Explore the vast range of titles available.

Scenarios for monitoring land cover on a large scale, involving large volumes of data, are becoming more prevalent in remote sensing applications. The accuracy of algorithms is important for environmental monitoring and assessments. Because they performed equally well throughout the various research regions and required little human involvement during the categorization process, they appear to be resilient and accurate for automated, big area change monitoring. Malekshahi City is one of the important and at the same time critical areas in terms of land use change and forest area reduction in Ilam Province. Therefore, this study aimed to compare the accuracy of nine different methods for identifying land use types in Malekshahi City located in Western Iran. Results revealed that the artificial neural network (ANN) algorithm with back-propagation algorithms could reach the highest accuracy and efficiency among the other methods with kappa coefficient and overall accuracy of approximately 0.94 and 96.5, respectively. Then, with an overall accuracy of about 91.35 and 90.0, respectively, the methods of Mahalanobis distance (MD) and minimum distance to mean (MDM) were introduced as the next priority to categorize land use. Further investigation of the classified land use showed that good results can be provided about the area of the land use classes of the region by applying the ANN algorithm due to high accuracy. According to those results, it can be concluded that this method is the best algorithm to extract land use maps in Malekshahi City because of high accuracy.

The classification accuracy of remotely sensed data and its sensitivity to classification algorithms have a critical importance for the geospatial community, as classified images provide the base layers for many applications and models. Support Vector Machines (SVMs), a non-parametric statistical learning method that has recently been used in numerous applications in image processing. The SVMs need user-defined parameters and each parameter has different impact on kernels hence the classification accuracy of SVMs is based upon the choice of the parameters and kernels. The objective of this study is to investigate the sensitivity of SVM architecture including internal parameters and kernel types on landuse classification accuracy of RapidEye imagery for the study area in Turkey. Four types of kernels (linear, polynomial, radial basis function, and sigmoid) were used for the SVM classification. A total of 63 different models were developed and implemented for sensitivity analysis of SVM architecture. The traditional Maximum Likelihood Classification (MLC) method was also performed for comparison. The classification accuracies of the best model for each kernel type and MLC are 85.63%, 83.94%, 83.94%, 83.82% and 81.64% for polynomial, linear, radial basis function, sigmoid kernels and MLC, respectively. The results suggest that the choice of model parameters and kernel types play an important role on SVMs classification accuracy. Best model of polynomial kernel outperformed all SVMs models and gave the highest classification accuracy of 85.63% with RapidEye imagery.

This paper evaluates accuracies of selected image classification strategies, as applied to Landsat imagery to assess urban impervious surfaces by comparing them to reference data manually delineated from high-resolution aerial photos. Our goal is to identify the most effective methods for delineating urban impervious surfaces using Landsat imagery, thereby guiding applications for selecting cost-effective delineation techniques. A high-resolution aerial photo was used to delineate impervious surfaces for selected census tracts for the City of Roanoke, Virginia. National Land Cover Database Impervious Surface data provided an overall accuracy benchmark at the city scale which was used to assess the Landsat classifications. Three different classification methods using three different band combinations provided overall accuracies in excess of 70% for the entire city. However, there were substantial variations in accuracy when the results were subdivided by census tract. No single classification method was found most effective across all census tracts; the best method for a specific tract depended on method, band combination, and physical characteristics of the area. These results highlight impacts of inherent local variability upon attempts to characterize physical structures of urban regions using a single metric, and the value of analysis at finer spatial scales.

The identification of the usage and coverage of the land is a major part of regional development. Crowdsourced geographic information systems provide valuable information about the land use of different regions. Although these data sources lack reliability and possess some limitations, they are useful in deriving building blocks for the usage of the land, where the manual surveys are not up‐to‐date, costly, and time consuming. At present, in the context of Sri Lanka, there is a lack of reliable and updated land‐use data. Moreover, there is a rapid growth in the construction industry, resulting in frequent changes in land‐use and land‐cover data. This paper presents a novel and an automated methodology based on learning models for identifying the usage and coverage of the land. The satellite imagery is used to identify the information related to land cover. They are integrated with Foursquare venue data, which is a popular crowdsourced geographic information, thus, enhancing the information level and the quality of land‐use visualization. The proposed methodology has shown a kappa coefficient of 74.03%, showing an average land‐use classification accuracy within a constrained environment.

The rapid urbanization and industrialization along with the expansion of cities in developing countries like Bangladesh converting vegetation and bare land into built-up area that remarkably boost up the land surface temperature (LST). This study has been conducted for correlating and monitoring the changes of landuse–landcover change (LULC) and LST of rapidly expanding Chittagong metropolitan area from 1989 to 2018 utilizing four Landsat satellite images (TM, ETM+, OLI, and TIRS). The Present study combines the techniques of remote sensing and geographic information system (GIS) to find out the spatial variation of LST and identify its relationship with LULC. Supervised classification technique has been employed in ERDAS IMAGINE 14.0 software to retrieve LULC data. The images of the study area were categorized into four different classes namely vegetation, urban structures, bared lands and water bodies. LSTs were estimated using the single thermal infrared band of Landsat TM, ETM+, and the band 10 and 11 of the TIRS sensor’s image for the split-window algorithm method. Concerning the relationship between LULC and LST, it has been found that vegetation and water bodies shows lowest LST while bared lands and urban structures indicates highest LST. LULC analysis shows a dramatic increase in urban structures (from 20.83 to 58.93%), decrease in vegetation (from 56.54 to 20.24%) and bared lands (from 16.67 to 11.90%) and a further small increase in water bodies after the 80s, because of digging new ponds. LST in the study area has been increasing as high-temperature LU types have increased and low temperature LU types have decreased. Consequently, the mean annual temperature showed 6.5 °C increase, the minimum and maximum LST increased by 9 °C and 4 °C throughout the study period. The highest maximum and lowest minimum LST has found 40°C during the years of 2010 to 2018 and 15 °C in the year of 1989, respectively. The study will assist the decision-maker to understand the impacts of unplanned urbanization for future city planning and urban management.

The escalating pace of urbanization and human activities presents formidable challenges to landuse patterns and ecological environments. Achieving a harmonious coexistence between humans and nature of high quality has emerged as a global imperative. Constructing an ecological security pattern has become an essential approach to mitigating the adverse ecological impacts of urban sprawl, safeguarding human well-being, and promoting the healthy development of ecosystems. Focusing on ecosystem sustainability, stability, and integrity, this study constructed the ecological security pattern in rapidly urbanizing areas, emphasizing achieving a well-balanced integration of urban expansion and ecological preservation. Ecological sources were identified by an evaluation system of “ecosystem service function–ecological sensitivity–landscape connectivity”. Resistance surfaces were constructed by integrating natural and human factors. Ecological corridors and nodes were extracted by methods such as the minimum cumulative resistance and gravity models. Taking Nanchang City as an example, the results show that there were 15 ecological sources, primarily woodland, displaying a distinct “island” phenomenon. Additionally, there were 41 ecological corridors with a combined length of 2170.54 km, exhibiting a dense distribution in the southwest and a sparse distribution in the northeast. The city was found to encompass 122 ecological nodes, predominantly situated along the corridors near the ecological sources, indicating a strong spatial aggregation pattern. An optimized ecological security pattern of “one ring, two belts, three zones, and multiple nodes” was proposed for synergizing ecological protection, restoration, and rapid urbanizing.

Detailed urban landuse information plays a fundamental role in smart city management. A sufficient sample size has been identified as a very crucial pre-request in machine learning algorithms for urban landuse classification. However, it is often difficult to recognize and label landuse categories from remote sensing images alone. Alternatively, field investigation is time-consuming with a high demand in human resources and monetary cost. Therefore, previous studies on urban landuse classification have often relied on a small size of labeled samples with very uneven spatial distribution. This study aims to explore the effectiveness of a semi-supervised classification framework with multi-source data for detailed urban landuse classification with a few labeled samples. A disagreement-based semi-supervised learning approach, the Co-Forest, was employed and compared with traditional supervised methods (e.g., random forest and XGBoost). Multi-source geospatial data were utilized including optical and nighttime light remote sensing and geospatial big data, which present the physical and socio-economic features of landuse categories. Taking urban landuse classification in Shenzhen City as a case, results show that the classification accuracy of the semi-supervised method are generally on par with that of traditional supervised methods, and less labeled samples are needed to achieve a comparable result under different training set ratios. Given a small sample size, the accuracy tends to be stable with training samples no less than 5% in total. Our results also indicate that the classification accuracy by using multi-source data is significantly higher than that with any single data source being applied. Among these data, map POI and high-resolution optical remote sensing data make larger contributions on the classification, followed by mobile data and nighttime light remote sensing data.

The article discusses a method for classifying land cover types in rural areas using a trained neural network. The focus is on distinguishing agriculturally cultivated areas and differentiating bare soil from quarry areas. This distinction is not present in publicly available databases like CORINE, UrbanAtlas, EuroSAT, or BigEarthNet. The research involves training a neural network on multi-temporal patches to classify Sentinel-2 images rapidly. This approach allows automated monitoring of cultivated areas, determining periods of bare soil vulnerability to erosion, and identifying open-pit areas with similar spectral characteristics to bare soil. After training the U-Net network, it achieved an average classification accuracy of 90% (OA) in the test areas, highlighting the importance of using OA for multi-class classifications, instead of ACC. Analysis of our main classes revealed high accuracy, 99.01% for quarries, 92.3% for bare soil, and an average of 94.8% for annual crops, demonstrating the model's capability to differentiate between crops at various growth stages and assess land cover categories effectively.

Historical land cover (LC) maps are an essential instrument for studying long-term spatio-temporal changes of the landscape. However, manual labelling on low-quality monochromatic historical orthophotos for semantic segmentation (pixel-level classification) is particularly challenging and time consuming. Therefore, this paper proposes a methodology for the automated extraction of very-high-resolution (VHR) multi-class LC maps from historical orthophotos under the absence of target-specific ground truth annotations. The methodology builds on recent evolutions in deep learning, leveraging domain adaptation and transfer learning. First, an unpaired image-to-image (I2I) translation between a source domain (recent RGB image of high quality, annotations available) and the target domain (historical monochromatic image of low quality, no annotations available) is learned using a conditional generative adversarial network (GAN). Second, a state-of-the-art fully convolutional network (FCN) for semantic segmentation is pre-trained on a large annotated RGB earth observation (EO) dataset that is converted to the target domain using the I2I function. Third, the FCN is fine-tuned using self-annotated data on a recent RGB orthophoto of the study area under consideration, after conversion using again the I2I function. The methodology is tested on a new custom dataset: the ‘Sagalassos historical land cover dataset’, which consists of three historical monochromatic orthophotos (1971, 1981, 1992) and one recent RGB orthophoto (2015) of VHR (0.3–0.84 m GSD) all capturing the same greater area around Sagalassos archaeological site (Turkey), and corresponding manually created annotations (2.7 km² per orthophoto) distinguishing 14 different LC classes. Furthermore, a comprehensive overview of open-source annotated EO datasets for multiclass semantic segmentation is provided, based on which an appropriate pretraining dataset can be selected. Results indicate that the proposed methodology is effective, increasing the mean intersection over union by 27.2% when using domain adaptation, and by 13.0% when using domain pretraining, and that transferring weights from a model pretrained on a dataset closer to the target domain is preferred.