annotations is expensive and arduous. The semi-supervised learning (SSL) strategy can enhance the generalization capability of a model with a small number of labeled samples. In this study, a novel semi-supervised adversarial semantic segmentation network is developed for remote sensing information extraction. A multiscale input convolution module (MICM) is designed to extract sufficient local features, while a Transformer module (TM) is applied for long-range dependency modeling. These modules are integrated to construct a segmentation network with a double-branch encoder. Additionally, a double-branch discriminator network with different convolution kernel sizes is proposed. The segmentation network and discriminator network are jointly trained under the semi-supervised adversarial learning (SSAL) framework to improve its segmentation accuracy in cases with small amounts of labeled data. Taking building extraction as a case study, experiments on three datasets with different resolutions are conducted to validate the proposed network. Semi-supervised semantic segmentation models, in which DeepLabv2, the pyramid scene parsing network (PSPNet), UNet and TransUNet are taken as backbone networks, are utilized for performance comparisons. The results suggest that the approach effectively improves the accuracy of semantic segmentation. The F1 and mean intersection over union (mIoU) accuracy measures are improved by 0.82–11.83% and 0.74–7.5%, respectively, over those of other methods.

Accurate, high-resolution global land use and land cover (LULC) mapping is crucial for environmental monitoring, but remains challenging when relying solely on multispectral data. Most existing global LULC mapping studies rely exclusively on multispectral observations, and even those that incorporate Synthetic Aperture Radar (SAR) data often fail to fully exploit the information it provides. SAR provides an all-weather sensing capability and is uniquely sensitive to surface structure, texture, and moisture—critical information for LULC classes that are often spectrally ambiguous. To address this data gap, we introduce the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Dynamic World+</i> dataset, a new global benchmark that expands the authoritative Dynamic World by aligning it with Sentinel-1 SAR data. Additionally, to facilitate the combination of multispectral and SAR data, we propose a lightweight transformer architecture termed <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">LULC-Former</i>. It incorporates two innovative modules, the Dual Modal Enhancement Module (DMEM) and Mutual Modal Aggregation Module (MMAM), designed to exploit cross-information between the two modalities in a split-fusion manner. These modules enable spectral features to guide the interpretation of SAR texture, and vice-versa, thereby improving the overall performance of global LULC semantic segmentation. Furthermore, we adopt an imbalanced parameter allocation strategy, which assigns parameters to different modalities based on the distinct physical information each provides for LULC characterization. Experiments demonstrate that our network outperforms existing transformer and CNN-based models, achieving a mean Intersection over Union (mIoU) of 59.58%, an Overall Accuracy (OA) of 79.48%, and an F1 Score of 71.68% with only 26.70 M parameters. Furthermore, the generated national-scale LULC maps across diverse regions demonstrate the effectiveness of the proposed dataset and network. The code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/uoe-haoyu/LULCFormer</uri>.

Keraniganj Upazila, adjacent to Dhaka city, has experienced significant land use and land cover (LULC) changes over the past three decades due to rapid urbanization and socio-economic transformation. This study aims to investigate the spatial and temporal dynamics of LULC in Keraniganj from 1989 to 2023 and assess the driving factors and environmental implications of these changes. To achieve this, multi-temporal Lands at imagery, remote sensing (RS), and geographic information systems (GIS) techniques were applied, complemented by socio-economic data. Supervised classification algorithms were used to generate accurate LULC maps, and the results were validated using confusion matrices and Kappa statistics. The analysis covered six major LULC categories: bare land/sand fill, urban settlements, cultivated land, rural settlement/homestead vegetation, water bodies, and wetlands/lowlands. The results reveal a substantial increase of 1,738.62 hectares in urban settlements, particularly between 2009 and 2023, primarily at the expense of cultivated land, rural settlements, and natural ecosystems. During the study period, Keraniganj lost 476.46 hectares of cultivated land, 480.51 hectares of rural settlement areas, 1,087.83 hectares of water bodies, and 166.68 hectares of wetlands. These trends indicate extensive land reclamation, unplanned urban growth, and environmental degradation, driven by the area’s proximity to Dhaka, improved transportation infrastructure, and lower living costs. The findings underscore the urgency of implementing sustainable land use policies and integrated urban planning to address the negative impacts of urban expansion, including habitat destruction, reduced biodiversity, heightened flood risks, and declining agricultural productivity. This research contributes valuable insights into LULC dynamics and provides a replicable framework for monitoring and managing land transformation in rapidly urbanizing regions globally. Jagannath University Journal of Life and Earth Sciences, 10(1) 91-122

Land use and land cover (LULC) change has become an important research topic for global environmental change and sustainable development. As an important part of worldwide land conservation, sustainable development and management of water resources, developing countries must ensure the use of innovative technology and tools that support their various decision making systems. This study provides the most recent LULC change analysis for the last six years (2015–2021) of Coatzacoalcos, Veracruz, Mexico, one of the most important petrochemical cities in the world and host of the ongoing Interoceanic Corridor project. The analysis was carried out using Landsat 8 Operational Land Imager (OLI) satellite images, ancillary data and ground-based surveys and the Normalized Difference Vegetation Index (NDVI) to identify and to ameliorate the discrimination between four main macro-classes and fourteen classes. The LULC classification was performed using the maximum likelihood classifier (MLC) to produce maps for each year, as it was found to be the best approach when compared to minimum distance (MDM) and spectral angle mapping (SAM) methods. The macro-classes were water, built-up, vegetation and bare soil, whereas the classes were an improved classification within those. Our study achieved both user accuracy (UA) and producer accuracy (PA) above 90% for the proposed macro-classes and classes. The average Kappa coefficient for macro-classes was 0.93, while for classes it was 0.96, both comparable to previous studies. The results from the LULC analysis show that residential, industry and commercial areas slowed down their growth throughout the study period. These changes were associated with socio-economical drivers such as insecurity and lack of economic investments. Groves and trees presented steady behaviors, with small increments during the five-year period. Swamps, on the other hand, significantly degraded, being about 2% of the study area in 2015 and 0.93% in 2021. Dunes and medium and high vegetation densities (∼80%) transitioned mostly to low vegetation densities. This behavior is associated with rainfall below the annual reference and increments of surface runoff due to the loss of vegetation cover. Lastly, the present study seeks to highlight the importance of remote sensing for a better understanding of the dynamics between human–nature interactions and to provide information to assist planners and decision-makers for more sustainable land development.

Land, sometimes referred to as dry land, is the solid surface of Earth that is not permanently covered by water. The study describes the land use and land cover (LULC) dynamics in Bokaro district at Jharkhand from 1991 to 2011 using Remote Sensing and Geographical Information System (GIS). Whatever, human activities is invented or generated begins from LAND. Therefore, its uses and disintegration by human throughout the period of time is very essential to be known. A city grows not only by population but also by change in its spatial dimensions. The present study investigated the change detection of Bokaro district, Jharkhand, India using LANDSAT satellite data for the year 1991, 2001 and 2011. Topo sheet and high resolution images from Google earth were also used for the study. The process used for the study is known as image processing under which supervised classification method was used using different classes for the study area. The results indicated that there is a drastic increase in agriculture and habitation whereas forest, water and fallow land reduced within the district boundary limit. Migration is still continuing in the district and afforestation/human activities done throughout the period effecting in the reduction of water. Keywords: Change detection, land use and land cover, LANDSAT, Geographical Information System (GIS), remote sensing Cite this Article Modi AK, Das P, Panda D. Land Use/Land Cover Change Dynamic of Bokaro District, East India Using Remote Sensing and GIS Techniques. Journal of Remote Sensing & GIS . 2018; 9(3): 16–22p.

&amp;lt;p&amp;gt;The land use and land cover (LULC) change induces hydrologic variability in a catchment and studying this variability is central to efficient water management practice in a catchment. The assessment of the alteration in hydrological processes due to LULC change and its influence on overall river ecosystem functioning is even more pertinent to developing nations that face the issue of water scarcity and pollution. In this work, we investigate the influence of the LULC change over a period of ~40 years (1970-2013) on the variability of natural or virgin flow in the Ramganga river, a major tributary of the Ganga river, India. For LULC change data, object-based image classification was performed on high-resolution satellite imageries acquired for the Ramganga river basin &amp;amp;#8211; CORONA (1970) and LISS IV (2013) images. The natural or virgin flows (i.e., the flow in the river without regulation practices such as construction of dams or barrages) were estimated by performing hydrological modeling using the Soil and Water Assessment Tool (SWAT). Initially, the SWAT model was set up, calibrated, and validated for the present flow scenario (i.e. with all management practices present) using LULC data of the year 2013. Natural flows were derived by removing all interventions and keeping agricultural practices only rain-fed. Next, keeping all parameters unchanged, the LULC data of the year 2013 was replaced by the LULC data of the year 1970. This enabled us to study the effects of LULC change on river hydrology between the period 1970-2013. The model showed good agreement between the observed and simulated flows with R&amp;lt;sup&amp;gt;2&amp;lt;/sup&amp;gt; values of 0.82 for the calibration period (2002-2014) and 0.68 for the validation period (1990-1999). The Nash-Sutcliffe efficiency values were 0.81 and 0.66 for calibration and validation periods respectively. The comparison of LULC data between the study period (1970 and 2013) reveals that land cover classes of agriculture, built-up, mixed forest, barren land, shrubs and bushes, and water areas were altered by nearly 6%, 102%, -7%, -59%, -75%, and -2% respectively (&amp;amp;#8216;-&amp;amp;#8217; sign indicates decrement in the land cover area). The influence of this LULC change was evident in the results from the hydrological model. For the years 2002-2013 (calibration period), the natural flows estimated using the LULC map of 2013 at the basin outlet were observed to be higher by 3-12% compared to flows estimated using the LULC input of 1970. The estimates of mean monthly flows for the years 2002-2013 at the basin outlet reveal that while the natural flows estimated using the LULC map of 2013 were higher compared to flow estimates using the LULC map of 1970 for most of the months, the flows during the dry months (May-July) were observed to be lower for the former compared to the latter. Our work provides valuable insights into hydrological variability in a major sub-basin of the Ganga river induced due to LULC changes and we advocate that alterations associated with LULC must be incorporated into water management strategies.&amp;lt;/p&amp;gt;

Recently, land cover and land use (LCLU) classification in remote sensing imagery has attracted research interest. The LCLU contains dynamic remote sensed images due to sensor technology ability, seasonal changes, and distance for resolution. Therefore, the deep learning-based LCLU classification system needs more investigation using deep learning techniques. Deep learning approaches have gotten more attention for their powerful performance improvements. Most recent studies have been performed on deep convolutional neural networks (CNNs) that have been trained on pre-trained networks in remote sensing classification. However, designing CNNs from scratch has not yet been widely investigated in remote sensed images as they need ample training time and a powerful processor. Therefore, we used hyperparameters and early stopping techniques to apply an end-to-end CNN feature extractor (CNN-FE) model for LCLU classification in the UC-Merced dataset. We approved the model's applicability in the domain area by retraining it on another dataset called SIRI-WHU and building the VGG19 pre-trained feature extractor model built on the same hyperparameters. The CNN-FE has outperformed the state-of-the-art baseline studies' accuracy and the VGG19 pre-trained model. Moreover, a better CNN-FE performance was achieved when trained in the UC-Merced dataset than the model performance when trained in the SIRI-WHU dataset.

As an important link between human and nature, land use/land cover change (LUCC) has become a research hotspot since the first day of the issue initiated. Remote sensing observation supplies important data source for LUCC research. With the constant improvement of the remote sensed image spatial resolution and the continuous development of computer scientific algorithms and models, the study of semantic segmentation in the field of automatic interpretation of remote sensed images has promoted quickly. There have been a number of achievements focusing on the one-class extraction of LUCC classification system, but limited progress in multi-class classifying mission when facing high resolution remote sensing data by using semantic segmentation model, because of its sophisticated structure, the phenomenon of synonyms spectrum and foreign body with spectrum in images. Additionally, the limitation of the semantic segmentation model is a crucial issue which needs to be solv for precision of classification in class recognition and the perfection of semantic segmentation on pixel level as well. The High-Resolution network (HRnet) model was used and improved in the study for semantic segmentation because it has the advantage of low resolution loss. In the LUCC classification work, the distribution of classes (such as cultivated land, woodland, grassland, other land, water, etc.) are usually extremely imbalanced, that leads to the limited classification recognition results of land types when using HRnet. So a combined Loss function of Lovasz Softmax Loss+Cross Entropy Loss was used in HRnet to classify the land cover of remote sensing images, to meet the requirements of high-resolution image classification not only in terms of low resolution loss but also in terms of class recognition. The combined loss function provided the direction of the model, and the appropriate loss function improved the performance of the model when faced with the imbalanced distribution of land classes. The results are as follows: 1) Compared with Deeplab and Unet, HRnet model performed better in segmentation results; 2) HRnet model with combined Loss function Lovasz Softmax Loss+Cross Entropy Loss (HRnet+) had better performance, and its test accuracy PA(Pixel Accuracy) value was 0.9577, MIOU (Mean Intersection over Union) value was 0.8801, and Kappa value was 0.9455. This indicated that the HRnet model with low loss of resolution could achieve better results in the automatic segmentation of high-resolution remote sensing images, and the combined loss function constructed in this study could effectively solve the problem of imbalanced distribution of classes in remote sensing images, and the edge optimization ability was also more significant.

Environmental monitoring, using the techniques of remote sensing (RS) and geographic information systems (GIS), allows the production of time efficient, cost-effective, and reliable surveillance and tracking data. Anthropogenic activities appear to be the major trigger of environmental changes, including land use and land cover (LULC) changes, while natural causes have only a minor impact in most cases. The Omayed Biosphere Reserve (OBR) stands as one of the Egyptian protected areas most highly affected by massive unplanned human activities. Thus, the main objective of this study is to determine the spatio-temporal changes in the OBR over a 35-year period using five Landsat (5 ETM images and 8 OLI-TIRS) imageries, with the specific aim of measuring change rates, trends, and magnitudes of LULC changes between 1984 and 2019 with the topography for planning and selection of developmental strategies. The Normalised Difference Vegetation Index is used to identify the vegetation characteristics of different eco-regions and delivers useful information for the study of vegetation health and density. Normalised Difference Built-up Index can likewise be used to quote built-up areas. Unsupervised classification was used to classify LULC patterns. Six classes were recognised: water bodies, coastal sand, urban areas, cultivated land, newly reclaimed areas, and bare soil. Our results reveal that about 33.55% of OBR land cover has transformed into other forms. Cultivated land and urban regions increased by about 143.5km2 and 56.17km2 from 1984 to 2019, respectively. Meanwhile, bare soil decreased to around 209.5km2 in 2019. In conclusion, the conversion of bare soil into urban land and cultivated areas is the major change in the last 35years in the OBR. Over the past three decades, the OBR has faced radical and imbalanced changes in its natural habitats. Therefore, monitoring and management of LULC changes are crucial for creating links between policy decisions, regulatory actions, and following LULC activities in the future, especially as many potential risks still exist in the remaining regions of the OBR.

Land use is an important factor in planning and managing land resources. Increasing pressure due to population and human resources in the world’s resources to meet growing needs contributes to significant land reform in various land uses. Remote sensing and GIS (geographic information system) techniques have been used to study land use change and land cover on the Tuticorin coast in Tamil Nadu. This study examines land use and land cover (LULC) changes from 2001 to 2017 for the coast of Tuticorin. The main objective of this study was to assess changes under the NRSC (National Remote Sensing Center) classification using Landsat ETM + and OLI images using visual interpretation with the help of image interpretation keys. The digitized land use and land cover features are categorized as aquaculture, built-up land, water bodies, cropland, fallow land, forest, a forest plantation, industrial area/mining, mangrove/swamp area, plantation, salt-affected land, saltpan, sandy areas, land with scrub, land without scrub, and waterlogged area. Apparently the whole study from 2001 to 2017 found that built-up land (+20.44 sq. km) and industrial/mining activities (+5.78 sq. km) were increased and cropland (−8.04 sq. km) and plantation (−7.62 sq. km) were decreased. The ground truth verification of the LULC features performed is made with an effective assessment of the changes. This study shows a significant environmental impact in the study area. In addition, it is crucial to strongly monitor the land use/land cover changes to maintain sustainable growth and in-depth coastal management requirements that can be taken to protect human health and property.