Search

Frequent mining activities create a series of geological and environmental problems resulting in an immeasurable loss of life and property. Adopting effective technologies that monitor the surface subsidence of mining areas reliably and accurately is necessary. Targeting problems associated with conventional distributed scatterers interferometric synthetic aperture radar (DS-InSAR) technology, we propose a DS-InSAR technology integrating unmanned aerial vehicle (UAV) photogrammetry products divided into two key technical contents: generating an external reference digital elevation model (DEM) fused with UAV DEM and refining distributed scatterers candidates (DSCs) fused with an UAV digital orthophoto map (DOM). We selected two mining areas, one in Wu'an, Hebei, and the other in Inner Mongolia, with different surface cover types, mining depths, and topographies as the research area. We used Sentinel-1A SAR images covering a mine in Wu'an from 4 November 2018 to 4 March 2019 and a mining area in Inner Mongolia from 11 June 2018 to 21 October 2018 to compare and analyze the subsidence results. We also combined these results with data from their respective field observation stations to assess accuracy. We could apply DS-InSAR technology integrated with UAV photogrammetry products to the subsidence monitoring of two mining areas with different landforms and mining characteristics. Comparing with the leveling and total station, the experimental results show that the RMSE was reduced by about 2 mm in both mining areas, and accuracy for the Wu'an region improved to a higher degree than Inner Mongolia did. Furthermore, the refinement method of DS eliminated 965 and 2948 lower-quality DSCs in the two mining areas. These demonstrate that our proposed method can effectively improve the accuracy and reliability of subsidence results from two mining areas. [ABSTRACT FROM AUTHOR]

While the protected area (PA) covers >15% of the planet's terrestrial land area and continues to expand, factors determining its effectiveness in conserving endangered species are being debated. We investigated the links between direct anthropogenic pressures, socioeconomic settings, and the coverage of vertebrate taxa by China's PA network, and indicated that high socioeconomic status and low levels of human pressure correlate with high species coverage, with threatened mammals more effectively conserved than reptiles or amphibians. Positive links between conservation outcomes and socioeconomic progress appear linked to local livelihood improvements triggering positive perceptions of local PAs—aided further by ecological compensation and tourism schemes introduced in wealthy areas and reinforced by continued positive conservation outcomes. Socioeconomic development of China's less developed regions might assist regional PA efficiency and achievement of the Kunming‐Montreal Global Biodiversity Framework, while also addressing potential shortcomings from an insufficient past focus on socioeconomic impacts for biodiversity conservation. [ABSTRACT FROM AUTHOR]

Background: Evidence regarding adverse pregnancy outcomes in patients with thyroid cancer has been conflicting, and the effect of thyroid dysfunction caused by thyroid hormone suppression therapy in terms of neonatal thyroid stimulating hormone (TSH) is unclear. This study aimed to investigate whether thyroid cancer was associated with adverse pregnancy outcomes and had an adverse effect on neonatal thyroid function. Methods: This was a retrospective study of 212 singleton pregnancies with thyroid cancer and 35,641 controls without thyroid cancer. Data on maternal pregnancy outcomes and neonatal outcomes were analyzed. Results: The median TSH level in the thyroid cancer group was significantly lower than that in the control group (0.87 µIU/mL vs. 1.17 µIU/mL; P < 0.001), while the FT4 level was higher than that in the control group (17.16 pmol/L vs. 16.33 pmol/L; P < 0.001). The percentage of thyroid peroxidase antibodies (TPOAb) positive in the thyroid cancer group was significantly higher than that in the control group (25.0% vs. 11.8%; P < 0.001). Pregnancies with thyroid cancer had a higher risk of late miscarriage (OR 7.166, 95% CI: 1.521, 33.775, P = 0.013), after adjusting maternal TPOAb positive, there was no statistical significance (OR 3.480, 95% CI: 0.423, 28.614, P = 0.246). Pregnancies with thyroid cancer had higher gestational weight gain (GWG) (14.0 kg vs. 13.0 kg, P < 0.001). Although there was no significant difference in the prevalence of gestational diabetes mellitus (GDM) (20.8% vs. 17.4%, P = 0.194), the oral glucose tolerance test (OGTT) showed that fasting plasma glucose and 2-hour value in the thyroid cancer group were higher than those in the control group (P = 0.020 and 0.004, respectively). There was no statistically significant difference in TSH between the thyroid cancer group and the control group, regardless of full-term newborns or preterm newborns. Conclusions: Thyroid cancer might not have substantial adverse effects on pregnancy outcomes except for excessive GWG. No adverse effect on neonatal TSH was found, but the effect on long-term thyroid function and neuropsychological function in offspring need further study. Trial registration: Beijing Birth Cohort Study (ChiCTR220058395). [ABSTRACT FROM AUTHOR]

Change detection (CD) methods using synthetic aperture radar (SAR) data have received significant attention in the field of remote sensing Earth observation, which mainly involves knowledge-driven and data-driven approaches. Knowledge-driven CD methods are based on the physical theoretical models with strong interpretability, but they lack the robust features of being deeply mined. In contrast, data-driven CD methods can extract deep features, but require abundant training samples, which are difficult to obtain for SAR data. To address these limitations, an end-to-end unsupervised CD network based on self-adaptive superpixel segmentation is proposed. Firstly, reliable training samples were selected using an unsupervised pre-task. Then, the superpixel generation and Siamese CD network were integrated into the unified framework to train them end-to-end until the global optimal parameters were obtained. Moreover, the backpropagation of the joint loss function promoted the adaptive adjustment of the superpixel. Finally, the binary change map was obtained. Several public SAR CD datasets were used to verify the effectiveness of the proposed method. The transfer learning experiment was implemented to further explore the ability to detect the changes and generalization performance of our network. The experimental results demonstrate that our proposed method achieved the most competitive results, outperforming seven other advanced deep-learning-based CD methods. Specifically, our method achieved the highest accuracy in OA, F1-score, and Kappa, and also showed superiority in suppressing speckle noise, refining change boundaries, and improving detection accuracy in a small area change. [ABSTRACT FROM AUTHOR]

China's adoption of an "ecological civilization" strategy since 2012 demonstrates its vision to move toward more sustainable development models while promoting green energy and strengthening environmental protection. However, China's continuing heavy reliance on fossil fuels, especially coal, masks the complex positive trends in energy use, more sustainable economic development, and environmental management. Through examining the evolving energy-GDP relationships in China over the past few decades, we elucidated the underlying complex nonlinear processes using environmental Kuznets curve analysis and showed that the energy–GDP relationship has fluctuated between positive and negative decoupling. Furthermore, fluctuating trajectories manifested as sharp regional divergences when early-transitioning coastal areas restructured toward services and exhibited robust decoupling, whereas the industrial interior remained fossil fuel reliant. We identified various forces related to these provincial disparities, where rapid urbanization, industrial realignment, and income growth were all important. By promoting service transitions and enabling low-carbon technologies in coastal hubs, these drivers also exposed infrastructure deficits and entrenched fossil fuel usage in the interior. These problems led to China's latest strategic plans emphasizing "higher quality" growth underpinned by clean energy transitions and coordinated regional policies. Ongoing declines in coal and the rise in renewable energy usage highlight promising momentum, but stable decoupling remains contingent on sustained policy coordination. The intricate balance between growth and environmental pressures in China provides essential lessons to help emerging economies achieve sustainability by catalyzing green transitions and attaining developmental milestones. [ABSTRACT FROM AUTHOR]

Multisource remote sensing images have rich features and high interpretability and are widely employed in many applications. However, highly unbalanced category distributions and complex backgrounds have created some difficulties in the application of remote sensing image semantic segmentation tasks, such as low accuracy of small target segmentation and inaccurate edge extraction. To solve these problems, in this paper, a feature map segmentation reconstruction module and an attention upsampling module are proposed. In the encoder part, the input feature map is equally segmented, and the segmented feature map is enlarged to effectively improve the small target feature information expression ability in the model. In the decoder part, the key segmentation and location information of shallow features are obtained using the global view. The deep semantic information and shallow spatial location information are fully combined to achieve a more refined upsampling operation. In addition, the attention mechanism of the spatial and channel squeeze and excitation block (scSE) is applied to pay more attention to important features and to suppress irrelevant background and redundant information. To verify the effectiveness of the proposed method, the WHU-OPT-SAR dataset and six state-of-the-art algorithms are utilized in comparative experiments. The experimental results show that our model has demonstrated the best performance and low computational complexity. With only approximately half the floating-point operation count and the number of model parameters of the MCANet model, which is specially designed for the dataset, our model surpasses MCANet by 1.52% and 1.53% in terms of mean intersection over union (mIoU) and F1 score, respectively. In particular, for small object regions such as roads and other categories, compared to the baseline model, the IoU and F1 score of our model are improved by 5.27% and 3.99% and by 5.68% and 5.65%, respectively. These results demonstrate the superior performance of our model in terms of accuracy and efficiency. [ABSTRACT FROM AUTHOR]

The shear strength of granitic gneiss residual soil (GGRS) determines the stability of colluvial landslides in the Huanggang area, China. It depends on several parameters that represent its structure and state as well as their interactions, and therefore requires accurate assessment. For an effective evaluation of shear strength parameters of GGRS based on these factors and their interactions, three parameters, namely, moisture content, bulk density, and fractal dimension of grain size, were selected as influencing factors in this study based on a thorough investigation of the survey data and physical property tests of landslides in the study area. The individual effects and interaction of the factors were then incorporated by implementing a series of direct shear tests employing the response surface methodology (RSM) into the regression model of the shear parameters. The results indicate that the factors affecting shear parameters in the order of greater to lower are bulk density, moisture content, and fractal dimension, and their interactions are insignificant. The proposed model was validated by applying it to soil specimens from other landslide sites with the same parent bedrock, showing the validity of the strength regression model. This study demonstrates that RSM can be applied for parameter estimation of soils and provide reliable performance, and is also significant for conducting landslide investigation, evaluation, and regional risk assessment. [ABSTRACT FROM AUTHOR]

Over the last four decades, China has experienced rapid parallel economic development and urbanization, leading to internal mass -migrations of its people from increasingly marginalized rural areas to urban centers where job opportunities and wealth are now concentrated. We compare the relative temporal growth trends in population-related and land-(i.e., area-) related urbanization systems to evaluate China's urbanization in the context of the 'New-Type' Urbanization Program (2014–2020). Based on coupling coordination models, we observed that the two systems were overall slightly decoupled since spatial urban expansion commonly outgrew urban population growth, but the degree of coordination between the two parameters was increasing. Employing exploratory spatial data analysis, we revealed that a high degree of coupling coordination has spread from Eastern to Western provinces. Urban planning and land policies have contributed to an increasing urban vegetation cover and the control of excessive urban land expansions. While China's urbanization appears to have become increasingly sustainable due to the increasing degree of coupling coordination between its subsystems, ongoing urban expansions require strong oversight to limit the environmental impacts of the country's sprawling mega-cities. • Population-related urbanization has an increasingly coupling coordination with land-related urbanization in China. • A more coordinated urbanization has spread from Eastern China to Western China. • Urban planning and land policies help control the excessive urban land expansions. • Urban quality of life still needs to be improved for a growing sustainability in China's urbanization. [ABSTRACT FROM AUTHOR]

Hyperspectral/multispectral and panchromatic of optical remote sensing images are commonly used for multisensor image fusion, which has been applied in various applications of Earth observation. However, the utilization of optical remote sensing data suffers from the limitation of bad weather and cloud contamination. To address aforementioned issue and enhance spatial details of polarimetric synthetic aperture radar (PolSAR) image, a novel dual-domain super-resolution image fusion method is proposed by combining improved spherically invariant random vector (ISIRV) model with generalized adaptive linear combination approximation (GALCA) technology in this study. The proposed method decomposes the task of image fusion into polarimetric and texture domain fusion by integrating polarimetric components of PolSAR image and texture detail component of panchromatic image, which can significantly improve spatial resolutions of the PolSAR image while preserving polarimetric information. The data fusion experiment is implemented with three data sets including panchromatic images of Gaofen-1 (GF-1) and Gaofen-2 (GF-2) and the quad-pol SAR data of Gaofen-3 (GF-3) and Radarsat-2. Results show that the proposed dual-domain image fusion method provides a better performance compared with state-of-the-art multisensor fusion methods (BT, PCA, GS, indusion, and PRACS) regarding qualitative and quantitative evaluations. In addition, results of image fusion are applied to image classification over agricultural and urban areas of China, which shows that classification accuracy is significantly improved when compared with the result using only the original image. [ABSTRACT FROM AUTHOR]

Earthworms can resist high levels of soil copper (Cu) contamination and play an essential role in absorbing them effectively. However, the molecular mechanisms underlying Cu tolerance in earthworms are poorly understood. To address this research gap, we studied alterations of Eisenia fetida in antioxidant enzymes, gut microbiota, metabolites, and genes under varying levels of Cu exposure soils (0, 67.58, 168.96, 337.92 mg/kg). Our results revealed a reduction in antioxidant enzyme activities across all treatment groups, indicating an adaptive response to alleviate Cu-induced oxidative stress. Analysis of gut microbiota revealed a significant increase in the abundance of bacteria associated with nutrient uptake and Cu2+ excretion under Cu stress. Furthermore, metabolomic analysis discovered an increase in certain metabolites associated with energy metabolism, such as pyruvic acid, L-malic acid, and fumaric acid, as Cu concentration escalated. These results suggested that enhanced energy supply contributes to the elevated tolerance of E. fetida towards Cu. Additionally, transcriptome analysis not only identified crucial detoxification genes (Hsp70 , CTSL , GST , CHAC , and GCLC), but also confirmed the critical role of glutathione metabolism as a key pathway in E. fetida Cu detoxification processes. These findings provide a new perspective on the molecular mechanisms of Cu tolerance in earthworms. [Display omitted] • The multi-omics of E. fetida under Cu stress were determined. • Antioxidant enzyme system in E. fetida responded to Cu stress. • Increased energy supply helps improve tolerance of E. fetida towards Cu. • Glutathione metabolic pathway play important roles in Cu detoxification. [ABSTRACT FROM AUTHOR]

Unsupervised change detection approaches, which are relatively straightforward and easy to implement and interpret, and which require no human intervention, are widely used in change detection. Polarimetric synthetic aperture radar (PolSAR), which has an all-weather response capability with increased polarimetric information, is a key tool for change detection. However, for PolSAR data, inadequate evaluation of the difference image (DI) map makes the threshold-based algorithms incompatible with the true distribution model, which causes the change detection results to be ineffective and inaccurate. In this paper, to solve these problems, we focus on the generation of the DI map and the selection of the optimal threshold. An omnibus test statistic is used to generate the DI map from multi-temporal PolSAR images, and an improved Kittler and Illingworth algorithm based on either Weibull or gamma distribution is used to obtain the optimal threshold for generating the change detection map. Multi-temporal PolSAR data obtained by the Radarsat-2 sensor over Wuhan in China are used to verify the efficiency of the proposed method. The experimental results using our approach obtained the best performance in East Lake and Yanxi Lake regions with false alarm rates of 1.59% and 1.80%, total errors of 2.73% and 4.33%, overall accuracy of 97.27% and 95.67%, and Kappa coefficients of 0.6486 and 0.6275, respectively. Our results demonstrated that the proposed method is more suitable than the other compared methods for multi-temporal PolSAR data, and it can obtain both effective and accurate results. [ABSTRACT FROM AUTHOR]

The traditional unsupervised change detection methods based on the pixel level can only detect the changes between two different times with same sensor, and the results are easily affected by speckle noise. In this paper, a novel method is proposed to detect change based on time-series data from different sensors. Firstly, the overall difference image of the time-series PolSAR is calculated by omnibus test statistics, and difference images between any two images in different times are acquired by Rj test statistics. Secondly, the difference images are segmented with a Generalized Statistical Region Merging (GSRM) algorithm which can suppress the effect of speckle noise. Generalized Gaussian Mixture Model (GGMM) is then used to obtain the time-series change detection maps in the final step of the proposed method. To verify the effectiveness of the proposed method, we carried out the experiment of change detection using time-series PolSAR images acquired by Radarsat-2 and Gaofen-3 over the city of Wuhan, in China. Results show that the proposed method can not only detect the time-series change from different sensors, but it can also better suppress the influence of speckle noise and improve the overall accuracy and Kappa coefficient. [ABSTRACT FROM AUTHOR]

Additive noise results from the thermal noise of the radar receiver or antenna, and often has an adverse effect on the subsequent applications. To weaken the influence of additive noise, it is crucial to know the actual additive noise level. However, the system noise level is not always provided in some single- or full-polarimetric synthetic aperture radar (PolSAR) products. As one of the typical low backscattering objects (LBOs) in PolSAR images, shadow occurs as soon as the depression angle of the synthetic aperture radar (SAR) system reaches a more value than the slope away from the radar illumination, and thus provides a feasible way to measure additive noise. This paper proposes a framework to automatically extract shadow regions by a trained model using the bag of words (BoW) model with data from different scenes. The additive noise can then be estimated across a large area based on the extracted shadow. The experiments based on single- and full-polarimetric Radarsat-2 data suggest that the shadow-based framework is an effective way to estimate additive noise. [ABSTRACT FROM AUTHOR]

• Model-based polarimetric decomposition method for soil moisture estimation (SME). • Dihedral scattering component is neglected in bare soil and sparse vegetation areas. • SME results of C- and L-band full polarimetric data are analyzed. • SME accuracies of bare soil, sparse, and dense grassland are evaluated. Polarimetric synthetic aperture radar (PolSAR), as an advanced active sensor, can penetrate vegetation to detect ground information. The multimode observation data can provide more information to interpret different scattering mechanisms and have great application potential in soil moisture estimation (SME). In this paper, a modified method of soil moisture inversion from PolSAR data based on a two-component polarimetric decomposition model is introduced. The proposed method neglects the dihedral scattering component in the case of sparse vegetation cover and no-building areas, compensates the orientation angle by considering the influence of surface roughness and terrain slope, and calculates the dielectric constant with the hybrid X-Bragg/Fresnel model. Gaofen-3 (C-band) data from Ordos, Inner Mongolia, and Soil Moisture Experiment 2003 (SMEX03) airborne synthetic aperture radar (AIRSAR) (C- and L-band) data from Southern Oklahoma were applied in the experimental research. Furthermore, the performance and validity limits were assessed by comparing the retrieval results with in situ measurements and Soil Moisture Active Passive (SMAP) soil moisture products. The results reveal a root-mean-square error (RMSE) of 8.66% with an inversion rate of up to 61% when using the Gaofen-3 data from Ordos. For the Southern Oklahoma study area, the inversion RMSE of the AIRSAR C- and L-band data is 8.57% and 8.95%, respectively. The tendency of the inversion results is also consistent with the change trend of the site observations. In the bare soil and sparse vegetation covered areas, the retrieval error is relatively small, while in dense vegetation covered areas, soil moisture is underestimated in both the C- and L-band data, and the former is more underestimated than the latter. [ABSTRACT FROM AUTHOR]

A new approach in polarimetric synthetic aperture radar (PolSAR) speckle filtering is proposed in this article. The proposed method preserves both point targets and dominant scattering mechanisms. The point targets are detected based on the span image, and they are then neither filtered nor involved in the other pixels’ filtering. To achieve the protection of the dominant scattering mechanism of each pixel, only pixels of the same dominant scattering mechanism as the centre pixel are included in the selection of the homogeneous pixels. Both point targets not being filtered and fact that only pixels of the same dominant scattering mechanism are included in the selection of the homogeneous pixels, which greatly improves the filtering efficiency. A likelihood-ratio test statistic based on the PolSAR covariance matrices is applied to determine the homogeneous pixels. Finally, the speckle filtering is processed using the weighted minimum mean square error estimator on the homogeneous pixels. We demonstrate the obvious advantages of the proposed method over other algorithms in the preservation of point targets and dominant scattering mechanisms, speckle suppression, protection of detail information, and maintenance of polarization information, by the use of both simulated and real PolSAR data. [ABSTRACT FROM AUTHOR]

The GaoFen-3 (GF-3) satellite is the first fully polarimetric synthetic aperture radar (SAR) satellite designed for civil use in China. The satellite operates in the C-band and has 12 imaging modes for various applications. Three fully polarimetric SAR (PolSAR) imaging modes are provided with a resolution of up to 8 m. Although polarimetric calibration (PolCAL) of the SAR system is periodically undertaken, there is still some residual distortion in the images. In order to assess the polarimetric accuracy of this satellite and improve the image quality, we analyzed the polarimetric distortion errors and performed a PolCAL experiment based on scattering properties and corner reflectors. The experiment indicates that the GF-3 images can meet the satellite’s polarimetric accuracy requirements, i.e., a channel imbalance of 0.5 dB in amplitude and ±10 degrees in phase and a crosstalk accuracy of -35 dB. However, some images still contain residual polarimetric distortion. The experiment also shows that the residual errors of the GF-3 standard images can be diminished after further PolCAL, with a channel imbalance of 0.26 dB in amplitude and ±0.2 degrees in phase and a crosstalk accuracy of -42 dB. [ABSTRACT FROM AUTHOR]

Vibrational spectroscopic techniques such as infrared, near-infrared and Raman spectroscopy have become popular in detecting and quantifying polymorphism of pharmaceutics since they are fast and non-destructive. This study assessed the ability of three vibrational spectroscopy combined with multivariate analysis to quantify a low-content undesired polymorph within a binary polymorphic mixture. Partial least squares (PLS) regression and support vector machine (SVM) regression were employed to build quantitative models. Fusidic acid, a steroidal antibiotic, was used as the model compound. It was found that PLS regression performed slightly better than SVM regression in all the three spectroscopic techniques. Root mean square errors of prediction (RMSEP) were ranging from 0.48% to 1.17% for diffuse reflectance FTIR spectroscopy and 1.60–1.93% for diffuse reflectance FT-NIR spectroscopy and 1.62–2.31% for Raman spectroscopy. The results indicate that diffuse reflectance FTIR spectroscopy offers significant advantages in providing accurate measurement of polymorphic content in the fusidic acid binary mixtures, while Raman spectroscopy is the least accurate technique for quantitative analysis of polymorphs. [ABSTRACT FROM AUTHOR]

Surface soil moisture (SM) retrieval over agricultural areas from polarimetric synthetic aperture radar (PolSAR) has long been restricted by vegetation attenuation, simplified polarimetric scattering modelling, and limited SAR measurements. This study proposes a modified polarimetric decomposition framework to retrieve SM from multi-incidence and multitemporal PolSAR observations. The framework is constructed by combining the X-Bragg model, the extended double Fresnel scattering model and the generalised volume scattering model (GVSM). Compared with traditional decomposition models, the proposed framework considers the depolarisation of dihedral scattering and the diverse vegetation contribution. Under the assumption that SM is invariant for the PolSAR observations at two different incidence angles and that vegetation scattering does not change between two consecutive measurements, analytical parameter solutions, including the dielectric constant of soil and crop stem, can be obtained by solving multivariable nonlinear equations. The proposed framework is applied to the time series of L-band uninhabited aerial vehicle synthetic aperture radar data acquired during the Soil Moisture Active Passive Validation Experiment in 2012. In this study, we assess retrieval performance by comparing the inversion results with in-situ measurements over bean, canola, corn, soybean, wheat and winter wheat areas and comparing the different performance of SM retrieval between the GVSM and Yamaguchi volume scattering models. Given that SM estimation is inherently influenced by crop phenology and empirical parameters which are introduced in the scattering models, we also investigate the influence of surface depolarisation angle and co-pol phase difference on SM estimation. Results show that the proposed retrieval framework provides an inversion accuracy of RMSE<6.0% and a correlation of R ≥0.6 with an inversion rate larger than 90%. Over wheat and winter wheat fields, a correlation of 0.8 between SM estimates and measurements is observed when the surface scattering is dominant. Specifically, stem permittivity, which is retrieved synchronously with SM also shows a linear relationship with crop biomass and plant water content over bean, corn, soybean and wheat fields. We also find that a priori knowledge of surface depolarisation angle, co-pol phase difference and adaptive volume scattering could help to improve the performance of the proposed SM retrieval framework. However, the GVSM model is still not fully adaptive because the co-pol power ratio of volume scattering is potentially influenced by ground scattering. • A new framework for soil moisture (SM) retrieval from multiple PolSAR observations. • The soil and crop stem permittivity are derived from the proposed framework. • SM estimates over selected crop fields show an accuracy of RMSE≤6.0%. • The surface depolarization angle influences the SM inversion performance. • The optimal selection of co-pol phase difference improves the SM estimation. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) is a significant application in maritime monitoring, which can provide SAR data throughout the day and in all weather conditions. With the development of artificial intelligence and big data technologies, the data-driven convolutional neural network (CNN) has become widely used in ship detection. However, the accuracy, feature visualization, and analysis of ship detection need to be improved further, when the CNN method is used. In this letter, we propose a two-stage ship detection for land-contained sea area without a traditional sea-land segmentation process. First, to decrease the possibly existing false alarms from the island, an island filter is used as the first step, and then threshold segmentation is used to quickly perform candidate detection. Second, a two-layer lightweight CNN model-based classifier is built to separate false alarms from the ship object. Finally, we discuss the CNN interpretation and visualize in detail when the ship is predicted in vertical–horizontal (VH) and vertical–vertical (VV) polarization. Experiments demonstrate that the proposed method can reach an accuracy of 99.4% and an F1 score of 0.99 based on the Sentinel-1 images for a ship with a size of less than 32 × 32. [ABSTRACT FROM AUTHOR]

Most of the traditional supervised classification methods using full-polarimetric synthetic aperture radar (PolSAR) imagery are dependent on sufficient training samples, whereas the results of pixel-based supervised classification methods show a high false alarm rate due to the influence of speckle noise. In this paper, to solve these problems, an object-based supervised classification method with an active learning (AL) method and random forest (RF) classifier is presented, which can enhance the classification performance for PolSAR imagery. The first step of the proposed method is used to reduce the influence of speckle noise through the generalized statistical region merging (GSRM) algorithm. A reliable training set is then selected from the different polarimetric features of the PolSAR imagery by the AL method. Finally, the RF classifier is applied to identify the different types of land cover in the three PolSAR images acquired by different sensors. The experimental results demonstrate that the proposed method can not only better suppress the influence of speckle noise, but can also significantly improve the overall accuracy and Kappa coefficient of the classification results, when compared with the traditional supervised classification methods. [ABSTRACT FROM AUTHOR]